Original Research

From Hassenfeld Children’s Hospital at NYU Langone Health, New York, NY (Dr Gallagher), T1D Exchange, Boston, MA (Saketh Rompicherla; Drs Ebekozien, Noor, Odugbesan, and Mungmode; Nicole Rioles, Emma Ospelt), University of Mississippi School of Population Health, Jackson, MS (Dr. Ebekozien), Icahn School of Medicine at Mount Sinai, New York, NY (Drs. Wilkes, O’Malley, and Rapaport), Weill Cornell Medicine, New York, NY (Drs. Antal and Feuer), NYU Long Island School of Medicine, Mineola, NY (Dr. Gabriel), NYU Langone Health, New York, NY (Dr. Golden), Barbara Davis Center, Aurora, CO (Dr. Alonso), Texas Children’s Hospital/Baylor College of Medicine, Houston, TX (Dr. Lyons), Stanford University, Stanford, CA (Dr. Prahalad), Children Mercy Kansas City, MO (Dr. Clements), Indiana University School of Medicine, IN (Dr. Neyman), Rady Children’s Hospital, University of California, San Diego, CA (Dr. Demeterco-Berggren).

Background: Patient outcomes of COVID-19 have improved throughout the pandemic. However, because it is not known whether outcomes of COVID-19 in the type 1 diabetes (T1D) population improved over time, we investigated differences in COVID-19 outcomes for patients with T1D in the United States.

Methods: We analyzed data collected via a registry of patients with T1D and COVID-19 from 56 sites between April 2020 and January 2021. We grouped cases into first surge (April 9, 2020, to July 31, 2020, n = 188) and late surge (August 1, 2020, to January 31, 2021, n = 410), and then compared outcomes between both groups using descriptive statistics and logistic regression models.

Results: Adverse outcomes were more frequent during the first surge, including diabetic ketoacidosis (32% vs 15%, P < .001), severe hypoglycemia (4% vs 1%, P = .04), and hospitalization (52% vs 22%, P < .001). Patients in the first surge were older (28 [SD,18.8] years vs 18.0 [SD, 11.1] years, P < .001), had higher median hemoglobin A1c levels (9.3 [interquartile range {IQR}, 4.0] vs 8.4 (IQR, 2.8), P < .001), and were more likely to use public insurance (107 [57%] vs 154 [38%], P < .001). The odds of hospitalization for adults in the first surge were 5 times higher compared to the late surge (odds ratio, 5.01; 95% CI, 2.11-12.63).

Conclusion: Patients with T1D who presented with COVID-19 during the first surge had a higher proportion of adverse outcomes than those who presented in a later surge.

Keywords: TD1, diabetic ketoacidosis, hypoglycemia.

After the World Health Organization declared the disease caused by the novel coronavirus SARS-CoV-2, COVID-19, a pandemic on March 11, 2020, the Centers for Disease Control and Prevention identified patients with diabetes as high risk for severe illness.^1-7 The case-fatality rate for COVID-19 has significantly improved over the past 2 years. Public health measures, less severe COVID-19 variants, increased access to testing, and new treatments for COVID-19 have contributed to improved outcomes.

The T1D Exchange has previously published findings on COVID-19 outcomes for patients with type 1 diabetes (T1D) using data from the T1D COVID-19 Surveillance Registry.^8-12 Given improved outcomes in COVID-19 in the general population, we sought to determine if outcomes for cases of COVID-19 reported to this registry changed over time.

 

Methods

This study was coordinated by the T1D Exchange and approved as nonhuman subject research by the Western Institutional Review Board. All participating centers also obtained local institutional review board approval. No identifiable patient information was collected as part of this noninterventional, cross-sectional study.

The T1D Exchange Multi-center COVID-19 Surveillance Study collected data from endocrinology clinics that completed a retrospective chart review and submitted information to T1D Exchange via an online questionnaire for all patients with T1D at their sites who tested positive for COVID-19.^13,14 The questionnaire was administered using the Qualtrics survey platform (www.qualtrics.com version XM) and contained 33 pre-coded and free-text response fields to collect patient and clinical attributes.

Each participating center identified 1 team member for reporting to avoid duplicate case submission. Each submitted case was reviewed for potential errors and incomplete information. The coordinating center verified the number of cases per site for data quality assurance.

Quantitative data were represented as mean (standard deviation) or median (interquartile range). Categorical data were described as the number (percentage) of patients. Summary statistics, including frequency and percentage for categorical variables, were calculated for all patient-related and clinical characteristics. The date August 1, 2021, was selected as the end of the first surge based on a review of national COVID-19 surges.

We used the Fisher’s exact test to assess associations between hospitalization and demographics, HbA1c, diabetes duration, symptoms, and adverse outcomes. In addition, multivariate logistic regression was used to calculate odds ratios (OR). Logistic regression models were used to determine the association between time of surge and hospitalization separately for both the pediatric and adult populations. Each model was adjusted for potential sociodemographic confounders, specifically age, sex, race, insurance, and HbA1c.

All tests were 2-sided, with type 1 error set at 5%. Fisher’s exact test and logistic regression were performed using statistical software R, version 3.6.2 (R Foundation for Statistical Computing).

Results

The characteristics of COVID-19 cases in patients with T1D that were reported early in the pandemic, before August 1, 2020 (first surge), compared with those of cases reported on and after August 1, 2020 (later surges) are shown in Table 1.

Patients with T1D who presented with COVID-19 during the first surge as compared to the later surges were older (mean age 28 [SD, 18.0] years vs 18.8 [SD, 11.1] years; P < .001) and had a longer duration of diabetes (P < .001). The first-surge group also had more patients with >20 years’ diabetes duration (20% vs 9%, P < .001). Obesity, hypertension, and chronic kidney disease were also more commonly reported in first-surge cases (all P < .001).

There was a significant difference in race and ethnicity reported in the first surge vs the later surge cases, with fewer patients identifying as non-Hispanic White (39% vs, 63%, P < .001) and more patients identifying as non-Hispanic Black (29% vs 12%, P < .001). The groups also differed significantly in terms of insurance type, with more people on public insurance in the first-surge group (57% vs 38%, P < .001). In addition, median HbA1c was higher (9.3% vs 8.4%, P < .001) and continuous glucose monitor and insulin pump use were less common (P = .02 and <.001, respectively) in the early surge.

All symptoms and adverse outcomes were reported more often in the first surge, including diabetic ketoacidosis (DKA; 32% vs 15%; P < .001) and severe hypoglycemia (4% vs 1%, P = .04). Hospitalization (52% vs 13%, P < .001) and ICU admission (24% vs 9%, P < .001) were reported more often in the first-surge group.

 

Regression Analyses

Table 2 shows the results of logistic regression analyses for hospitalization in the pediatric (≤19 years of age) and adult (>19 years of age) groups, along with the odds of hospitalization during the first vs late surge among COVID-positive people with T1D. Adult patients who tested positive in the first surge were about 5 times more likely to be hospitalized than adults who tested positive for infection in the late surge after adjusting for age, insurance type, sex, race, and HbA1c levels. Pediatric patients also had an increased odds for hospitalization during the first surge, but this increase was not statistically significant.

Discussion

Our analysis of COVID-19 cases in patients with T1D reported by diabetes providers across the United States found that adverse outcomes were more prevalent early in the pandemic. There may be a number of reasons for this difference in outcomes between patients who presented in the first surge vs a later surge. First, because testing for COVID-19 was extremely limited and reserved for hospitalized patients early in the pandemic, the first-surge patients with confirmed COVID-19 likely represent a skewed population of higher-acuity patients. This may also explain the relative paucity of cases in younger patients reported early in the pandemic. Second, worse outcomes in the early surge may also have been associated with overwhelmed hospitals in New York City at the start of the outbreak. According to Cummings et al, the abrupt surge of critically ill patients hospitalized with severe acute respiratory distress syndrome initially outpaced their capacity to provide prone-positioning ventilation, which has been expanded since then.¹⁵ While there was very little hypertension, cardiovascular disease, or kidney disease reported in the pediatric groups, there was a higher prevalence of obesity in the pediatric group from the mid-Atlantic region. Obesity has been associated with a worse prognosis for COVID-19 illness in children.¹⁶ Finally, there were 5 deaths reported in this study, all of which were reported during the first surge. Older age and increased rates of cardiovascular disease and chronic kidney disease in the first surge cases likely contributed to worse outcomes for adults in mid-Atlantic region relative to the other regions. Minority race and the use of public insurance, risk factors for more severe outcomes in all regions, were also more common in cases reported from the mid-Atlantic region.

This study has several limitations. First, it is a cross-sectional study that relies upon voluntary provider reports. Second, availability of COVID-19 testing was limited in all regions in spring 2020. Third, different regions of the country experienced subsequent surges at different times within the reported timeframes in this analysis. Fourth, this report time period does not include the impact of the newer COVID-19 variants. Finally, trends in COVID-19 outcomes were affected by the evolution of care that developed throughout 2020.

Conclusion

Adult patients with T1D and COVID-19 who reported during the first surge had about 5 times higher hospitalization odds than those who presented in a later surge.

Corresponding author: Osagie Ebekozien, MD, MPH, 11 Avenue de Lafayette, Boston, MA 02111; oebekozien@t1dexchange.org

Disclosures: Dr Ebekozien reports receiving research grants from Medtronic Diabetes, Eli Lilly, and Dexcom, and receiving honoraria from Medtronic Diabetes.

From Hassenfeld Children’s Hospital at NYU Langone Health, New York, NY (Dr Gallagher), T1D Exchange, Boston, MA (Saketh Rompicherla; Drs Ebekozien, Noor, Odugbesan, and Mungmode; Nicole Rioles, Emma Ospelt), University of Mississippi School of Population Health, Jackson, MS (Dr. Ebekozien), Icahn School of Medicine at Mount Sinai, New York, NY (Drs. Wilkes, O’Malley, and Rapaport), Weill Cornell Medicine, New York, NY (Drs. Antal and Feuer), NYU Long Island School of Medicine, Mineola, NY (Dr. Gabriel), NYU Langone Health, New York, NY (Dr. Golden), Barbara Davis Center, Aurora, CO (Dr. Alonso), Texas Children’s Hospital/Baylor College of Medicine, Houston, TX (Dr. Lyons), Stanford University, Stanford, CA (Dr. Prahalad), Children Mercy Kansas City, MO (Dr. Clements), Indiana University School of Medicine, IN (Dr. Neyman), Rady Children’s Hospital, University of California, San Diego, CA (Dr. Demeterco-Berggren).

Background: Patient outcomes of COVID-19 have improved throughout the pandemic. However, because it is not known whether outcomes of COVID-19 in the type 1 diabetes (T1D) population improved over time, we investigated differences in COVID-19 outcomes for patients with T1D in the United States.

Methods: We analyzed data collected via a registry of patients with T1D and COVID-19 from 56 sites between April 2020 and January 2021. We grouped cases into first surge (April 9, 2020, to July 31, 2020, n = 188) and late surge (August 1, 2020, to January 31, 2021, n = 410), and then compared outcomes between both groups using descriptive statistics and logistic regression models.

Results: Adverse outcomes were more frequent during the first surge, including diabetic ketoacidosis (32% vs 15%, P < .001), severe hypoglycemia (4% vs 1%, P = .04), and hospitalization (52% vs 22%, P < .001). Patients in the first surge were older (28 [SD,18.8] years vs 18.0 [SD, 11.1] years, P < .001), had higher median hemoglobin A1c levels (9.3 [interquartile range {IQR}, 4.0] vs 8.4 (IQR, 2.8), P < .001), and were more likely to use public insurance (107 [57%] vs 154 [38%], P < .001). The odds of hospitalization for adults in the first surge were 5 times higher compared to the late surge (odds ratio, 5.01; 95% CI, 2.11-12.63).

Conclusion: Patients with T1D who presented with COVID-19 during the first surge had a higher proportion of adverse outcomes than those who presented in a later surge.

Keywords: TD1, diabetic ketoacidosis, hypoglycemia.

After the World Health Organization declared the disease caused by the novel coronavirus SARS-CoV-2, COVID-19, a pandemic on March 11, 2020, the Centers for Disease Control and Prevention identified patients with diabetes as high risk for severe illness.^1-7 The case-fatality rate for COVID-19 has significantly improved over the past 2 years. Public health measures, less severe COVID-19 variants, increased access to testing, and new treatments for COVID-19 have contributed to improved outcomes.

The T1D Exchange has previously published findings on COVID-19 outcomes for patients with type 1 diabetes (T1D) using data from the T1D COVID-19 Surveillance Registry.^8-12 Given improved outcomes in COVID-19 in the general population, we sought to determine if outcomes for cases of COVID-19 reported to this registry changed over time.

 

Methods

This study was coordinated by the T1D Exchange and approved as nonhuman subject research by the Western Institutional Review Board. All participating centers also obtained local institutional review board approval. No identifiable patient information was collected as part of this noninterventional, cross-sectional study.

The T1D Exchange Multi-center COVID-19 Surveillance Study collected data from endocrinology clinics that completed a retrospective chart review and submitted information to T1D Exchange via an online questionnaire for all patients with T1D at their sites who tested positive for COVID-19.^13,14 The questionnaire was administered using the Qualtrics survey platform (www.qualtrics.com version XM) and contained 33 pre-coded and free-text response fields to collect patient and clinical attributes.

Each participating center identified 1 team member for reporting to avoid duplicate case submission. Each submitted case was reviewed for potential errors and incomplete information. The coordinating center verified the number of cases per site for data quality assurance.

Quantitative data were represented as mean (standard deviation) or median (interquartile range). Categorical data were described as the number (percentage) of patients. Summary statistics, including frequency and percentage for categorical variables, were calculated for all patient-related and clinical characteristics. The date August 1, 2021, was selected as the end of the first surge based on a review of national COVID-19 surges.

We used the Fisher’s exact test to assess associations between hospitalization and demographics, HbA1c, diabetes duration, symptoms, and adverse outcomes. In addition, multivariate logistic regression was used to calculate odds ratios (OR). Logistic regression models were used to determine the association between time of surge and hospitalization separately for both the pediatric and adult populations. Each model was adjusted for potential sociodemographic confounders, specifically age, sex, race, insurance, and HbA1c.

All tests were 2-sided, with type 1 error set at 5%. Fisher’s exact test and logistic regression were performed using statistical software R, version 3.6.2 (R Foundation for Statistical Computing).

Results

The characteristics of COVID-19 cases in patients with T1D that were reported early in the pandemic, before August 1, 2020 (first surge), compared with those of cases reported on and after August 1, 2020 (later surges) are shown in Table 1.

Patients with T1D who presented with COVID-19 during the first surge as compared to the later surges were older (mean age 28 [SD, 18.0] years vs 18.8 [SD, 11.1] years; P < .001) and had a longer duration of diabetes (P < .001). The first-surge group also had more patients with >20 years’ diabetes duration (20% vs 9%, P < .001). Obesity, hypertension, and chronic kidney disease were also more commonly reported in first-surge cases (all P < .001).

There was a significant difference in race and ethnicity reported in the first surge vs the later surge cases, with fewer patients identifying as non-Hispanic White (39% vs, 63%, P < .001) and more patients identifying as non-Hispanic Black (29% vs 12%, P < .001). The groups also differed significantly in terms of insurance type, with more people on public insurance in the first-surge group (57% vs 38%, P < .001). In addition, median HbA1c was higher (9.3% vs 8.4%, P < .001) and continuous glucose monitor and insulin pump use were less common (P = .02 and <.001, respectively) in the early surge.

All symptoms and adverse outcomes were reported more often in the first surge, including diabetic ketoacidosis (DKA; 32% vs 15%; P < .001) and severe hypoglycemia (4% vs 1%, P = .04). Hospitalization (52% vs 13%, P < .001) and ICU admission (24% vs 9%, P < .001) were reported more often in the first-surge group.

 

Regression Analyses

Table 2 shows the results of logistic regression analyses for hospitalization in the pediatric (≤19 years of age) and adult (>19 years of age) groups, along with the odds of hospitalization during the first vs late surge among COVID-positive people with T1D. Adult patients who tested positive in the first surge were about 5 times more likely to be hospitalized than adults who tested positive for infection in the late surge after adjusting for age, insurance type, sex, race, and HbA1c levels. Pediatric patients also had an increased odds for hospitalization during the first surge, but this increase was not statistically significant.

Discussion

Our analysis of COVID-19 cases in patients with T1D reported by diabetes providers across the United States found that adverse outcomes were more prevalent early in the pandemic. There may be a number of reasons for this difference in outcomes between patients who presented in the first surge vs a later surge. First, because testing for COVID-19 was extremely limited and reserved for hospitalized patients early in the pandemic, the first-surge patients with confirmed COVID-19 likely represent a skewed population of higher-acuity patients. This may also explain the relative paucity of cases in younger patients reported early in the pandemic. Second, worse outcomes in the early surge may also have been associated with overwhelmed hospitals in New York City at the start of the outbreak. According to Cummings et al, the abrupt surge of critically ill patients hospitalized with severe acute respiratory distress syndrome initially outpaced their capacity to provide prone-positioning ventilation, which has been expanded since then.¹⁵ While there was very little hypertension, cardiovascular disease, or kidney disease reported in the pediatric groups, there was a higher prevalence of obesity in the pediatric group from the mid-Atlantic region. Obesity has been associated with a worse prognosis for COVID-19 illness in children.¹⁶ Finally, there were 5 deaths reported in this study, all of which were reported during the first surge. Older age and increased rates of cardiovascular disease and chronic kidney disease in the first surge cases likely contributed to worse outcomes for adults in mid-Atlantic region relative to the other regions. Minority race and the use of public insurance, risk factors for more severe outcomes in all regions, were also more common in cases reported from the mid-Atlantic region.

This study has several limitations. First, it is a cross-sectional study that relies upon voluntary provider reports. Second, availability of COVID-19 testing was limited in all regions in spring 2020. Third, different regions of the country experienced subsequent surges at different times within the reported timeframes in this analysis. Fourth, this report time period does not include the impact of the newer COVID-19 variants. Finally, trends in COVID-19 outcomes were affected by the evolution of care that developed throughout 2020.

Conclusion

Adult patients with T1D and COVID-19 who reported during the first surge had about 5 times higher hospitalization odds than those who presented in a later surge.

Corresponding author: Osagie Ebekozien, MD, MPH, 11 Avenue de Lafayette, Boston, MA 02111; oebekozien@t1dexchange.org

Disclosures: Dr Ebekozien reports receiving research grants from Medtronic Diabetes, Eli Lilly, and Dexcom, and receiving honoraria from Medtronic Diabetes.

From Hassenfeld Children’s Hospital at NYU Langone Health, New York, NY (Dr Gallagher), T1D Exchange, Boston, MA (Saketh Rompicherla; Drs Ebekozien, Noor, Odugbesan, and Mungmode; Nicole Rioles, Emma Ospelt), University of Mississippi School of Population Health, Jackson, MS (Dr. Ebekozien), Icahn School of Medicine at Mount Sinai, New York, NY (Drs. Wilkes, O’Malley, and Rapaport), Weill Cornell Medicine, New York, NY (Drs. Antal and Feuer), NYU Long Island School of Medicine, Mineola, NY (Dr. Gabriel), NYU Langone Health, New York, NY (Dr. Golden), Barbara Davis Center, Aurora, CO (Dr. Alonso), Texas Children’s Hospital/Baylor College of Medicine, Houston, TX (Dr. Lyons), Stanford University, Stanford, CA (Dr. Prahalad), Children Mercy Kansas City, MO (Dr. Clements), Indiana University School of Medicine, IN (Dr. Neyman), Rady Children’s Hospital, University of California, San Diego, CA (Dr. Demeterco-Berggren).

Background: Patient outcomes of COVID-19 have improved throughout the pandemic. However, because it is not known whether outcomes of COVID-19 in the type 1 diabetes (T1D) population improved over time, we investigated differences in COVID-19 outcomes for patients with T1D in the United States.

Methods: We analyzed data collected via a registry of patients with T1D and COVID-19 from 56 sites between April 2020 and January 2021. We grouped cases into first surge (April 9, 2020, to July 31, 2020, n = 188) and late surge (August 1, 2020, to January 31, 2021, n = 410), and then compared outcomes between both groups using descriptive statistics and logistic regression models.

Results: Adverse outcomes were more frequent during the first surge, including diabetic ketoacidosis (32% vs 15%, P < .001), severe hypoglycemia (4% vs 1%, P = .04), and hospitalization (52% vs 22%, P < .001). Patients in the first surge were older (28 [SD,18.8] years vs 18.0 [SD, 11.1] years, P < .001), had higher median hemoglobin A1c levels (9.3 [interquartile range {IQR}, 4.0] vs 8.4 (IQR, 2.8), P < .001), and were more likely to use public insurance (107 [57%] vs 154 [38%], P < .001). The odds of hospitalization for adults in the first surge were 5 times higher compared to the late surge (odds ratio, 5.01; 95% CI, 2.11-12.63).

Conclusion: Patients with T1D who presented with COVID-19 during the first surge had a higher proportion of adverse outcomes than those who presented in a later surge.

Keywords: TD1, diabetic ketoacidosis, hypoglycemia.

After the World Health Organization declared the disease caused by the novel coronavirus SARS-CoV-2, COVID-19, a pandemic on March 11, 2020, the Centers for Disease Control and Prevention identified patients with diabetes as high risk for severe illness.^1-7 The case-fatality rate for COVID-19 has significantly improved over the past 2 years. Public health measures, less severe COVID-19 variants, increased access to testing, and new treatments for COVID-19 have contributed to improved outcomes.

The T1D Exchange has previously published findings on COVID-19 outcomes for patients with type 1 diabetes (T1D) using data from the T1D COVID-19 Surveillance Registry.^8-12 Given improved outcomes in COVID-19 in the general population, we sought to determine if outcomes for cases of COVID-19 reported to this registry changed over time.

 

Methods

This study was coordinated by the T1D Exchange and approved as nonhuman subject research by the Western Institutional Review Board. All participating centers also obtained local institutional review board approval. No identifiable patient information was collected as part of this noninterventional, cross-sectional study.

The T1D Exchange Multi-center COVID-19 Surveillance Study collected data from endocrinology clinics that completed a retrospective chart review and submitted information to T1D Exchange via an online questionnaire for all patients with T1D at their sites who tested positive for COVID-19.^13,14 The questionnaire was administered using the Qualtrics survey platform (www.qualtrics.com version XM) and contained 33 pre-coded and free-text response fields to collect patient and clinical attributes.

Each participating center identified 1 team member for reporting to avoid duplicate case submission. Each submitted case was reviewed for potential errors and incomplete information. The coordinating center verified the number of cases per site for data quality assurance.

Quantitative data were represented as mean (standard deviation) or median (interquartile range). Categorical data were described as the number (percentage) of patients. Summary statistics, including frequency and percentage for categorical variables, were calculated for all patient-related and clinical characteristics. The date August 1, 2021, was selected as the end of the first surge based on a review of national COVID-19 surges.

We used the Fisher’s exact test to assess associations between hospitalization and demographics, HbA1c, diabetes duration, symptoms, and adverse outcomes. In addition, multivariate logistic regression was used to calculate odds ratios (OR). Logistic regression models were used to determine the association between time of surge and hospitalization separately for both the pediatric and adult populations. Each model was adjusted for potential sociodemographic confounders, specifically age, sex, race, insurance, and HbA1c.

All tests were 2-sided, with type 1 error set at 5%. Fisher’s exact test and logistic regression were performed using statistical software R, version 3.6.2 (R Foundation for Statistical Computing).

Results

The characteristics of COVID-19 cases in patients with T1D that were reported early in the pandemic, before August 1, 2020 (first surge), compared with those of cases reported on and after August 1, 2020 (later surges) are shown in Table 1.

Patients with T1D who presented with COVID-19 during the first surge as compared to the later surges were older (mean age 28 [SD, 18.0] years vs 18.8 [SD, 11.1] years; P < .001) and had a longer duration of diabetes (P < .001). The first-surge group also had more patients with >20 years’ diabetes duration (20% vs 9%, P < .001). Obesity, hypertension, and chronic kidney disease were also more commonly reported in first-surge cases (all P < .001).

There was a significant difference in race and ethnicity reported in the first surge vs the later surge cases, with fewer patients identifying as non-Hispanic White (39% vs, 63%, P < .001) and more patients identifying as non-Hispanic Black (29% vs 12%, P < .001). The groups also differed significantly in terms of insurance type, with more people on public insurance in the first-surge group (57% vs 38%, P < .001). In addition, median HbA1c was higher (9.3% vs 8.4%, P < .001) and continuous glucose monitor and insulin pump use were less common (P = .02 and <.001, respectively) in the early surge.

All symptoms and adverse outcomes were reported more often in the first surge, including diabetic ketoacidosis (DKA; 32% vs 15%; P < .001) and severe hypoglycemia (4% vs 1%, P = .04). Hospitalization (52% vs 13%, P < .001) and ICU admission (24% vs 9%, P < .001) were reported more often in the first-surge group.

 

Regression Analyses

Table 2 shows the results of logistic regression analyses for hospitalization in the pediatric (≤19 years of age) and adult (>19 years of age) groups, along with the odds of hospitalization during the first vs late surge among COVID-positive people with T1D. Adult patients who tested positive in the first surge were about 5 times more likely to be hospitalized than adults who tested positive for infection in the late surge after adjusting for age, insurance type, sex, race, and HbA1c levels. Pediatric patients also had an increased odds for hospitalization during the first surge, but this increase was not statistically significant.

Discussion

Our analysis of COVID-19 cases in patients with T1D reported by diabetes providers across the United States found that adverse outcomes were more prevalent early in the pandemic. There may be a number of reasons for this difference in outcomes between patients who presented in the first surge vs a later surge. First, because testing for COVID-19 was extremely limited and reserved for hospitalized patients early in the pandemic, the first-surge patients with confirmed COVID-19 likely represent a skewed population of higher-acuity patients. This may also explain the relative paucity of cases in younger patients reported early in the pandemic. Second, worse outcomes in the early surge may also have been associated with overwhelmed hospitals in New York City at the start of the outbreak. According to Cummings et al, the abrupt surge of critically ill patients hospitalized with severe acute respiratory distress syndrome initially outpaced their capacity to provide prone-positioning ventilation, which has been expanded since then.¹⁵ While there was very little hypertension, cardiovascular disease, or kidney disease reported in the pediatric groups, there was a higher prevalence of obesity in the pediatric group from the mid-Atlantic region. Obesity has been associated with a worse prognosis for COVID-19 illness in children.¹⁶ Finally, there were 5 deaths reported in this study, all of which were reported during the first surge. Older age and increased rates of cardiovascular disease and chronic kidney disease in the first surge cases likely contributed to worse outcomes for adults in mid-Atlantic region relative to the other regions. Minority race and the use of public insurance, risk factors for more severe outcomes in all regions, were also more common in cases reported from the mid-Atlantic region.

This study has several limitations. First, it is a cross-sectional study that relies upon voluntary provider reports. Second, availability of COVID-19 testing was limited in all regions in spring 2020. Third, different regions of the country experienced subsequent surges at different times within the reported timeframes in this analysis. Fourth, this report time period does not include the impact of the newer COVID-19 variants. Finally, trends in COVID-19 outcomes were affected by the evolution of care that developed throughout 2020.

Conclusion

Adult patients with T1D and COVID-19 who reported during the first surge had about 5 times higher hospitalization odds than those who presented in a later surge.

Corresponding author: Osagie Ebekozien, MD, MPH, 11 Avenue de Lafayette, Boston, MA 02111; oebekozien@t1dexchange.org

Disclosures: Dr Ebekozien reports receiving research grants from Medtronic Diabetes, Eli Lilly, and Dexcom, and receiving honoraria from Medtronic Diabetes.

From the Department of Emergency Medicine, Santa Croce e Carle Hospital, Cuneo, Italy (Drs. Abram, Tosello, Emanuele Bernardi, Allione, Cavalot, Dutto, Corsini, Martini, Sciolla, Sara Bernardi, and Lauria). From the School of Emergency Medicine, University of Turin, Turin, Italy (Drs. Paglietta and Giamello).

Objective: This retrospective and prospective cohort study was designed to describe the characteristics, treatments, and outcomes of patients with SARS-CoV-2 infection (COVID-19) admitted to subintensive care units (SICU) and to identify the variables associated with outcomes. SICUs have been extremely stressed during the pandemic, but most data regarding critically ill COVID-19 patients come from intensive care units (ICUs). Studies about COVID-19 patients in SICUs are lacking.

Setting and participants: The study included 88 COVID-19 patients admitted to our SICU in Cuneo, Italy, between March and May 2020.

Measurements: Clinical and ventilatory data were collected, and patients were divided by outcome. Multivariable logistic regression analysis examined the variables associated with negative outcomes (transfer to the ICU, palliation, or death in a SICU).

Results: A total of 60 patients (68%) had a positive outcome, and 28 patients (32%) had a negative outcome; 69 patients (78%) underwent continuous positive airway pressure (CPAP). Pronation (n = 37 [42%]) had been more frequently adopted in patients who had a positive outcome vs a negative outcome (n = 30 [50%] vs n = 7 [25%]; P = .048), and the median (interquartile range) Pao₂/Fio₂ ratio after 6 hours of prone positioning was lower in patients who had a negative outcome vs a positive outcome (144 [140-168] vs 249 [195-268], P = .006). Independent predictors of a negative outcome were diabetes (odds ratio [OR], 8.22; 95% CI, 1.50-44.70; P = .015), higher D-dimer (OR, 1.28; 95% CI, 1.04-1.57; P = .019), higher lactate dehydrogenase level (OR, 1.003; 95% CI, 1.000-1.006; P = .039), and lower lymphocytes count (OR, 0.996; 95% CI, 0.993-0.999; P = .004).

Conclusion: SICUs have a fundamental role in the treatment of critically ill patients with COVID-19, who require long-term CPAP and pronation cycles. Diabetes, lymphopenia, and high D-dimer and LDH levels are associated with negative outcomes.

Keywords: emergency medicine, noninvasive ventilation, prone position, continuous positive airway pressure.

The COVID-19 pandemic has led to large increases in hospital admissions. Subintensive care units (SICUs) are among the wards most under pressure worldwide,¹ dealing with the increased number of critically ill patients who need noninvasive ventilation, as well as serving as the best alternative to overfilled intensive care units (ICUs). In Italy, SICUs are playing a fundamental role in the management of COVID-19 patients, providing early treatment of respiratory failure by continuous noninvasive ventilation in order to reduce the need for intubation.^2-5 Nevertheless, the great majority of available data about critically ill COVID-19 patients comes from ICUs. Full studies about outcomes of patients in SICUs are lacking and need to be conducted.

We sought to evaluate the characteristics and outcomes of patients admitted to our SICU for COVID-19 to describe the treatments they needed and their impact on prognosis, and to identify the variables associated with patient outcomes.

Methods

Study Design

This cohort study used data from patients who were admitted in the very first weeks of the pandemic. Data were collected retrospectively as well as prospectively, since the ethical committee approved our project. The quality and quantity of data in the 2 groups were comparable.

Data were collected from electronic and written medical records gathered during the patient’s entire stay in our SICU. Data were entered in a database with limited and controlled access. This study complied with the Declaration of Helsinki and was approved by the local ethics committees (ID: MEDURG10).

Study Population

We studied 88 consecutive patients admitted to the SICU of the Santa Croce e Carle Teaching Hospital, Cuneo, Italy, for COVID-19, from March 8 to May 1, 2020. The diagnosis was based on acute respiratory failure associated with SARS-CoV-2 RNA detection on nasopharyngeal swab or tracheal aspirate and/or typical COVID-19 features on a pulmonary computed tomography (CT) scan.⁶ Exclusion criteria were age younger than 18 years and patient denial of permission to use their data for research purposes (the great majority of patients could actively give consent; for patients who were too sick to do so, family members were asked whether they were aware of any reason why the patient would deny consent).

 

 

Clinical Data

The past medical history and recent symptoms description were obtained by manually reviewing medical records. Epidemiological exposure was defined as contact with SARS-CoV-2–positive people or staying in an epidemic outbreak area. Initial vital parameters, venous blood tests, arterial blood gas analysis, chest x-ray, as well as the result of the nasopharyngeal swab were gathered from the emergency department (ED) examination. (Additional swabs could be requested when the first one was negative but clinical suspicion for COVID-19 was high.) Upon admission to the SICU, a standardized panel of blood tests was performed, which was repeated the next day and then every 48 hours. Arterial blood gas analysis was performed when clinically indicated, at least twice a day, or following a scheduled time in patients undergoing pronation. Charlson Comorbidity Index⁷ and MuLBSTA score⁸ were calculated based on the collected data.

Imaging

Chest ultrasonography was performed in the ED at the time of hospitalization and once a day in the SICU. Pulmonary high-resolution computed tomography (HRCT) was performed when clinically indicated or when the results of nasopharyngeal swabs and/or x-ray results were discordant with COVID-19 clinical suspicion. Contrast CT was performed when pulmonary embolism was suspected.

Medical Therapy

Hydroxychloroquine, antiviral agents, tocilizumab, and ruxolitinib were used in the early phase of the pandemic, then were dismissed after evidence of no efficacy.^9-11 Steroids and low-molecular-weight heparin were used afterward. Enoxaparin was used at the standard prophylactic dosage, and 70% of the anticoagulant dosage was also adopted in patients with moderate-to-severe COVID-19 and D-dimer values >3 times the normal value.^12-14 Antibiotics were given when a bacterial superinfection was suspected.

Oxygen and Ventilatory Therapy

Oxygen support or noninvasive ventilation were started based on patients’ respiratory efficacy, estimated by respiratory rate and the ratio of partial pressure of arterial oxygen and fraction of inspired oxygen (P/F ratio).^15,16 Oxygen support was delivered through nasal cannula, Venturi mask, or reservoir mask. Noninvasive ventilation was performed by continuous positive airway pressure (CPAP) when the P/F ratio was <250 or the respiratory rate was >25 breaths per minute, using the helmet interface.^5,17 Prone positioning during CPAP^18-20 was adopted in patients meeting the acute respiratory distress syndrome (ARDS) criteria²¹ and having persistence of respiratory distress and P/F <300 after a 1-hour trial of CPAP.

The prone position was maintained based on patient tolerance. P/F ratio was measured before pronation (T0), after 1 hour of prone position (T1), before resupination (T2), and 6 hours after resupination (T3). With the same timing, the patient was asked to rate their comfort in each position, from 0 (lack of comfort) to 10 (optimal comfort). Delta P/F was defined as the difference between P/F at T3 and basal P/F at T0.

Outcomes

Positive outcomes were defined as patient discharge from the SICU or transfer to a lower-intensity care ward for treatment continuation. Negative outcomes were defined as need for transfer to the ICU, transfer to another ward for palliation, or death in the SICU.

Statistical Analysis

Continuous data are reported as median and interquartile range (IQR); normal distribution of variables was tested using the Shapiro-Wilk test. Categorical variables were reported as absolute number and percentage. The Mann-Whitney test was used to compare continuous variables between groups, and chi-square test with continuity correction was used for categorical variables. The variables that were most significantly associated with a negative outcome on the univariate analysis were included in a stepwise logistic regression analysis, in order to identify independent predictors of patient outcome. Statistical analysis was performed using JASP (JASP Team) software.

 

 

Results

Study Population

Of the 88 patients included in the study, 70% were male; the median age was 66 years (IQR, 60-77). In most patients, the diagnosis of COVID-19 was derived from a positive SARS-CoV-2 nasopharyngeal swab. Six patients, however, maintained a negative swab at all determinations but had clinical and imaging features strongly suggesting COVID-19. No patients met the exclusion criteria. Most patients came from the ED (n = 58 [66%]) or general wards (n = 22 [25%]), while few were transferred from the ICU (n = 8 [9%]). The median length of stay in the SICU was 4 days (IQR, 2-7). An epidemiological link to affected persons or a known virus exposure was identifiable in 37 patients (42%).

Clinical, Laboratory, and Imaging Data

The clinical and anthropometric characteristics of patients are shown in Table 1. Hypertension and smoking habits were prevalent in our population, and the median Charlson Comorbidity Index was 3. Most patients experienced fever, dyspnea, and cough during the days before hospitalization.

Laboratory data showed a marked inflammatory milieu in all studied patients, both at baseline and after 24 and 72 hours. Lymphopenia was observed, along with a significant increase of lactate dehydrogenase (LDH), C-reactive protein (CPR), and D-dimer, and a mild increase of procalcitonin. N-terminal pro-brain natriuretic peptide (NT-proBNP) values were also increased, with normal troponin I values (Table 2).

Chest x-rays were obtained in almost all patients, while HRCT was performed in nearly half of patients. Complete bedside pulmonary ultrasonography data were available for 64 patients. Heterogeneous pulmonary alterations were found, regardless of the radiological technique, and multilobe infiltrates were the prevalent radiological pattern (73%) (Table 3). Seven patients (8%) were diagnosed with associated pulmonary embolism.

 

Medical Therapy

Most patients (89%) received hydroxychloroquine, whereas steroids were used in one-third of the population (36%). Immunomodulators (tocilizumab and ruxolitinib) were restricted to 12 patients (14%). Empirical antiviral therapy was introduced in the first 41 patients (47%). Enoxaparin was the default agent for thromboembolism prophylaxis, and 6 patients (7%) received 70% of the anticoagulating dose.

Oxygen and Ventilatory Therapy

Basal median P/F ratio was 253 (IQR, 218-291), and respiratory rate at triage was 20 breaths/min (IQR, 16-28), underlining a moderate-to-severe respiratory insufficiency at presentation. A total of 69 patients (78%) underwent CPAP, with a median positive end-expiratory pressure (PEEP) of 10.0 cm H₂O (IQR, 7.5-10.0) and fraction of inspired oxygen (Fio₂) of 0.40 (IQR, 0.40-0.50). In 37 patients (42%) who received ongoing CPAP, prone positioning was adopted. In this subgroup, respiratory rate was not significantly different from baseline to resupination (24 vs 25 breaths/min). The median P/F improved from 197 (IQR, 154-236) at baseline to 217 (IQR, 180-262) after pronation (the duration of the prone position was variable, depending on patients’ tolerance: 1 to 6 hours or every pronation cycle). The median delta P/F ratio was 39.4 (IQR, –17.0 to 78.0).

Outcomes

A total of 28 patients (32%) had a negative outcome in the SICU: 8 patients (9%) died, having no clinical indication for higher-intensity care; 6 patients (7%) were transferred to general wards for palliation; and 14 patients (16%) needed an upgrade of cure intensity and were transferred to the ICU. Of these 14 patients, 9 died in the ICU. The total in-hospital mortality of COVID-19 patients, including patients transferred from the SICU to general wards in fair condition, was 27% (n = 24). Clinical, laboratory, and therapeutic characteristics between the 2 groups are shown in Table 4.

Patients who had a negative outcome were significantly older and had more comorbidities, as suggested by a significantly higher prevalence of diabetes and higher Charlson Comorbidity scores (reflecting the mortality risk based on age and comorbidities). The median MuLBSTA score, which estimates the 90-day mortality risk from viral pneumonia, was also higher in patients who had a negative outcome (9.33%). Symptom occurrence was not different in patients with a negative outcome (apart from cough, which was less frequent), but these patients underwent hospitalization earlier—since the appearance of their first COVID-19 symptoms—compared to patients who had a positive outcome. No difference was found in antihypertensive therapy with angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers among outcome groups.

More pronounced laboratory abnormalities were found in patients who had a negative outcome, compared to patients who had a positive outcome: lower lymphocytes and higher C-reactive protein (CRP), procalcitonin, D-dimer, LDH, and NT-proBNP. We found no differences in the radiological distribution of pulmonary involvement in patients who had negative or positive outcomes, nor in the adopted medical treatment.

Data showed no difference in CPAP implementation in the 2 groups. However, prone positioning had been more frequently adopted in the group of patients who had a positive outcome, compared with patients who had a negative outcome. No differences of basal P/F were found in patients who had a negative or positive outcome, but the median P/F after 6 hours of prone position was significantly lower in patients who had a negative outcome. The delta P/F ratio did not differ in the 2 groups of patients.

Multivariate Analysis

A logistic regression model was created, including the variables significantly associated with outcomes in the univariate analysis (age, sex, history of diabetes, lymphocytes, CRP, procalcitonin, LDH, NT-proBNP, and D-dimer). In the multivariate analysis, independent predictors of a negative outcome were history of diabetes (odds ratio [OR], 8.22; 95% CI, 1.50-44.70; P =.015), high D-dimer values (OR, 1.28; CI, 1.04-1.57; P = .019), high LDH values (OR, 1.003; CI, 1.000-1.006; P = .039), and low lymphocytes count (OR, 0.996; CI, 0.993-0.999; P = .004).

 

Discussion

Role of Subintensive Units and Mortality

The novelty of our report is its attempt to investigate the specific group of COVID-19 patients admitted to a SICU. In Italy, SICUs receive acutely ill, spontaneously breathing patients who need (invasive) hemodynamic monitoring, vasoactive medication, renal replacement therapy, chest- tube placement, thrombolysis, and respiratory noninvasive support. The nurse-to-patient ratio is higher than for general wards (usually 1 nurse to every 4 or 5 patients), though lower than for ICUs. In northern Italy, a great number of COVID-19 patients have required this kind of high-intensity care during the pandemic: Noninvasive ventilation support had to be maintained for several days, pronation maneuvers required a high number of people 2 or 3 times a day, and strict monitoring had to be assured. The SICU setting allows patients to buy time as a bridge to progressive reduction of pulmonary involvement, sometimes preventing the need for intubation.

The high prevalence of negative outcomes in the SICU underlines the complexity of COVID-19 patients in this setting. In fact, published data about mortality for patients with severe COVID-19 pneumonia are similar to ours.^22,23

Clinical, Laboratory, and Imaging Data

Our analysis confirmed a high rate of comorbidities in COVID-19 patients²⁴ and their prognostic role with age.^25,26 A marked inflammatory milieu was a negative prognostic indicator, and associated concomitant bacterial superinfection could have led to a worse prognosis (procalcitonin was associated with negative outcomes).²⁷ The cardiovascular system was nevertheless stressed, as suggested by higher values of NT-proBNP in patients with negative outcomes, which could reflect sepsis-related systemic involvement.²⁸

It is known that the pulmonary damage caused by SARS-CoV-2 has a dynamic radiological and clinical course, with early areas of subsegmental consolidation, and bilateral ground-glass opacities predominating later in the course of the disease.²⁹ This could explain why in our population we found no specific radiological pattern leading to a worse outcome.

Medical Therapy

No specific pharmacological therapy was found to be associated with a positive outcome in our study, just like antiviral and immunomodulator therapies failed to demonstrate effectiveness in subsequent pandemic surges. The low statistical power of our study did not allow us to give insight into the effectiveness of steroids and heparin at any dosage.

PEEP Support and Prone Positioning

Continuous positive airway pressure was initiated in the majority of patients and maintained for several days. This was an absolute novelty, because we rarely had to keep patients in helmets for long. This was feasible thanks to the SICU’s high nurse-to-patient ratio and the possibility of providing monitored sedation. Patients who could no longer tolerate CPAP helmets or did not improve with CPAP support were evaluated with anesthetists for programming further management. No initial data on respiratory rate, level of hypoxemia, or oxygen support need (level of PEEP and Fio₂) could discriminate between outcomes.

Prone positioning during CPAP was implemented in 42% of our study population: P/F ratio amelioration after prone positioning was highly variable, ranging from very good P/F ratio improvements to few responses or no response. No significantly greater delta P/F ratio was seen after the first prone positioning cycle in patients who had a positive outcome, probably due to the small size of our population, but we observed a clear positive trend. Interestingly, patients showing a negative outcome had a lower percentage of long-term responses to prone positioning: 6 hours after resupination, they lost the benefit of prone positioning in terms of P/F ratio amelioration. Similarly, a greater number of patients tolerating prone positioning had a positive outcome. These data give insight on the possible benefits of prone positioning in a noninvasively supported cohort of patients, which has been mentioned in previous studies.^30,31

 

Outcomes and Variables Associated With Negative Outcomes

After correction for age and sex, we found in multiple regression analysis that higher D-dimer and LDH values, lymphopenia, and history of diabetes were independently associated with a worse outcome. Although our results had low statistical significance, we consider the trend of the obtained odds ratios important from a clinical point of view. These results could lead to greater attention being placed on COVID-19 patients who present with these characteristics upon their arrival to the ED because they have increased risk of death or intensive care need. Clinicians should consider SICU admission for these patients in order to guarantee closer monitoring and possibly more aggressive ventilatory treatments, earlier pronation, or earlier transfer to the ICU.

Limitations

The major limitation to our study is undoubtedly its statistical power, due to its relatively low patient population. Particularly, the small number of patients who underwent pronation did not allow speculation about the efficacy of this technique, although preliminary data seem promising. However, ours is among the first studies regarding patients with COVID-19 admitted to a SICU, and these preliminary data truthfully describe the Italian, and perhaps international, experience with the first surge of the pandemic.

Conclusions

Our data highlight the primary role of the SICU in COVID-19 in adequately treating critically ill patients who have high care needs different from intubation, and who require noninvasive ventilation for prolonged times as well as frequent pronation cycles. This setting of care may represent a valid, reliable, and effective option for critically ill respiratory patients. History of diabetes, lymphopenia, and high D-dimer and LDH values are independently associated with negative outcomes, and patients presenting with these characteristics should be strictly monitored.

Acknowledgments: The authors thank the Informatica System S.R.L., as well as Allessando Mendolia for the pro bono creation of the ISCovidCollect data collecting app.

Corresponding author: Sara Abram, MD, via Coppino, 12100 Cuneo, Italy; sara.abram84@gmail.com.

Disclosures: None.

From the Department of Emergency Medicine, Santa Croce e Carle Hospital, Cuneo, Italy (Drs. Abram, Tosello, Emanuele Bernardi, Allione, Cavalot, Dutto, Corsini, Martini, Sciolla, Sara Bernardi, and Lauria). From the School of Emergency Medicine, University of Turin, Turin, Italy (Drs. Paglietta and Giamello).

Objective: This retrospective and prospective cohort study was designed to describe the characteristics, treatments, and outcomes of patients with SARS-CoV-2 infection (COVID-19) admitted to subintensive care units (SICU) and to identify the variables associated with outcomes. SICUs have been extremely stressed during the pandemic, but most data regarding critically ill COVID-19 patients come from intensive care units (ICUs). Studies about COVID-19 patients in SICUs are lacking.

Setting and participants: The study included 88 COVID-19 patients admitted to our SICU in Cuneo, Italy, between March and May 2020.

Measurements: Clinical and ventilatory data were collected, and patients were divided by outcome. Multivariable logistic regression analysis examined the variables associated with negative outcomes (transfer to the ICU, palliation, or death in a SICU).

Results: A total of 60 patients (68%) had a positive outcome, and 28 patients (32%) had a negative outcome; 69 patients (78%) underwent continuous positive airway pressure (CPAP). Pronation (n = 37 [42%]) had been more frequently adopted in patients who had a positive outcome vs a negative outcome (n = 30 [50%] vs n = 7 [25%]; P = .048), and the median (interquartile range) Pao₂/Fio₂ ratio after 6 hours of prone positioning was lower in patients who had a negative outcome vs a positive outcome (144 [140-168] vs 249 [195-268], P = .006). Independent predictors of a negative outcome were diabetes (odds ratio [OR], 8.22; 95% CI, 1.50-44.70; P = .015), higher D-dimer (OR, 1.28; 95% CI, 1.04-1.57; P = .019), higher lactate dehydrogenase level (OR, 1.003; 95% CI, 1.000-1.006; P = .039), and lower lymphocytes count (OR, 0.996; 95% CI, 0.993-0.999; P = .004).

Conclusion: SICUs have a fundamental role in the treatment of critically ill patients with COVID-19, who require long-term CPAP and pronation cycles. Diabetes, lymphopenia, and high D-dimer and LDH levels are associated with negative outcomes.

Keywords: emergency medicine, noninvasive ventilation, prone position, continuous positive airway pressure.

The COVID-19 pandemic has led to large increases in hospital admissions. Subintensive care units (SICUs) are among the wards most under pressure worldwide,¹ dealing with the increased number of critically ill patients who need noninvasive ventilation, as well as serving as the best alternative to overfilled intensive care units (ICUs). In Italy, SICUs are playing a fundamental role in the management of COVID-19 patients, providing early treatment of respiratory failure by continuous noninvasive ventilation in order to reduce the need for intubation.^2-5 Nevertheless, the great majority of available data about critically ill COVID-19 patients comes from ICUs. Full studies about outcomes of patients in SICUs are lacking and need to be conducted.

We sought to evaluate the characteristics and outcomes of patients admitted to our SICU for COVID-19 to describe the treatments they needed and their impact on prognosis, and to identify the variables associated with patient outcomes.

Methods

Study Design

This cohort study used data from patients who were admitted in the very first weeks of the pandemic. Data were collected retrospectively as well as prospectively, since the ethical committee approved our project. The quality and quantity of data in the 2 groups were comparable.

Data were collected from electronic and written medical records gathered during the patient’s entire stay in our SICU. Data were entered in a database with limited and controlled access. This study complied with the Declaration of Helsinki and was approved by the local ethics committees (ID: MEDURG10).

Study Population

We studied 88 consecutive patients admitted to the SICU of the Santa Croce e Carle Teaching Hospital, Cuneo, Italy, for COVID-19, from March 8 to May 1, 2020. The diagnosis was based on acute respiratory failure associated with SARS-CoV-2 RNA detection on nasopharyngeal swab or tracheal aspirate and/or typical COVID-19 features on a pulmonary computed tomography (CT) scan.⁶ Exclusion criteria were age younger than 18 years and patient denial of permission to use their data for research purposes (the great majority of patients could actively give consent; for patients who were too sick to do so, family members were asked whether they were aware of any reason why the patient would deny consent).

 

 

Clinical Data

The past medical history and recent symptoms description were obtained by manually reviewing medical records. Epidemiological exposure was defined as contact with SARS-CoV-2–positive people or staying in an epidemic outbreak area. Initial vital parameters, venous blood tests, arterial blood gas analysis, chest x-ray, as well as the result of the nasopharyngeal swab were gathered from the emergency department (ED) examination. (Additional swabs could be requested when the first one was negative but clinical suspicion for COVID-19 was high.) Upon admission to the SICU, a standardized panel of blood tests was performed, which was repeated the next day and then every 48 hours. Arterial blood gas analysis was performed when clinically indicated, at least twice a day, or following a scheduled time in patients undergoing pronation. Charlson Comorbidity Index⁷ and MuLBSTA score⁸ were calculated based on the collected data.

Imaging

Chest ultrasonography was performed in the ED at the time of hospitalization and once a day in the SICU. Pulmonary high-resolution computed tomography (HRCT) was performed when clinically indicated or when the results of nasopharyngeal swabs and/or x-ray results were discordant with COVID-19 clinical suspicion. Contrast CT was performed when pulmonary embolism was suspected.

Medical Therapy

Hydroxychloroquine, antiviral agents, tocilizumab, and ruxolitinib were used in the early phase of the pandemic, then were dismissed after evidence of no efficacy.^9-11 Steroids and low-molecular-weight heparin were used afterward. Enoxaparin was used at the standard prophylactic dosage, and 70% of the anticoagulant dosage was also adopted in patients with moderate-to-severe COVID-19 and D-dimer values >3 times the normal value.^12-14 Antibiotics were given when a bacterial superinfection was suspected.

Oxygen and Ventilatory Therapy

Oxygen support or noninvasive ventilation were started based on patients’ respiratory efficacy, estimated by respiratory rate and the ratio of partial pressure of arterial oxygen and fraction of inspired oxygen (P/F ratio).^15,16 Oxygen support was delivered through nasal cannula, Venturi mask, or reservoir mask. Noninvasive ventilation was performed by continuous positive airway pressure (CPAP) when the P/F ratio was <250 or the respiratory rate was >25 breaths per minute, using the helmet interface.^5,17 Prone positioning during CPAP^18-20 was adopted in patients meeting the acute respiratory distress syndrome (ARDS) criteria²¹ and having persistence of respiratory distress and P/F <300 after a 1-hour trial of CPAP.

The prone position was maintained based on patient tolerance. P/F ratio was measured before pronation (T0), after 1 hour of prone position (T1), before resupination (T2), and 6 hours after resupination (T3). With the same timing, the patient was asked to rate their comfort in each position, from 0 (lack of comfort) to 10 (optimal comfort). Delta P/F was defined as the difference between P/F at T3 and basal P/F at T0.

Outcomes

Positive outcomes were defined as patient discharge from the SICU or transfer to a lower-intensity care ward for treatment continuation. Negative outcomes were defined as need for transfer to the ICU, transfer to another ward for palliation, or death in the SICU.

Statistical Analysis

Continuous data are reported as median and interquartile range (IQR); normal distribution of variables was tested using the Shapiro-Wilk test. Categorical variables were reported as absolute number and percentage. The Mann-Whitney test was used to compare continuous variables between groups, and chi-square test with continuity correction was used for categorical variables. The variables that were most significantly associated with a negative outcome on the univariate analysis were included in a stepwise logistic regression analysis, in order to identify independent predictors of patient outcome. Statistical analysis was performed using JASP (JASP Team) software.

 

 

Results

Study Population

Of the 88 patients included in the study, 70% were male; the median age was 66 years (IQR, 60-77). In most patients, the diagnosis of COVID-19 was derived from a positive SARS-CoV-2 nasopharyngeal swab. Six patients, however, maintained a negative swab at all determinations but had clinical and imaging features strongly suggesting COVID-19. No patients met the exclusion criteria. Most patients came from the ED (n = 58 [66%]) or general wards (n = 22 [25%]), while few were transferred from the ICU (n = 8 [9%]). The median length of stay in the SICU was 4 days (IQR, 2-7). An epidemiological link to affected persons or a known virus exposure was identifiable in 37 patients (42%).

Clinical, Laboratory, and Imaging Data

The clinical and anthropometric characteristics of patients are shown in Table 1. Hypertension and smoking habits were prevalent in our population, and the median Charlson Comorbidity Index was 3. Most patients experienced fever, dyspnea, and cough during the days before hospitalization.

Laboratory data showed a marked inflammatory milieu in all studied patients, both at baseline and after 24 and 72 hours. Lymphopenia was observed, along with a significant increase of lactate dehydrogenase (LDH), C-reactive protein (CPR), and D-dimer, and a mild increase of procalcitonin. N-terminal pro-brain natriuretic peptide (NT-proBNP) values were also increased, with normal troponin I values (Table 2).

Chest x-rays were obtained in almost all patients, while HRCT was performed in nearly half of patients. Complete bedside pulmonary ultrasonography data were available for 64 patients. Heterogeneous pulmonary alterations were found, regardless of the radiological technique, and multilobe infiltrates were the prevalent radiological pattern (73%) (Table 3). Seven patients (8%) were diagnosed with associated pulmonary embolism.

 

Medical Therapy

Most patients (89%) received hydroxychloroquine, whereas steroids were used in one-third of the population (36%). Immunomodulators (tocilizumab and ruxolitinib) were restricted to 12 patients (14%). Empirical antiviral therapy was introduced in the first 41 patients (47%). Enoxaparin was the default agent for thromboembolism prophylaxis, and 6 patients (7%) received 70% of the anticoagulating dose.

Oxygen and Ventilatory Therapy

Basal median P/F ratio was 253 (IQR, 218-291), and respiratory rate at triage was 20 breaths/min (IQR, 16-28), underlining a moderate-to-severe respiratory insufficiency at presentation. A total of 69 patients (78%) underwent CPAP, with a median positive end-expiratory pressure (PEEP) of 10.0 cm H₂O (IQR, 7.5-10.0) and fraction of inspired oxygen (Fio₂) of 0.40 (IQR, 0.40-0.50). In 37 patients (42%) who received ongoing CPAP, prone positioning was adopted. In this subgroup, respiratory rate was not significantly different from baseline to resupination (24 vs 25 breaths/min). The median P/F improved from 197 (IQR, 154-236) at baseline to 217 (IQR, 180-262) after pronation (the duration of the prone position was variable, depending on patients’ tolerance: 1 to 6 hours or every pronation cycle). The median delta P/F ratio was 39.4 (IQR, –17.0 to 78.0).

Outcomes

A total of 28 patients (32%) had a negative outcome in the SICU: 8 patients (9%) died, having no clinical indication for higher-intensity care; 6 patients (7%) were transferred to general wards for palliation; and 14 patients (16%) needed an upgrade of cure intensity and were transferred to the ICU. Of these 14 patients, 9 died in the ICU. The total in-hospital mortality of COVID-19 patients, including patients transferred from the SICU to general wards in fair condition, was 27% (n = 24). Clinical, laboratory, and therapeutic characteristics between the 2 groups are shown in Table 4.

Patients who had a negative outcome were significantly older and had more comorbidities, as suggested by a significantly higher prevalence of diabetes and higher Charlson Comorbidity scores (reflecting the mortality risk based on age and comorbidities). The median MuLBSTA score, which estimates the 90-day mortality risk from viral pneumonia, was also higher in patients who had a negative outcome (9.33%). Symptom occurrence was not different in patients with a negative outcome (apart from cough, which was less frequent), but these patients underwent hospitalization earlier—since the appearance of their first COVID-19 symptoms—compared to patients who had a positive outcome. No difference was found in antihypertensive therapy with angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers among outcome groups.

More pronounced laboratory abnormalities were found in patients who had a negative outcome, compared to patients who had a positive outcome: lower lymphocytes and higher C-reactive protein (CRP), procalcitonin, D-dimer, LDH, and NT-proBNP. We found no differences in the radiological distribution of pulmonary involvement in patients who had negative or positive outcomes, nor in the adopted medical treatment.

Data showed no difference in CPAP implementation in the 2 groups. However, prone positioning had been more frequently adopted in the group of patients who had a positive outcome, compared with patients who had a negative outcome. No differences of basal P/F were found in patients who had a negative or positive outcome, but the median P/F after 6 hours of prone position was significantly lower in patients who had a negative outcome. The delta P/F ratio did not differ in the 2 groups of patients.

Multivariate Analysis

A logistic regression model was created, including the variables significantly associated with outcomes in the univariate analysis (age, sex, history of diabetes, lymphocytes, CRP, procalcitonin, LDH, NT-proBNP, and D-dimer). In the multivariate analysis, independent predictors of a negative outcome were history of diabetes (odds ratio [OR], 8.22; 95% CI, 1.50-44.70; P =.015), high D-dimer values (OR, 1.28; CI, 1.04-1.57; P = .019), high LDH values (OR, 1.003; CI, 1.000-1.006; P = .039), and low lymphocytes count (OR, 0.996; CI, 0.993-0.999; P = .004).

 

Discussion

Role of Subintensive Units and Mortality

The novelty of our report is its attempt to investigate the specific group of COVID-19 patients admitted to a SICU. In Italy, SICUs receive acutely ill, spontaneously breathing patients who need (invasive) hemodynamic monitoring, vasoactive medication, renal replacement therapy, chest- tube placement, thrombolysis, and respiratory noninvasive support. The nurse-to-patient ratio is higher than for general wards (usually 1 nurse to every 4 or 5 patients), though lower than for ICUs. In northern Italy, a great number of COVID-19 patients have required this kind of high-intensity care during the pandemic: Noninvasive ventilation support had to be maintained for several days, pronation maneuvers required a high number of people 2 or 3 times a day, and strict monitoring had to be assured. The SICU setting allows patients to buy time as a bridge to progressive reduction of pulmonary involvement, sometimes preventing the need for intubation.

The high prevalence of negative outcomes in the SICU underlines the complexity of COVID-19 patients in this setting. In fact, published data about mortality for patients with severe COVID-19 pneumonia are similar to ours.^22,23

Clinical, Laboratory, and Imaging Data

Our analysis confirmed a high rate of comorbidities in COVID-19 patients²⁴ and their prognostic role with age.^25,26 A marked inflammatory milieu was a negative prognostic indicator, and associated concomitant bacterial superinfection could have led to a worse prognosis (procalcitonin was associated with negative outcomes).²⁷ The cardiovascular system was nevertheless stressed, as suggested by higher values of NT-proBNP in patients with negative outcomes, which could reflect sepsis-related systemic involvement.²⁸

It is known that the pulmonary damage caused by SARS-CoV-2 has a dynamic radiological and clinical course, with early areas of subsegmental consolidation, and bilateral ground-glass opacities predominating later in the course of the disease.²⁹ This could explain why in our population we found no specific radiological pattern leading to a worse outcome.

Medical Therapy

No specific pharmacological therapy was found to be associated with a positive outcome in our study, just like antiviral and immunomodulator therapies failed to demonstrate effectiveness in subsequent pandemic surges. The low statistical power of our study did not allow us to give insight into the effectiveness of steroids and heparin at any dosage.

PEEP Support and Prone Positioning

Continuous positive airway pressure was initiated in the majority of patients and maintained for several days. This was an absolute novelty, because we rarely had to keep patients in helmets for long. This was feasible thanks to the SICU’s high nurse-to-patient ratio and the possibility of providing monitored sedation. Patients who could no longer tolerate CPAP helmets or did not improve with CPAP support were evaluated with anesthetists for programming further management. No initial data on respiratory rate, level of hypoxemia, or oxygen support need (level of PEEP and Fio₂) could discriminate between outcomes.

Prone positioning during CPAP was implemented in 42% of our study population: P/F ratio amelioration after prone positioning was highly variable, ranging from very good P/F ratio improvements to few responses or no response. No significantly greater delta P/F ratio was seen after the first prone positioning cycle in patients who had a positive outcome, probably due to the small size of our population, but we observed a clear positive trend. Interestingly, patients showing a negative outcome had a lower percentage of long-term responses to prone positioning: 6 hours after resupination, they lost the benefit of prone positioning in terms of P/F ratio amelioration. Similarly, a greater number of patients tolerating prone positioning had a positive outcome. These data give insight on the possible benefits of prone positioning in a noninvasively supported cohort of patients, which has been mentioned in previous studies.^30,31

 

Outcomes and Variables Associated With Negative Outcomes

After correction for age and sex, we found in multiple regression analysis that higher D-dimer and LDH values, lymphopenia, and history of diabetes were independently associated with a worse outcome. Although our results had low statistical significance, we consider the trend of the obtained odds ratios important from a clinical point of view. These results could lead to greater attention being placed on COVID-19 patients who present with these characteristics upon their arrival to the ED because they have increased risk of death or intensive care need. Clinicians should consider SICU admission for these patients in order to guarantee closer monitoring and possibly more aggressive ventilatory treatments, earlier pronation, or earlier transfer to the ICU.

Limitations

The major limitation to our study is undoubtedly its statistical power, due to its relatively low patient population. Particularly, the small number of patients who underwent pronation did not allow speculation about the efficacy of this technique, although preliminary data seem promising. However, ours is among the first studies regarding patients with COVID-19 admitted to a SICU, and these preliminary data truthfully describe the Italian, and perhaps international, experience with the first surge of the pandemic.

Conclusions

Our data highlight the primary role of the SICU in COVID-19 in adequately treating critically ill patients who have high care needs different from intubation, and who require noninvasive ventilation for prolonged times as well as frequent pronation cycles. This setting of care may represent a valid, reliable, and effective option for critically ill respiratory patients. History of diabetes, lymphopenia, and high D-dimer and LDH values are independently associated with negative outcomes, and patients presenting with these characteristics should be strictly monitored.

Acknowledgments: The authors thank the Informatica System S.R.L., as well as Allessando Mendolia for the pro bono creation of the ISCovidCollect data collecting app.

Corresponding author: Sara Abram, MD, via Coppino, 12100 Cuneo, Italy; sara.abram84@gmail.com.

Disclosures: None.

From the Department of Emergency Medicine, Santa Croce e Carle Hospital, Cuneo, Italy (Drs. Abram, Tosello, Emanuele Bernardi, Allione, Cavalot, Dutto, Corsini, Martini, Sciolla, Sara Bernardi, and Lauria). From the School of Emergency Medicine, University of Turin, Turin, Italy (Drs. Paglietta and Giamello).

Objective: This retrospective and prospective cohort study was designed to describe the characteristics, treatments, and outcomes of patients with SARS-CoV-2 infection (COVID-19) admitted to subintensive care units (SICU) and to identify the variables associated with outcomes. SICUs have been extremely stressed during the pandemic, but most data regarding critically ill COVID-19 patients come from intensive care units (ICUs). Studies about COVID-19 patients in SICUs are lacking.

Setting and participants: The study included 88 COVID-19 patients admitted to our SICU in Cuneo, Italy, between March and May 2020.

Measurements: Clinical and ventilatory data were collected, and patients were divided by outcome. Multivariable logistic regression analysis examined the variables associated with negative outcomes (transfer to the ICU, palliation, or death in a SICU).

Results: A total of 60 patients (68%) had a positive outcome, and 28 patients (32%) had a negative outcome; 69 patients (78%) underwent continuous positive airway pressure (CPAP). Pronation (n = 37 [42%]) had been more frequently adopted in patients who had a positive outcome vs a negative outcome (n = 30 [50%] vs n = 7 [25%]; P = .048), and the median (interquartile range) Pao₂/Fio₂ ratio after 6 hours of prone positioning was lower in patients who had a negative outcome vs a positive outcome (144 [140-168] vs 249 [195-268], P = .006). Independent predictors of a negative outcome were diabetes (odds ratio [OR], 8.22; 95% CI, 1.50-44.70; P = .015), higher D-dimer (OR, 1.28; 95% CI, 1.04-1.57; P = .019), higher lactate dehydrogenase level (OR, 1.003; 95% CI, 1.000-1.006; P = .039), and lower lymphocytes count (OR, 0.996; 95% CI, 0.993-0.999; P = .004).

Conclusion: SICUs have a fundamental role in the treatment of critically ill patients with COVID-19, who require long-term CPAP and pronation cycles. Diabetes, lymphopenia, and high D-dimer and LDH levels are associated with negative outcomes.

Keywords: emergency medicine, noninvasive ventilation, prone position, continuous positive airway pressure.

The COVID-19 pandemic has led to large increases in hospital admissions. Subintensive care units (SICUs) are among the wards most under pressure worldwide,¹ dealing with the increased number of critically ill patients who need noninvasive ventilation, as well as serving as the best alternative to overfilled intensive care units (ICUs). In Italy, SICUs are playing a fundamental role in the management of COVID-19 patients, providing early treatment of respiratory failure by continuous noninvasive ventilation in order to reduce the need for intubation.^2-5 Nevertheless, the great majority of available data about critically ill COVID-19 patients comes from ICUs. Full studies about outcomes of patients in SICUs are lacking and need to be conducted.

We sought to evaluate the characteristics and outcomes of patients admitted to our SICU for COVID-19 to describe the treatments they needed and their impact on prognosis, and to identify the variables associated with patient outcomes.

Methods

Study Design

This cohort study used data from patients who were admitted in the very first weeks of the pandemic. Data were collected retrospectively as well as prospectively, since the ethical committee approved our project. The quality and quantity of data in the 2 groups were comparable.

Data were collected from electronic and written medical records gathered during the patient’s entire stay in our SICU. Data were entered in a database with limited and controlled access. This study complied with the Declaration of Helsinki and was approved by the local ethics committees (ID: MEDURG10).

Study Population

We studied 88 consecutive patients admitted to the SICU of the Santa Croce e Carle Teaching Hospital, Cuneo, Italy, for COVID-19, from March 8 to May 1, 2020. The diagnosis was based on acute respiratory failure associated with SARS-CoV-2 RNA detection on nasopharyngeal swab or tracheal aspirate and/or typical COVID-19 features on a pulmonary computed tomography (CT) scan.⁶ Exclusion criteria were age younger than 18 years and patient denial of permission to use their data for research purposes (the great majority of patients could actively give consent; for patients who were too sick to do so, family members were asked whether they were aware of any reason why the patient would deny consent).

 

 

Clinical Data

The past medical history and recent symptoms description were obtained by manually reviewing medical records. Epidemiological exposure was defined as contact with SARS-CoV-2–positive people or staying in an epidemic outbreak area. Initial vital parameters, venous blood tests, arterial blood gas analysis, chest x-ray, as well as the result of the nasopharyngeal swab were gathered from the emergency department (ED) examination. (Additional swabs could be requested when the first one was negative but clinical suspicion for COVID-19 was high.) Upon admission to the SICU, a standardized panel of blood tests was performed, which was repeated the next day and then every 48 hours. Arterial blood gas analysis was performed when clinically indicated, at least twice a day, or following a scheduled time in patients undergoing pronation. Charlson Comorbidity Index⁷ and MuLBSTA score⁸ were calculated based on the collected data.

Imaging

Chest ultrasonography was performed in the ED at the time of hospitalization and once a day in the SICU. Pulmonary high-resolution computed tomography (HRCT) was performed when clinically indicated or when the results of nasopharyngeal swabs and/or x-ray results were discordant with COVID-19 clinical suspicion. Contrast CT was performed when pulmonary embolism was suspected.

Medical Therapy

Hydroxychloroquine, antiviral agents, tocilizumab, and ruxolitinib were used in the early phase of the pandemic, then were dismissed after evidence of no efficacy.^9-11 Steroids and low-molecular-weight heparin were used afterward. Enoxaparin was used at the standard prophylactic dosage, and 70% of the anticoagulant dosage was also adopted in patients with moderate-to-severe COVID-19 and D-dimer values >3 times the normal value.^12-14 Antibiotics were given when a bacterial superinfection was suspected.

Oxygen and Ventilatory Therapy

Oxygen support or noninvasive ventilation were started based on patients’ respiratory efficacy, estimated by respiratory rate and the ratio of partial pressure of arterial oxygen and fraction of inspired oxygen (P/F ratio).^15,16 Oxygen support was delivered through nasal cannula, Venturi mask, or reservoir mask. Noninvasive ventilation was performed by continuous positive airway pressure (CPAP) when the P/F ratio was <250 or the respiratory rate was >25 breaths per minute, using the helmet interface.^5,17 Prone positioning during CPAP^18-20 was adopted in patients meeting the acute respiratory distress syndrome (ARDS) criteria²¹ and having persistence of respiratory distress and P/F <300 after a 1-hour trial of CPAP.

The prone position was maintained based on patient tolerance. P/F ratio was measured before pronation (T0), after 1 hour of prone position (T1), before resupination (T2), and 6 hours after resupination (T3). With the same timing, the patient was asked to rate their comfort in each position, from 0 (lack of comfort) to 10 (optimal comfort). Delta P/F was defined as the difference between P/F at T3 and basal P/F at T0.

Outcomes

Positive outcomes were defined as patient discharge from the SICU or transfer to a lower-intensity care ward for treatment continuation. Negative outcomes were defined as need for transfer to the ICU, transfer to another ward for palliation, or death in the SICU.

Statistical Analysis

Continuous data are reported as median and interquartile range (IQR); normal distribution of variables was tested using the Shapiro-Wilk test. Categorical variables were reported as absolute number and percentage. The Mann-Whitney test was used to compare continuous variables between groups, and chi-square test with continuity correction was used for categorical variables. The variables that were most significantly associated with a negative outcome on the univariate analysis were included in a stepwise logistic regression analysis, in order to identify independent predictors of patient outcome. Statistical analysis was performed using JASP (JASP Team) software.

 

 

Results

Study Population

Of the 88 patients included in the study, 70% were male; the median age was 66 years (IQR, 60-77). In most patients, the diagnosis of COVID-19 was derived from a positive SARS-CoV-2 nasopharyngeal swab. Six patients, however, maintained a negative swab at all determinations but had clinical and imaging features strongly suggesting COVID-19. No patients met the exclusion criteria. Most patients came from the ED (n = 58 [66%]) or general wards (n = 22 [25%]), while few were transferred from the ICU (n = 8 [9%]). The median length of stay in the SICU was 4 days (IQR, 2-7). An epidemiological link to affected persons or a known virus exposure was identifiable in 37 patients (42%).

Clinical, Laboratory, and Imaging Data

The clinical and anthropometric characteristics of patients are shown in Table 1. Hypertension and smoking habits were prevalent in our population, and the median Charlson Comorbidity Index was 3. Most patients experienced fever, dyspnea, and cough during the days before hospitalization.

Laboratory data showed a marked inflammatory milieu in all studied patients, both at baseline and after 24 and 72 hours. Lymphopenia was observed, along with a significant increase of lactate dehydrogenase (LDH), C-reactive protein (CPR), and D-dimer, and a mild increase of procalcitonin. N-terminal pro-brain natriuretic peptide (NT-proBNP) values were also increased, with normal troponin I values (Table 2).

Chest x-rays were obtained in almost all patients, while HRCT was performed in nearly half of patients. Complete bedside pulmonary ultrasonography data were available for 64 patients. Heterogeneous pulmonary alterations were found, regardless of the radiological technique, and multilobe infiltrates were the prevalent radiological pattern (73%) (Table 3). Seven patients (8%) were diagnosed with associated pulmonary embolism.

 

Medical Therapy

Most patients (89%) received hydroxychloroquine, whereas steroids were used in one-third of the population (36%). Immunomodulators (tocilizumab and ruxolitinib) were restricted to 12 patients (14%). Empirical antiviral therapy was introduced in the first 41 patients (47%). Enoxaparin was the default agent for thromboembolism prophylaxis, and 6 patients (7%) received 70% of the anticoagulating dose.

Oxygen and Ventilatory Therapy

Basal median P/F ratio was 253 (IQR, 218-291), and respiratory rate at triage was 20 breaths/min (IQR, 16-28), underlining a moderate-to-severe respiratory insufficiency at presentation. A total of 69 patients (78%) underwent CPAP, with a median positive end-expiratory pressure (PEEP) of 10.0 cm H₂O (IQR, 7.5-10.0) and fraction of inspired oxygen (Fio₂) of 0.40 (IQR, 0.40-0.50). In 37 patients (42%) who received ongoing CPAP, prone positioning was adopted. In this subgroup, respiratory rate was not significantly different from baseline to resupination (24 vs 25 breaths/min). The median P/F improved from 197 (IQR, 154-236) at baseline to 217 (IQR, 180-262) after pronation (the duration of the prone position was variable, depending on patients’ tolerance: 1 to 6 hours or every pronation cycle). The median delta P/F ratio was 39.4 (IQR, –17.0 to 78.0).

Outcomes

A total of 28 patients (32%) had a negative outcome in the SICU: 8 patients (9%) died, having no clinical indication for higher-intensity care; 6 patients (7%) were transferred to general wards for palliation; and 14 patients (16%) needed an upgrade of cure intensity and were transferred to the ICU. Of these 14 patients, 9 died in the ICU. The total in-hospital mortality of COVID-19 patients, including patients transferred from the SICU to general wards in fair condition, was 27% (n = 24). Clinical, laboratory, and therapeutic characteristics between the 2 groups are shown in Table 4.

Patients who had a negative outcome were significantly older and had more comorbidities, as suggested by a significantly higher prevalence of diabetes and higher Charlson Comorbidity scores (reflecting the mortality risk based on age and comorbidities). The median MuLBSTA score, which estimates the 90-day mortality risk from viral pneumonia, was also higher in patients who had a negative outcome (9.33%). Symptom occurrence was not different in patients with a negative outcome (apart from cough, which was less frequent), but these patients underwent hospitalization earlier—since the appearance of their first COVID-19 symptoms—compared to patients who had a positive outcome. No difference was found in antihypertensive therapy with angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers among outcome groups.

More pronounced laboratory abnormalities were found in patients who had a negative outcome, compared to patients who had a positive outcome: lower lymphocytes and higher C-reactive protein (CRP), procalcitonin, D-dimer, LDH, and NT-proBNP. We found no differences in the radiological distribution of pulmonary involvement in patients who had negative or positive outcomes, nor in the adopted medical treatment.

Data showed no difference in CPAP implementation in the 2 groups. However, prone positioning had been more frequently adopted in the group of patients who had a positive outcome, compared with patients who had a negative outcome. No differences of basal P/F were found in patients who had a negative or positive outcome, but the median P/F after 6 hours of prone position was significantly lower in patients who had a negative outcome. The delta P/F ratio did not differ in the 2 groups of patients.

Multivariate Analysis

A logistic regression model was created, including the variables significantly associated with outcomes in the univariate analysis (age, sex, history of diabetes, lymphocytes, CRP, procalcitonin, LDH, NT-proBNP, and D-dimer). In the multivariate analysis, independent predictors of a negative outcome were history of diabetes (odds ratio [OR], 8.22; 95% CI, 1.50-44.70; P =.015), high D-dimer values (OR, 1.28; CI, 1.04-1.57; P = .019), high LDH values (OR, 1.003; CI, 1.000-1.006; P = .039), and low lymphocytes count (OR, 0.996; CI, 0.993-0.999; P = .004).

 

Discussion

Role of Subintensive Units and Mortality

The novelty of our report is its attempt to investigate the specific group of COVID-19 patients admitted to a SICU. In Italy, SICUs receive acutely ill, spontaneously breathing patients who need (invasive) hemodynamic monitoring, vasoactive medication, renal replacement therapy, chest- tube placement, thrombolysis, and respiratory noninvasive support. The nurse-to-patient ratio is higher than for general wards (usually 1 nurse to every 4 or 5 patients), though lower than for ICUs. In northern Italy, a great number of COVID-19 patients have required this kind of high-intensity care during the pandemic: Noninvasive ventilation support had to be maintained for several days, pronation maneuvers required a high number of people 2 or 3 times a day, and strict monitoring had to be assured. The SICU setting allows patients to buy time as a bridge to progressive reduction of pulmonary involvement, sometimes preventing the need for intubation.

The high prevalence of negative outcomes in the SICU underlines the complexity of COVID-19 patients in this setting. In fact, published data about mortality for patients with severe COVID-19 pneumonia are similar to ours.^22,23

Clinical, Laboratory, and Imaging Data

Our analysis confirmed a high rate of comorbidities in COVID-19 patients²⁴ and their prognostic role with age.^25,26 A marked inflammatory milieu was a negative prognostic indicator, and associated concomitant bacterial superinfection could have led to a worse prognosis (procalcitonin was associated with negative outcomes).²⁷ The cardiovascular system was nevertheless stressed, as suggested by higher values of NT-proBNP in patients with negative outcomes, which could reflect sepsis-related systemic involvement.²⁸

It is known that the pulmonary damage caused by SARS-CoV-2 has a dynamic radiological and clinical course, with early areas of subsegmental consolidation, and bilateral ground-glass opacities predominating later in the course of the disease.²⁹ This could explain why in our population we found no specific radiological pattern leading to a worse outcome.

Medical Therapy

No specific pharmacological therapy was found to be associated with a positive outcome in our study, just like antiviral and immunomodulator therapies failed to demonstrate effectiveness in subsequent pandemic surges. The low statistical power of our study did not allow us to give insight into the effectiveness of steroids and heparin at any dosage.

PEEP Support and Prone Positioning

Continuous positive airway pressure was initiated in the majority of patients and maintained for several days. This was an absolute novelty, because we rarely had to keep patients in helmets for long. This was feasible thanks to the SICU’s high nurse-to-patient ratio and the possibility of providing monitored sedation. Patients who could no longer tolerate CPAP helmets or did not improve with CPAP support were evaluated with anesthetists for programming further management. No initial data on respiratory rate, level of hypoxemia, or oxygen support need (level of PEEP and Fio₂) could discriminate between outcomes.

Prone positioning during CPAP was implemented in 42% of our study population: P/F ratio amelioration after prone positioning was highly variable, ranging from very good P/F ratio improvements to few responses or no response. No significantly greater delta P/F ratio was seen after the first prone positioning cycle in patients who had a positive outcome, probably due to the small size of our population, but we observed a clear positive trend. Interestingly, patients showing a negative outcome had a lower percentage of long-term responses to prone positioning: 6 hours after resupination, they lost the benefit of prone positioning in terms of P/F ratio amelioration. Similarly, a greater number of patients tolerating prone positioning had a positive outcome. These data give insight on the possible benefits of prone positioning in a noninvasively supported cohort of patients, which has been mentioned in previous studies.^30,31

 

Outcomes and Variables Associated With Negative Outcomes

After correction for age and sex, we found in multiple regression analysis that higher D-dimer and LDH values, lymphopenia, and history of diabetes were independently associated with a worse outcome. Although our results had low statistical significance, we consider the trend of the obtained odds ratios important from a clinical point of view. These results could lead to greater attention being placed on COVID-19 patients who present with these characteristics upon their arrival to the ED because they have increased risk of death or intensive care need. Clinicians should consider SICU admission for these patients in order to guarantee closer monitoring and possibly more aggressive ventilatory treatments, earlier pronation, or earlier transfer to the ICU.

Limitations

The major limitation to our study is undoubtedly its statistical power, due to its relatively low patient population. Particularly, the small number of patients who underwent pronation did not allow speculation about the efficacy of this technique, although preliminary data seem promising. However, ours is among the first studies regarding patients with COVID-19 admitted to a SICU, and these preliminary data truthfully describe the Italian, and perhaps international, experience with the first surge of the pandemic.

Conclusions

Our data highlight the primary role of the SICU in COVID-19 in adequately treating critically ill patients who have high care needs different from intubation, and who require noninvasive ventilation for prolonged times as well as frequent pronation cycles. This setting of care may represent a valid, reliable, and effective option for critically ill respiratory patients. History of diabetes, lymphopenia, and high D-dimer and LDH values are independently associated with negative outcomes, and patients presenting with these characteristics should be strictly monitored.

Acknowledgments: The authors thank the Informatica System S.R.L., as well as Allessando Mendolia for the pro bono creation of the ISCovidCollect data collecting app.

Corresponding author: Sara Abram, MD, via Coppino, 12100 Cuneo, Italy; sara.abram84@gmail.com.

Disclosures: None.

Lower back pain (LBP) affects an estimated 9.4% of the global population and has resulted in more years lived with disability than any other health condition.¹ LBP affects a wide range of populations, but US veterans have been shown to have significantly higher rates of back pain than nonveterans. The National Institutes of Health reports that 65.6% of veterans experience chronic pain; 9.1% of veterans experience severe, chronic pain.² Chronic back pain is treated by a range of methods, including medications, surgery, physical therapy (PT), patient education, and behavioral therapy.³ However, chronic neuropathic back pain has been shown to have limited responsiveness to medication.⁴

Neuropathic pain is caused by lesions in the somatosensory nervous system, resulting in spontaneous pain and amplified pain responses to both painful and nonpainful stimuli.⁵ The most common location for neuropathic pain is the back and legs. Between 10% and 40% of people who undergo lumbosacral spine surgery to treat neuropathic radicular pain will experience further neuropathic pain.⁶ This condition is referred to as failed back surgery syndrome or postlaminectomy syndrome (PLS). While neuropathic back pain has had limited responsiveness to medication and repeated lumbosacral spine surgery, spinal cord stimulation (SCS) has shown promise as an effective form of pain treatment for those experiencing PLS and other spine disorders.^7-10 In addition, SCS therapy has had a very low incidence of complications, which may be on the decline with recent technological advancements.¹¹ Patients with a diagnosis of PLS, LBP, or complex regional pain syndrome (CRPS) who have not responded to medications, therapy, and/or injections for ≥ 6 months were eligible for a trial of SCS therapy. Trial leads were placed via the percutaneous route with the battery strapped to the waistline for 3 to 5 days and were removed in clinic. Patients who experienced > 60% pain relief and functional improvement received a SCS implant.

The effectiveness of SCS has been demonstrated in a nonveteran population, but it has not been studied in a veteran population.¹² US Department of Veterans Affairs (VA) health care coverage is different from Medicare and private insurance in that it is classified as a benefit and not insurance. The goals of treatment at the VA may include considerations in addition to feeling better, and patient presentations may not align with those in the private sector.

We hypothesize that SCS is both a safe and beneficial treatment option for veterans with chronic intractable spine and/or extremity pain. The purpose of this study was to determine the efficacy and safety of SCS in a veteran population.

Methods

The efficacy and safety of SCS was determined via a retrospective study. Inclusion criteria for the study consisted of any Southeastern Louisiana Veterans Health Care System (SLVHCS) patient who had an SCS trial and/or implant from 2008 to 2020. Eligible veterans must have had chronic pain for at least 6 months and had previously tried multiple medications, PT, transcutaneous nerve stimulation, facet injections, epidural steroid injections, or surgery without success. For medication therapy to be considered unsuccessful, it must have included acetaminophen, nonsteroidal anti-inflammatory drugs, and ≥ 1 adjuvant medication (gabapentin, duloxetine, amitriptyline, lidocaine, and menthol). A diagnosis of chronic LBP, PLS, cervical or lumbar spondylosis with radiculopathy, complex regional pain syndrome, or chronic pain syndrome was required for eligibility. Patients whose pain decreased by > 60% and had functional improvement in a 3- to 5-day trial received SCS implantation with percutaneous leads by a pain physician or paddle lead by a neurosurgeon.

The SLVHCS Institutional Review Board approved this study. Electronic health records were reviewed to determine patient age, anthropometric data, and date of SCS implantation. Patients were then called and interviewed to complete a survey. After obtaining verbal consent to the study, subjects were surveyed regarding whether the patient would recommend the procedure to peers, adverse effects (AEs) or complications, and the ability to decrease opiates if applicable. A verbal Pain Outcome Questionnaire (POQ) assessment of activities of daily living also was given during the phone interview regarding pain levels before SCS and at the time of the phone interview.¹³ (eAppendix available at doi:10.12788/fp.0204) Following the survey, a chart review was performed to corroborate the given AEs or complications and opiate use information. Before and after results of the POQ were compared via a paired sample t test, and P values < .05 were considered significant. Analyses were performed by IBM SPSS, version 26.

The primary outcome measure for this study was whether veterans would recommend SCS to their peers; in our view, this categorical outcome measure seemed to be more valuable to share with future patients who might be candidates for SCS. Since VA health care coverage and goals of treatment may be different from a nonveteran population, we opted to use this primary measure to decrease the possibility of confounding variables.

Secondary outcome measures included changes in POC scores, improvements in activities of daily living, and decreases in use of opioid pain medications.

POQ responses were recorded during the telephone interviews (0 to 10 scale). A paired sample t test was conducted to compare pain levels before and after SCS implant. Pain levels were gathered in the single phone call. Patient opioid usage, if applicable, was assessed by converting medications to morphine milligram equivalent dosing (MMED). Since patients who were on chronic opioids took multiple formulations, we changed the total daily dose to all morphine; for this study, morphine was considered equivalent to hydrocodone, and oxycodone was 1.5x morphine.

 

Results

Of the 90 SLVHCS patients who received an SCS implant between 2008 and 2020, 76 were reached by telephone and 65 had their responses recorded in the study. Of the 11 patients who were not included, 5 had the SCS removed; it is unclear whether these veterans would have recommended the treatment. Four were unable to quantify pain and/or SCS effects, and 2 were excluded due to a dementia diagnosis years after the implant. The mean (SD) age of participants was 63.9 (10.3) years. Forty percent of patients had a diabetes mellitus diagnosis and 1 had prediabetes. Patients’ most common qualifying diagnosis for SCS was PLS (47.7%) followed by chronic LBP (26.2%). A percutaneous 2-lead technique was the most common type of SCS type used (60.0%) followed by 1-lead (21.5%). The most common SCS manufacturer was Boston Scientific (87.7%)(Table 1). Most veterans (76.9%) recommended SCS to their peers; 13.8% did not recommend SCS; 9.2% were undecided and stated that they were unable to recommend because they did not want to persuade a peer to get SCS (Figure).

There was a statistically significant decrease in opioid use for the 40 veterans for whom pain medication was converted (P < .001)(Table 2). Six patients reported using opioids at some point but could not remember their dose, and no records were found in their chart review, so they were not included in the MMED analysis. In that group, 4 patients reported using opioids before SCS but discontinued the opioid use after SCS implantation, and 2 patients noted using opioids before SCS and concomitantly. Eighteen subjects reported no opioid use at any point before or after SCS (Table 3).

There were few life-threatening complications of SCS. Three veterans developed skin dehiscence; 2 had dehiscence at the battery/generator site, and 1 had dehiscence at the lead anchor site. Two patients with dehiscence also had morbid obesity, and the third had postoperative malnourishment. The dehiscence occurred 3 and 8 months postoperation. All 3 patients with dehiscence had the SCS explanted, though they were eager to get a new SCS implanted as soon as possible because SCS was their most successful treatment to date.

Twenty of the 64 veterans surveyed reported other complications of SCS, including lead migration, lack of pain coverage, paresthesia and numbness, soreness around generator site, SCS shocking patient when performing full thoracic spine flexion, and shingles at the battery site (Table 4). There were 11 explants among the 76 veterans contacted. The primary reason for explant was lack of pain coverage.

Patient concerns included pain with sitting in chairs due to tenderness around the implant, SCS helping with physical pain but not mental pain, SCS only working during the day and not helping with sleep, and patients lacking education regarding possible complications of SCS.

Discussion

In this nonrandomized retrospective review, SCS was shown to be an effective treatment for intractable spine and/or extremity pain. Veterans’ pain levels were significantly reduced following SCS implantation, and more than three-fourths of veterans recommended SCS to their peers. We used the recommendation of SCS to peers as the most important metric regarding the effectiveness of SCS, as this measure was felt to be more valuable to share with future patients; furthermore, categorical analysis has been shown to be more valuable than ordinal pain scales to measure pain.¹⁴ In addition to wanting to expand the available research to the general public, we wanted a measure that we could easily relay to our patient population regarding SCS.

The explant rate of 14.5% among surveyed veterans falls at the higher end of the normal ranges found in previous studies of long-term SCS outcomes.^15-17 One possible reason for the higher rate is that we did not differentiate based on the reason for the explant (ie, no benefit, further surgery needed for underlying medical condition, or SCS-specific complications). Another possible contributing factor to the higher than expected explant rate is the geographic location in the New Orleans metro area; New Orleans is considered to have one of the highest rates of obesity in the United States and obesity typically has other diseases associated with it such as hypertension and diabetes mellitus.

 

Limitations

Limitations of the study include the relatively low number of subjects, subjective nature of the interview questions, and the patients’ answers. Typically the POQ has been used as a prospective assessment of pain; whether it is valid in a retrospective analysis is not clear. While there was a statistically significant decrease of opioid use after getting SCS, this study can only show correlation, not causation. During the study period, there has been a drastic change in opioid prescribing patterns and efforts to decrease the amount of opioids prescribed.

Subjects also were asked to rate their pain and quality of life before SCS. Some subjects had SCS implantation up to 10 years prior to the phone interview. The variable amount of time between SCS implantation and interview likely affected subjects’ responses. Chronic pain is a moving target. Patients have good days and bad days that would likely change opinions on SCS benefits on a single phone interview. Some patients needed battery replacements at the time of the interview (battery life averaged about 3 to 5 years in our study population) and were asked to report current levels of pain from the perspective of when their batteries were still functional, further affecting results.

Conclusions

SCS was shown to improve the quality of life of US veterans at SLVHCS across a wide variety of metrics, including activities of daily living, as well as mental and physical health. For veterans with chronic intractable pain who have tried and failed more conservative treatments, SCS is a great treatment.

Lower back pain (LBP) affects an estimated 9.4% of the global population and has resulted in more years lived with disability than any other health condition.¹ LBP affects a wide range of populations, but US veterans have been shown to have significantly higher rates of back pain than nonveterans. The National Institutes of Health reports that 65.6% of veterans experience chronic pain; 9.1% of veterans experience severe, chronic pain.² Chronic back pain is treated by a range of methods, including medications, surgery, physical therapy (PT), patient education, and behavioral therapy.³ However, chronic neuropathic back pain has been shown to have limited responsiveness to medication.⁴

Neuropathic pain is caused by lesions in the somatosensory nervous system, resulting in spontaneous pain and amplified pain responses to both painful and nonpainful stimuli.⁵ The most common location for neuropathic pain is the back and legs. Between 10% and 40% of people who undergo lumbosacral spine surgery to treat neuropathic radicular pain will experience further neuropathic pain.⁶ This condition is referred to as failed back surgery syndrome or postlaminectomy syndrome (PLS). While neuropathic back pain has had limited responsiveness to medication and repeated lumbosacral spine surgery, spinal cord stimulation (SCS) has shown promise as an effective form of pain treatment for those experiencing PLS and other spine disorders.^7-10 In addition, SCS therapy has had a very low incidence of complications, which may be on the decline with recent technological advancements.¹¹ Patients with a diagnosis of PLS, LBP, or complex regional pain syndrome (CRPS) who have not responded to medications, therapy, and/or injections for ≥ 6 months were eligible for a trial of SCS therapy. Trial leads were placed via the percutaneous route with the battery strapped to the waistline for 3 to 5 days and were removed in clinic. Patients who experienced > 60% pain relief and functional improvement received a SCS implant.

The effectiveness of SCS has been demonstrated in a nonveteran population, but it has not been studied in a veteran population.¹² US Department of Veterans Affairs (VA) health care coverage is different from Medicare and private insurance in that it is classified as a benefit and not insurance. The goals of treatment at the VA may include considerations in addition to feeling better, and patient presentations may not align with those in the private sector.

We hypothesize that SCS is both a safe and beneficial treatment option for veterans with chronic intractable spine and/or extremity pain. The purpose of this study was to determine the efficacy and safety of SCS in a veteran population.

Methods

The efficacy and safety of SCS was determined via a retrospective study. Inclusion criteria for the study consisted of any Southeastern Louisiana Veterans Health Care System (SLVHCS) patient who had an SCS trial and/or implant from 2008 to 2020. Eligible veterans must have had chronic pain for at least 6 months and had previously tried multiple medications, PT, transcutaneous nerve stimulation, facet injections, epidural steroid injections, or surgery without success. For medication therapy to be considered unsuccessful, it must have included acetaminophen, nonsteroidal anti-inflammatory drugs, and ≥ 1 adjuvant medication (gabapentin, duloxetine, amitriptyline, lidocaine, and menthol). A diagnosis of chronic LBP, PLS, cervical or lumbar spondylosis with radiculopathy, complex regional pain syndrome, or chronic pain syndrome was required for eligibility. Patients whose pain decreased by > 60% and had functional improvement in a 3- to 5-day trial received SCS implantation with percutaneous leads by a pain physician or paddle lead by a neurosurgeon.

The SLVHCS Institutional Review Board approved this study. Electronic health records were reviewed to determine patient age, anthropometric data, and date of SCS implantation. Patients were then called and interviewed to complete a survey. After obtaining verbal consent to the study, subjects were surveyed regarding whether the patient would recommend the procedure to peers, adverse effects (AEs) or complications, and the ability to decrease opiates if applicable. A verbal Pain Outcome Questionnaire (POQ) assessment of activities of daily living also was given during the phone interview regarding pain levels before SCS and at the time of the phone interview.¹³ (eAppendix available at doi:10.12788/fp.0204) Following the survey, a chart review was performed to corroborate the given AEs or complications and opiate use information. Before and after results of the POQ were compared via a paired sample t test, and P values < .05 were considered significant. Analyses were performed by IBM SPSS, version 26.

The primary outcome measure for this study was whether veterans would recommend SCS to their peers; in our view, this categorical outcome measure seemed to be more valuable to share with future patients who might be candidates for SCS. Since VA health care coverage and goals of treatment may be different from a nonveteran population, we opted to use this primary measure to decrease the possibility of confounding variables.

Secondary outcome measures included changes in POC scores, improvements in activities of daily living, and decreases in use of opioid pain medications.

POQ responses were recorded during the telephone interviews (0 to 10 scale). A paired sample t test was conducted to compare pain levels before and after SCS implant. Pain levels were gathered in the single phone call. Patient opioid usage, if applicable, was assessed by converting medications to morphine milligram equivalent dosing (MMED). Since patients who were on chronic opioids took multiple formulations, we changed the total daily dose to all morphine; for this study, morphine was considered equivalent to hydrocodone, and oxycodone was 1.5x morphine.

 

Results

Of the 90 SLVHCS patients who received an SCS implant between 2008 and 2020, 76 were reached by telephone and 65 had their responses recorded in the study. Of the 11 patients who were not included, 5 had the SCS removed; it is unclear whether these veterans would have recommended the treatment. Four were unable to quantify pain and/or SCS effects, and 2 were excluded due to a dementia diagnosis years after the implant. The mean (SD) age of participants was 63.9 (10.3) years. Forty percent of patients had a diabetes mellitus diagnosis and 1 had prediabetes. Patients’ most common qualifying diagnosis for SCS was PLS (47.7%) followed by chronic LBP (26.2%). A percutaneous 2-lead technique was the most common type of SCS type used (60.0%) followed by 1-lead (21.5%). The most common SCS manufacturer was Boston Scientific (87.7%)(Table 1). Most veterans (76.9%) recommended SCS to their peers; 13.8% did not recommend SCS; 9.2% were undecided and stated that they were unable to recommend because they did not want to persuade a peer to get SCS (Figure).

There was a statistically significant decrease in opioid use for the 40 veterans for whom pain medication was converted (P < .001)(Table 2). Six patients reported using opioids at some point but could not remember their dose, and no records were found in their chart review, so they were not included in the MMED analysis. In that group, 4 patients reported using opioids before SCS but discontinued the opioid use after SCS implantation, and 2 patients noted using opioids before SCS and concomitantly. Eighteen subjects reported no opioid use at any point before or after SCS (Table 3).

There were few life-threatening complications of SCS. Three veterans developed skin dehiscence; 2 had dehiscence at the battery/generator site, and 1 had dehiscence at the lead anchor site. Two patients with dehiscence also had morbid obesity, and the third had postoperative malnourishment. The dehiscence occurred 3 and 8 months postoperation. All 3 patients with dehiscence had the SCS explanted, though they were eager to get a new SCS implanted as soon as possible because SCS was their most successful treatment to date.

Twenty of the 64 veterans surveyed reported other complications of SCS, including lead migration, lack of pain coverage, paresthesia and numbness, soreness around generator site, SCS shocking patient when performing full thoracic spine flexion, and shingles at the battery site (Table 4). There were 11 explants among the 76 veterans contacted. The primary reason for explant was lack of pain coverage.

Patient concerns included pain with sitting in chairs due to tenderness around the implant, SCS helping with physical pain but not mental pain, SCS only working during the day and not helping with sleep, and patients lacking education regarding possible complications of SCS.

Discussion

In this nonrandomized retrospective review, SCS was shown to be an effective treatment for intractable spine and/or extremity pain. Veterans’ pain levels were significantly reduced following SCS implantation, and more than three-fourths of veterans recommended SCS to their peers. We used the recommendation of SCS to peers as the most important metric regarding the effectiveness of SCS, as this measure was felt to be more valuable to share with future patients; furthermore, categorical analysis has been shown to be more valuable than ordinal pain scales to measure pain.¹⁴ In addition to wanting to expand the available research to the general public, we wanted a measure that we could easily relay to our patient population regarding SCS.

The explant rate of 14.5% among surveyed veterans falls at the higher end of the normal ranges found in previous studies of long-term SCS outcomes.^15-17 One possible reason for the higher rate is that we did not differentiate based on the reason for the explant (ie, no benefit, further surgery needed for underlying medical condition, or SCS-specific complications). Another possible contributing factor to the higher than expected explant rate is the geographic location in the New Orleans metro area; New Orleans is considered to have one of the highest rates of obesity in the United States and obesity typically has other diseases associated with it such as hypertension and diabetes mellitus.

 

Limitations

Limitations of the study include the relatively low number of subjects, subjective nature of the interview questions, and the patients’ answers. Typically the POQ has been used as a prospective assessment of pain; whether it is valid in a retrospective analysis is not clear. While there was a statistically significant decrease of opioid use after getting SCS, this study can only show correlation, not causation. During the study period, there has been a drastic change in opioid prescribing patterns and efforts to decrease the amount of opioids prescribed.

Subjects also were asked to rate their pain and quality of life before SCS. Some subjects had SCS implantation up to 10 years prior to the phone interview. The variable amount of time between SCS implantation and interview likely affected subjects’ responses. Chronic pain is a moving target. Patients have good days and bad days that would likely change opinions on SCS benefits on a single phone interview. Some patients needed battery replacements at the time of the interview (battery life averaged about 3 to 5 years in our study population) and were asked to report current levels of pain from the perspective of when their batteries were still functional, further affecting results.

Conclusions

SCS was shown to improve the quality of life of US veterans at SLVHCS across a wide variety of metrics, including activities of daily living, as well as mental and physical health. For veterans with chronic intractable pain who have tried and failed more conservative treatments, SCS is a great treatment.

Lower back pain (LBP) affects an estimated 9.4% of the global population and has resulted in more years lived with disability than any other health condition.¹ LBP affects a wide range of populations, but US veterans have been shown to have significantly higher rates of back pain than nonveterans. The National Institutes of Health reports that 65.6% of veterans experience chronic pain; 9.1% of veterans experience severe, chronic pain.² Chronic back pain is treated by a range of methods, including medications, surgery, physical therapy (PT), patient education, and behavioral therapy.³ However, chronic neuropathic back pain has been shown to have limited responsiveness to medication.⁴

Neuropathic pain is caused by lesions in the somatosensory nervous system, resulting in spontaneous pain and amplified pain responses to both painful and nonpainful stimuli.⁵ The most common location for neuropathic pain is the back and legs. Between 10% and 40% of people who undergo lumbosacral spine surgery to treat neuropathic radicular pain will experience further neuropathic pain.⁶ This condition is referred to as failed back surgery syndrome or postlaminectomy syndrome (PLS). While neuropathic back pain has had limited responsiveness to medication and repeated lumbosacral spine surgery, spinal cord stimulation (SCS) has shown promise as an effective form of pain treatment for those experiencing PLS and other spine disorders.^7-10 In addition, SCS therapy has had a very low incidence of complications, which may be on the decline with recent technological advancements.¹¹ Patients with a diagnosis of PLS, LBP, or complex regional pain syndrome (CRPS) who have not responded to medications, therapy, and/or injections for ≥ 6 months were eligible for a trial of SCS therapy. Trial leads were placed via the percutaneous route with the battery strapped to the waistline for 3 to 5 days and were removed in clinic. Patients who experienced > 60% pain relief and functional improvement received a SCS implant.

The effectiveness of SCS has been demonstrated in a nonveteran population, but it has not been studied in a veteran population.¹² US Department of Veterans Affairs (VA) health care coverage is different from Medicare and private insurance in that it is classified as a benefit and not insurance. The goals of treatment at the VA may include considerations in addition to feeling better, and patient presentations may not align with those in the private sector.

We hypothesize that SCS is both a safe and beneficial treatment option for veterans with chronic intractable spine and/or extremity pain. The purpose of this study was to determine the efficacy and safety of SCS in a veteran population.

Methods

The efficacy and safety of SCS was determined via a retrospective study. Inclusion criteria for the study consisted of any Southeastern Louisiana Veterans Health Care System (SLVHCS) patient who had an SCS trial and/or implant from 2008 to 2020. Eligible veterans must have had chronic pain for at least 6 months and had previously tried multiple medications, PT, transcutaneous nerve stimulation, facet injections, epidural steroid injections, or surgery without success. For medication therapy to be considered unsuccessful, it must have included acetaminophen, nonsteroidal anti-inflammatory drugs, and ≥ 1 adjuvant medication (gabapentin, duloxetine, amitriptyline, lidocaine, and menthol). A diagnosis of chronic LBP, PLS, cervical or lumbar spondylosis with radiculopathy, complex regional pain syndrome, or chronic pain syndrome was required for eligibility. Patients whose pain decreased by > 60% and had functional improvement in a 3- to 5-day trial received SCS implantation with percutaneous leads by a pain physician or paddle lead by a neurosurgeon.

The SLVHCS Institutional Review Board approved this study. Electronic health records were reviewed to determine patient age, anthropometric data, and date of SCS implantation. Patients were then called and interviewed to complete a survey. After obtaining verbal consent to the study, subjects were surveyed regarding whether the patient would recommend the procedure to peers, adverse effects (AEs) or complications, and the ability to decrease opiates if applicable. A verbal Pain Outcome Questionnaire (POQ) assessment of activities of daily living also was given during the phone interview regarding pain levels before SCS and at the time of the phone interview.¹³ (eAppendix available at doi:10.12788/fp.0204) Following the survey, a chart review was performed to corroborate the given AEs or complications and opiate use information. Before and after results of the POQ were compared via a paired sample t test, and P values < .05 were considered significant. Analyses were performed by IBM SPSS, version 26.

The primary outcome measure for this study was whether veterans would recommend SCS to their peers; in our view, this categorical outcome measure seemed to be more valuable to share with future patients who might be candidates for SCS. Since VA health care coverage and goals of treatment may be different from a nonveteran population, we opted to use this primary measure to decrease the possibility of confounding variables.

Secondary outcome measures included changes in POC scores, improvements in activities of daily living, and decreases in use of opioid pain medications.

POQ responses were recorded during the telephone interviews (0 to 10 scale). A paired sample t test was conducted to compare pain levels before and after SCS implant. Pain levels were gathered in the single phone call. Patient opioid usage, if applicable, was assessed by converting medications to morphine milligram equivalent dosing (MMED). Since patients who were on chronic opioids took multiple formulations, we changed the total daily dose to all morphine; for this study, morphine was considered equivalent to hydrocodone, and oxycodone was 1.5x morphine.

 

Results

Of the 90 SLVHCS patients who received an SCS implant between 2008 and 2020, 76 were reached by telephone and 65 had their responses recorded in the study. Of the 11 patients who were not included, 5 had the SCS removed; it is unclear whether these veterans would have recommended the treatment. Four were unable to quantify pain and/or SCS effects, and 2 were excluded due to a dementia diagnosis years after the implant. The mean (SD) age of participants was 63.9 (10.3) years. Forty percent of patients had a diabetes mellitus diagnosis and 1 had prediabetes. Patients’ most common qualifying diagnosis for SCS was PLS (47.7%) followed by chronic LBP (26.2%). A percutaneous 2-lead technique was the most common type of SCS type used (60.0%) followed by 1-lead (21.5%). The most common SCS manufacturer was Boston Scientific (87.7%)(Table 1). Most veterans (76.9%) recommended SCS to their peers; 13.8% did not recommend SCS; 9.2% were undecided and stated that they were unable to recommend because they did not want to persuade a peer to get SCS (Figure).

There was a statistically significant decrease in opioid use for the 40 veterans for whom pain medication was converted (P < .001)(Table 2). Six patients reported using opioids at some point but could not remember their dose, and no records were found in their chart review, so they were not included in the MMED analysis. In that group, 4 patients reported using opioids before SCS but discontinued the opioid use after SCS implantation, and 2 patients noted using opioids before SCS and concomitantly. Eighteen subjects reported no opioid use at any point before or after SCS (Table 3).

There were few life-threatening complications of SCS. Three veterans developed skin dehiscence; 2 had dehiscence at the battery/generator site, and 1 had dehiscence at the lead anchor site. Two patients with dehiscence also had morbid obesity, and the third had postoperative malnourishment. The dehiscence occurred 3 and 8 months postoperation. All 3 patients with dehiscence had the SCS explanted, though they were eager to get a new SCS implanted as soon as possible because SCS was their most successful treatment to date.

Twenty of the 64 veterans surveyed reported other complications of SCS, including lead migration, lack of pain coverage, paresthesia and numbness, soreness around generator site, SCS shocking patient when performing full thoracic spine flexion, and shingles at the battery site (Table 4). There were 11 explants among the 76 veterans contacted. The primary reason for explant was lack of pain coverage.

Patient concerns included pain with sitting in chairs due to tenderness around the implant, SCS helping with physical pain but not mental pain, SCS only working during the day and not helping with sleep, and patients lacking education regarding possible complications of SCS.

Discussion

In this nonrandomized retrospective review, SCS was shown to be an effective treatment for intractable spine and/or extremity pain. Veterans’ pain levels were significantly reduced following SCS implantation, and more than three-fourths of veterans recommended SCS to their peers. We used the recommendation of SCS to peers as the most important metric regarding the effectiveness of SCS, as this measure was felt to be more valuable to share with future patients; furthermore, categorical analysis has been shown to be more valuable than ordinal pain scales to measure pain.¹⁴ In addition to wanting to expand the available research to the general public, we wanted a measure that we could easily relay to our patient population regarding SCS.

The explant rate of 14.5% among surveyed veterans falls at the higher end of the normal ranges found in previous studies of long-term SCS outcomes.^15-17 One possible reason for the higher rate is that we did not differentiate based on the reason for the explant (ie, no benefit, further surgery needed for underlying medical condition, or SCS-specific complications). Another possible contributing factor to the higher than expected explant rate is the geographic location in the New Orleans metro area; New Orleans is considered to have one of the highest rates of obesity in the United States and obesity typically has other diseases associated with it such as hypertension and diabetes mellitus.

 

Limitations

Limitations of the study include the relatively low number of subjects, subjective nature of the interview questions, and the patients’ answers. Typically the POQ has been used as a prospective assessment of pain; whether it is valid in a retrospective analysis is not clear. While there was a statistically significant decrease of opioid use after getting SCS, this study can only show correlation, not causation. During the study period, there has been a drastic change in opioid prescribing patterns and efforts to decrease the amount of opioids prescribed.

Subjects also were asked to rate their pain and quality of life before SCS. Some subjects had SCS implantation up to 10 years prior to the phone interview. The variable amount of time between SCS implantation and interview likely affected subjects’ responses. Chronic pain is a moving target. Patients have good days and bad days that would likely change opinions on SCS benefits on a single phone interview. Some patients needed battery replacements at the time of the interview (battery life averaged about 3 to 5 years in our study population) and were asked to report current levels of pain from the perspective of when their batteries were still functional, further affecting results.

Conclusions

SCS was shown to improve the quality of life of US veterans at SLVHCS across a wide variety of metrics, including activities of daily living, as well as mental and physical health. For veterans with chronic intractable pain who have tried and failed more conservative treatments, SCS is a great treatment.

Suicide is a national public health concern that affects thousands of US individuals and families, with repercussions that reverberate through entire communities. In 2019, there were 47,500 US deaths by suicide, which accounted for about 1 death every 11 minutes.¹ Suicide remains the tenth leading cause of death in the United States and has been part of the top 12 leading causes of death since 1975.² Unfortunately, this trend has worsened; suicide rates have increased by 35% from 1999 to 2018.³ One particularly vulnerable population is US veterans who accounted for 13.8% of all suicide deaths in 2018.⁴ Among veterans, the suicide death average increased from 16.6 per day in 2005 to 17.6 in 2018.⁴ Furthermore, veterans experiencing homelessness are 5 times more likely to attempt suicide and 2.5 times more likely to have suicidal ideation compared with veterans without a history of homelessness.⁴ Suicide is a significant issue among veterans experiencing homelessness: Veterans account for about 11% of the overall US homeless population.⁵

Recent data suggest opportunities for suicide risk assessment in the primary care setting. A study from the Veterans Health Administration (VHA) Office for Suicide Prevention found that in 2014 an average of 20 veterans died by suicide every day and 6 of the 20 (30%) on average used VHA services within the prior year.⁶ Similarly, a review of 40 studies on suicide found that 45% of suicide victims had contact with their primary care practitioner (PCP) within 1 month of suicide, and 75% of victims had contact within the year of suicide.⁷ An analysis of depression screening in 2008/2009 using Patient Health Questionnaire-2 (PHQ-2) or Patient Health Questionnaire-9 (PHQ-9) at 3 large western US Department of Veterans Affairs (VA) medical centers found that 55% were screened for depression.⁸ The VA has made suicide prevention a top priority and supports the established US goal of reducing annual suicide deaths by 20% by 2025.⁹ Given key opportunities for suicide risk assessment in the primary care setting, the VHA Office of Mental Health and Suicide Prevention implemented a national, standardized process for suicide risk assessment on October 1, 2018.^10,11

The VA approach to suicide screening, evaluation, and documentation has evolved over time. Between October 2018 and December 2020, the process was augmented to include 3 stages embedded into the electronic health record (EHR): a primary screen (PHQ-2 with Item 9 from the PHQ-9 [PHQ-2+I9]), a secondary screen (Columbia-Suicide Severity Rating Scale [C-SSRS]), and a tertiary screen (Comprehensive Suicide Risk Evaluation [CSRE]). The primary screen consisted of the depression screening using the PHQ-2 with the addition of I9 asking about suicidal ideation. The secondary screening, or C-SSRS, included 8 questions to elaborate on suicidal ideation, intent, plan, and any history of suicidal attempts or preparatory behaviors. The tertiary screen consisted of the CSRE, a questionnaire developed internally by the VA in 2018 to further evaluate the veteran’s suicidal thoughts, attempts, warning signs, risk factors, protective factors, and reasons for living. The goal of the screenings was to identify veterans at risk of suicide, assess risk severity, and to individually tailor risk mitigation strategies for safe disposition. These risk categories were developed by the regional Mental Illness Research, Education and Clinical Center, which suggested treatment strategies, such as hospitalization or close outpatient follow-up.^12,13

The Homeless Patient Aligned Care Team (HPACT) clinic at the West Los Angeles VA Medical Center (WLAVAMC) in California, one of the largest VA homeless clinics in the country and 1 of 7 national VA Office of Academic Affiliation Centers of Excellence in Primary Care Education training programs implemented the standardized tools for suicide risk screening and quality improvement (QI). The HPACT clinic is an interprofessional team, including primary care, mental health, social work, pharmacy, and peer support, that is adjacent to the WLAVAMC general primary care clinics. The team collaboratively addresses both medical and psychosocial needs of veterans with a focus on the Housing First Model, an approach that prioritizes ending homelessness while addressing all factors associated with veterans' health and well-being. After 1 year of stable housing, veterans graduate to the WLAVAMC general primary care clinics.

Given the vulnerability of veterans experiencing homelessness, the clinic leadership identified suicide risk screening as a high priority initiative and created a taskforce to oversee effective implementation of clinic screening efforts. An interprofessional team of nurse practitioners (NPs), pharmacists, physicians, psychologists, social workers (SWs), and trainees formed to improve screening efforts and use the QI principles to guide analysis and intervention. The team wrote the following SMART (Specific, Measurable, Achievable, Relevant, and Time-bound) Aim statements: (1) ensure > 90% of eligible patients receive a primary screen; (2) ensure > 90% of positive primary I9 screens receive subsequent screenings within 24 hours; and (3) increase staff comfort and familiarity using the screening tools. This article examines the results of the screening initiative 1-year postimplementation, describes difficulties faced, and suggests strategies that might be used to overcome those challenges.

Methods

This QI analysis was exempt from institutional review board review. Prior to the standardized national suicide risk assessment rollout of October 1, 2018, the QI team met to review and understand the workflow to be implemented into the HPACT clinic. To describe the initial screening process, the new suicide risk assessment consisted of primary, secondary, and tertiary screens that would warrant subsequent intervention by clinicians if positive (Figure 1). The primary screen included the PHQ-2+I9 questionnaire (PHQ-2 for depression and I9 for suicidal ideation). If either were positive, follow-up questionnaires were required. Of note, patients with a prior depression diagnosis, cognitive impairment defined at a severity of moderate or greater based on clinician evaluation and judgement, or life expectancy < 6 months were exempt from screening because, by definition, they had theoretically already been screened and classified as under surveillance.

A positive I9 response prompted a secondary screen using C-SSRS. A positive secondary screen prompted a tertiary screen using CSRE. If the PHQ-2 screening was positive but I9 was negative, the standard follow-up depression clinical reminder was used. Any clinical staff member could perform the primary screen, including licensed vocational nurses (LVNs), registered nurses (RNs), and SWs in any setting (eg, emergency department, primary care, inpatient services). The secondary and tertiary screens required completion by a licensed clinician. RNs were able to perform the secondary screen but not the tertiary screen.

The HPACT clinic serves approximately 3000 patients by 50 staff and trainees divided into 2 teams. LVNs and RNs were tasked to conduct the primary screen as part of their initial clinic check-in. If the primary screen was positive for scheduled patients, LVNs notified a PCP to complete the secondary screen. For unscheduled patients, RNs conducted a primary screen and, if positive, a secondary screen. If the secondary screen was positive, a tertiary screen was performed by mental health practitioners or SWs, or PCPs if the former were unavailable. SWs, mental health practitioners, and PCPs were colocated in the clinic, which allowed for safe and convenient warm handoffs between clinicians.

 

During this process, the interprofessional team overseeing the suicide screening implementation efforts in the HPACT clinic met in-person biweekly beginning 1 month prior to the October 1, 2018 implementation. QI tools, including flowcharts and root cause analyses, were used to analyze feedback on efficient workflow and optimize staff responsibilities. A survey assessed staff comfort and familiarity using the suicide screening tools. Informal interviews were conducted with a representative from each stage of patient care to facilitate interprofessional participation and to troubleshoot any issues. Process flowcharts that clearly delineated staff roles based on current clinic workflow and the recommendations set forth by the new process were distributed at an initial staff meeting. The process flowchart was updated after staff feedback and distributed again along with a review of the C-SSRS and CSRE at an all-staff meeting in February 2019. The QI team continued to meet to formally evaluate their SMART Aims and to identify factors driving the success and failure of the implementation.

The VA Informatics and Computing Infrastructure (VINCI) provided project data after a formal request was submitted for this analysis. At the direction of the local QI team, the VINCI team provided aggregate patient counts derived from individual patient data in the VA Corporate Data Warehouse. The data analyzed are frequencies and proportions; no bivariate or multivariate statistics were performed.

Results

During the project year, the HPACT clinic had 2932 unique patients assigned to primary care. Of those veterans, 533 (18%) were exempt from screening by protocol. Of the remainder, staff screened 1876 (64%) of eligible veterans for suicide risk (Figure 2), which did not meet the SMART Aim of screening > 90% of eligible veterans. For the follow-up screens, using a QI dashboard designed for reviewing I9 and C-SSRS results, the QI team reviewed a convenience sample of 5 provider panels and identified 34 positive I9 screens. Twenty of those 34 patients (59%) received a C-SSRS within 24 hours of the positive I9, which did not meet the SMART Aim of ensuring > 90% of primary I9 screens had subsequent C-SSRS screening within 24 hours.

Of the veterans screened, 1,271 (43%) had their screening performed outside of the HPACT primary care team assigned, while 605 (21%) patients had their screening performed by an HPACT member. Most of the screening that occurred outside of the assigned primary care team occurred in other physical settings, including other VA facilities.

Of the 523 (18%) patients who were not screened, 331 (11%) patients had no visit to the HPACT clinic and 132 (5%) empaneled patients did not visit any VA site within the 1-year period. There were 192 (7%) patients who were not screened that had a visit to HPACT while 19 (1%) of those patients declined screening. A total of 184 (6%) patients were not screened and thus were considered true missed opportunities. This group of patients were eligible for screening but did not undergo screening in the HPACT clinic or any other VA setting despite visiting the VA.

The QI team created a fishbone diagram to identify opportunities to improve screening rates and patient care (Figure 3). Using the fishbone tool, the QI team identified 5 main categories limiting complete uptake of suicide risk assessment at the HPACT clinic: health record factors, communication, clinician buy-in, system factors, and patient factors. Among the most salient barriers to use of the screening tool, the EHR system needed to be refreshed after a positive screen to be reminded of the next step, requiring close communication during patient handoffs. Handoff was confusing as there was no dedicated process to communicate positive screen information. Clinicians were concerned that completing the process, especially the tertiary screen, would be time consuming and burdensome in an already busy clinic; some clinicians were uncomfortable discussing the topic of suicide as they did not feel they had the expertise to address a positive screen. In addition, some patients were reluctant to answer the screen honestly due to past hospitalizations or concerns about stigma.

Discussion

Though the QI project failed to meet the SMART Aim of ensuring > 90% of eligible patients received a primary screen for suicide risk and > 90% of positive primary I9 screens received subsequent screenings within 24 hours, the results highlight effective practices and barriers for implementation of wide-scale EHR-based interventions for suicide assessment. Most missed screening opportunities were due to patients being lost to follow-up over the duration of the project, which is a challenge faced in this patient population. A recent analysis of the national rollout of this screening program found that 95% of eligible veterans with a visit to the VA in the first year of the program received screening.¹⁴ In a post hoc analysis using the same eligibility criteria, the rate of screening for this project was 83%. Reflecting on the data from this national cohort compared with the HPACT clinic, this brings to light potential circumstances that may be unique to veterans experiencing homelessness compared with the general veteran population, for instance, the level of engagement may be lower among veterans experiencing homelessness, though this is beyond the scope of this article. Nonetheless, promoting interprofessional collaboration, visualizing effective process flows, establishing clear lines of communication and roles for involved staff, and opening avenues for continuous feedback and troubleshooting are all potentially effective interventions to improve suicide screening rates within the veteran population.

This HPACT clinic initiative aimed to determine how a new screening process would be implemented while identifying potential areas for improvement. Surprisingly, 43% of patients who were screened had their screening performed outside of the HPACT clinic, most often in the inpatient setting at other WLAVAMC clinics or other VA systems. It is possible that due to the nature of the patient population that the HPACT clinic serves with intensive service needs, these patients have wider geographic and clinical location use than most clinic populations due to the transient nature of patients with housing insecurity. What is encouraging, however, is that through this systemwide initiative, there is an impetus to screen veterans, regardless of who performs the screening. This is particularly meaningful given that rates of depression screening may be as low as 4% among PCPs.¹⁵ During implementation, the QI team learned that nearly 18% of the empaneled HPACT patients were exempt from screening. The exempt patients do not have an active clinical reminder for depression screens. Instead, these patients are receiving mental health surveillance and specialty treatment, during which continuous monitoring and assessment for suicidal ideation and risk of suicide are performed. Additionally, an EHR-based factor that also may limit appropriate follow-up and contribute to missed opportunities is that secondary and tertiary screens do not populate until the EHR was refreshed after positive primary screens, which introduces human error in a process that could be automated. Both RNs and PCPs may occasionally miss secondary and tertiary screens due to this issue, which continues to be a barrier. Given the high risk HPACT clinic population, the QI team encouraged staff members to frequently screen patients for suicidal ideation regardless of clinical reminders. A consideration for the future would be to identify optimal frequency for screening and to continue to validate assessment methods.

 

Finally, while the percentage of patients who were considered missed opportunities (visited the HPACT clinic but were not screened) was relatively small at 6% of the total panel of patients, this number theoretically should be zero. Though this project was not designed to identify the specific causes for missed opportunities, future QI efforts may consider evaluating for other potential reasons. These may include differing process flows for various encounters (same-day care visits, scheduled primary care visit, RN-only visit), screening not activating at time of visit, time constraints, or other unseen reasons. Another important population is the 11% of patients who were otherwise eligible for screening but did not visit the HPACT clinic, and in some cases, no other VA location. There are a few explanatory reasons centered on the mobility of this population between health systems. However, this patient population also may be among the most vulnerable and at risk: 62% of veteran suicides in 2017 had not had a VA encounter that year.¹³ While there is no requirement that the veteran visit the HPACT clinic annually, future efforts may focus on increasing engagement through other means of outreach, including site visits and community care involvement, knowing the nature of the sporadic follow-up patterns in this patient population. Future work may also involve examining suicide rates by primary care clinic and triage patterns between interprofessional staff.

Limitations

Due to the limited sample size, findings cannot be generalized to all VA sites. The QI team used retrospective, administrative data. Additionally, since this is a primary care clinic focused on a specialized population, this result may not be generalizable to all primary care settings, other primary care populations, or even other homeless primary care clinics, though it may establish a benchmark when other clinics internally examine their data and processes.

Conclusions

Improving screening protocols can lead to identification of at-risk individuals who would not have otherwise been identified.^16,17 As the US continues to grapple with mental health and suicide, efforts toward addressing this important issue among veterans remains a top priority.

Acknowledgments

Thank you to the VAGLAHS Center of Excellence in Primary Care Education faculty and trainees, the HPACT staff, and the VA Informatics and Computing Infrastructure (VINCI) for data support.

Suicide is a national public health concern that affects thousands of US individuals and families, with repercussions that reverberate through entire communities. In 2019, there were 47,500 US deaths by suicide, which accounted for about 1 death every 11 minutes.¹ Suicide remains the tenth leading cause of death in the United States and has been part of the top 12 leading causes of death since 1975.² Unfortunately, this trend has worsened; suicide rates have increased by 35% from 1999 to 2018.³ One particularly vulnerable population is US veterans who accounted for 13.8% of all suicide deaths in 2018.⁴ Among veterans, the suicide death average increased from 16.6 per day in 2005 to 17.6 in 2018.⁴ Furthermore, veterans experiencing homelessness are 5 times more likely to attempt suicide and 2.5 times more likely to have suicidal ideation compared with veterans without a history of homelessness.⁴ Suicide is a significant issue among veterans experiencing homelessness: Veterans account for about 11% of the overall US homeless population.⁵

Recent data suggest opportunities for suicide risk assessment in the primary care setting. A study from the Veterans Health Administration (VHA) Office for Suicide Prevention found that in 2014 an average of 20 veterans died by suicide every day and 6 of the 20 (30%) on average used VHA services within the prior year.⁶ Similarly, a review of 40 studies on suicide found that 45% of suicide victims had contact with their primary care practitioner (PCP) within 1 month of suicide, and 75% of victims had contact within the year of suicide.⁷ An analysis of depression screening in 2008/2009 using Patient Health Questionnaire-2 (PHQ-2) or Patient Health Questionnaire-9 (PHQ-9) at 3 large western US Department of Veterans Affairs (VA) medical centers found that 55% were screened for depression.⁸ The VA has made suicide prevention a top priority and supports the established US goal of reducing annual suicide deaths by 20% by 2025.⁹ Given key opportunities for suicide risk assessment in the primary care setting, the VHA Office of Mental Health and Suicide Prevention implemented a national, standardized process for suicide risk assessment on October 1, 2018.^10,11

The VA approach to suicide screening, evaluation, and documentation has evolved over time. Between October 2018 and December 2020, the process was augmented to include 3 stages embedded into the electronic health record (EHR): a primary screen (PHQ-2 with Item 9 from the PHQ-9 [PHQ-2+I9]), a secondary screen (Columbia-Suicide Severity Rating Scale [C-SSRS]), and a tertiary screen (Comprehensive Suicide Risk Evaluation [CSRE]). The primary screen consisted of the depression screening using the PHQ-2 with the addition of I9 asking about suicidal ideation. The secondary screening, or C-SSRS, included 8 questions to elaborate on suicidal ideation, intent, plan, and any history of suicidal attempts or preparatory behaviors. The tertiary screen consisted of the CSRE, a questionnaire developed internally by the VA in 2018 to further evaluate the veteran’s suicidal thoughts, attempts, warning signs, risk factors, protective factors, and reasons for living. The goal of the screenings was to identify veterans at risk of suicide, assess risk severity, and to individually tailor risk mitigation strategies for safe disposition. These risk categories were developed by the regional Mental Illness Research, Education and Clinical Center, which suggested treatment strategies, such as hospitalization or close outpatient follow-up.^12,13

The Homeless Patient Aligned Care Team (HPACT) clinic at the West Los Angeles VA Medical Center (WLAVAMC) in California, one of the largest VA homeless clinics in the country and 1 of 7 national VA Office of Academic Affiliation Centers of Excellence in Primary Care Education training programs implemented the standardized tools for suicide risk screening and quality improvement (QI). The HPACT clinic is an interprofessional team, including primary care, mental health, social work, pharmacy, and peer support, that is adjacent to the WLAVAMC general primary care clinics. The team collaboratively addresses both medical and psychosocial needs of veterans with a focus on the Housing First Model, an approach that prioritizes ending homelessness while addressing all factors associated with veterans' health and well-being. After 1 year of stable housing, veterans graduate to the WLAVAMC general primary care clinics.

Given the vulnerability of veterans experiencing homelessness, the clinic leadership identified suicide risk screening as a high priority initiative and created a taskforce to oversee effective implementation of clinic screening efforts. An interprofessional team of nurse practitioners (NPs), pharmacists, physicians, psychologists, social workers (SWs), and trainees formed to improve screening efforts and use the QI principles to guide analysis and intervention. The team wrote the following SMART (Specific, Measurable, Achievable, Relevant, and Time-bound) Aim statements: (1) ensure > 90% of eligible patients receive a primary screen; (2) ensure > 90% of positive primary I9 screens receive subsequent screenings within 24 hours; and (3) increase staff comfort and familiarity using the screening tools. This article examines the results of the screening initiative 1-year postimplementation, describes difficulties faced, and suggests strategies that might be used to overcome those challenges.

Methods

This QI analysis was exempt from institutional review board review. Prior to the standardized national suicide risk assessment rollout of October 1, 2018, the QI team met to review and understand the workflow to be implemented into the HPACT clinic. To describe the initial screening process, the new suicide risk assessment consisted of primary, secondary, and tertiary screens that would warrant subsequent intervention by clinicians if positive (Figure 1). The primary screen included the PHQ-2+I9 questionnaire (PHQ-2 for depression and I9 for suicidal ideation). If either were positive, follow-up questionnaires were required. Of note, patients with a prior depression diagnosis, cognitive impairment defined at a severity of moderate or greater based on clinician evaluation and judgement, or life expectancy < 6 months were exempt from screening because, by definition, they had theoretically already been screened and classified as under surveillance.

A positive I9 response prompted a secondary screen using C-SSRS. A positive secondary screen prompted a tertiary screen using CSRE. If the PHQ-2 screening was positive but I9 was negative, the standard follow-up depression clinical reminder was used. Any clinical staff member could perform the primary screen, including licensed vocational nurses (LVNs), registered nurses (RNs), and SWs in any setting (eg, emergency department, primary care, inpatient services). The secondary and tertiary screens required completion by a licensed clinician. RNs were able to perform the secondary screen but not the tertiary screen.

The HPACT clinic serves approximately 3000 patients by 50 staff and trainees divided into 2 teams. LVNs and RNs were tasked to conduct the primary screen as part of their initial clinic check-in. If the primary screen was positive for scheduled patients, LVNs notified a PCP to complete the secondary screen. For unscheduled patients, RNs conducted a primary screen and, if positive, a secondary screen. If the secondary screen was positive, a tertiary screen was performed by mental health practitioners or SWs, or PCPs if the former were unavailable. SWs, mental health practitioners, and PCPs were colocated in the clinic, which allowed for safe and convenient warm handoffs between clinicians.

 

During this process, the interprofessional team overseeing the suicide screening implementation efforts in the HPACT clinic met in-person biweekly beginning 1 month prior to the October 1, 2018 implementation. QI tools, including flowcharts and root cause analyses, were used to analyze feedback on efficient workflow and optimize staff responsibilities. A survey assessed staff comfort and familiarity using the suicide screening tools. Informal interviews were conducted with a representative from each stage of patient care to facilitate interprofessional participation and to troubleshoot any issues. Process flowcharts that clearly delineated staff roles based on current clinic workflow and the recommendations set forth by the new process were distributed at an initial staff meeting. The process flowchart was updated after staff feedback and distributed again along with a review of the C-SSRS and CSRE at an all-staff meeting in February 2019. The QI team continued to meet to formally evaluate their SMART Aims and to identify factors driving the success and failure of the implementation.

The VA Informatics and Computing Infrastructure (VINCI) provided project data after a formal request was submitted for this analysis. At the direction of the local QI team, the VINCI team provided aggregate patient counts derived from individual patient data in the VA Corporate Data Warehouse. The data analyzed are frequencies and proportions; no bivariate or multivariate statistics were performed.

Results

During the project year, the HPACT clinic had 2932 unique patients assigned to primary care. Of those veterans, 533 (18%) were exempt from screening by protocol. Of the remainder, staff screened 1876 (64%) of eligible veterans for suicide risk (Figure 2), which did not meet the SMART Aim of screening > 90% of eligible veterans. For the follow-up screens, using a QI dashboard designed for reviewing I9 and C-SSRS results, the QI team reviewed a convenience sample of 5 provider panels and identified 34 positive I9 screens. Twenty of those 34 patients (59%) received a C-SSRS within 24 hours of the positive I9, which did not meet the SMART Aim of ensuring > 90% of primary I9 screens had subsequent C-SSRS screening within 24 hours.

Of the veterans screened, 1,271 (43%) had their screening performed outside of the HPACT primary care team assigned, while 605 (21%) patients had their screening performed by an HPACT member. Most of the screening that occurred outside of the assigned primary care team occurred in other physical settings, including other VA facilities.

Of the 523 (18%) patients who were not screened, 331 (11%) patients had no visit to the HPACT clinic and 132 (5%) empaneled patients did not visit any VA site within the 1-year period. There were 192 (7%) patients who were not screened that had a visit to HPACT while 19 (1%) of those patients declined screening. A total of 184 (6%) patients were not screened and thus were considered true missed opportunities. This group of patients were eligible for screening but did not undergo screening in the HPACT clinic or any other VA setting despite visiting the VA.

The QI team created a fishbone diagram to identify opportunities to improve screening rates and patient care (Figure 3). Using the fishbone tool, the QI team identified 5 main categories limiting complete uptake of suicide risk assessment at the HPACT clinic: health record factors, communication, clinician buy-in, system factors, and patient factors. Among the most salient barriers to use of the screening tool, the EHR system needed to be refreshed after a positive screen to be reminded of the next step, requiring close communication during patient handoffs. Handoff was confusing as there was no dedicated process to communicate positive screen information. Clinicians were concerned that completing the process, especially the tertiary screen, would be time consuming and burdensome in an already busy clinic; some clinicians were uncomfortable discussing the topic of suicide as they did not feel they had the expertise to address a positive screen. In addition, some patients were reluctant to answer the screen honestly due to past hospitalizations or concerns about stigma.

Discussion

Though the QI project failed to meet the SMART Aim of ensuring > 90% of eligible patients received a primary screen for suicide risk and > 90% of positive primary I9 screens received subsequent screenings within 24 hours, the results highlight effective practices and barriers for implementation of wide-scale EHR-based interventions for suicide assessment. Most missed screening opportunities were due to patients being lost to follow-up over the duration of the project, which is a challenge faced in this patient population. A recent analysis of the national rollout of this screening program found that 95% of eligible veterans with a visit to the VA in the first year of the program received screening.¹⁴ In a post hoc analysis using the same eligibility criteria, the rate of screening for this project was 83%. Reflecting on the data from this national cohort compared with the HPACT clinic, this brings to light potential circumstances that may be unique to veterans experiencing homelessness compared with the general veteran population, for instance, the level of engagement may be lower among veterans experiencing homelessness, though this is beyond the scope of this article. Nonetheless, promoting interprofessional collaboration, visualizing effective process flows, establishing clear lines of communication and roles for involved staff, and opening avenues for continuous feedback and troubleshooting are all potentially effective interventions to improve suicide screening rates within the veteran population.

This HPACT clinic initiative aimed to determine how a new screening process would be implemented while identifying potential areas for improvement. Surprisingly, 43% of patients who were screened had their screening performed outside of the HPACT clinic, most often in the inpatient setting at other WLAVAMC clinics or other VA systems. It is possible that due to the nature of the patient population that the HPACT clinic serves with intensive service needs, these patients have wider geographic and clinical location use than most clinic populations due to the transient nature of patients with housing insecurity. What is encouraging, however, is that through this systemwide initiative, there is an impetus to screen veterans, regardless of who performs the screening. This is particularly meaningful given that rates of depression screening may be as low as 4% among PCPs.¹⁵ During implementation, the QI team learned that nearly 18% of the empaneled HPACT patients were exempt from screening. The exempt patients do not have an active clinical reminder for depression screens. Instead, these patients are receiving mental health surveillance and specialty treatment, during which continuous monitoring and assessment for suicidal ideation and risk of suicide are performed. Additionally, an EHR-based factor that also may limit appropriate follow-up and contribute to missed opportunities is that secondary and tertiary screens do not populate until the EHR was refreshed after positive primary screens, which introduces human error in a process that could be automated. Both RNs and PCPs may occasionally miss secondary and tertiary screens due to this issue, which continues to be a barrier. Given the high risk HPACT clinic population, the QI team encouraged staff members to frequently screen patients for suicidal ideation regardless of clinical reminders. A consideration for the future would be to identify optimal frequency for screening and to continue to validate assessment methods.

 

Finally, while the percentage of patients who were considered missed opportunities (visited the HPACT clinic but were not screened) was relatively small at 6% of the total panel of patients, this number theoretically should be zero. Though this project was not designed to identify the specific causes for missed opportunities, future QI efforts may consider evaluating for other potential reasons. These may include differing process flows for various encounters (same-day care visits, scheduled primary care visit, RN-only visit), screening not activating at time of visit, time constraints, or other unseen reasons. Another important population is the 11% of patients who were otherwise eligible for screening but did not visit the HPACT clinic, and in some cases, no other VA location. There are a few explanatory reasons centered on the mobility of this population between health systems. However, this patient population also may be among the most vulnerable and at risk: 62% of veteran suicides in 2017 had not had a VA encounter that year.¹³ While there is no requirement that the veteran visit the HPACT clinic annually, future efforts may focus on increasing engagement through other means of outreach, including site visits and community care involvement, knowing the nature of the sporadic follow-up patterns in this patient population. Future work may also involve examining suicide rates by primary care clinic and triage patterns between interprofessional staff.

Limitations

Due to the limited sample size, findings cannot be generalized to all VA sites. The QI team used retrospective, administrative data. Additionally, since this is a primary care clinic focused on a specialized population, this result may not be generalizable to all primary care settings, other primary care populations, or even other homeless primary care clinics, though it may establish a benchmark when other clinics internally examine their data and processes.

Conclusions

Improving screening protocols can lead to identification of at-risk individuals who would not have otherwise been identified.^16,17 As the US continues to grapple with mental health and suicide, efforts toward addressing this important issue among veterans remains a top priority.

Acknowledgments

Thank you to the VAGLAHS Center of Excellence in Primary Care Education faculty and trainees, the HPACT staff, and the VA Informatics and Computing Infrastructure (VINCI) for data support.

Suicide is a national public health concern that affects thousands of US individuals and families, with repercussions that reverberate through entire communities. In 2019, there were 47,500 US deaths by suicide, which accounted for about 1 death every 11 minutes.¹ Suicide remains the tenth leading cause of death in the United States and has been part of the top 12 leading causes of death since 1975.² Unfortunately, this trend has worsened; suicide rates have increased by 35% from 1999 to 2018.³ One particularly vulnerable population is US veterans who accounted for 13.8% of all suicide deaths in 2018.⁴ Among veterans, the suicide death average increased from 16.6 per day in 2005 to 17.6 in 2018.⁴ Furthermore, veterans experiencing homelessness are 5 times more likely to attempt suicide and 2.5 times more likely to have suicidal ideation compared with veterans without a history of homelessness.⁴ Suicide is a significant issue among veterans experiencing homelessness: Veterans account for about 11% of the overall US homeless population.⁵

Recent data suggest opportunities for suicide risk assessment in the primary care setting. A study from the Veterans Health Administration (VHA) Office for Suicide Prevention found that in 2014 an average of 20 veterans died by suicide every day and 6 of the 20 (30%) on average used VHA services within the prior year.⁶ Similarly, a review of 40 studies on suicide found that 45% of suicide victims had contact with their primary care practitioner (PCP) within 1 month of suicide, and 75% of victims had contact within the year of suicide.⁷ An analysis of depression screening in 2008/2009 using Patient Health Questionnaire-2 (PHQ-2) or Patient Health Questionnaire-9 (PHQ-9) at 3 large western US Department of Veterans Affairs (VA) medical centers found that 55% were screened for depression.⁸ The VA has made suicide prevention a top priority and supports the established US goal of reducing annual suicide deaths by 20% by 2025.⁹ Given key opportunities for suicide risk assessment in the primary care setting, the VHA Office of Mental Health and Suicide Prevention implemented a national, standardized process for suicide risk assessment on October 1, 2018.^10,11

The VA approach to suicide screening, evaluation, and documentation has evolved over time. Between October 2018 and December 2020, the process was augmented to include 3 stages embedded into the electronic health record (EHR): a primary screen (PHQ-2 with Item 9 from the PHQ-9 [PHQ-2+I9]), a secondary screen (Columbia-Suicide Severity Rating Scale [C-SSRS]), and a tertiary screen (Comprehensive Suicide Risk Evaluation [CSRE]). The primary screen consisted of the depression screening using the PHQ-2 with the addition of I9 asking about suicidal ideation. The secondary screening, or C-SSRS, included 8 questions to elaborate on suicidal ideation, intent, plan, and any history of suicidal attempts or preparatory behaviors. The tertiary screen consisted of the CSRE, a questionnaire developed internally by the VA in 2018 to further evaluate the veteran’s suicidal thoughts, attempts, warning signs, risk factors, protective factors, and reasons for living. The goal of the screenings was to identify veterans at risk of suicide, assess risk severity, and to individually tailor risk mitigation strategies for safe disposition. These risk categories were developed by the regional Mental Illness Research, Education and Clinical Center, which suggested treatment strategies, such as hospitalization or close outpatient follow-up.^12,13

The Homeless Patient Aligned Care Team (HPACT) clinic at the West Los Angeles VA Medical Center (WLAVAMC) in California, one of the largest VA homeless clinics in the country and 1 of 7 national VA Office of Academic Affiliation Centers of Excellence in Primary Care Education training programs implemented the standardized tools for suicide risk screening and quality improvement (QI). The HPACT clinic is an interprofessional team, including primary care, mental health, social work, pharmacy, and peer support, that is adjacent to the WLAVAMC general primary care clinics. The team collaboratively addresses both medical and psychosocial needs of veterans with a focus on the Housing First Model, an approach that prioritizes ending homelessness while addressing all factors associated with veterans' health and well-being. After 1 year of stable housing, veterans graduate to the WLAVAMC general primary care clinics.

Given the vulnerability of veterans experiencing homelessness, the clinic leadership identified suicide risk screening as a high priority initiative and created a taskforce to oversee effective implementation of clinic screening efforts. An interprofessional team of nurse practitioners (NPs), pharmacists, physicians, psychologists, social workers (SWs), and trainees formed to improve screening efforts and use the QI principles to guide analysis and intervention. The team wrote the following SMART (Specific, Measurable, Achievable, Relevant, and Time-bound) Aim statements: (1) ensure > 90% of eligible patients receive a primary screen; (2) ensure > 90% of positive primary I9 screens receive subsequent screenings within 24 hours; and (3) increase staff comfort and familiarity using the screening tools. This article examines the results of the screening initiative 1-year postimplementation, describes difficulties faced, and suggests strategies that might be used to overcome those challenges.

Methods

This QI analysis was exempt from institutional review board review. Prior to the standardized national suicide risk assessment rollout of October 1, 2018, the QI team met to review and understand the workflow to be implemented into the HPACT clinic. To describe the initial screening process, the new suicide risk assessment consisted of primary, secondary, and tertiary screens that would warrant subsequent intervention by clinicians if positive (Figure 1). The primary screen included the PHQ-2+I9 questionnaire (PHQ-2 for depression and I9 for suicidal ideation). If either were positive, follow-up questionnaires were required. Of note, patients with a prior depression diagnosis, cognitive impairment defined at a severity of moderate or greater based on clinician evaluation and judgement, or life expectancy < 6 months were exempt from screening because, by definition, they had theoretically already been screened and classified as under surveillance.

A positive I9 response prompted a secondary screen using C-SSRS. A positive secondary screen prompted a tertiary screen using CSRE. If the PHQ-2 screening was positive but I9 was negative, the standard follow-up depression clinical reminder was used. Any clinical staff member could perform the primary screen, including licensed vocational nurses (LVNs), registered nurses (RNs), and SWs in any setting (eg, emergency department, primary care, inpatient services). The secondary and tertiary screens required completion by a licensed clinician. RNs were able to perform the secondary screen but not the tertiary screen.

The HPACT clinic serves approximately 3000 patients by 50 staff and trainees divided into 2 teams. LVNs and RNs were tasked to conduct the primary screen as part of their initial clinic check-in. If the primary screen was positive for scheduled patients, LVNs notified a PCP to complete the secondary screen. For unscheduled patients, RNs conducted a primary screen and, if positive, a secondary screen. If the secondary screen was positive, a tertiary screen was performed by mental health practitioners or SWs, or PCPs if the former were unavailable. SWs, mental health practitioners, and PCPs were colocated in the clinic, which allowed for safe and convenient warm handoffs between clinicians.

 

During this process, the interprofessional team overseeing the suicide screening implementation efforts in the HPACT clinic met in-person biweekly beginning 1 month prior to the October 1, 2018 implementation. QI tools, including flowcharts and root cause analyses, were used to analyze feedback on efficient workflow and optimize staff responsibilities. A survey assessed staff comfort and familiarity using the suicide screening tools. Informal interviews were conducted with a representative from each stage of patient care to facilitate interprofessional participation and to troubleshoot any issues. Process flowcharts that clearly delineated staff roles based on current clinic workflow and the recommendations set forth by the new process were distributed at an initial staff meeting. The process flowchart was updated after staff feedback and distributed again along with a review of the C-SSRS and CSRE at an all-staff meeting in February 2019. The QI team continued to meet to formally evaluate their SMART Aims and to identify factors driving the success and failure of the implementation.

The VA Informatics and Computing Infrastructure (VINCI) provided project data after a formal request was submitted for this analysis. At the direction of the local QI team, the VINCI team provided aggregate patient counts derived from individual patient data in the VA Corporate Data Warehouse. The data analyzed are frequencies and proportions; no bivariate or multivariate statistics were performed.

Results

During the project year, the HPACT clinic had 2932 unique patients assigned to primary care. Of those veterans, 533 (18%) were exempt from screening by protocol. Of the remainder, staff screened 1876 (64%) of eligible veterans for suicide risk (Figure 2), which did not meet the SMART Aim of screening > 90% of eligible veterans. For the follow-up screens, using a QI dashboard designed for reviewing I9 and C-SSRS results, the QI team reviewed a convenience sample of 5 provider panels and identified 34 positive I9 screens. Twenty of those 34 patients (59%) received a C-SSRS within 24 hours of the positive I9, which did not meet the SMART Aim of ensuring > 90% of primary I9 screens had subsequent C-SSRS screening within 24 hours.

Of the veterans screened, 1,271 (43%) had their screening performed outside of the HPACT primary care team assigned, while 605 (21%) patients had their screening performed by an HPACT member. Most of the screening that occurred outside of the assigned primary care team occurred in other physical settings, including other VA facilities.

Of the 523 (18%) patients who were not screened, 331 (11%) patients had no visit to the HPACT clinic and 132 (5%) empaneled patients did not visit any VA site within the 1-year period. There were 192 (7%) patients who were not screened that had a visit to HPACT while 19 (1%) of those patients declined screening. A total of 184 (6%) patients were not screened and thus were considered true missed opportunities. This group of patients were eligible for screening but did not undergo screening in the HPACT clinic or any other VA setting despite visiting the VA.

The QI team created a fishbone diagram to identify opportunities to improve screening rates and patient care (Figure 3). Using the fishbone tool, the QI team identified 5 main categories limiting complete uptake of suicide risk assessment at the HPACT clinic: health record factors, communication, clinician buy-in, system factors, and patient factors. Among the most salient barriers to use of the screening tool, the EHR system needed to be refreshed after a positive screen to be reminded of the next step, requiring close communication during patient handoffs. Handoff was confusing as there was no dedicated process to communicate positive screen information. Clinicians were concerned that completing the process, especially the tertiary screen, would be time consuming and burdensome in an already busy clinic; some clinicians were uncomfortable discussing the topic of suicide as they did not feel they had the expertise to address a positive screen. In addition, some patients were reluctant to answer the screen honestly due to past hospitalizations or concerns about stigma.

Discussion

Though the QI project failed to meet the SMART Aim of ensuring > 90% of eligible patients received a primary screen for suicide risk and > 90% of positive primary I9 screens received subsequent screenings within 24 hours, the results highlight effective practices and barriers for implementation of wide-scale EHR-based interventions for suicide assessment. Most missed screening opportunities were due to patients being lost to follow-up over the duration of the project, which is a challenge faced in this patient population. A recent analysis of the national rollout of this screening program found that 95% of eligible veterans with a visit to the VA in the first year of the program received screening.¹⁴ In a post hoc analysis using the same eligibility criteria, the rate of screening for this project was 83%. Reflecting on the data from this national cohort compared with the HPACT clinic, this brings to light potential circumstances that may be unique to veterans experiencing homelessness compared with the general veteran population, for instance, the level of engagement may be lower among veterans experiencing homelessness, though this is beyond the scope of this article. Nonetheless, promoting interprofessional collaboration, visualizing effective process flows, establishing clear lines of communication and roles for involved staff, and opening avenues for continuous feedback and troubleshooting are all potentially effective interventions to improve suicide screening rates within the veteran population.

This HPACT clinic initiative aimed to determine how a new screening process would be implemented while identifying potential areas for improvement. Surprisingly, 43% of patients who were screened had their screening performed outside of the HPACT clinic, most often in the inpatient setting at other WLAVAMC clinics or other VA systems. It is possible that due to the nature of the patient population that the HPACT clinic serves with intensive service needs, these patients have wider geographic and clinical location use than most clinic populations due to the transient nature of patients with housing insecurity. What is encouraging, however, is that through this systemwide initiative, there is an impetus to screen veterans, regardless of who performs the screening. This is particularly meaningful given that rates of depression screening may be as low as 4% among PCPs.¹⁵ During implementation, the QI team learned that nearly 18% of the empaneled HPACT patients were exempt from screening. The exempt patients do not have an active clinical reminder for depression screens. Instead, these patients are receiving mental health surveillance and specialty treatment, during which continuous monitoring and assessment for suicidal ideation and risk of suicide are performed. Additionally, an EHR-based factor that also may limit appropriate follow-up and contribute to missed opportunities is that secondary and tertiary screens do not populate until the EHR was refreshed after positive primary screens, which introduces human error in a process that could be automated. Both RNs and PCPs may occasionally miss secondary and tertiary screens due to this issue, which continues to be a barrier. Given the high risk HPACT clinic population, the QI team encouraged staff members to frequently screen patients for suicidal ideation regardless of clinical reminders. A consideration for the future would be to identify optimal frequency for screening and to continue to validate assessment methods.

 

Finally, while the percentage of patients who were considered missed opportunities (visited the HPACT clinic but were not screened) was relatively small at 6% of the total panel of patients, this number theoretically should be zero. Though this project was not designed to identify the specific causes for missed opportunities, future QI efforts may consider evaluating for other potential reasons. These may include differing process flows for various encounters (same-day care visits, scheduled primary care visit, RN-only visit), screening not activating at time of visit, time constraints, or other unseen reasons. Another important population is the 11% of patients who were otherwise eligible for screening but did not visit the HPACT clinic, and in some cases, no other VA location. There are a few explanatory reasons centered on the mobility of this population between health systems. However, this patient population also may be among the most vulnerable and at risk: 62% of veteran suicides in 2017 had not had a VA encounter that year.¹³ While there is no requirement that the veteran visit the HPACT clinic annually, future efforts may focus on increasing engagement through other means of outreach, including site visits and community care involvement, knowing the nature of the sporadic follow-up patterns in this patient population. Future work may also involve examining suicide rates by primary care clinic and triage patterns between interprofessional staff.

Limitations

Due to the limited sample size, findings cannot be generalized to all VA sites. The QI team used retrospective, administrative data. Additionally, since this is a primary care clinic focused on a specialized population, this result may not be generalizable to all primary care settings, other primary care populations, or even other homeless primary care clinics, though it may establish a benchmark when other clinics internally examine their data and processes.

Conclusions

Improving screening protocols can lead to identification of at-risk individuals who would not have otherwise been identified.^16,17 As the US continues to grapple with mental health and suicide, efforts toward addressing this important issue among veterans remains a top priority.

Acknowledgments

Thank you to the VAGLAHS Center of Excellence in Primary Care Education faculty and trainees, the HPACT staff, and the VA Informatics and Computing Infrastructure (VINCI) for data support.

Electronic consultation (e-consult) is designed to increase access to specialty care by facilitating communication between primary care and specialty clinicians without the need for outpatient face-to-face encounters.^1–4 In 2011, the US Department of Veterans Affairs (VA) implemented an e-consult program as a component of its overall strategy to increase access to specialty services, reduce costs of care, and reduce appointment travel burden on patients.

E-consult has substantially increased within the VA since its implementation.^5,6 Consistent with limited evaluations from other health care systems, evaluations of the VA e-consult program demonstrated reduced costs, reduced travel time for patients, and improved access to specialty care.^2,5–11 However, there is wide variation in e-consult use across VA specialties, facilities, and regions.^5,6,12,13 For example, hematology, preoperative evaluation, neurosurgery, endocrinology, and infectious diseases use e-consults more frequently when compared with in-person consults in the VA.⁶ Reasons for this variation or specific barriers and facilitators of using e-consults have not been described.

Prior qualitative studies report that primary care practitioners (PCPs) describe e-consults as convenient, educational, beneficial for patient care, and useful for improving patient access to specialty care.^8,14,15 One study identified limited PCP knowledge of e-consults as a barrier to use.¹⁶ Specialists have reported that e-consult improves clinical communication, but increases their workload.^1,14,17,18 These studies did not assess perspectives from both clinicians who initiate e-consults and those who respond to them. This is the first qualitative study to assess e-consult perceptions from perspectives of both PCPs and specialists among a large, national sample of VA clinicians who use e-consults. The objective of this study was to understand perspectives of e-consults between PCPs and specialists that may be relevant to increasing adoption in the VA.

Methods

The team (CL, ML, PG, 2 analysts under the guidance of GS and JS and support from RRK, and a biostatistician) conducted semistructured interviews with PCPs, specialists, and specialty division leaders who were employed by VA in 2016 and 2017. Specialties of interest were identified by the VA Office of Specialty Care and included cardiology, endocrinology, gastroenterology, and hematology.

E-Consult Procedures

Within the VA, the specific procedures used to initiate, triage and manage e-consults are coordinated at VA medical centers (VAMCs) and at the Veterans Integrated Service Network (VISN) regional level. E-consult can be requested by any clinician. Generally, e-consults are initiated by PCPs through standardized, specialty-specific templates. Recipients, typically specialists, respond by answering questions, suggesting additional testing and evaluation, or requesting an in-person visit. Communication is documented in the patient’s electronic health record (EHR). Specialists receive different levels of workload credit for responding to e-consults similar to a relative value unit reimbursement model. Training in the use of e-consults is available to practitioners but may vary at local and regional levels.

Recruitment

Our sample included PCPs, specialists, and specialty care division leaders. We first quantified e-consult rates (e-consults per 100 patient visits) between July 2016 and June 2017 at VA facilities within primary care and the 4 priority specialties and identified the 30 sites with the highest e-consult rates and 30 sites with the lowest e-consult rates. Sites with < 500 total visits, < 3 specialties, or without any e-consult visit during the study period were excluded. E-consult rates at community-based outpatient clinics were included with associated VAMCs. We then stratified PCPs by whether they were high or low users of e-consults (determined by the top and bottom users within each site) and credentials (MD vs nurse practitioner [NP] or physician assistant [PA]). Specialists were sampled based on their rate of use relative to colleagues within their site and the use rate of their division. We sampled division chiefs and individuals who had > 300 total visits and 1 e-consult during the study period. To recruit participants, the primary investigator sent an initial email and 2 reminder emails. The team followed up with respondents to schedule an interview.

Interview guides were designed to elicit rich descriptions of barriers and facilitators to e-consult use (eAppendix available at doi:10.12788/fp.0214). The team used the Practical Robust Implementation and Sustainability Model (PRISM), which considers factors along 6 domains for intervention planning, implementation, and sustainment.¹⁹ Telephone interviews lasted about 20 minutes and were conducted between September 2017 and March 2018. Interviews were recorded and transcribed verbatim.

Analysis

The team used an iterative, team-based, inductive/deductive approach to conventional content analysis.^20,21 Initial code categories were created so that we could identify e-consult best practices—facilitators of e-consult that were recommended by both PCPs and specialists. Inductive codes or labels applied to identify meaningful quotations, phrases, or key terms were used to identify emergent ideas and were added throughout coding after discussion among team members. Consensus was reached using a team-based approach.²¹ Four analysts independently coded the same 3 transcripts and met to discuss points of divergence and convergence. Analyses continued with emergent themes, categories, and conclusions. Atlas.ti. v.7 was used for coding and data management.²²

 

Results

We conducted 34 interviews with clinicians (Table 1) from 13 VISNs. Four best-practice themes emerged among both PCPs and specialists, including that e-consults (1) are best suited for certain clinical questions and patients; (2) require relevant background information from requesting clinicians and clear recommendations from responding clinicians; (3) are a novel opportunity to provide efficient, transparent care; and (4) may not be fully adopted due to low awareness. Supporting quotations for the following findings are provided in Table 2.

Specific Clinical Questions and Patients

PCPs described specific patients and questions for which they most frequently used e-consults, such as for medication changes (Q1), determining treatment steps (Q2,3), and or clarifying laboratory or imaging findings. PCPs frequently used e-consults for patients who did not require a physical examination or when specialists could make recommendations without seeing patients face-to-face (Q3). An important use of e-consults described by PCPs was for treating conditions they could manage within primary care if additional guidance were available (Q4). Several PCPs and specialists also noted that e-consults were particularly useful for patients who were unable to travel or did not want face-to-face appointments (Q5). Notably, PCPs and specialists mentioned situations for which e-consults were inappropriate, including when a detailed history or physical examination was needed, or if a complex condition was suspected (Q6).

Background Data and Clear Recommendations

Participants described necessary data that should be included in high-quality e-consults. Specialists voiced frustration in time-consuming chart reviews that were often necessary when these data were not provided by the requestor. In some cases, specialists were unable to access necessary EHR data, which delayed responses (Q7). PCPs noted that the most useful responses carefully considered the question, used current patient information to determine treatments, provided clear recommendations, and defined who was responsible for next steps (Q8). PCPs and specialists stated that e-consult templates that required relevant information facilitated high-quality e-consults. Neither wanted to waste the other clinician's time (Q8).

A Novel Opportunity

Many PCPs felt that e-consults improved communication (eg, efficiency, response time), established new communication between clinicians, and reduced patients’ appointment burden (Q10, Q11). Many specialists felt that e-consults improved documentation of communication between clinicians and increased transparency of clinical decisions (Q12). Additionally, many specialists mentioned that e-consults capture previously informal curbside consults, enabling them to receive workload credit (Q13).

Lack of Awareness

Some noted that the biggest barrier to e-consults was not being aware of them generally, or which specialties offer e-consults (Q14). One PCP described e-consults as the best kept secret and found value in sharing the utility of e-consults with colleagues (Q15). All participants, including those who did not frequently use e-consults, felt that e-consults improved the quality of care by providing more timely care or better answers to clinical questions (Q16). Several practitioners also felt that e-consults increased access to specialty care. For example, specialists reported that e-consults enabled them to better manage patient load by using e-consults to answer relatively simple questions, reserving face-to-face consults for more complex patients (Q17).

Discussion

The objective of this study was to identify potential best practices for e-consults that may help increase their quality and use within the VA. We built on prior studies that offered insights on PCP and specialists’ overall satisfaction with e-consult by identifying several themes relevant to the further adoption of e-consults in the VA and elsewhere without a face-to-face visit.^{8,13,14,16–18} Future work may be beneficial in identifying whether the study themes identified can explain variation in e-consult use or whether addressing these factors might lead to increased or higher quality e-consult use. We are unaware of any qualitative study of comparable scale in a different health care system. Further, this is the first study to assess perspectives on e-consults among those who initiate and respond to them within the same health care system. Perhaps the most important finding from this study is that e-consults are generally viewed favorably, which is a necessary leverage point to increase their adoption within the system.

 

Clinicians reported several benefits to e-consults, including timely responses to clinical questions, efficient communication, allow for documentation of specialist recommendations, and help capture workload. These benefits are consistent with prior literature that indicates both PCPs and specialists in the VA and other health care systems feel that e-consults improves communication, decreases unnecessary visits, and improves quality of care.^1,14,17,18 In particular, clinicians reported that e-consults improve their practice efficiency and efficacy. This is of critical importance given the pressures of providing timely access to primary and specialty care within the VA. Interestingly, many VA practitioners were unaware which specialties offered e-consults within their facilities, reflecting previous work showing that PCPs are often unaware of e-consult options.¹⁶ This may partially explain variation in e-consult use. Increasing awareness and educating clinicians on the benefits of e-consults may help promote use among non- and low users.

A common theme reported by both groups was the importance of providing necessary information within e-consult questions and responses. Specialists felt there was a need to ensure that PCPs provide relevant and patient-specific information that would enable them to efficiently and accurately answer questions without the need for extensive EHR review. This reflects previous work showing that specialists are often unable to respond to e-consult requests because they do not contain sufficient information.²² PCPs described a need to ensure that specialists’ responses included information that was detailed enough to make clinical decisions without the need for a reconsult. This highlights a common challenge to medical consultation, in that necessary or relevant information may not be apparent to all clinicians. To address this, there may be a role in developing enhanced, flexible templating that elicits necessary patient-specific information. Such a template may automatically pull relevant data from the EHR and prompt clinicians to provide important information. We did not assess how perspectives of templates varied, and further work could help define precisely what constitutes an effective template, including how it should capture appropriate patient data and how this impacts acceptability or use of e-consults generally. Collaboratively developed service agreements and e-consult templates could help guide PCPs and specialists to engage in efficient communication.

Another theme among both groups was that e-consult is most appropriate within specific clinical scenarios. Examples included review of laboratory results, questions about medication changes, or for patients who were reluctant to travel to appointments. Identifying and promoting specific opportunities for e-consults may help increase their use and align e-consult practices with scenarios that are likely to provide the most benefit to patients. For example, it could be helpful to understand the distance patients must travel for specialty care. Providing that information during clinical encounters could trigger clinicians to consider e-consults as an option. Future work might aim to identify clinical scenarios that clinicians feel are not well suited for e-consults and determine how to adapt them for those scenarios.

Limitations

Generalizability of these findings is limited given the qualitative study design. Participants’ descriptions of experiences with e-consults reflect the experiences of clinicians in the VA and may not reflect clinicians in other settings. We also interviewed a sample of clinicians who were already using e-consults. Important information could be learned from future work with those who have not yet adopted e-consult procedures or adopted and abandoned them.

Conclusions

E-consult is perceived as beneficial by VA PCPs and specialists. Participants suggested using e-consults for appropriate questions or patients and including necessary information and next steps in both the initial e-consult and response. Finding ways to facilitate e-consults with these suggestions in mind may increase delivery of high-quality e-consults. Future work could compare the findings of this work to similar work assessing clinicians perceptions of e-consults outside of the VA.

Electronic consultation (e-consult) is designed to increase access to specialty care by facilitating communication between primary care and specialty clinicians without the need for outpatient face-to-face encounters.^1–4 In 2011, the US Department of Veterans Affairs (VA) implemented an e-consult program as a component of its overall strategy to increase access to specialty services, reduce costs of care, and reduce appointment travel burden on patients.

E-consult has substantially increased within the VA since its implementation.^5,6 Consistent with limited evaluations from other health care systems, evaluations of the VA e-consult program demonstrated reduced costs, reduced travel time for patients, and improved access to specialty care.^2,5–11 However, there is wide variation in e-consult use across VA specialties, facilities, and regions.^5,6,12,13 For example, hematology, preoperative evaluation, neurosurgery, endocrinology, and infectious diseases use e-consults more frequently when compared with in-person consults in the VA.⁶ Reasons for this variation or specific barriers and facilitators of using e-consults have not been described.

Prior qualitative studies report that primary care practitioners (PCPs) describe e-consults as convenient, educational, beneficial for patient care, and useful for improving patient access to specialty care.^8,14,15 One study identified limited PCP knowledge of e-consults as a barrier to use.¹⁶ Specialists have reported that e-consult improves clinical communication, but increases their workload.^1,14,17,18 These studies did not assess perspectives from both clinicians who initiate e-consults and those who respond to them. This is the first qualitative study to assess e-consult perceptions from perspectives of both PCPs and specialists among a large, national sample of VA clinicians who use e-consults. The objective of this study was to understand perspectives of e-consults between PCPs and specialists that may be relevant to increasing adoption in the VA.

Methods

The team (CL, ML, PG, 2 analysts under the guidance of GS and JS and support from RRK, and a biostatistician) conducted semistructured interviews with PCPs, specialists, and specialty division leaders who were employed by VA in 2016 and 2017. Specialties of interest were identified by the VA Office of Specialty Care and included cardiology, endocrinology, gastroenterology, and hematology.

E-Consult Procedures

Within the VA, the specific procedures used to initiate, triage and manage e-consults are coordinated at VA medical centers (VAMCs) and at the Veterans Integrated Service Network (VISN) regional level. E-consult can be requested by any clinician. Generally, e-consults are initiated by PCPs through standardized, specialty-specific templates. Recipients, typically specialists, respond by answering questions, suggesting additional testing and evaluation, or requesting an in-person visit. Communication is documented in the patient’s electronic health record (EHR). Specialists receive different levels of workload credit for responding to e-consults similar to a relative value unit reimbursement model. Training in the use of e-consults is available to practitioners but may vary at local and regional levels.

Recruitment

Our sample included PCPs, specialists, and specialty care division leaders. We first quantified e-consult rates (e-consults per 100 patient visits) between July 2016 and June 2017 at VA facilities within primary care and the 4 priority specialties and identified the 30 sites with the highest e-consult rates and 30 sites with the lowest e-consult rates. Sites with < 500 total visits, < 3 specialties, or without any e-consult visit during the study period were excluded. E-consult rates at community-based outpatient clinics were included with associated VAMCs. We then stratified PCPs by whether they were high or low users of e-consults (determined by the top and bottom users within each site) and credentials (MD vs nurse practitioner [NP] or physician assistant [PA]). Specialists were sampled based on their rate of use relative to colleagues within their site and the use rate of their division. We sampled division chiefs and individuals who had > 300 total visits and 1 e-consult during the study period. To recruit participants, the primary investigator sent an initial email and 2 reminder emails. The team followed up with respondents to schedule an interview.

Interview guides were designed to elicit rich descriptions of barriers and facilitators to e-consult use (eAppendix available at doi:10.12788/fp.0214). The team used the Practical Robust Implementation and Sustainability Model (PRISM), which considers factors along 6 domains for intervention planning, implementation, and sustainment.¹⁹ Telephone interviews lasted about 20 minutes and were conducted between September 2017 and March 2018. Interviews were recorded and transcribed verbatim.

Analysis

The team used an iterative, team-based, inductive/deductive approach to conventional content analysis.^20,21 Initial code categories were created so that we could identify e-consult best practices—facilitators of e-consult that were recommended by both PCPs and specialists. Inductive codes or labels applied to identify meaningful quotations, phrases, or key terms were used to identify emergent ideas and were added throughout coding after discussion among team members. Consensus was reached using a team-based approach.²¹ Four analysts independently coded the same 3 transcripts and met to discuss points of divergence and convergence. Analyses continued with emergent themes, categories, and conclusions. Atlas.ti. v.7 was used for coding and data management.²²

 

Results

We conducted 34 interviews with clinicians (Table 1) from 13 VISNs. Four best-practice themes emerged among both PCPs and specialists, including that e-consults (1) are best suited for certain clinical questions and patients; (2) require relevant background information from requesting clinicians and clear recommendations from responding clinicians; (3) are a novel opportunity to provide efficient, transparent care; and (4) may not be fully adopted due to low awareness. Supporting quotations for the following findings are provided in Table 2.

Specific Clinical Questions and Patients

PCPs described specific patients and questions for which they most frequently used e-consults, such as for medication changes (Q1), determining treatment steps (Q2,3), and or clarifying laboratory or imaging findings. PCPs frequently used e-consults for patients who did not require a physical examination or when specialists could make recommendations without seeing patients face-to-face (Q3). An important use of e-consults described by PCPs was for treating conditions they could manage within primary care if additional guidance were available (Q4). Several PCPs and specialists also noted that e-consults were particularly useful for patients who were unable to travel or did not want face-to-face appointments (Q5). Notably, PCPs and specialists mentioned situations for which e-consults were inappropriate, including when a detailed history or physical examination was needed, or if a complex condition was suspected (Q6).

Background Data and Clear Recommendations

Participants described necessary data that should be included in high-quality e-consults. Specialists voiced frustration in time-consuming chart reviews that were often necessary when these data were not provided by the requestor. In some cases, specialists were unable to access necessary EHR data, which delayed responses (Q7). PCPs noted that the most useful responses carefully considered the question, used current patient information to determine treatments, provided clear recommendations, and defined who was responsible for next steps (Q8). PCPs and specialists stated that e-consult templates that required relevant information facilitated high-quality e-consults. Neither wanted to waste the other clinician's time (Q8).

A Novel Opportunity

Many PCPs felt that e-consults improved communication (eg, efficiency, response time), established new communication between clinicians, and reduced patients’ appointment burden (Q10, Q11). Many specialists felt that e-consults improved documentation of communication between clinicians and increased transparency of clinical decisions (Q12). Additionally, many specialists mentioned that e-consults capture previously informal curbside consults, enabling them to receive workload credit (Q13).

Lack of Awareness

Some noted that the biggest barrier to e-consults was not being aware of them generally, or which specialties offer e-consults (Q14). One PCP described e-consults as the best kept secret and found value in sharing the utility of e-consults with colleagues (Q15). All participants, including those who did not frequently use e-consults, felt that e-consults improved the quality of care by providing more timely care or better answers to clinical questions (Q16). Several practitioners also felt that e-consults increased access to specialty care. For example, specialists reported that e-consults enabled them to better manage patient load by using e-consults to answer relatively simple questions, reserving face-to-face consults for more complex patients (Q17).

Discussion

The objective of this study was to identify potential best practices for e-consults that may help increase their quality and use within the VA. We built on prior studies that offered insights on PCP and specialists’ overall satisfaction with e-consult by identifying several themes relevant to the further adoption of e-consults in the VA and elsewhere without a face-to-face visit.^{8,13,14,16–18} Future work may be beneficial in identifying whether the study themes identified can explain variation in e-consult use or whether addressing these factors might lead to increased or higher quality e-consult use. We are unaware of any qualitative study of comparable scale in a different health care system. Further, this is the first study to assess perspectives on e-consults among those who initiate and respond to them within the same health care system. Perhaps the most important finding from this study is that e-consults are generally viewed favorably, which is a necessary leverage point to increase their adoption within the system.

 

Clinicians reported several benefits to e-consults, including timely responses to clinical questions, efficient communication, allow for documentation of specialist recommendations, and help capture workload. These benefits are consistent with prior literature that indicates both PCPs and specialists in the VA and other health care systems feel that e-consults improves communication, decreases unnecessary visits, and improves quality of care.^1,14,17,18 In particular, clinicians reported that e-consults improve their practice efficiency and efficacy. This is of critical importance given the pressures of providing timely access to primary and specialty care within the VA. Interestingly, many VA practitioners were unaware which specialties offered e-consults within their facilities, reflecting previous work showing that PCPs are often unaware of e-consult options.¹⁶ This may partially explain variation in e-consult use. Increasing awareness and educating clinicians on the benefits of e-consults may help promote use among non- and low users.

A common theme reported by both groups was the importance of providing necessary information within e-consult questions and responses. Specialists felt there was a need to ensure that PCPs provide relevant and patient-specific information that would enable them to efficiently and accurately answer questions without the need for extensive EHR review. This reflects previous work showing that specialists are often unable to respond to e-consult requests because they do not contain sufficient information.²² PCPs described a need to ensure that specialists’ responses included information that was detailed enough to make clinical decisions without the need for a reconsult. This highlights a common challenge to medical consultation, in that necessary or relevant information may not be apparent to all clinicians. To address this, there may be a role in developing enhanced, flexible templating that elicits necessary patient-specific information. Such a template may automatically pull relevant data from the EHR and prompt clinicians to provide important information. We did not assess how perspectives of templates varied, and further work could help define precisely what constitutes an effective template, including how it should capture appropriate patient data and how this impacts acceptability or use of e-consults generally. Collaboratively developed service agreements and e-consult templates could help guide PCPs and specialists to engage in efficient communication.

Another theme among both groups was that e-consult is most appropriate within specific clinical scenarios. Examples included review of laboratory results, questions about medication changes, or for patients who were reluctant to travel to appointments. Identifying and promoting specific opportunities for e-consults may help increase their use and align e-consult practices with scenarios that are likely to provide the most benefit to patients. For example, it could be helpful to understand the distance patients must travel for specialty care. Providing that information during clinical encounters could trigger clinicians to consider e-consults as an option. Future work might aim to identify clinical scenarios that clinicians feel are not well suited for e-consults and determine how to adapt them for those scenarios.

Limitations

Generalizability of these findings is limited given the qualitative study design. Participants’ descriptions of experiences with e-consults reflect the experiences of clinicians in the VA and may not reflect clinicians in other settings. We also interviewed a sample of clinicians who were already using e-consults. Important information could be learned from future work with those who have not yet adopted e-consult procedures or adopted and abandoned them.

Conclusions

E-consult is perceived as beneficial by VA PCPs and specialists. Participants suggested using e-consults for appropriate questions or patients and including necessary information and next steps in both the initial e-consult and response. Finding ways to facilitate e-consults with these suggestions in mind may increase delivery of high-quality e-consults. Future work could compare the findings of this work to similar work assessing clinicians perceptions of e-consults outside of the VA.

Electronic consultation (e-consult) is designed to increase access to specialty care by facilitating communication between primary care and specialty clinicians without the need for outpatient face-to-face encounters.^1–4 In 2011, the US Department of Veterans Affairs (VA) implemented an e-consult program as a component of its overall strategy to increase access to specialty services, reduce costs of care, and reduce appointment travel burden on patients.

E-consult has substantially increased within the VA since its implementation.^5,6 Consistent with limited evaluations from other health care systems, evaluations of the VA e-consult program demonstrated reduced costs, reduced travel time for patients, and improved access to specialty care.^2,5–11 However, there is wide variation in e-consult use across VA specialties, facilities, and regions.^5,6,12,13 For example, hematology, preoperative evaluation, neurosurgery, endocrinology, and infectious diseases use e-consults more frequently when compared with in-person consults in the VA.⁶ Reasons for this variation or specific barriers and facilitators of using e-consults have not been described.

Prior qualitative studies report that primary care practitioners (PCPs) describe e-consults as convenient, educational, beneficial for patient care, and useful for improving patient access to specialty care.^8,14,15 One study identified limited PCP knowledge of e-consults as a barrier to use.¹⁶ Specialists have reported that e-consult improves clinical communication, but increases their workload.^1,14,17,18 These studies did not assess perspectives from both clinicians who initiate e-consults and those who respond to them. This is the first qualitative study to assess e-consult perceptions from perspectives of both PCPs and specialists among a large, national sample of VA clinicians who use e-consults. The objective of this study was to understand perspectives of e-consults between PCPs and specialists that may be relevant to increasing adoption in the VA.

Methods

The team (CL, ML, PG, 2 analysts under the guidance of GS and JS and support from RRK, and a biostatistician) conducted semistructured interviews with PCPs, specialists, and specialty division leaders who were employed by VA in 2016 and 2017. Specialties of interest were identified by the VA Office of Specialty Care and included cardiology, endocrinology, gastroenterology, and hematology.

E-Consult Procedures

Within the VA, the specific procedures used to initiate, triage and manage e-consults are coordinated at VA medical centers (VAMCs) and at the Veterans Integrated Service Network (VISN) regional level. E-consult can be requested by any clinician. Generally, e-consults are initiated by PCPs through standardized, specialty-specific templates. Recipients, typically specialists, respond by answering questions, suggesting additional testing and evaluation, or requesting an in-person visit. Communication is documented in the patient’s electronic health record (EHR). Specialists receive different levels of workload credit for responding to e-consults similar to a relative value unit reimbursement model. Training in the use of e-consults is available to practitioners but may vary at local and regional levels.

Recruitment

Our sample included PCPs, specialists, and specialty care division leaders. We first quantified e-consult rates (e-consults per 100 patient visits) between July 2016 and June 2017 at VA facilities within primary care and the 4 priority specialties and identified the 30 sites with the highest e-consult rates and 30 sites with the lowest e-consult rates. Sites with < 500 total visits, < 3 specialties, or without any e-consult visit during the study period were excluded. E-consult rates at community-based outpatient clinics were included with associated VAMCs. We then stratified PCPs by whether they were high or low users of e-consults (determined by the top and bottom users within each site) and credentials (MD vs nurse practitioner [NP] or physician assistant [PA]). Specialists were sampled based on their rate of use relative to colleagues within their site and the use rate of their division. We sampled division chiefs and individuals who had > 300 total visits and 1 e-consult during the study period. To recruit participants, the primary investigator sent an initial email and 2 reminder emails. The team followed up with respondents to schedule an interview.

Interview guides were designed to elicit rich descriptions of barriers and facilitators to e-consult use (eAppendix available at doi:10.12788/fp.0214). The team used the Practical Robust Implementation and Sustainability Model (PRISM), which considers factors along 6 domains for intervention planning, implementation, and sustainment.¹⁹ Telephone interviews lasted about 20 minutes and were conducted between September 2017 and March 2018. Interviews were recorded and transcribed verbatim.

Analysis

The team used an iterative, team-based, inductive/deductive approach to conventional content analysis.^20,21 Initial code categories were created so that we could identify e-consult best practices—facilitators of e-consult that were recommended by both PCPs and specialists. Inductive codes or labels applied to identify meaningful quotations, phrases, or key terms were used to identify emergent ideas and were added throughout coding after discussion among team members. Consensus was reached using a team-based approach.²¹ Four analysts independently coded the same 3 transcripts and met to discuss points of divergence and convergence. Analyses continued with emergent themes, categories, and conclusions. Atlas.ti. v.7 was used for coding and data management.²²

 

Results

We conducted 34 interviews with clinicians (Table 1) from 13 VISNs. Four best-practice themes emerged among both PCPs and specialists, including that e-consults (1) are best suited for certain clinical questions and patients; (2) require relevant background information from requesting clinicians and clear recommendations from responding clinicians; (3) are a novel opportunity to provide efficient, transparent care; and (4) may not be fully adopted due to low awareness. Supporting quotations for the following findings are provided in Table 2.

Specific Clinical Questions and Patients

PCPs described specific patients and questions for which they most frequently used e-consults, such as for medication changes (Q1), determining treatment steps (Q2,3), and or clarifying laboratory or imaging findings. PCPs frequently used e-consults for patients who did not require a physical examination or when specialists could make recommendations without seeing patients face-to-face (Q3). An important use of e-consults described by PCPs was for treating conditions they could manage within primary care if additional guidance were available (Q4). Several PCPs and specialists also noted that e-consults were particularly useful for patients who were unable to travel or did not want face-to-face appointments (Q5). Notably, PCPs and specialists mentioned situations for which e-consults were inappropriate, including when a detailed history or physical examination was needed, or if a complex condition was suspected (Q6).

Background Data and Clear Recommendations

Participants described necessary data that should be included in high-quality e-consults. Specialists voiced frustration in time-consuming chart reviews that were often necessary when these data were not provided by the requestor. In some cases, specialists were unable to access necessary EHR data, which delayed responses (Q7). PCPs noted that the most useful responses carefully considered the question, used current patient information to determine treatments, provided clear recommendations, and defined who was responsible for next steps (Q8). PCPs and specialists stated that e-consult templates that required relevant information facilitated high-quality e-consults. Neither wanted to waste the other clinician's time (Q8).

A Novel Opportunity

Many PCPs felt that e-consults improved communication (eg, efficiency, response time), established new communication between clinicians, and reduced patients’ appointment burden (Q10, Q11). Many specialists felt that e-consults improved documentation of communication between clinicians and increased transparency of clinical decisions (Q12). Additionally, many specialists mentioned that e-consults capture previously informal curbside consults, enabling them to receive workload credit (Q13).

Lack of Awareness

Some noted that the biggest barrier to e-consults was not being aware of them generally, or which specialties offer e-consults (Q14). One PCP described e-consults as the best kept secret and found value in sharing the utility of e-consults with colleagues (Q15). All participants, including those who did not frequently use e-consults, felt that e-consults improved the quality of care by providing more timely care or better answers to clinical questions (Q16). Several practitioners also felt that e-consults increased access to specialty care. For example, specialists reported that e-consults enabled them to better manage patient load by using e-consults to answer relatively simple questions, reserving face-to-face consults for more complex patients (Q17).

Discussion

The objective of this study was to identify potential best practices for e-consults that may help increase their quality and use within the VA. We built on prior studies that offered insights on PCP and specialists’ overall satisfaction with e-consult by identifying several themes relevant to the further adoption of e-consults in the VA and elsewhere without a face-to-face visit.^{8,13,14,16–18} Future work may be beneficial in identifying whether the study themes identified can explain variation in e-consult use or whether addressing these factors might lead to increased or higher quality e-consult use. We are unaware of any qualitative study of comparable scale in a different health care system. Further, this is the first study to assess perspectives on e-consults among those who initiate and respond to them within the same health care system. Perhaps the most important finding from this study is that e-consults are generally viewed favorably, which is a necessary leverage point to increase their adoption within the system.

 

Clinicians reported several benefits to e-consults, including timely responses to clinical questions, efficient communication, allow for documentation of specialist recommendations, and help capture workload. These benefits are consistent with prior literature that indicates both PCPs and specialists in the VA and other health care systems feel that e-consults improves communication, decreases unnecessary visits, and improves quality of care.^1,14,17,18 In particular, clinicians reported that e-consults improve their practice efficiency and efficacy. This is of critical importance given the pressures of providing timely access to primary and specialty care within the VA. Interestingly, many VA practitioners were unaware which specialties offered e-consults within their facilities, reflecting previous work showing that PCPs are often unaware of e-consult options.¹⁶ This may partially explain variation in e-consult use. Increasing awareness and educating clinicians on the benefits of e-consults may help promote use among non- and low users.

A common theme reported by both groups was the importance of providing necessary information within e-consult questions and responses. Specialists felt there was a need to ensure that PCPs provide relevant and patient-specific information that would enable them to efficiently and accurately answer questions without the need for extensive EHR review. This reflects previous work showing that specialists are often unable to respond to e-consult requests because they do not contain sufficient information.²² PCPs described a need to ensure that specialists’ responses included information that was detailed enough to make clinical decisions without the need for a reconsult. This highlights a common challenge to medical consultation, in that necessary or relevant information may not be apparent to all clinicians. To address this, there may be a role in developing enhanced, flexible templating that elicits necessary patient-specific information. Such a template may automatically pull relevant data from the EHR and prompt clinicians to provide important information. We did not assess how perspectives of templates varied, and further work could help define precisely what constitutes an effective template, including how it should capture appropriate patient data and how this impacts acceptability or use of e-consults generally. Collaboratively developed service agreements and e-consult templates could help guide PCPs and specialists to engage in efficient communication.

Another theme among both groups was that e-consult is most appropriate within specific clinical scenarios. Examples included review of laboratory results, questions about medication changes, or for patients who were reluctant to travel to appointments. Identifying and promoting specific opportunities for e-consults may help increase their use and align e-consult practices with scenarios that are likely to provide the most benefit to patients. For example, it could be helpful to understand the distance patients must travel for specialty care. Providing that information during clinical encounters could trigger clinicians to consider e-consults as an option. Future work might aim to identify clinical scenarios that clinicians feel are not well suited for e-consults and determine how to adapt them for those scenarios.

Limitations

Generalizability of these findings is limited given the qualitative study design. Participants’ descriptions of experiences with e-consults reflect the experiences of clinicians in the VA and may not reflect clinicians in other settings. We also interviewed a sample of clinicians who were already using e-consults. Important information could be learned from future work with those who have not yet adopted e-consult procedures or adopted and abandoned them.

Conclusions

E-consult is perceived as beneficial by VA PCPs and specialists. Participants suggested using e-consults for appropriate questions or patients and including necessary information and next steps in both the initial e-consult and response. Finding ways to facilitate e-consults with these suggestions in mind may increase delivery of high-quality e-consults. Future work could compare the findings of this work to similar work assessing clinicians perceptions of e-consults outside of the VA.

Vitiligo is a chronic dermatologic condition that negatively affects quality of life (QOL), with substantial burden on the psychosocial well-being of patients.¹ There is no cure, and current treatment modalities are aimed at controlling the chronic relapsing condition.^1-3 Despite topical and cosmetic treatments for stabilization and repigmentation, vitiligo remains unpredictable.³

All genders, races, ethnicities, and socioeconomic classes are equally affected.⁴ The underlying etiology of vitiligo remains unknown to a great extent and is more poorly understood by the general public compared with other skin diseases (eg, acne).⁵ Patients with vitiligo experience social withdrawal, decreased sense of self-esteem, anxiety, depression, and suicidal ideation.^5,6 Stigmatization has the greatest impact on QOL, with strong correlations between avoidance behaviors and lesion concealment.^6-8 Although the condition is especially disfiguring for darker skin types, lighter skin types also are substantially affected, with similar overall self-reported stress.^6,7

Individuals with chronic illnesses such as vitiligo turn to online communities for health information and social support, commiserating with others who have the same condition.^9,10 Online forums are platforms for asynchronous peer-to-peer exchange of disease-related information for better management of long-term disease.¹¹ Moreover, of all available internet resources, online forum posts are the most commonly accessed source of information (91%) for patients following visits with their doctors.¹²

Qualitative research involving chronic skin conditions and the information exchanged in online forums has been conducted for patients with acne, psoriasis, and atopic dermatitis, but not for patients with vitiligo.^13-16 Although online questionnaires have been administered to patients with vitiligo, the content within online forums is not well characterized.^2,17

The purpose of this qualitative study was to evaluate the online content exchanged by individuals with vitiligo to better understand the general attitudes and help-seeking behaviors in online forums.

Methods

Study Design—This qualitative study sought to investigate health beliefs and messages about vitiligo posted by users in US-based online discussion forums. An interpretive research paradigm was utilized so that all content collected in online forums were the views expressed by individuals.^18-20 An integrated approach was used in the development of the coding manual, with pre-established major themes and subthemes as a guiding framework.^16,21,22 We adhered to an inductive grounded method by means of de novo line-by-line coding, such that we had flexibility for new subthemes to emerge throughout the duration of the entire coding process.²³

Individual posts and subsequent replies embedded within public online forums were used as the collected data source. Google was utilized as the primary search engine to identify forums pertaining to vitiligo, as 80% of US adults with chronic disease report that their inquiries for health information start with Google, Bing, or Yahoo.²⁴ The institutional review board at the Wake Forest School of Medicine (Winston-Salem, North Carolina) granted approval of the study (IRB00063073). Online forums were considered “property” of the public domain and were accessible to all, eliminating the need for written informed consent.^24-26

 

Search Criteria—We conducted our forum search in February 2020 with a systematic approach using predetermined phrases—online forum vitiligo support, vitiligo online message board, and vitiligo forums—which yielded more than 358,171 total results (eTable 1). Threads were identified in chronological order (from newest to oldest) based on how they appeared during each internet search, and all Google results for the respective search phrases were reviewed. Dates of selected threads ranged from 2005 to 2020. Only sites with US domains were included. Posts that either included views and understandings of vitiligo or belonged to a thread that contained a vitiligo discussion were deemed relevant for inclusion. Forums were excluded if registration or means of payment was required to view posts, if there were fewer than 2 user replies to a thread, if threads contained patient photographs, or if no posts had been made in the last 2 years (rendering the thread inactive). No social media platforms, such as Facebook, or formal online platforms, such as MyVitiligoTeam, were included in the search. A no-fee-for-access was chosen for this study, as the majority of those with a chronic condition who encounter a required paywall find the information elsewhere.²⁵

Data Analysis—A total of 39 online forums were deemed relevant to the topic of vitiligo; 9 of them met inclusion criteria (eTable 2). The messages within the forums were copied verbatim into a password-encrypted text document, and usernames in the threads were de-identified, ensuring user confidentiality.

An inductive thematic analysis was utilized to explore the views and beliefs of online forum users discussing vitiligo. One author (M.B.G.) read the extracted message threads, developed an initial codebook, and established a finalized version with the agreement of another author (A.M.B.)(eTable 3). The forums were independently coded (M.B.G. and A.M.B.) in a line-by-line manner according to the codebook. Discrepancies were documented and resolved. Data saturation was adequately achieved, such that no new themes emerged during the iterative coding process. NVivo was used for qualitative analysis.

Results

Nine forums met inclusion criteria, comprising 105 pages of text. There were 61 total discussion threads, with 382 anonymous contributing users. Most users initiated a thread by posting either a question, an advice statement, or a request for help. The psychosocial impact of the disease permeated multiple domains,including personal relationships and daily life. Several threads discussed treatment, including effective camouflage and makeup, as well as peer validation of physician-prescribed treatments, along with threads dedicated to “cures” or homeopathy regimens. In several instances, commercial product endorsement, testimonials, and marketing links were reposted by the same user multiple times.

Inductive thematic analysis highlighted diverse themes and subthemes related to the beliefs and perspectives of users with vitiligo or with relatives or friends with vitiligo: psychosocial impact, disease management and camouflage/concealment, alternative medicine/homeopathy/cures, interactions with the public and health care providers, and skin tone and race. Quotes from individuals were included to demonstrate themes and subthemes.

Psychosocial Impact: QOL, Sources of Support, and Coping—There was a broad range of comments on how patients cope with and view their vitiligo. Some individuals felt vitiligo made them special, and others were at peace with and accepted their condition. In contrast, others reported the disease had devastated them and interfered with relationships. Individuals shared their stories of grief and hardships through childhood and adulthood and their concerns, especially on affected visible areas or the potential for disease progression. Users were vocal about how vitiligo affected their daily routines and lives, sharing how they felt uncomfortable outside the home, no longer engaged in swimming or exposing their legs, and preferred to stay inside instead. Some users adopted a “tough love” approach to coping, sharing how they have learned to either embrace their vitiligo or “live with it.” Some examples include:

“My best advice is go with the flow, vitiligo is not the worst thing that can happen.”

 

“I hate my life with vitiligo yet really I feel so selfish that there is much worse suffering in the world than a few white patches.”

Other advice was very practical:

“I hope it isn’t vanity that is tearing you apart because that is only skin deep. Make a fashion statement with hats.”

Some users acknowledged and adopted the mantra that vitiligo is not a somatic condition or “physical ailment,” while others emphasized its pervasive psychological burden:

“I still deal with this psychologically . . . You must keep a positive attitude and frame of mind . . . Vitiligo will not kill you, but you do need to stay strong and keep your head up emotionally.”

“I am just really thankful that I have a disease that will not kill me or that has [not] affected me physically at all. I consider myself lucky.”

Disease Management: Treatment, Vitiligo Course, Advice-Seeking, Camouflage—The range of information discussed for treatment was highly variable. There were many accounts in which users advised others to seek professional help, namely that of a dermatologist, for a formal assessment. Many expressed frustrations with treatments and their ineffectiveness, to which the majority of users said to consult with a professional and to remain patient and hopeful/optimistic:

“The best thing to do would be to take an appointment with a dermatologist and have the discoloration checked out. That’s the only way to know whether it is vitiligo or not.”

“My way of dealing with it is to gain control by camouflage.”

“The calming effect of being in control of my vitiligo, whether with concealers, self-tan or anything else, has stopped my feelings of despair.”

 

Beliefs on Alternative Medicine: Homeopathy and Alternative Regimens—Although some threads started with a post asking for the best treatments, others initiated a discussion by posting “best herbal treatments for cure” or “how to cure my vitiligo,” emphasizing the beliefs and wishes for a cure for vitiligo. Alternative therapies that users endorsed included apple cider vinegar, toothpaste, vitamins, and Ayurvedic treatment, among others. Dietary plans were popular, with users claiming success with dietary alterations in stopping and preventing lesion progression. For example, individuals felt that avoidance of sugar, meat, dairy, and citrus fruits or drinks and consumption of only filtered water were crucial to preventing further lesion spread and resulted in their “cure”:

“Don’t eat chocolate, wine (made of grapes), coffee, or tea if you don’t want to have vitiligo or let it get worse. Take Vitamin B, biotin, and nuts for Vitamin E.”

Other dangerous messages pitted treatments by health professionals against beliefs in homeopathy:

“I feel that vitiligo treatment is all in your diet and vitamins. All that medicine and UV lights is a no-no . . .w ith every medicine there is a side effect. The doctors could be healing your vitiligo and severely damaging you inside and out, and you won’t know until years later.”

There was a minor presence of users advising against homeopathy and the associated misinformation and inaccurate claims on curing vitiligo, though this group was small in comparison to the number of users posting outlandish claims on cure:

“There is no cure . . . It’s where your immune system attacks your skin cells causing loss of pigmentation. The skin that has lost the pigmentation can’t be reversed.”

Interactions With the Public and Health Care Providers—Those with vitiligo encounter unique situations in public and in their daily lives. Many of the accounts shared anecdotal stories on how patients have handled the stigma and discrimination faced:

“I have had to face discrimination at school, public places, college, functions, and every new person I have met has asked me this: ‘how did this happen?’”

Those with vitiligo even stated how they wished others would deal with their condition out in public, hoping that others would directly ask what the lesions were instead of the more hurtful staring. There were many stories in which users said others feel vitiligo was contagious or “dirty” and stressed that the condition is not infectious:

“I refer to myself as ‘camo-man’ and reassure people I come into contact with that it is not contagious.”

“Once I was eating at a restaurant . . . and a little girl said to her mom, ‘Look, Mom, that lady doesn’t wash her arms, look how dirty they are.’ That just broke my heart.”

 

Skin Tone and Implications—The belief that vitiligo lesions are less dramatic or less anxiety provoking for individuals with lighter skin was noted by users themselves and by health care providers in certain cases. Skin tone and its impact on QOL was confusing and contentious. Some users with fair skin stated their vitiligo was “less of an annoyance” or “less obvious” compared with individuals with darker complexions. Conversely, other accounts of self-reported White users vehemently stressed the anxieties felt by depigmented lesions, despite being “already white at baseline.”

“Was told by my dermatologist (upon diagnosis) that ‘You’re lucky you’re not African American—it shows up on them much worse. You’re so fair, it doesn’t really matter.’ ”

“You didn’t say what race you are. I could imagine it has a bigger impact if you are anything other than White.”

Comment

Patients Looking for Cures—The general attitude within the forums was uplifting and encouraging, with users detailing how they respond to others in public and sharing their personal perspectives. We found a mix of information regarding disease management and treatment of vitiligo. Overall, there was uncertainty about treatments, with individuals expressing concern that their treatments were ineffective or had failed or that better alternatives would be more suitable for their condition. We found many anecdotal endorsements of homeopathic remedies for vitiligo, with users boasting that their disease had not only been cured but had never returned. Some users completely denounced these statements, while other threads seemed to revolve completely around “cure” discussions with no dissenting voices. The number of discussions related to homeopathy was concerning. Furthermore, there often were no moderators within threads to remove cure-related content, whether commercially endorsed or anecdotal. It is plausible that supplements and vitamins recommended by some physicians may be incorrectly interpreted as a “cure” in online discussions. Our findings are consistent with prior reports that forums are a platform to express dissatisfaction with treatment and the need for additional treatment options.^15,22

Concern Expressed by Health Care Providers—Prior qualitative research has described how patients with chronic dermatologic conditions believe that health care providers minimize patients’ psychological distress.^27,28 We found several accounts in which an individual had explicitly stated their provider had “belittled” the extent and impact of vitiligo when comparing skin phototypes. This suggests either that physicians underestimate the impact of vitiligo on their patients or that physicians are not expressing enough empathic concern about the impact the condition has on those affected.

Cosmetic Aspects of Vitiligo—Few clinical trials have investigated QOL and cosmetic acceptability of treatments as outcome measures.²⁹ We found several instances in which users with vitiligo had reported being dismissed as having a “cosmetic disease,” consistent with other work demonstrating the negative impact on such dismissals.²² Moreover, concealment and camouflage techniques frequently were discussed, demonstrating the relevance of cosmetic management as an important research topic.

Trustworthy Sources of Health Information—Patients still view physicians as trustworthy and a key source of health care information and advice.^30-32 Patients with vitiligo who have been directed to reliable information sources often express gratitude²² and want health professionals to remain an important source in their health information-seeking.³¹ Given the range in information discussed online, it may be valuable to invite patients to share what information they have encountered online.

 

Our study highlights the conflicting health information and advice shared by users in online forums, complicating an already psychologically burdensome condition. Guiding patients to credible, moderated sites and resources that are accurate, understandable, and easy to access may help dispel the conflicting messages and stories discussed in the online community.

Study Strengths and Limitations—Limitations included reporting bias and reliance on self-reported information on the diagnosis and extent of individuals’ vitiligo. Excluding social media websites and platforms from the data collection is a limitation to comprehensively assessing the topic of internet users with vitiligo. Many social media platforms direct patients and their family members to support groups and therefore may have excluded these particular individuals. Social media platforms were excluded from our research owing to the prerequisite of creating user accounts or registering as an online member. Our inclusion criteria were specific to forums that did not require registering or creating an account and were therefore freely accessible to all internet viewers. There is an inherent lack of context present in online forums, preventing data collection on individuals’ demographics and socioeconomic backgrounds. However, anonymity may have allowed individuals to express their thoughts more freely.

An integrated approach, along with our sampling method of online forums not requiring registration, allows for greater transferability and understanding of the health needs of the general public with vitiligo.

Conclusion

Individuals with vitiligo continue to seek peer psychosocial support for the physical and emotional management of their disease. Counseling those with vitiligo about cosmetic concealment options, homeopathy, and treatment scams remains paramount. Directing patients to evidence-based resources, along with providing structured sources of support, may help to improve the psychosocial burden and QOL experienced by patients with vitiligo. Connecting patients with local and national support groups moderated by physicians, such as the Global Vitiligo Foundation (https://globalvitiligofoundation.org/), may provide benefit to patients with vitiligo.

Vitiligo is a chronic dermatologic condition that negatively affects quality of life (QOL), with substantial burden on the psychosocial well-being of patients.¹ There is no cure, and current treatment modalities are aimed at controlling the chronic relapsing condition.^1-3 Despite topical and cosmetic treatments for stabilization and repigmentation, vitiligo remains unpredictable.³

All genders, races, ethnicities, and socioeconomic classes are equally affected.⁴ The underlying etiology of vitiligo remains unknown to a great extent and is more poorly understood by the general public compared with other skin diseases (eg, acne).⁵ Patients with vitiligo experience social withdrawal, decreased sense of self-esteem, anxiety, depression, and suicidal ideation.^5,6 Stigmatization has the greatest impact on QOL, with strong correlations between avoidance behaviors and lesion concealment.^6-8 Although the condition is especially disfiguring for darker skin types, lighter skin types also are substantially affected, with similar overall self-reported stress.^6,7

Individuals with chronic illnesses such as vitiligo turn to online communities for health information and social support, commiserating with others who have the same condition.^9,10 Online forums are platforms for asynchronous peer-to-peer exchange of disease-related information for better management of long-term disease.¹¹ Moreover, of all available internet resources, online forum posts are the most commonly accessed source of information (91%) for patients following visits with their doctors.¹²

Qualitative research involving chronic skin conditions and the information exchanged in online forums has been conducted for patients with acne, psoriasis, and atopic dermatitis, but not for patients with vitiligo.^13-16 Although online questionnaires have been administered to patients with vitiligo, the content within online forums is not well characterized.^2,17

The purpose of this qualitative study was to evaluate the online content exchanged by individuals with vitiligo to better understand the general attitudes and help-seeking behaviors in online forums.

Methods

Study Design—This qualitative study sought to investigate health beliefs and messages about vitiligo posted by users in US-based online discussion forums. An interpretive research paradigm was utilized so that all content collected in online forums were the views expressed by individuals.^18-20 An integrated approach was used in the development of the coding manual, with pre-established major themes and subthemes as a guiding framework.^16,21,22 We adhered to an inductive grounded method by means of de novo line-by-line coding, such that we had flexibility for new subthemes to emerge throughout the duration of the entire coding process.²³

Individual posts and subsequent replies embedded within public online forums were used as the collected data source. Google was utilized as the primary search engine to identify forums pertaining to vitiligo, as 80% of US adults with chronic disease report that their inquiries for health information start with Google, Bing, or Yahoo.²⁴ The institutional review board at the Wake Forest School of Medicine (Winston-Salem, North Carolina) granted approval of the study (IRB00063073). Online forums were considered “property” of the public domain and were accessible to all, eliminating the need for written informed consent.^24-26

 

Search Criteria—We conducted our forum search in February 2020 with a systematic approach using predetermined phrases—online forum vitiligo support, vitiligo online message board, and vitiligo forums—which yielded more than 358,171 total results (eTable 1). Threads were identified in chronological order (from newest to oldest) based on how they appeared during each internet search, and all Google results for the respective search phrases were reviewed. Dates of selected threads ranged from 2005 to 2020. Only sites with US domains were included. Posts that either included views and understandings of vitiligo or belonged to a thread that contained a vitiligo discussion were deemed relevant for inclusion. Forums were excluded if registration or means of payment was required to view posts, if there were fewer than 2 user replies to a thread, if threads contained patient photographs, or if no posts had been made in the last 2 years (rendering the thread inactive). No social media platforms, such as Facebook, or formal online platforms, such as MyVitiligoTeam, were included in the search. A no-fee-for-access was chosen for this study, as the majority of those with a chronic condition who encounter a required paywall find the information elsewhere.²⁵

Data Analysis—A total of 39 online forums were deemed relevant to the topic of vitiligo; 9 of them met inclusion criteria (eTable 2). The messages within the forums were copied verbatim into a password-encrypted text document, and usernames in the threads were de-identified, ensuring user confidentiality.

An inductive thematic analysis was utilized to explore the views and beliefs of online forum users discussing vitiligo. One author (M.B.G.) read the extracted message threads, developed an initial codebook, and established a finalized version with the agreement of another author (A.M.B.)(eTable 3). The forums were independently coded (M.B.G. and A.M.B.) in a line-by-line manner according to the codebook. Discrepancies were documented and resolved. Data saturation was adequately achieved, such that no new themes emerged during the iterative coding process. NVivo was used for qualitative analysis.

Results

Nine forums met inclusion criteria, comprising 105 pages of text. There were 61 total discussion threads, with 382 anonymous contributing users. Most users initiated a thread by posting either a question, an advice statement, or a request for help. The psychosocial impact of the disease permeated multiple domains,including personal relationships and daily life. Several threads discussed treatment, including effective camouflage and makeup, as well as peer validation of physician-prescribed treatments, along with threads dedicated to “cures” or homeopathy regimens. In several instances, commercial product endorsement, testimonials, and marketing links were reposted by the same user multiple times.

Inductive thematic analysis highlighted diverse themes and subthemes related to the beliefs and perspectives of users with vitiligo or with relatives or friends with vitiligo: psychosocial impact, disease management and camouflage/concealment, alternative medicine/homeopathy/cures, interactions with the public and health care providers, and skin tone and race. Quotes from individuals were included to demonstrate themes and subthemes.

Psychosocial Impact: QOL, Sources of Support, and Coping—There was a broad range of comments on how patients cope with and view their vitiligo. Some individuals felt vitiligo made them special, and others were at peace with and accepted their condition. In contrast, others reported the disease had devastated them and interfered with relationships. Individuals shared their stories of grief and hardships through childhood and adulthood and their concerns, especially on affected visible areas or the potential for disease progression. Users were vocal about how vitiligo affected their daily routines and lives, sharing how they felt uncomfortable outside the home, no longer engaged in swimming or exposing their legs, and preferred to stay inside instead. Some users adopted a “tough love” approach to coping, sharing how they have learned to either embrace their vitiligo or “live with it.” Some examples include:

“My best advice is go with the flow, vitiligo is not the worst thing that can happen.”

 

“I hate my life with vitiligo yet really I feel so selfish that there is much worse suffering in the world than a few white patches.”

Other advice was very practical:

“I hope it isn’t vanity that is tearing you apart because that is only skin deep. Make a fashion statement with hats.”

Some users acknowledged and adopted the mantra that vitiligo is not a somatic condition or “physical ailment,” while others emphasized its pervasive psychological burden:

“I still deal with this psychologically . . . You must keep a positive attitude and frame of mind . . . Vitiligo will not kill you, but you do need to stay strong and keep your head up emotionally.”

“I am just really thankful that I have a disease that will not kill me or that has [not] affected me physically at all. I consider myself lucky.”

Disease Management: Treatment, Vitiligo Course, Advice-Seeking, Camouflage—The range of information discussed for treatment was highly variable. There were many accounts in which users advised others to seek professional help, namely that of a dermatologist, for a formal assessment. Many expressed frustrations with treatments and their ineffectiveness, to which the majority of users said to consult with a professional and to remain patient and hopeful/optimistic:

“The best thing to do would be to take an appointment with a dermatologist and have the discoloration checked out. That’s the only way to know whether it is vitiligo or not.”

“My way of dealing with it is to gain control by camouflage.”

“The calming effect of being in control of my vitiligo, whether with concealers, self-tan or anything else, has stopped my feelings of despair.”

 

Beliefs on Alternative Medicine: Homeopathy and Alternative Regimens—Although some threads started with a post asking for the best treatments, others initiated a discussion by posting “best herbal treatments for cure” or “how to cure my vitiligo,” emphasizing the beliefs and wishes for a cure for vitiligo. Alternative therapies that users endorsed included apple cider vinegar, toothpaste, vitamins, and Ayurvedic treatment, among others. Dietary plans were popular, with users claiming success with dietary alterations in stopping and preventing lesion progression. For example, individuals felt that avoidance of sugar, meat, dairy, and citrus fruits or drinks and consumption of only filtered water were crucial to preventing further lesion spread and resulted in their “cure”:

“Don’t eat chocolate, wine (made of grapes), coffee, or tea if you don’t want to have vitiligo or let it get worse. Take Vitamin B, biotin, and nuts for Vitamin E.”

Other dangerous messages pitted treatments by health professionals against beliefs in homeopathy:

“I feel that vitiligo treatment is all in your diet and vitamins. All that medicine and UV lights is a no-no . . .w ith every medicine there is a side effect. The doctors could be healing your vitiligo and severely damaging you inside and out, and you won’t know until years later.”

There was a minor presence of users advising against homeopathy and the associated misinformation and inaccurate claims on curing vitiligo, though this group was small in comparison to the number of users posting outlandish claims on cure:

“There is no cure . . . It’s where your immune system attacks your skin cells causing loss of pigmentation. The skin that has lost the pigmentation can’t be reversed.”

Interactions With the Public and Health Care Providers—Those with vitiligo encounter unique situations in public and in their daily lives. Many of the accounts shared anecdotal stories on how patients have handled the stigma and discrimination faced:

“I have had to face discrimination at school, public places, college, functions, and every new person I have met has asked me this: ‘how did this happen?’”

Those with vitiligo even stated how they wished others would deal with their condition out in public, hoping that others would directly ask what the lesions were instead of the more hurtful staring. There were many stories in which users said others feel vitiligo was contagious or “dirty” and stressed that the condition is not infectious:

“I refer to myself as ‘camo-man’ and reassure people I come into contact with that it is not contagious.”

“Once I was eating at a restaurant . . . and a little girl said to her mom, ‘Look, Mom, that lady doesn’t wash her arms, look how dirty they are.’ That just broke my heart.”

 

Skin Tone and Implications—The belief that vitiligo lesions are less dramatic or less anxiety provoking for individuals with lighter skin was noted by users themselves and by health care providers in certain cases. Skin tone and its impact on QOL was confusing and contentious. Some users with fair skin stated their vitiligo was “less of an annoyance” or “less obvious” compared with individuals with darker complexions. Conversely, other accounts of self-reported White users vehemently stressed the anxieties felt by depigmented lesions, despite being “already white at baseline.”

“Was told by my dermatologist (upon diagnosis) that ‘You’re lucky you’re not African American—it shows up on them much worse. You’re so fair, it doesn’t really matter.’ ”

“You didn’t say what race you are. I could imagine it has a bigger impact if you are anything other than White.”

Comment

Patients Looking for Cures—The general attitude within the forums was uplifting and encouraging, with users detailing how they respond to others in public and sharing their personal perspectives. We found a mix of information regarding disease management and treatment of vitiligo. Overall, there was uncertainty about treatments, with individuals expressing concern that their treatments were ineffective or had failed or that better alternatives would be more suitable for their condition. We found many anecdotal endorsements of homeopathic remedies for vitiligo, with users boasting that their disease had not only been cured but had never returned. Some users completely denounced these statements, while other threads seemed to revolve completely around “cure” discussions with no dissenting voices. The number of discussions related to homeopathy was concerning. Furthermore, there often were no moderators within threads to remove cure-related content, whether commercially endorsed or anecdotal. It is plausible that supplements and vitamins recommended by some physicians may be incorrectly interpreted as a “cure” in online discussions. Our findings are consistent with prior reports that forums are a platform to express dissatisfaction with treatment and the need for additional treatment options.^15,22

Concern Expressed by Health Care Providers—Prior qualitative research has described how patients with chronic dermatologic conditions believe that health care providers minimize patients’ psychological distress.^27,28 We found several accounts in which an individual had explicitly stated their provider had “belittled” the extent and impact of vitiligo when comparing skin phototypes. This suggests either that physicians underestimate the impact of vitiligo on their patients or that physicians are not expressing enough empathic concern about the impact the condition has on those affected.

Cosmetic Aspects of Vitiligo—Few clinical trials have investigated QOL and cosmetic acceptability of treatments as outcome measures.²⁹ We found several instances in which users with vitiligo had reported being dismissed as having a “cosmetic disease,” consistent with other work demonstrating the negative impact on such dismissals.²² Moreover, concealment and camouflage techniques frequently were discussed, demonstrating the relevance of cosmetic management as an important research topic.

Trustworthy Sources of Health Information—Patients still view physicians as trustworthy and a key source of health care information and advice.^30-32 Patients with vitiligo who have been directed to reliable information sources often express gratitude²² and want health professionals to remain an important source in their health information-seeking.³¹ Given the range in information discussed online, it may be valuable to invite patients to share what information they have encountered online.

 

Our study highlights the conflicting health information and advice shared by users in online forums, complicating an already psychologically burdensome condition. Guiding patients to credible, moderated sites and resources that are accurate, understandable, and easy to access may help dispel the conflicting messages and stories discussed in the online community.

Study Strengths and Limitations—Limitations included reporting bias and reliance on self-reported information on the diagnosis and extent of individuals’ vitiligo. Excluding social media websites and platforms from the data collection is a limitation to comprehensively assessing the topic of internet users with vitiligo. Many social media platforms direct patients and their family members to support groups and therefore may have excluded these particular individuals. Social media platforms were excluded from our research owing to the prerequisite of creating user accounts or registering as an online member. Our inclusion criteria were specific to forums that did not require registering or creating an account and were therefore freely accessible to all internet viewers. There is an inherent lack of context present in online forums, preventing data collection on individuals’ demographics and socioeconomic backgrounds. However, anonymity may have allowed individuals to express their thoughts more freely.

An integrated approach, along with our sampling method of online forums not requiring registration, allows for greater transferability and understanding of the health needs of the general public with vitiligo.

Conclusion

Individuals with vitiligo continue to seek peer psychosocial support for the physical and emotional management of their disease. Counseling those with vitiligo about cosmetic concealment options, homeopathy, and treatment scams remains paramount. Directing patients to evidence-based resources, along with providing structured sources of support, may help to improve the psychosocial burden and QOL experienced by patients with vitiligo. Connecting patients with local and national support groups moderated by physicians, such as the Global Vitiligo Foundation (https://globalvitiligofoundation.org/), may provide benefit to patients with vitiligo.

Vitiligo is a chronic dermatologic condition that negatively affects quality of life (QOL), with substantial burden on the psychosocial well-being of patients.¹ There is no cure, and current treatment modalities are aimed at controlling the chronic relapsing condition.^1-3 Despite topical and cosmetic treatments for stabilization and repigmentation, vitiligo remains unpredictable.³

All genders, races, ethnicities, and socioeconomic classes are equally affected.⁴ The underlying etiology of vitiligo remains unknown to a great extent and is more poorly understood by the general public compared with other skin diseases (eg, acne).⁵ Patients with vitiligo experience social withdrawal, decreased sense of self-esteem, anxiety, depression, and suicidal ideation.^5,6 Stigmatization has the greatest impact on QOL, with strong correlations between avoidance behaviors and lesion concealment.^6-8 Although the condition is especially disfiguring for darker skin types, lighter skin types also are substantially affected, with similar overall self-reported stress.^6,7

Individuals with chronic illnesses such as vitiligo turn to online communities for health information and social support, commiserating with others who have the same condition.^9,10 Online forums are platforms for asynchronous peer-to-peer exchange of disease-related information for better management of long-term disease.¹¹ Moreover, of all available internet resources, online forum posts are the most commonly accessed source of information (91%) for patients following visits with their doctors.¹²

Qualitative research involving chronic skin conditions and the information exchanged in online forums has been conducted for patients with acne, psoriasis, and atopic dermatitis, but not for patients with vitiligo.^13-16 Although online questionnaires have been administered to patients with vitiligo, the content within online forums is not well characterized.^2,17

The purpose of this qualitative study was to evaluate the online content exchanged by individuals with vitiligo to better understand the general attitudes and help-seeking behaviors in online forums.

Methods

Study Design—This qualitative study sought to investigate health beliefs and messages about vitiligo posted by users in US-based online discussion forums. An interpretive research paradigm was utilized so that all content collected in online forums were the views expressed by individuals.^18-20 An integrated approach was used in the development of the coding manual, with pre-established major themes and subthemes as a guiding framework.^16,21,22 We adhered to an inductive grounded method by means of de novo line-by-line coding, such that we had flexibility for new subthemes to emerge throughout the duration of the entire coding process.²³

Individual posts and subsequent replies embedded within public online forums were used as the collected data source. Google was utilized as the primary search engine to identify forums pertaining to vitiligo, as 80% of US adults with chronic disease report that their inquiries for health information start with Google, Bing, or Yahoo.²⁴ The institutional review board at the Wake Forest School of Medicine (Winston-Salem, North Carolina) granted approval of the study (IRB00063073). Online forums were considered “property” of the public domain and were accessible to all, eliminating the need for written informed consent.^24-26

 

Search Criteria—We conducted our forum search in February 2020 with a systematic approach using predetermined phrases—online forum vitiligo support, vitiligo online message board, and vitiligo forums—which yielded more than 358,171 total results (eTable 1). Threads were identified in chronological order (from newest to oldest) based on how they appeared during each internet search, and all Google results for the respective search phrases were reviewed. Dates of selected threads ranged from 2005 to 2020. Only sites with US domains were included. Posts that either included views and understandings of vitiligo or belonged to a thread that contained a vitiligo discussion were deemed relevant for inclusion. Forums were excluded if registration or means of payment was required to view posts, if there were fewer than 2 user replies to a thread, if threads contained patient photographs, or if no posts had been made in the last 2 years (rendering the thread inactive). No social media platforms, such as Facebook, or formal online platforms, such as MyVitiligoTeam, were included in the search. A no-fee-for-access was chosen for this study, as the majority of those with a chronic condition who encounter a required paywall find the information elsewhere.²⁵

Data Analysis—A total of 39 online forums were deemed relevant to the topic of vitiligo; 9 of them met inclusion criteria (eTable 2). The messages within the forums were copied verbatim into a password-encrypted text document, and usernames in the threads were de-identified, ensuring user confidentiality.

An inductive thematic analysis was utilized to explore the views and beliefs of online forum users discussing vitiligo. One author (M.B.G.) read the extracted message threads, developed an initial codebook, and established a finalized version with the agreement of another author (A.M.B.)(eTable 3). The forums were independently coded (M.B.G. and A.M.B.) in a line-by-line manner according to the codebook. Discrepancies were documented and resolved. Data saturation was adequately achieved, such that no new themes emerged during the iterative coding process. NVivo was used for qualitative analysis.

Results

Nine forums met inclusion criteria, comprising 105 pages of text. There were 61 total discussion threads, with 382 anonymous contributing users. Most users initiated a thread by posting either a question, an advice statement, or a request for help. The psychosocial impact of the disease permeated multiple domains,including personal relationships and daily life. Several threads discussed treatment, including effective camouflage and makeup, as well as peer validation of physician-prescribed treatments, along with threads dedicated to “cures” or homeopathy regimens. In several instances, commercial product endorsement, testimonials, and marketing links were reposted by the same user multiple times.

Inductive thematic analysis highlighted diverse themes and subthemes related to the beliefs and perspectives of users with vitiligo or with relatives or friends with vitiligo: psychosocial impact, disease management and camouflage/concealment, alternative medicine/homeopathy/cures, interactions with the public and health care providers, and skin tone and race. Quotes from individuals were included to demonstrate themes and subthemes.

Psychosocial Impact: QOL, Sources of Support, and Coping—There was a broad range of comments on how patients cope with and view their vitiligo. Some individuals felt vitiligo made them special, and others were at peace with and accepted their condition. In contrast, others reported the disease had devastated them and interfered with relationships. Individuals shared their stories of grief and hardships through childhood and adulthood and their concerns, especially on affected visible areas or the potential for disease progression. Users were vocal about how vitiligo affected their daily routines and lives, sharing how they felt uncomfortable outside the home, no longer engaged in swimming or exposing their legs, and preferred to stay inside instead. Some users adopted a “tough love” approach to coping, sharing how they have learned to either embrace their vitiligo or “live with it.” Some examples include:

“My best advice is go with the flow, vitiligo is not the worst thing that can happen.”

 

“I hate my life with vitiligo yet really I feel so selfish that there is much worse suffering in the world than a few white patches.”

Other advice was very practical:

“I hope it isn’t vanity that is tearing you apart because that is only skin deep. Make a fashion statement with hats.”

Some users acknowledged and adopted the mantra that vitiligo is not a somatic condition or “physical ailment,” while others emphasized its pervasive psychological burden:

“I still deal with this psychologically . . . You must keep a positive attitude and frame of mind . . . Vitiligo will not kill you, but you do need to stay strong and keep your head up emotionally.”

“I am just really thankful that I have a disease that will not kill me or that has [not] affected me physically at all. I consider myself lucky.”

Disease Management: Treatment, Vitiligo Course, Advice-Seeking, Camouflage—The range of information discussed for treatment was highly variable. There were many accounts in which users advised others to seek professional help, namely that of a dermatologist, for a formal assessment. Many expressed frustrations with treatments and their ineffectiveness, to which the majority of users said to consult with a professional and to remain patient and hopeful/optimistic:

“The best thing to do would be to take an appointment with a dermatologist and have the discoloration checked out. That’s the only way to know whether it is vitiligo or not.”

“My way of dealing with it is to gain control by camouflage.”

“The calming effect of being in control of my vitiligo, whether with concealers, self-tan or anything else, has stopped my feelings of despair.”

 

Beliefs on Alternative Medicine: Homeopathy and Alternative Regimens—Although some threads started with a post asking for the best treatments, others initiated a discussion by posting “best herbal treatments for cure” or “how to cure my vitiligo,” emphasizing the beliefs and wishes for a cure for vitiligo. Alternative therapies that users endorsed included apple cider vinegar, toothpaste, vitamins, and Ayurvedic treatment, among others. Dietary plans were popular, with users claiming success with dietary alterations in stopping and preventing lesion progression. For example, individuals felt that avoidance of sugar, meat, dairy, and citrus fruits or drinks and consumption of only filtered water were crucial to preventing further lesion spread and resulted in their “cure”:

“Don’t eat chocolate, wine (made of grapes), coffee, or tea if you don’t want to have vitiligo or let it get worse. Take Vitamin B, biotin, and nuts for Vitamin E.”

Other dangerous messages pitted treatments by health professionals against beliefs in homeopathy:

“I feel that vitiligo treatment is all in your diet and vitamins. All that medicine and UV lights is a no-no . . .w ith every medicine there is a side effect. The doctors could be healing your vitiligo and severely damaging you inside and out, and you won’t know until years later.”

There was a minor presence of users advising against homeopathy and the associated misinformation and inaccurate claims on curing vitiligo, though this group was small in comparison to the number of users posting outlandish claims on cure:

“There is no cure . . . It’s where your immune system attacks your skin cells causing loss of pigmentation. The skin that has lost the pigmentation can’t be reversed.”

Interactions With the Public and Health Care Providers—Those with vitiligo encounter unique situations in public and in their daily lives. Many of the accounts shared anecdotal stories on how patients have handled the stigma and discrimination faced:

“I have had to face discrimination at school, public places, college, functions, and every new person I have met has asked me this: ‘how did this happen?’”

Those with vitiligo even stated how they wished others would deal with their condition out in public, hoping that others would directly ask what the lesions were instead of the more hurtful staring. There were many stories in which users said others feel vitiligo was contagious or “dirty” and stressed that the condition is not infectious:

“I refer to myself as ‘camo-man’ and reassure people I come into contact with that it is not contagious.”

“Once I was eating at a restaurant . . . and a little girl said to her mom, ‘Look, Mom, that lady doesn’t wash her arms, look how dirty they are.’ That just broke my heart.”

 

Skin Tone and Implications—The belief that vitiligo lesions are less dramatic or less anxiety provoking for individuals with lighter skin was noted by users themselves and by health care providers in certain cases. Skin tone and its impact on QOL was confusing and contentious. Some users with fair skin stated their vitiligo was “less of an annoyance” or “less obvious” compared with individuals with darker complexions. Conversely, other accounts of self-reported White users vehemently stressed the anxieties felt by depigmented lesions, despite being “already white at baseline.”

“Was told by my dermatologist (upon diagnosis) that ‘You’re lucky you’re not African American—it shows up on them much worse. You’re so fair, it doesn’t really matter.’ ”

“You didn’t say what race you are. I could imagine it has a bigger impact if you are anything other than White.”

Comment

Patients Looking for Cures—The general attitude within the forums was uplifting and encouraging, with users detailing how they respond to others in public and sharing their personal perspectives. We found a mix of information regarding disease management and treatment of vitiligo. Overall, there was uncertainty about treatments, with individuals expressing concern that their treatments were ineffective or had failed or that better alternatives would be more suitable for their condition. We found many anecdotal endorsements of homeopathic remedies for vitiligo, with users boasting that their disease had not only been cured but had never returned. Some users completely denounced these statements, while other threads seemed to revolve completely around “cure” discussions with no dissenting voices. The number of discussions related to homeopathy was concerning. Furthermore, there often were no moderators within threads to remove cure-related content, whether commercially endorsed or anecdotal. It is plausible that supplements and vitamins recommended by some physicians may be incorrectly interpreted as a “cure” in online discussions. Our findings are consistent with prior reports that forums are a platform to express dissatisfaction with treatment and the need for additional treatment options.^15,22

Concern Expressed by Health Care Providers—Prior qualitative research has described how patients with chronic dermatologic conditions believe that health care providers minimize patients’ psychological distress.^27,28 We found several accounts in which an individual had explicitly stated their provider had “belittled” the extent and impact of vitiligo when comparing skin phototypes. This suggests either that physicians underestimate the impact of vitiligo on their patients or that physicians are not expressing enough empathic concern about the impact the condition has on those affected.

Cosmetic Aspects of Vitiligo—Few clinical trials have investigated QOL and cosmetic acceptability of treatments as outcome measures.²⁹ We found several instances in which users with vitiligo had reported being dismissed as having a “cosmetic disease,” consistent with other work demonstrating the negative impact on such dismissals.²² Moreover, concealment and camouflage techniques frequently were discussed, demonstrating the relevance of cosmetic management as an important research topic.

Trustworthy Sources of Health Information—Patients still view physicians as trustworthy and a key source of health care information and advice.^30-32 Patients with vitiligo who have been directed to reliable information sources often express gratitude²² and want health professionals to remain an important source in their health information-seeking.³¹ Given the range in information discussed online, it may be valuable to invite patients to share what information they have encountered online.

 

Our study highlights the conflicting health information and advice shared by users in online forums, complicating an already psychologically burdensome condition. Guiding patients to credible, moderated sites and resources that are accurate, understandable, and easy to access may help dispel the conflicting messages and stories discussed in the online community.

Study Strengths and Limitations—Limitations included reporting bias and reliance on self-reported information on the diagnosis and extent of individuals’ vitiligo. Excluding social media websites and platforms from the data collection is a limitation to comprehensively assessing the topic of internet users with vitiligo. Many social media platforms direct patients and their family members to support groups and therefore may have excluded these particular individuals. Social media platforms were excluded from our research owing to the prerequisite of creating user accounts or registering as an online member. Our inclusion criteria were specific to forums that did not require registering or creating an account and were therefore freely accessible to all internet viewers. There is an inherent lack of context present in online forums, preventing data collection on individuals’ demographics and socioeconomic backgrounds. However, anonymity may have allowed individuals to express their thoughts more freely.

An integrated approach, along with our sampling method of online forums not requiring registration, allows for greater transferability and understanding of the health needs of the general public with vitiligo.

Conclusion

Individuals with vitiligo continue to seek peer psychosocial support for the physical and emotional management of their disease. Counseling those with vitiligo about cosmetic concealment options, homeopathy, and treatment scams remains paramount. Directing patients to evidence-based resources, along with providing structured sources of support, may help to improve the psychosocial burden and QOL experienced by patients with vitiligo. Connecting patients with local and national support groups moderated by physicians, such as the Global Vitiligo Foundation (https://globalvitiligofoundation.org/), may provide benefit to patients with vitiligo.

Medication nonadherence is common with oral antipsychotic formulations, resulting in relapse, increased morbidity, and more frequent psychiatric hospitalization.^1-7 Psychiatric hospitalization and illness decompensation is costly to health care systems and leads to reduced quality of life for veterans and families.^6,7 Long-acting injectable antipsychotics (LAIAs) were developed to enhance antipsychotic adherence and improve patient outcomes, including reduced psychiatric hospitalization.^8-12

Little outcomes data exist comparing LAIAs, including biweekly risperidone microspheres and monthly paliperidone palmitate.^10-13 Risperidone microspheres require a 3-week oral crossover and are administered every 2 weeks, whereas paliperidone palmitate does not require an oral crossover and is administered every 4 weeks. The paliperidone palmitate loading regimen replaces an oral crossover.

The primary objective of this study was to compare the number of psychiatric hospitalizations between veterans administered risperidone microspheres and those on paliperidone palmitate pre- and post-LAIA initiation. Secondary objectives were to assess rehospitalization rates between patients taking risperidone microspheres and paliperidone palmitate, reduction in pre- and posthospitalization rates with LAIAs, and medication adherence.

Methods

This observational study with a retrospective cohort design was conducted at the Veterans Affairs Loma Linda Healthcare System (VALLHS) in California. We examined veterans who were initiated on LAIAs risperidone microspheres or paliperidone palmitate from January 01, 2016 through December 31, 2018. Veterans who were aged ≥ 18 years and received ≥ 2 injections of either risperidone microspheres or paliperidone palmitate during the study period were included. Veterans were excluded if they had received < 2 doses of either LAIA, received the LAIA outside of the review period, were nonadherent to risperidone crossover if they received risperidone microspheres, or transferred their care to another facility. At VALLHS, LAIA injections are administered by a nurse, and veterans must travel to the facility to receive the injections.

Extracted patient chart elements included participant demographics; diagnoses; comorbid alcohol, nicotine, opioid, or other substance use; duration on LAIA; psychiatric hospitalizations pre- and postinitiation of the LAIA; medication adherence; and medication discontinuation based on clinician documentation and clinic orders (Table 1).

Nonadherence to LAIA was defined as missing an injection by > 3 days for risperidone microspheres and > 7 days for paliperidone palmitate. This time frame was based on pharmacokinetic information listed in the products’ package inserts.^14,15 Nonadherence to oral risperidone crossover with risperidone microspheres was defined as ≤ 80% of days covered.

Data Analysis

Patient demographics were analyzed using descriptive statistics and experimental comparisons between the risperidone microspheres and paliperidone palmitate groups to assess baseline differences between groups. Psychiatric hospitalizations pre- and post-LAIA were analyzed with parallel group (between veterans–independent groups) and pre-post (within veterans–dependent groups) designs. Index hospitalizations were examined for a period equivalent to the length of time veterans were on the LAIA. Psychiatric rehospitalization rates were analyzed for patients who had index hospitalizations and were rehospitalized for any period when they were receiving the LAIA. Incidences of pre- and post-LAIA hospitalizations were calculated in 100 person-years.

Parallel-group analysis was analyzed using the χ² and Mann-Whitney U tests. Pre-post analyses were analyzed using the Wilcoxon rank sum test. P was set at < .05 for statistical significance.

 

Results

We screened 111 veterans, and 97 were included in this study (risperidone microspheres, 44; paliperidone palmitate, 53). Mean (SD) age was 46 (13.8) years, 92% were male, 38% were White, 94% were diagnosed with schizophrenia or schizoaffective disorder, and 11% were homeless. Substance use was documented as 52% for nicotine products, 40% for alcohol, 31% for cannabis, 27% for methamphetamine, 7% for cocaine, and 3% for opioids. Cannabis, methamphetamine, cocaine, and opioid use were based on clinician documentation and listed as active diagnoses at the time of LAIA initiation. Statistical significance was found in index hospitalizations P = .009) and history of cocaine use disorder (6.8% vs 7.5%, P < .001).

Veterans administered risperidone microspheres had fewer mean (SD) post-LAIA hospitalizations (0.4 [1.0] vs 0.9 [1.5]; P = .02) and were less likely to be rehospitalized (22.7% vs 47.2%, P = .01) compared with paliperidone palmitate. However, veterans taking risperidone microspheres had a shorter mean (SD) treatment duration (41.6 [40.2] vs 58.2 [45.7] weeks, P = .04) compared with paliperidone palmitate, mainly because patients switched to a different LAIA or oral antipsychotic. No differences were detected in nonadherence and discontinuation between risperidone microspheres and paliperidone palmitate. All veterans in the risperidone microspheres group adhered to oral risperidone crossover with an average 87.8% days covered (Table 2).

The average maintenance dose of risperidone microspheres was 42 mg every 2 weeks and 153 mg every 4 weeks for paliperidone palmitate.

Across the sample, 84% of veterans had a previous psychiatric hospitalization, although veterans initiated on risperidone microspheres had significantly higher mean (SD) index hospitalizations than those started on paliperidone palmitate (3.2 [2.6] risperidone microspheres vs 2.1 [1.9] paliperidone palmitate, P = .009). Both groups had significant decreases in mean (SD) hospitalizations (3.2 [2.6] to 0.4 [1.0], risperidone microspheres vs 2.1 [1.9] to 0.9 [1.5] paliperidone palmitate). The risperidone microspheres group had a larger decrease in mean (SD) hospitalizations post-LAIA (2.8 [2.9] risperidone microspheres vs 1.3 [1.7] paliperidone palmitate, P = .001) (Table 3).

Differences in incidence per 100 person-years between pre- and post-LAIA hospitalizations were larger in risperidone microspheres users than in paliperidone palmitate (73.8 vs 33.7, P = .01) (Figure). No differences between risperidone microspheres and paliperidone palmitate were detected when looking at incidence pre-LAIA (102.2 vs 75.8, P = .22) and post-LAIA (28.4 vs 42.1, P = .38) separately.

Thirty veterans in the risperidone microspheres group discontinued LAIA: 11 were nonadherent, 5 experienced adverse effects (AEs), and 14 discontinued due to inconvenience. Among 33 veterans in the paliperidone palmitate group who discontinued the LAIA, 15 were nonadherent, 11 experienced AEs, 4 stopped due to of inconvenience, and 3 switched to a less frequently administered LAIA. The most common AEs reported were injection site reactions, cholinergic AEs (salivation, lacrimation, urination), orthostasis, and weight gain.

Discussion

The main finding of this study was that initiation of LAIAs significantly reduced hospitalizations. Veterans taking risperidone microspheres had higher index hospitalizations and lower posttreatment hospitalizations compared with paliperidone palmitate. We found that patients initiated on risperidone microspheres had more hospitalizations before use of a LAIA than those initiated on paliperidone palmitate. Risperidone microspheres reduced the number of hospitalization post-LAIA significantly more than paliperidone palmitate. We also found that veterans taking risperidone microspheres were on the medication for less mean (SD) time than those on paliperidone palmitate (41.6 [40.2] vs 58.2 [45.7] weeks; P = .04).

To our knowledge, this is one of the few studies that compared outcomes of psychiatric hospitalizations, medication adherence, and treatment discontinuation between risperidone microspheres and paliperidone palmitate, specifically in a veteran population.^16-19 Limosin and colleagues aimed to compare length of stay during the initial hospitalization, rehospitalization risk, and treatment duration between risperidone microspheres and paliperidone palmitate in patients with schizophrenia.¹⁶ These researchers detected no differences in initial hospitalization duration and time to rehospitalization between risperidone microspheres and paliperidone palmitate.¹⁶ The study revealed a more favorable trend in time to discontinuation for paliperidone palmitate, but switching between LAIAs might have confounded the data.¹⁶ The authors note that their study lacked power, and patients on paliperidone palmitate had significantly more nonpsychiatric comorbidities.¹⁶ Joshi and colleagues looked at adherence, medication discontinuation, hospitalization rates, emergency department visits, and hospitalization costs between risperidone microspheres and paliperidone palmitate in patients identified in Truven MarketScan Commercial, Medicare Supplemental, and Medicaid Multi-State insurance databases.¹⁷ The authors found paliperidone palmitate to be superior in all objectives with better adherence, lower discontinuation rates, less likelihood of hospitalization, fewer emergency department visits, and lower hospitalization costs compared with risperidone microspheres.¹⁷ Korell and colleagues aimed to establish reference ranges for plasma concentrations of risperidone and paliperidone among adherent patients.¹⁸

 

The researchers established reference ranges for risperidone and paliperidone plasma concentrations that represented expected variability within a population and were derived from population pharmacokinetic models.¹⁸ Gopal and colleagues conducted a post hoc comparison between paliperidone palmitate and oral risperidone during initiation of long-acting injectable risperidone in patients with acute schizophrenia.¹⁹ The researchers found that during the first month after initiating long-acting injectable risperidone, paliperidone palmitate without oral supplementation had similar efficacy and safety to oral risperidone among these patients.¹⁹

LAIAs can create a steadier drug plasma concentration compared with oral antipsychotics and do not need to be taken daily. These agents improve adherence by reducing the frequency of medication administrations.^20-24 Assessing nonadherence is easier with LAIAs by counting missed injections compared with oral antipsychotics that require calculation of percentage of days covered.²⁵

The results in our study are somewhat unexpected in part because of the close relationship between risperidone and paliperidone. Risperidone is converted to paliperidone (9-OH-risperidone) via hepatic cytochrome P450 2D6. Although the molecules do not have identical pharmacologic profiles, it is accepted that they are similar enough that risperidone can establish oral tolerability when transitioning therapy to paliperidone palmitate and vice versa.²⁴ Although the active moiety in risperidone microspheres and paliperidone palmitate is similar, the dosing interval for risperidone microspheres is 2 weeks compared with 4 weeks with paliperidone palmitate. One potential explanation as to why veterans started on risperidone microspheres experienced better outcomes is because they had twice as many office visits with the health care team. Facility procedures dictate veterans receive the LAIA at an on-site clinic. During the visits, a licensed vocational nurse administers the injection and monitors the patient for 15 to 30 minutes afterward.

Despite new LAIAs coming to market, high-quality data examining potential differences in treatment outcomes among agents are limited. This is problematic for clinicians who want to optimize care by understanding how administration schedules or other aspects of LAIA use could modify treatment outcomes. Our results suggest that an advantage might exist in selecting an agent with a more frequent administration schedule, at least initially. This could allow for close monitoring and regular therapeutic contact, which could improve short-term outcomes. This conclusion is supported by meta-analyses, randomized controlled trials, and conceptual articles conducted by Wehring and colleagues, Berwaerts and colleagues, and Parellada and colleagues, respectively, who examined patients on different LAIAs and contact with health care professionals as part of their research.^26-28 These researchers concluded that patients who had regular contact with a health care professional had better outcomes when initiated on a LAIA.^26-28

Limitations

There are several limitations in this study. Retrospective and observational methods introduce risks of bias and confounding variables. Sample size might have limited statistical power to detect differences. Veterans might have had undocumented pre- or posthospitalizations at other institutions, which was not accounted for and lack of rehospitalization is not conclusive of a positive outcome. Institutions could improve on our study and help to fill gaps in comparative data by conducting larger analyses over longer periods and including more LAIA agents.

Conclusions

Although veterans that were administered risperidone microspheres had a shorter treatment duration, they were less likely to be rehospitalized, had a fewer mean number of post-LAIA hospitalizations, and had a larger difference in incidence in 100 person-years compared with veterans on paliperidone palmitate. Nonadherence and discontinuation rates were comparable between risperidone microspheres and paliperidone palmitate. Future studies could aim to further clarify differences in outcomes among agents or administration schedules.

Medication nonadherence is common with oral antipsychotic formulations, resulting in relapse, increased morbidity, and more frequent psychiatric hospitalization.^1-7 Psychiatric hospitalization and illness decompensation is costly to health care systems and leads to reduced quality of life for veterans and families.^6,7 Long-acting injectable antipsychotics (LAIAs) were developed to enhance antipsychotic adherence and improve patient outcomes, including reduced psychiatric hospitalization.^8-12

Little outcomes data exist comparing LAIAs, including biweekly risperidone microspheres and monthly paliperidone palmitate.^10-13 Risperidone microspheres require a 3-week oral crossover and are administered every 2 weeks, whereas paliperidone palmitate does not require an oral crossover and is administered every 4 weeks. The paliperidone palmitate loading regimen replaces an oral crossover.

The primary objective of this study was to compare the number of psychiatric hospitalizations between veterans administered risperidone microspheres and those on paliperidone palmitate pre- and post-LAIA initiation. Secondary objectives were to assess rehospitalization rates between patients taking risperidone microspheres and paliperidone palmitate, reduction in pre- and posthospitalization rates with LAIAs, and medication adherence.

Methods

This observational study with a retrospective cohort design was conducted at the Veterans Affairs Loma Linda Healthcare System (VALLHS) in California. We examined veterans who were initiated on LAIAs risperidone microspheres or paliperidone palmitate from January 01, 2016 through December 31, 2018. Veterans who were aged ≥ 18 years and received ≥ 2 injections of either risperidone microspheres or paliperidone palmitate during the study period were included. Veterans were excluded if they had received < 2 doses of either LAIA, received the LAIA outside of the review period, were nonadherent to risperidone crossover if they received risperidone microspheres, or transferred their care to another facility. At VALLHS, LAIA injections are administered by a nurse, and veterans must travel to the facility to receive the injections.

Extracted patient chart elements included participant demographics; diagnoses; comorbid alcohol, nicotine, opioid, or other substance use; duration on LAIA; psychiatric hospitalizations pre- and postinitiation of the LAIA; medication adherence; and medication discontinuation based on clinician documentation and clinic orders (Table 1).

Nonadherence to LAIA was defined as missing an injection by > 3 days for risperidone microspheres and > 7 days for paliperidone palmitate. This time frame was based on pharmacokinetic information listed in the products’ package inserts.^14,15 Nonadherence to oral risperidone crossover with risperidone microspheres was defined as ≤ 80% of days covered.

Data Analysis

Patient demographics were analyzed using descriptive statistics and experimental comparisons between the risperidone microspheres and paliperidone palmitate groups to assess baseline differences between groups. Psychiatric hospitalizations pre- and post-LAIA were analyzed with parallel group (between veterans–independent groups) and pre-post (within veterans–dependent groups) designs. Index hospitalizations were examined for a period equivalent to the length of time veterans were on the LAIA. Psychiatric rehospitalization rates were analyzed for patients who had index hospitalizations and were rehospitalized for any period when they were receiving the LAIA. Incidences of pre- and post-LAIA hospitalizations were calculated in 100 person-years.

Parallel-group analysis was analyzed using the χ² and Mann-Whitney U tests. Pre-post analyses were analyzed using the Wilcoxon rank sum test. P was set at < .05 for statistical significance.

 

Results

We screened 111 veterans, and 97 were included in this study (risperidone microspheres, 44; paliperidone palmitate, 53). Mean (SD) age was 46 (13.8) years, 92% were male, 38% were White, 94% were diagnosed with schizophrenia or schizoaffective disorder, and 11% were homeless. Substance use was documented as 52% for nicotine products, 40% for alcohol, 31% for cannabis, 27% for methamphetamine, 7% for cocaine, and 3% for opioids. Cannabis, methamphetamine, cocaine, and opioid use were based on clinician documentation and listed as active diagnoses at the time of LAIA initiation. Statistical significance was found in index hospitalizations P = .009) and history of cocaine use disorder (6.8% vs 7.5%, P < .001).

Veterans administered risperidone microspheres had fewer mean (SD) post-LAIA hospitalizations (0.4 [1.0] vs 0.9 [1.5]; P = .02) and were less likely to be rehospitalized (22.7% vs 47.2%, P = .01) compared with paliperidone palmitate. However, veterans taking risperidone microspheres had a shorter mean (SD) treatment duration (41.6 [40.2] vs 58.2 [45.7] weeks, P = .04) compared with paliperidone palmitate, mainly because patients switched to a different LAIA or oral antipsychotic. No differences were detected in nonadherence and discontinuation between risperidone microspheres and paliperidone palmitate. All veterans in the risperidone microspheres group adhered to oral risperidone crossover with an average 87.8% days covered (Table 2).

The average maintenance dose of risperidone microspheres was 42 mg every 2 weeks and 153 mg every 4 weeks for paliperidone palmitate.

Across the sample, 84% of veterans had a previous psychiatric hospitalization, although veterans initiated on risperidone microspheres had significantly higher mean (SD) index hospitalizations than those started on paliperidone palmitate (3.2 [2.6] risperidone microspheres vs 2.1 [1.9] paliperidone palmitate, P = .009). Both groups had significant decreases in mean (SD) hospitalizations (3.2 [2.6] to 0.4 [1.0], risperidone microspheres vs 2.1 [1.9] to 0.9 [1.5] paliperidone palmitate). The risperidone microspheres group had a larger decrease in mean (SD) hospitalizations post-LAIA (2.8 [2.9] risperidone microspheres vs 1.3 [1.7] paliperidone palmitate, P = .001) (Table 3).

Differences in incidence per 100 person-years between pre- and post-LAIA hospitalizations were larger in risperidone microspheres users than in paliperidone palmitate (73.8 vs 33.7, P = .01) (Figure). No differences between risperidone microspheres and paliperidone palmitate were detected when looking at incidence pre-LAIA (102.2 vs 75.8, P = .22) and post-LAIA (28.4 vs 42.1, P = .38) separately.

Thirty veterans in the risperidone microspheres group discontinued LAIA: 11 were nonadherent, 5 experienced adverse effects (AEs), and 14 discontinued due to inconvenience. Among 33 veterans in the paliperidone palmitate group who discontinued the LAIA, 15 were nonadherent, 11 experienced AEs, 4 stopped due to of inconvenience, and 3 switched to a less frequently administered LAIA. The most common AEs reported were injection site reactions, cholinergic AEs (salivation, lacrimation, urination), orthostasis, and weight gain.

Discussion

The main finding of this study was that initiation of LAIAs significantly reduced hospitalizations. Veterans taking risperidone microspheres had higher index hospitalizations and lower posttreatment hospitalizations compared with paliperidone palmitate. We found that patients initiated on risperidone microspheres had more hospitalizations before use of a LAIA than those initiated on paliperidone palmitate. Risperidone microspheres reduced the number of hospitalization post-LAIA significantly more than paliperidone palmitate. We also found that veterans taking risperidone microspheres were on the medication for less mean (SD) time than those on paliperidone palmitate (41.6 [40.2] vs 58.2 [45.7] weeks; P = .04).

To our knowledge, this is one of the few studies that compared outcomes of psychiatric hospitalizations, medication adherence, and treatment discontinuation between risperidone microspheres and paliperidone palmitate, specifically in a veteran population.^16-19 Limosin and colleagues aimed to compare length of stay during the initial hospitalization, rehospitalization risk, and treatment duration between risperidone microspheres and paliperidone palmitate in patients with schizophrenia.¹⁶ These researchers detected no differences in initial hospitalization duration and time to rehospitalization between risperidone microspheres and paliperidone palmitate.¹⁶ The study revealed a more favorable trend in time to discontinuation for paliperidone palmitate, but switching between LAIAs might have confounded the data.¹⁶ The authors note that their study lacked power, and patients on paliperidone palmitate had significantly more nonpsychiatric comorbidities.¹⁶ Joshi and colleagues looked at adherence, medication discontinuation, hospitalization rates, emergency department visits, and hospitalization costs between risperidone microspheres and paliperidone palmitate in patients identified in Truven MarketScan Commercial, Medicare Supplemental, and Medicaid Multi-State insurance databases.¹⁷ The authors found paliperidone palmitate to be superior in all objectives with better adherence, lower discontinuation rates, less likelihood of hospitalization, fewer emergency department visits, and lower hospitalization costs compared with risperidone microspheres.¹⁷ Korell and colleagues aimed to establish reference ranges for plasma concentrations of risperidone and paliperidone among adherent patients.¹⁸

 

The researchers established reference ranges for risperidone and paliperidone plasma concentrations that represented expected variability within a population and were derived from population pharmacokinetic models.¹⁸ Gopal and colleagues conducted a post hoc comparison between paliperidone palmitate and oral risperidone during initiation of long-acting injectable risperidone in patients with acute schizophrenia.¹⁹ The researchers found that during the first month after initiating long-acting injectable risperidone, paliperidone palmitate without oral supplementation had similar efficacy and safety to oral risperidone among these patients.¹⁹

LAIAs can create a steadier drug plasma concentration compared with oral antipsychotics and do not need to be taken daily. These agents improve adherence by reducing the frequency of medication administrations.^20-24 Assessing nonadherence is easier with LAIAs by counting missed injections compared with oral antipsychotics that require calculation of percentage of days covered.²⁵

The results in our study are somewhat unexpected in part because of the close relationship between risperidone and paliperidone. Risperidone is converted to paliperidone (9-OH-risperidone) via hepatic cytochrome P450 2D6. Although the molecules do not have identical pharmacologic profiles, it is accepted that they are similar enough that risperidone can establish oral tolerability when transitioning therapy to paliperidone palmitate and vice versa.²⁴ Although the active moiety in risperidone microspheres and paliperidone palmitate is similar, the dosing interval for risperidone microspheres is 2 weeks compared with 4 weeks with paliperidone palmitate. One potential explanation as to why veterans started on risperidone microspheres experienced better outcomes is because they had twice as many office visits with the health care team. Facility procedures dictate veterans receive the LAIA at an on-site clinic. During the visits, a licensed vocational nurse administers the injection and monitors the patient for 15 to 30 minutes afterward.

Despite new LAIAs coming to market, high-quality data examining potential differences in treatment outcomes among agents are limited. This is problematic for clinicians who want to optimize care by understanding how administration schedules or other aspects of LAIA use could modify treatment outcomes. Our results suggest that an advantage might exist in selecting an agent with a more frequent administration schedule, at least initially. This could allow for close monitoring and regular therapeutic contact, which could improve short-term outcomes. This conclusion is supported by meta-analyses, randomized controlled trials, and conceptual articles conducted by Wehring and colleagues, Berwaerts and colleagues, and Parellada and colleagues, respectively, who examined patients on different LAIAs and contact with health care professionals as part of their research.^26-28 These researchers concluded that patients who had regular contact with a health care professional had better outcomes when initiated on a LAIA.^26-28

Limitations

There are several limitations in this study. Retrospective and observational methods introduce risks of bias and confounding variables. Sample size might have limited statistical power to detect differences. Veterans might have had undocumented pre- or posthospitalizations at other institutions, which was not accounted for and lack of rehospitalization is not conclusive of a positive outcome. Institutions could improve on our study and help to fill gaps in comparative data by conducting larger analyses over longer periods and including more LAIA agents.

Conclusions

Although veterans that were administered risperidone microspheres had a shorter treatment duration, they were less likely to be rehospitalized, had a fewer mean number of post-LAIA hospitalizations, and had a larger difference in incidence in 100 person-years compared with veterans on paliperidone palmitate. Nonadherence and discontinuation rates were comparable between risperidone microspheres and paliperidone palmitate. Future studies could aim to further clarify differences in outcomes among agents or administration schedules.

Medication nonadherence is common with oral antipsychotic formulations, resulting in relapse, increased morbidity, and more frequent psychiatric hospitalization.^1-7 Psychiatric hospitalization and illness decompensation is costly to health care systems and leads to reduced quality of life for veterans and families.^6,7 Long-acting injectable antipsychotics (LAIAs) were developed to enhance antipsychotic adherence and improve patient outcomes, including reduced psychiatric hospitalization.^8-12

Little outcomes data exist comparing LAIAs, including biweekly risperidone microspheres and monthly paliperidone palmitate.^10-13 Risperidone microspheres require a 3-week oral crossover and are administered every 2 weeks, whereas paliperidone palmitate does not require an oral crossover and is administered every 4 weeks. The paliperidone palmitate loading regimen replaces an oral crossover.

The primary objective of this study was to compare the number of psychiatric hospitalizations between veterans administered risperidone microspheres and those on paliperidone palmitate pre- and post-LAIA initiation. Secondary objectives were to assess rehospitalization rates between patients taking risperidone microspheres and paliperidone palmitate, reduction in pre- and posthospitalization rates with LAIAs, and medication adherence.

Methods

This observational study with a retrospective cohort design was conducted at the Veterans Affairs Loma Linda Healthcare System (VALLHS) in California. We examined veterans who were initiated on LAIAs risperidone microspheres or paliperidone palmitate from January 01, 2016 through December 31, 2018. Veterans who were aged ≥ 18 years and received ≥ 2 injections of either risperidone microspheres or paliperidone palmitate during the study period were included. Veterans were excluded if they had received < 2 doses of either LAIA, received the LAIA outside of the review period, were nonadherent to risperidone crossover if they received risperidone microspheres, or transferred their care to another facility. At VALLHS, LAIA injections are administered by a nurse, and veterans must travel to the facility to receive the injections.

Extracted patient chart elements included participant demographics; diagnoses; comorbid alcohol, nicotine, opioid, or other substance use; duration on LAIA; psychiatric hospitalizations pre- and postinitiation of the LAIA; medication adherence; and medication discontinuation based on clinician documentation and clinic orders (Table 1).

Nonadherence to LAIA was defined as missing an injection by > 3 days for risperidone microspheres and > 7 days for paliperidone palmitate. This time frame was based on pharmacokinetic information listed in the products’ package inserts.^14,15 Nonadherence to oral risperidone crossover with risperidone microspheres was defined as ≤ 80% of days covered.

Data Analysis

Patient demographics were analyzed using descriptive statistics and experimental comparisons between the risperidone microspheres and paliperidone palmitate groups to assess baseline differences between groups. Psychiatric hospitalizations pre- and post-LAIA were analyzed with parallel group (between veterans–independent groups) and pre-post (within veterans–dependent groups) designs. Index hospitalizations were examined for a period equivalent to the length of time veterans were on the LAIA. Psychiatric rehospitalization rates were analyzed for patients who had index hospitalizations and were rehospitalized for any period when they were receiving the LAIA. Incidences of pre- and post-LAIA hospitalizations were calculated in 100 person-years.

Parallel-group analysis was analyzed using the χ² and Mann-Whitney U tests. Pre-post analyses were analyzed using the Wilcoxon rank sum test. P was set at < .05 for statistical significance.

 

Results

We screened 111 veterans, and 97 were included in this study (risperidone microspheres, 44; paliperidone palmitate, 53). Mean (SD) age was 46 (13.8) years, 92% were male, 38% were White, 94% were diagnosed with schizophrenia or schizoaffective disorder, and 11% were homeless. Substance use was documented as 52% for nicotine products, 40% for alcohol, 31% for cannabis, 27% for methamphetamine, 7% for cocaine, and 3% for opioids. Cannabis, methamphetamine, cocaine, and opioid use were based on clinician documentation and listed as active diagnoses at the time of LAIA initiation. Statistical significance was found in index hospitalizations P = .009) and history of cocaine use disorder (6.8% vs 7.5%, P < .001).

Veterans administered risperidone microspheres had fewer mean (SD) post-LAIA hospitalizations (0.4 [1.0] vs 0.9 [1.5]; P = .02) and were less likely to be rehospitalized (22.7% vs 47.2%, P = .01) compared with paliperidone palmitate. However, veterans taking risperidone microspheres had a shorter mean (SD) treatment duration (41.6 [40.2] vs 58.2 [45.7] weeks, P = .04) compared with paliperidone palmitate, mainly because patients switched to a different LAIA or oral antipsychotic. No differences were detected in nonadherence and discontinuation between risperidone microspheres and paliperidone palmitate. All veterans in the risperidone microspheres group adhered to oral risperidone crossover with an average 87.8% days covered (Table 2).

The average maintenance dose of risperidone microspheres was 42 mg every 2 weeks and 153 mg every 4 weeks for paliperidone palmitate.

Across the sample, 84% of veterans had a previous psychiatric hospitalization, although veterans initiated on risperidone microspheres had significantly higher mean (SD) index hospitalizations than those started on paliperidone palmitate (3.2 [2.6] risperidone microspheres vs 2.1 [1.9] paliperidone palmitate, P = .009). Both groups had significant decreases in mean (SD) hospitalizations (3.2 [2.6] to 0.4 [1.0], risperidone microspheres vs 2.1 [1.9] to 0.9 [1.5] paliperidone palmitate). The risperidone microspheres group had a larger decrease in mean (SD) hospitalizations post-LAIA (2.8 [2.9] risperidone microspheres vs 1.3 [1.7] paliperidone palmitate, P = .001) (Table 3).

Differences in incidence per 100 person-years between pre- and post-LAIA hospitalizations were larger in risperidone microspheres users than in paliperidone palmitate (73.8 vs 33.7, P = .01) (Figure). No differences between risperidone microspheres and paliperidone palmitate were detected when looking at incidence pre-LAIA (102.2 vs 75.8, P = .22) and post-LAIA (28.4 vs 42.1, P = .38) separately.

Thirty veterans in the risperidone microspheres group discontinued LAIA: 11 were nonadherent, 5 experienced adverse effects (AEs), and 14 discontinued due to inconvenience. Among 33 veterans in the paliperidone palmitate group who discontinued the LAIA, 15 were nonadherent, 11 experienced AEs, 4 stopped due to of inconvenience, and 3 switched to a less frequently administered LAIA. The most common AEs reported were injection site reactions, cholinergic AEs (salivation, lacrimation, urination), orthostasis, and weight gain.

Discussion

The main finding of this study was that initiation of LAIAs significantly reduced hospitalizations. Veterans taking risperidone microspheres had higher index hospitalizations and lower posttreatment hospitalizations compared with paliperidone palmitate. We found that patients initiated on risperidone microspheres had more hospitalizations before use of a LAIA than those initiated on paliperidone palmitate. Risperidone microspheres reduced the number of hospitalization post-LAIA significantly more than paliperidone palmitate. We also found that veterans taking risperidone microspheres were on the medication for less mean (SD) time than those on paliperidone palmitate (41.6 [40.2] vs 58.2 [45.7] weeks; P = .04).

To our knowledge, this is one of the few studies that compared outcomes of psychiatric hospitalizations, medication adherence, and treatment discontinuation between risperidone microspheres and paliperidone palmitate, specifically in a veteran population.^16-19 Limosin and colleagues aimed to compare length of stay during the initial hospitalization, rehospitalization risk, and treatment duration between risperidone microspheres and paliperidone palmitate in patients with schizophrenia.¹⁶ These researchers detected no differences in initial hospitalization duration and time to rehospitalization between risperidone microspheres and paliperidone palmitate.¹⁶ The study revealed a more favorable trend in time to discontinuation for paliperidone palmitate, but switching between LAIAs might have confounded the data.¹⁶ The authors note that their study lacked power, and patients on paliperidone palmitate had significantly more nonpsychiatric comorbidities.¹⁶ Joshi and colleagues looked at adherence, medication discontinuation, hospitalization rates, emergency department visits, and hospitalization costs between risperidone microspheres and paliperidone palmitate in patients identified in Truven MarketScan Commercial, Medicare Supplemental, and Medicaid Multi-State insurance databases.¹⁷ The authors found paliperidone palmitate to be superior in all objectives with better adherence, lower discontinuation rates, less likelihood of hospitalization, fewer emergency department visits, and lower hospitalization costs compared with risperidone microspheres.¹⁷ Korell and colleagues aimed to establish reference ranges for plasma concentrations of risperidone and paliperidone among adherent patients.¹⁸

 

The researchers established reference ranges for risperidone and paliperidone plasma concentrations that represented expected variability within a population and were derived from population pharmacokinetic models.¹⁸ Gopal and colleagues conducted a post hoc comparison between paliperidone palmitate and oral risperidone during initiation of long-acting injectable risperidone in patients with acute schizophrenia.¹⁹ The researchers found that during the first month after initiating long-acting injectable risperidone, paliperidone palmitate without oral supplementation had similar efficacy and safety to oral risperidone among these patients.¹⁹

LAIAs can create a steadier drug plasma concentration compared with oral antipsychotics and do not need to be taken daily. These agents improve adherence by reducing the frequency of medication administrations.^20-24 Assessing nonadherence is easier with LAIAs by counting missed injections compared with oral antipsychotics that require calculation of percentage of days covered.²⁵

The results in our study are somewhat unexpected in part because of the close relationship between risperidone and paliperidone. Risperidone is converted to paliperidone (9-OH-risperidone) via hepatic cytochrome P450 2D6. Although the molecules do not have identical pharmacologic profiles, it is accepted that they are similar enough that risperidone can establish oral tolerability when transitioning therapy to paliperidone palmitate and vice versa.²⁴ Although the active moiety in risperidone microspheres and paliperidone palmitate is similar, the dosing interval for risperidone microspheres is 2 weeks compared with 4 weeks with paliperidone palmitate. One potential explanation as to why veterans started on risperidone microspheres experienced better outcomes is because they had twice as many office visits with the health care team. Facility procedures dictate veterans receive the LAIA at an on-site clinic. During the visits, a licensed vocational nurse administers the injection and monitors the patient for 15 to 30 minutes afterward.

Despite new LAIAs coming to market, high-quality data examining potential differences in treatment outcomes among agents are limited. This is problematic for clinicians who want to optimize care by understanding how administration schedules or other aspects of LAIA use could modify treatment outcomes. Our results suggest that an advantage might exist in selecting an agent with a more frequent administration schedule, at least initially. This could allow for close monitoring and regular therapeutic contact, which could improve short-term outcomes. This conclusion is supported by meta-analyses, randomized controlled trials, and conceptual articles conducted by Wehring and colleagues, Berwaerts and colleagues, and Parellada and colleagues, respectively, who examined patients on different LAIAs and contact with health care professionals as part of their research.^26-28 These researchers concluded that patients who had regular contact with a health care professional had better outcomes when initiated on a LAIA.^26-28

Limitations

There are several limitations in this study. Retrospective and observational methods introduce risks of bias and confounding variables. Sample size might have limited statistical power to detect differences. Veterans might have had undocumented pre- or posthospitalizations at other institutions, which was not accounted for and lack of rehospitalization is not conclusive of a positive outcome. Institutions could improve on our study and help to fill gaps in comparative data by conducting larger analyses over longer periods and including more LAIA agents.

Conclusions

Although veterans that were administered risperidone microspheres had a shorter treatment duration, they were less likely to be rehospitalized, had a fewer mean number of post-LAIA hospitalizations, and had a larger difference in incidence in 100 person-years compared with veterans on paliperidone palmitate. Nonadherence and discontinuation rates were comparable between risperidone microspheres and paliperidone palmitate. Future studies could aim to further clarify differences in outcomes among agents or administration schedules.

Ease of access to opioids in the perioperative period is a risk factor for opioid misuse and has been identified as a strong risk factor for heroin use.^1,2 Three-quarters of today’s heroin users were introduced to opioids through prescription medications.² The United States accounts for about 80% of the global opioid supply consumption, and deaths from opioid overdose are increasing: 70,630 deaths in 2019 alone.^3,4

The Centers for Disease Control and Prevention (CDC) has called for changes in opioid prescribing. The American Academy of Orthopaedic Surgeons (AAOS) also has published an information statement with strategies to decrease opioid misuse and abuse.^5,6 Arthroplasty surgeons have recently focused on decreasing use of opioids in total knee arthroplasty (TKA), a procedure traditionally associated with high levels of opioid consumption and historical reliance on opioid monotherapy for postoperative analgesia.^7,8 From a clinical perspective, prolonged postoperative opioid use contributes to poorer surgical outcomes due to increased risk of complications, including stiffness, infection, and revision TKA.⁹

Multimodal pain regimens are increasingly being used to control postoperative pain as data supports their efficacy.^10,11 Previous studies have found that simultaneous modulation of multiple pain pathways decreases narcotics consumption and improves patient outcomes.^12,13 Along with other adjuvant therapies, peripheral nerve blocks, such as adductor canal block (ACB) and femoral nerve block (FNB), have been used to decrease postoperative pain.¹⁴ Studies have shown that ACB has fewer complications and shorter functional recovery times compared with FNB.^15,16 The distribution of the ACB excludes the femoral nerve, thus preserving greater quadriceps strength while providing equivalent levels of analgesia compared with FNB.^15,17,18 The ACB has shown decreased near-fall events and improved balance scores in the immediate postoperative period.¹⁹

Our study analyzed opioid consumption patterns of TKA patients from a US Department of Veterans Affairs (VA) medical center before and after the institution of a multimodal analgesic protocol using ACB. The primary purpose of this study was to determine whether a protocol that included intraoperative spinal anesthesia with a postoperative multimodal analgesic regimen and ACB was associated with a decreased postoperative opioid requirement when compared with patients who received intraoperative general anesthesia and a traditional opioid regimen. Secondary outcomes included the effect of opioid consumption on range of motion on postoperative day (POD) 1 and number of opioid prescriptions written at the first postoperative clinic visit.

Methods

Approval for the study was obtained from the institutional review board at the Dayton Veterans Affairs Medical Center (DVAMC) in Ohio. A retrospective chart review was performed to collect data from all patients undergoing TKA at DVAMC from June 1, 2011, through December 31, 2015. Exclusion criteria included multiple surgeries in the study time frame, documented chronic pain, allergy to local anesthetics, daily preoperative use of opioids, and incomplete data in the health record.

All surgeries were performed by 2 staff arthroplasty surgeons at a single VAMC. All patients attended a preoperative visit where a history, physical, and anesthesia evaluation were performed, and watched an educational video detailing surgical indications and postoperative rehabilitation. All surgeries were performed with tourniquets and a periarticular injection was performed at the conclusion of each case. Surgeon 1 treatment of choice was 10 mL 0.5% bupivacaine, whereas surgeon 2 performed a posterior capsular injection of 30 mL 0.25% bupivacaine and a periarticular injection of 30 mg ketorolac in 10 mL 0.25% bupivacaine with epinephrine.

Prior to August 2014, general endotracheal anesthesia was used intraoperatively. A patient-controlled analgesia (PCA) pump of morphine or hydromorphone and additional oral oxycodone or hydrocodone was used for postoperative pain. PCA pumps were patient dependent. In the control group, 245 patients received the morphine PCA while 61 received the hydromorphone PCA. Morphine PCA dosing consisted of 1-mg doses every 10 minutes with potential baseline infusion rates of 0.5 to 1.0 mg/h and a 4-hour limit of 20 mg. Hydromorphone PCA dosing consisted of 0.2 to 0.4-mg doses with a potential continuous dose of 0.2 to 0.4 mg/h and a 4-hour limit of 4 mg.

 

In August 2014, a new analgesic protocol was adopted for TKA consisting of intraoperative spinal anesthesia (0.75% bupivacaine) with IV sedation (propofol), a postoperative multimodal analgesic regimen, an ACB performed in the postanesthesia care unit (PACU), and opioids as needed (protocol group). The ACB catheter was a 0.5% ropivo caine hydrochloride injection. It was attached to a local anesthetic fixed flow rate pump that administers 0.5% ropivacaine without epinephrine at 8 mL/h and was removed on POD 5 by the patient. The multimodal medication regimen included IV ketorolac 15 mg every 6 hours for 3 doses, gabapentin 300 mg every 8 hours, acetaminophen 975 mg every 8 hours, meloxicam 7.5 mg daily, tramadol 50 mg every 6 hours, oxycodone 5 mg 1 to 2 tabs every 4 hours as needed, and IV hydromorphone 0.5 mg every 4 hours as needed for breakthrough pain.

Preoperative demographic characteristics were collected (Table 1). Data on all IV and oral opioid requirements were collected for both groups, converted to morphine milligram equivalents (MME), and a total morphine equivalent dose (MED) was calculated.^20,21

In April 2015, a separate protocol change occurred at the DVAMC with the goal of discharge on POD 1. To standardize outcomes before and after this change, data collection regarding opioid requirements was concluded at midnight on POD 1. If a patient was discharged before midnight on POD 1, opioid requirement through the time of discharge was collected. All surgeries were performed in the morning to early afternoon; however, specific surgical times were not collected. Patients were also evaluated by a physical therapist on POD 0, and maximal knee flexion and extension were measured on POD 1. Patients were discharged with prescriptions for oxycodone/acetaminophen and tramadol and were seen 3 weeks later for their first postoperative visit. Opioid refills at the first postoperative visit were recorded. All statistical analyses were performed in SAS 9.4 with significance set to α = 0.05. Between-groups differences in preoperative and perioperative characteristics as well as postoperative outcomes were analyzed using independent samples t tests for continuous variables and Fisher exact tests for dichotomous discrete variables. Where groups differed for a pre- or perioperative variable, linear mixed models analysis was used to determine whether IV, oral, and total MEDs were significantly affected by the interaction between the pre- or perioperative variable with analgesia group. For refills at the postoperative visit, the effects of pre- or perioperative differences were tested using χ² tests. Effect sizes for outcome variables were estimated using Cohen d and probability of superiority (Δ) for continuous variables, and relative risk (RR) in the case of discrete variables.²²

Results

During the study period from June 1, 2011, through December 31, 2015, 533 eligible TKAs were performed, 306 in the control group and 227 in the protocol group. The groups had similar sex distribution; body mass index; knee range of motion; diagnoses of diabetes mellitus, coronary artery disease, and chronic kidney disease; and history of deep vein thrombosis (DVT) or pulmonary embolism (P ≥ .05). The protocol group was significantly older (P = .04) and had a significantly higher rate of chronic obstructive pulmonary disease (COPD) (P = .002). There were no significant differences between number of procedures performed by surgeon (P = .48) or total tourniquet time (P = .13) (Table 2). Mean (SD) length of stay was significantly greater in the control group compared with the protocol group (2.5 [1.3] vs 1.4 [0.7] days, P < .001).

Figure 1 shows the distributions of each type of opioid used. Compared with the control group, the protocol group had a significantly lower mean (SD) IV opioid use: 178.2 (98.0) MED vs 12.0 (24.6) MED (P < .001; d = 2.19; Δ = 0.94) and mean (SD) total opioid use: 241.7 (120.1) MED vs 74.8 (42.7) MED (P < .001; d = 1.76; Δ = 0.89). Mean (SD) oral opioid use did not differ between groups (control, 63.6 [45.4] MED; protocol, 62.9 [31.4] MED; P = .85; d = 0.02; Δ = 0.51). A significantly lower percentage of patients in the protocol group received additional opioids at the 3-week follow-up when compared to the control group: 46.7% vs 61.3%, respectively (P < .001; RR, 0.76; 95% CI, 0.65-0.90).

There were no significant differences in postoperative mean (SD) maximum knee flexion (control, 67.2 [15.7]°; protocol, 67.8 [19.2]°; P = .72; d = 0.03; Δ = 0.51) or mean (SD) total flexion/extension arc (control, 66.2 [15.9]°; protocol, 67.9 [19.4]°; P = .32; d = 0.10; Δ = 0.53). Mean (SD) postoperative maximum knee extension was significantly higher in the protocol group compared with the control group (-0.1 [2.1]° vs 1.0 [3.7]°; P < .001; d = 0.35; Δ = 0.60). More patients in the protocol group (92.5%) were discharged to home compared with the control group (86.6%) (P = .02; RR, 1.07; 95% CI, 1.01-1.13).

 

Because age and rates of COPD differed between groups, sensitivity analyses were conducted to determine whether these variables influenced postoperative opioid use. The relationship between age and group was significant for IV (P < .001) and total opioid use (P < .001). Younger patients received higher MED doses than older patients within the control group, while dosages were fairly consistent regardless of age in the protocol group (Figure 2). There was no significance in age interaction effect with regard to oral opioids (P = .83) nor opioid refills at 3-week follow-up (P = .24).

The sensitivity analysis for COPD found that a diagnosis of COPD did not significantly influence utilization of IV opioids (P = .10), or total opioids (P = .68). There was a significant interaction effect for oral opioids (Figure 3). Patients in the control group with COPD required significantly higher mean (SD) oral opioids than patients without COPD (91.5 [123.9] MED and 62.0 [36.0] MED, respectively; P = .03). In the control group, the χ² test was significant regarding opioid prescription refills at the 3-week visit (P = .004) with 62.4% of patients with COPD requiring refills vs 44.4% without COPD (P = .004). There was no difference in refills in the protocol group (46.4% vs 48.4%).

Finally, 2-sided independent samples t test evaluated total MED use between the 2 surgeons. There was no difference in total MED per patient for the surgeons. In the control group, mean (SD) total MED for surgeon 1 was 232.9 (118.7) MED vs 252.8 (121.5) MED for surgeon 2 (P = .18). In the protocol group, the mean (SD) total MED was 72.5 (43.2) and 77.4 (42.1) for surgeon 1 and surgeon 2, respectively (P = .39).

Discussion

Coordinated efforts with major medical organizations are being made to decrease opioid prescriptions and exposure.^5,6 To our knowledge, no study has quantified a decrease in opioid requirement in a VA population after implementation of a protocol that includes intraoperative spinal anesthesia and a postoperative multimodal analgesic regimen including ACB after TKA. The analgesic protocol described in this study aligns with recommendations from both the CDC and the AAOS to decrease opioid use and misuse by maximizing nonopioid medications and limiting the size and number of opioid prescriptions. However, public and medical opinion of opioids as well as prescribing practices have changed over time with a trend toward lower opioid use. The interventions, as part of the described protocol, are a result of these changes and attempt to minimize opioid use while maximizing postoperative analgesia.

Our data showed a significant decrease in total opioid use through POD 1, IV opioid use, and opioid prescriptions provided at the first postoperative visit. The protocol group used only 6.7% of the IV opioids and 30.9% of the total opioids that were used by the control group. The substantial difference in IV opioid requirement, 166.2 MED, is equivalent to 8 mg of IV hydromorphone or 55 mg of IV morphine. The difference in total opioid requirement was similar at 166.9 MED, equivalent to 111 mg of oral oxycodone.

Decreasing opioid use has the additional benefit of improving outcomes, as higher doses of opioids have been associated with increased length of stay, greater rates of DVT, and postoperative infection.²³ These complications occurred in a stepwise manner, suggesting a dose-response gradient that makes the sizable decrease noted in our data of greater relevance.²³ While the adverse effects (AEs) of opioids are well known, there are limited data on opioid dosing and its effect on perioperative outcomes.²³

A significant decrease in the percentage of patients receiving an opioid prescription at the first postoperative visit suggests a decrease in the number of patients on prolonged opioids after TKA with implementation of modern analgesic modalities. The duration of postoperative opioid use has been found to be the strongest predictor of misuse, and each postoperative refill increases the probability of misuse by 44%.²⁴ In addition, opioid use for > 3 months after TKA is associated with increased risk of periprosthetic infection, increased overall revision rate, and stiffness at 1 year postoperatively.⁹ While not entirely under the control of the surgeon, measures to decrease the number of postoperative opioid refills may lead to a decrease in opioid misuse.

 

In the control group, older patients tended to receive less opioids. This is likely due to physiologic changes in opioid metabolism associated with aging, including decreased renal and hepatic opioid metabolism and alterations in overall body composition that increase relative potency and duration of action of opioids in a geriatric population.^25,26 No difference in opioid use by age was found for the protocol group.

Patients in the protocol group demonstrated significantly greater maximal knee extension on POD 1 compared with the control group. No difference in maximal flexion was found. This difference in extension may partially be explained by the use of an ACB. One benefit of ACB is greater quadriceps strength and fewer near-fall events when compared with FNB.^15,19

Our results corroborate the findings of similar studies. A randomized controlled trial comparing a multimodal analgesic regimen with a periarticular injection without a postoperative ACB to a hydromorphone PCA revealed a significant decrease in opioid use in the multimodal analgesic group.²⁷ Along with lower opioid requirements, the multimodal analgesic group had lower visual analog scale pain scores, fewer AEs, faster progression to physical therapy milestones, and higher satisfaction.²⁷ Recent guidelines from the French Society of Anaesthesia and Intensive Care Medicine recommend against the use of gabapentin as a method of postoperative pain control. However, this specifically refers to the preoperative administration of gabapentin. This same set of guidelines later cites a high level of evidence suggesting patients undergoing arthroplasty benefit more from gabapentinoids.²⁸ Multiple analgesic protocols that include gabapentin as a part of a multimodal approach have been shown to have positive results.^13,29

In our study, patients receiving the multimodal analgesic regimen were significantly more likely to be discharged home rather than to postacute care facilities, which have been associated with increased rates of major complications, 30-day readmission, and 30-day reoperation.^30,31 In addition, discharge to an inpatient rehabilitation or skilled nursing facility has not been found to result in higher functional outcomes, despite $3.2 billion spent yearly on rehabilitation services after primary TKA.^32,33

A component of our described analgesic protocol included spinal anesthesia intraoperatively. The differences between groups regarding anesthesia type can be attributed to this protocol change. A significantly greater percentage of patients in the protocol group received spinal anesthesia, while more patients in the control group received general anesthesia. While patients who received spinal anesthesia may have enhanced analgesia in the immediate postoperative period, no differences in opioid outcomes were seen based on anesthesia type. Known benefits of intraoperative spinal anesthesia include decreased perioperative blood loss and a smaller decrease in hemoglobin postoperatively, as well as lower rates of in-hospital complications, including pulmonary embolism, pneumonia, cerebrovascular events, and acute renal failure.³⁴

Limitations

A number of limitations of this study should be noted. One was a protocol change regarding length of stay, which occurred during the study period and resulted in a significantly shorter length of stay in the protocol group. As a result, opioid use data were analyzed only through midnight at the end of POD 1. Patients who were discharged on POD 1 did not have opioid use data available for the full duration of the first POD, which may exaggerate the decrease in opioid requirements, as opioids used after discharge but prior to midnight on POD 1 were not recorded. However, opioids taken at home are oral with a low MME compared with IV opioids received by hospitalized patients in the control group. In addition, if taken as prescribed, patients at home would only have enough time to take a few doses of opioids prior to the midnight cutoff. We do not believe this difference in time of opioid use meaningfully affected the data. An additional limitation includes the variability between periarticular injections between surgeons. While the percentage of patients that received injections from surgeon 1 vs surgeon 2 were similar, it cannot be ruled out as a potential confounding factor. Other limitations include a lack of pain scores to compare subjective pain ratings, the retrospective nature of the study, and a largely homogenous male VA population.

Conclusions

Ease of access to opioids is a risk factor for opioid abuse, which itself is a risk factor for subsequent heroin use.^1,2 The CDC and AAOS have thus published recommendations regarding opioid prescribing practices to decrease opioid use and abuse.^5,6 Our described protocol, which aligns with these recommendations, resulted in a significant decrease in IV opioid requirement, total opioid requirement, and lower rates of opioid prescriptions provided at the first postoperative visit. These promising findings demonstrate a lower percentage of patients on long-term opioids after TKA and a significantly decreased cumulative opioid exposure.

Ease of access to opioids in the perioperative period is a risk factor for opioid misuse and has been identified as a strong risk factor for heroin use.^1,2 Three-quarters of today’s heroin users were introduced to opioids through prescription medications.² The United States accounts for about 80% of the global opioid supply consumption, and deaths from opioid overdose are increasing: 70,630 deaths in 2019 alone.^3,4

The Centers for Disease Control and Prevention (CDC) has called for changes in opioid prescribing. The American Academy of Orthopaedic Surgeons (AAOS) also has published an information statement with strategies to decrease opioid misuse and abuse.^5,6 Arthroplasty surgeons have recently focused on decreasing use of opioids in total knee arthroplasty (TKA), a procedure traditionally associated with high levels of opioid consumption and historical reliance on opioid monotherapy for postoperative analgesia.^7,8 From a clinical perspective, prolonged postoperative opioid use contributes to poorer surgical outcomes due to increased risk of complications, including stiffness, infection, and revision TKA.⁹

Multimodal pain regimens are increasingly being used to control postoperative pain as data supports their efficacy.^10,11 Previous studies have found that simultaneous modulation of multiple pain pathways decreases narcotics consumption and improves patient outcomes.^12,13 Along with other adjuvant therapies, peripheral nerve blocks, such as adductor canal block (ACB) and femoral nerve block (FNB), have been used to decrease postoperative pain.¹⁴ Studies have shown that ACB has fewer complications and shorter functional recovery times compared with FNB.^15,16 The distribution of the ACB excludes the femoral nerve, thus preserving greater quadriceps strength while providing equivalent levels of analgesia compared with FNB.^15,17,18 The ACB has shown decreased near-fall events and improved balance scores in the immediate postoperative period.¹⁹

Our study analyzed opioid consumption patterns of TKA patients from a US Department of Veterans Affairs (VA) medical center before and after the institution of a multimodal analgesic protocol using ACB. The primary purpose of this study was to determine whether a protocol that included intraoperative spinal anesthesia with a postoperative multimodal analgesic regimen and ACB was associated with a decreased postoperative opioid requirement when compared with patients who received intraoperative general anesthesia and a traditional opioid regimen. Secondary outcomes included the effect of opioid consumption on range of motion on postoperative day (POD) 1 and number of opioid prescriptions written at the first postoperative clinic visit.

Methods

Approval for the study was obtained from the institutional review board at the Dayton Veterans Affairs Medical Center (DVAMC) in Ohio. A retrospective chart review was performed to collect data from all patients undergoing TKA at DVAMC from June 1, 2011, through December 31, 2015. Exclusion criteria included multiple surgeries in the study time frame, documented chronic pain, allergy to local anesthetics, daily preoperative use of opioids, and incomplete data in the health record.

All surgeries were performed by 2 staff arthroplasty surgeons at a single VAMC. All patients attended a preoperative visit where a history, physical, and anesthesia evaluation were performed, and watched an educational video detailing surgical indications and postoperative rehabilitation. All surgeries were performed with tourniquets and a periarticular injection was performed at the conclusion of each case. Surgeon 1 treatment of choice was 10 mL 0.5% bupivacaine, whereas surgeon 2 performed a posterior capsular injection of 30 mL 0.25% bupivacaine and a periarticular injection of 30 mg ketorolac in 10 mL 0.25% bupivacaine with epinephrine.

Prior to August 2014, general endotracheal anesthesia was used intraoperatively. A patient-controlled analgesia (PCA) pump of morphine or hydromorphone and additional oral oxycodone or hydrocodone was used for postoperative pain. PCA pumps were patient dependent. In the control group, 245 patients received the morphine PCA while 61 received the hydromorphone PCA. Morphine PCA dosing consisted of 1-mg doses every 10 minutes with potential baseline infusion rates of 0.5 to 1.0 mg/h and a 4-hour limit of 20 mg. Hydromorphone PCA dosing consisted of 0.2 to 0.4-mg doses with a potential continuous dose of 0.2 to 0.4 mg/h and a 4-hour limit of 4 mg.

 

In August 2014, a new analgesic protocol was adopted for TKA consisting of intraoperative spinal anesthesia (0.75% bupivacaine) with IV sedation (propofol), a postoperative multimodal analgesic regimen, an ACB performed in the postanesthesia care unit (PACU), and opioids as needed (protocol group). The ACB catheter was a 0.5% ropivo caine hydrochloride injection. It was attached to a local anesthetic fixed flow rate pump that administers 0.5% ropivacaine without epinephrine at 8 mL/h and was removed on POD 5 by the patient. The multimodal medication regimen included IV ketorolac 15 mg every 6 hours for 3 doses, gabapentin 300 mg every 8 hours, acetaminophen 975 mg every 8 hours, meloxicam 7.5 mg daily, tramadol 50 mg every 6 hours, oxycodone 5 mg 1 to 2 tabs every 4 hours as needed, and IV hydromorphone 0.5 mg every 4 hours as needed for breakthrough pain.

Preoperative demographic characteristics were collected (Table 1). Data on all IV and oral opioid requirements were collected for both groups, converted to morphine milligram equivalents (MME), and a total morphine equivalent dose (MED) was calculated.^20,21

In April 2015, a separate protocol change occurred at the DVAMC with the goal of discharge on POD 1. To standardize outcomes before and after this change, data collection regarding opioid requirements was concluded at midnight on POD 1. If a patient was discharged before midnight on POD 1, opioid requirement through the time of discharge was collected. All surgeries were performed in the morning to early afternoon; however, specific surgical times were not collected. Patients were also evaluated by a physical therapist on POD 0, and maximal knee flexion and extension were measured on POD 1. Patients were discharged with prescriptions for oxycodone/acetaminophen and tramadol and were seen 3 weeks later for their first postoperative visit. Opioid refills at the first postoperative visit were recorded. All statistical analyses were performed in SAS 9.4 with significance set to α = 0.05. Between-groups differences in preoperative and perioperative characteristics as well as postoperative outcomes were analyzed using independent samples t tests for continuous variables and Fisher exact tests for dichotomous discrete variables. Where groups differed for a pre- or perioperative variable, linear mixed models analysis was used to determine whether IV, oral, and total MEDs were significantly affected by the interaction between the pre- or perioperative variable with analgesia group. For refills at the postoperative visit, the effects of pre- or perioperative differences were tested using χ² tests. Effect sizes for outcome variables were estimated using Cohen d and probability of superiority (Δ) for continuous variables, and relative risk (RR) in the case of discrete variables.²²

Results

During the study period from June 1, 2011, through December 31, 2015, 533 eligible TKAs were performed, 306 in the control group and 227 in the protocol group. The groups had similar sex distribution; body mass index; knee range of motion; diagnoses of diabetes mellitus, coronary artery disease, and chronic kidney disease; and history of deep vein thrombosis (DVT) or pulmonary embolism (P ≥ .05). The protocol group was significantly older (P = .04) and had a significantly higher rate of chronic obstructive pulmonary disease (COPD) (P = .002). There were no significant differences between number of procedures performed by surgeon (P = .48) or total tourniquet time (P = .13) (Table 2). Mean (SD) length of stay was significantly greater in the control group compared with the protocol group (2.5 [1.3] vs 1.4 [0.7] days, P < .001).

Figure 1 shows the distributions of each type of opioid used. Compared with the control group, the protocol group had a significantly lower mean (SD) IV opioid use: 178.2 (98.0) MED vs 12.0 (24.6) MED (P < .001; d = 2.19; Δ = 0.94) and mean (SD) total opioid use: 241.7 (120.1) MED vs 74.8 (42.7) MED (P < .001; d = 1.76; Δ = 0.89). Mean (SD) oral opioid use did not differ between groups (control, 63.6 [45.4] MED; protocol, 62.9 [31.4] MED; P = .85; d = 0.02; Δ = 0.51). A significantly lower percentage of patients in the protocol group received additional opioids at the 3-week follow-up when compared to the control group: 46.7% vs 61.3%, respectively (P < .001; RR, 0.76; 95% CI, 0.65-0.90).

There were no significant differences in postoperative mean (SD) maximum knee flexion (control, 67.2 [15.7]°; protocol, 67.8 [19.2]°; P = .72; d = 0.03; Δ = 0.51) or mean (SD) total flexion/extension arc (control, 66.2 [15.9]°; protocol, 67.9 [19.4]°; P = .32; d = 0.10; Δ = 0.53). Mean (SD) postoperative maximum knee extension was significantly higher in the protocol group compared with the control group (-0.1 [2.1]° vs 1.0 [3.7]°; P < .001; d = 0.35; Δ = 0.60). More patients in the protocol group (92.5%) were discharged to home compared with the control group (86.6%) (P = .02; RR, 1.07; 95% CI, 1.01-1.13).

 

Because age and rates of COPD differed between groups, sensitivity analyses were conducted to determine whether these variables influenced postoperative opioid use. The relationship between age and group was significant for IV (P < .001) and total opioid use (P < .001). Younger patients received higher MED doses than older patients within the control group, while dosages were fairly consistent regardless of age in the protocol group (Figure 2). There was no significance in age interaction effect with regard to oral opioids (P = .83) nor opioid refills at 3-week follow-up (P = .24).

The sensitivity analysis for COPD found that a diagnosis of COPD did not significantly influence utilization of IV opioids (P = .10), or total opioids (P = .68). There was a significant interaction effect for oral opioids (Figure 3). Patients in the control group with COPD required significantly higher mean (SD) oral opioids than patients without COPD (91.5 [123.9] MED and 62.0 [36.0] MED, respectively; P = .03). In the control group, the χ² test was significant regarding opioid prescription refills at the 3-week visit (P = .004) with 62.4% of patients with COPD requiring refills vs 44.4% without COPD (P = .004). There was no difference in refills in the protocol group (46.4% vs 48.4%).

Finally, 2-sided independent samples t test evaluated total MED use between the 2 surgeons. There was no difference in total MED per patient for the surgeons. In the control group, mean (SD) total MED for surgeon 1 was 232.9 (118.7) MED vs 252.8 (121.5) MED for surgeon 2 (P = .18). In the protocol group, the mean (SD) total MED was 72.5 (43.2) and 77.4 (42.1) for surgeon 1 and surgeon 2, respectively (P = .39).

Discussion

Coordinated efforts with major medical organizations are being made to decrease opioid prescriptions and exposure.^5,6 To our knowledge, no study has quantified a decrease in opioid requirement in a VA population after implementation of a protocol that includes intraoperative spinal anesthesia and a postoperative multimodal analgesic regimen including ACB after TKA. The analgesic protocol described in this study aligns with recommendations from both the CDC and the AAOS to decrease opioid use and misuse by maximizing nonopioid medications and limiting the size and number of opioid prescriptions. However, public and medical opinion of opioids as well as prescribing practices have changed over time with a trend toward lower opioid use. The interventions, as part of the described protocol, are a result of these changes and attempt to minimize opioid use while maximizing postoperative analgesia.

Our data showed a significant decrease in total opioid use through POD 1, IV opioid use, and opioid prescriptions provided at the first postoperative visit. The protocol group used only 6.7% of the IV opioids and 30.9% of the total opioids that were used by the control group. The substantial difference in IV opioid requirement, 166.2 MED, is equivalent to 8 mg of IV hydromorphone or 55 mg of IV morphine. The difference in total opioid requirement was similar at 166.9 MED, equivalent to 111 mg of oral oxycodone.

Decreasing opioid use has the additional benefit of improving outcomes, as higher doses of opioids have been associated with increased length of stay, greater rates of DVT, and postoperative infection.²³ These complications occurred in a stepwise manner, suggesting a dose-response gradient that makes the sizable decrease noted in our data of greater relevance.²³ While the adverse effects (AEs) of opioids are well known, there are limited data on opioid dosing and its effect on perioperative outcomes.²³

A significant decrease in the percentage of patients receiving an opioid prescription at the first postoperative visit suggests a decrease in the number of patients on prolonged opioids after TKA with implementation of modern analgesic modalities. The duration of postoperative opioid use has been found to be the strongest predictor of misuse, and each postoperative refill increases the probability of misuse by 44%.²⁴ In addition, opioid use for > 3 months after TKA is associated with increased risk of periprosthetic infection, increased overall revision rate, and stiffness at 1 year postoperatively.⁹ While not entirely under the control of the surgeon, measures to decrease the number of postoperative opioid refills may lead to a decrease in opioid misuse.

 

In the control group, older patients tended to receive less opioids. This is likely due to physiologic changes in opioid metabolism associated with aging, including decreased renal and hepatic opioid metabolism and alterations in overall body composition that increase relative potency and duration of action of opioids in a geriatric population.^25,26 No difference in opioid use by age was found for the protocol group.

Patients in the protocol group demonstrated significantly greater maximal knee extension on POD 1 compared with the control group. No difference in maximal flexion was found. This difference in extension may partially be explained by the use of an ACB. One benefit of ACB is greater quadriceps strength and fewer near-fall events when compared with FNB.^15,19

Our results corroborate the findings of similar studies. A randomized controlled trial comparing a multimodal analgesic regimen with a periarticular injection without a postoperative ACB to a hydromorphone PCA revealed a significant decrease in opioid use in the multimodal analgesic group.²⁷ Along with lower opioid requirements, the multimodal analgesic group had lower visual analog scale pain scores, fewer AEs, faster progression to physical therapy milestones, and higher satisfaction.²⁷ Recent guidelines from the French Society of Anaesthesia and Intensive Care Medicine recommend against the use of gabapentin as a method of postoperative pain control. However, this specifically refers to the preoperative administration of gabapentin. This same set of guidelines later cites a high level of evidence suggesting patients undergoing arthroplasty benefit more from gabapentinoids.²⁸ Multiple analgesic protocols that include gabapentin as a part of a multimodal approach have been shown to have positive results.^13,29

In our study, patients receiving the multimodal analgesic regimen were significantly more likely to be discharged home rather than to postacute care facilities, which have been associated with increased rates of major complications, 30-day readmission, and 30-day reoperation.^30,31 In addition, discharge to an inpatient rehabilitation or skilled nursing facility has not been found to result in higher functional outcomes, despite $3.2 billion spent yearly on rehabilitation services after primary TKA.^32,33

A component of our described analgesic protocol included spinal anesthesia intraoperatively. The differences between groups regarding anesthesia type can be attributed to this protocol change. A significantly greater percentage of patients in the protocol group received spinal anesthesia, while more patients in the control group received general anesthesia. While patients who received spinal anesthesia may have enhanced analgesia in the immediate postoperative period, no differences in opioid outcomes were seen based on anesthesia type. Known benefits of intraoperative spinal anesthesia include decreased perioperative blood loss and a smaller decrease in hemoglobin postoperatively, as well as lower rates of in-hospital complications, including pulmonary embolism, pneumonia, cerebrovascular events, and acute renal failure.³⁴

Limitations

A number of limitations of this study should be noted. One was a protocol change regarding length of stay, which occurred during the study period and resulted in a significantly shorter length of stay in the protocol group. As a result, opioid use data were analyzed only through midnight at the end of POD 1. Patients who were discharged on POD 1 did not have opioid use data available for the full duration of the first POD, which may exaggerate the decrease in opioid requirements, as opioids used after discharge but prior to midnight on POD 1 were not recorded. However, opioids taken at home are oral with a low MME compared with IV opioids received by hospitalized patients in the control group. In addition, if taken as prescribed, patients at home would only have enough time to take a few doses of opioids prior to the midnight cutoff. We do not believe this difference in time of opioid use meaningfully affected the data. An additional limitation includes the variability between periarticular injections between surgeons. While the percentage of patients that received injections from surgeon 1 vs surgeon 2 were similar, it cannot be ruled out as a potential confounding factor. Other limitations include a lack of pain scores to compare subjective pain ratings, the retrospective nature of the study, and a largely homogenous male VA population.

Conclusions

Ease of access to opioids is a risk factor for opioid abuse, which itself is a risk factor for subsequent heroin use.^1,2 The CDC and AAOS have thus published recommendations regarding opioid prescribing practices to decrease opioid use and abuse.^5,6 Our described protocol, which aligns with these recommendations, resulted in a significant decrease in IV opioid requirement, total opioid requirement, and lower rates of opioid prescriptions provided at the first postoperative visit. These promising findings demonstrate a lower percentage of patients on long-term opioids after TKA and a significantly decreased cumulative opioid exposure.

Ease of access to opioids in the perioperative period is a risk factor for opioid misuse and has been identified as a strong risk factor for heroin use.^1,2 Three-quarters of today’s heroin users were introduced to opioids through prescription medications.² The United States accounts for about 80% of the global opioid supply consumption, and deaths from opioid overdose are increasing: 70,630 deaths in 2019 alone.^3,4

The Centers for Disease Control and Prevention (CDC) has called for changes in opioid prescribing. The American Academy of Orthopaedic Surgeons (AAOS) also has published an information statement with strategies to decrease opioid misuse and abuse.^5,6 Arthroplasty surgeons have recently focused on decreasing use of opioids in total knee arthroplasty (TKA), a procedure traditionally associated with high levels of opioid consumption and historical reliance on opioid monotherapy for postoperative analgesia.^7,8 From a clinical perspective, prolonged postoperative opioid use contributes to poorer surgical outcomes due to increased risk of complications, including stiffness, infection, and revision TKA.⁹

Multimodal pain regimens are increasingly being used to control postoperative pain as data supports their efficacy.^10,11 Previous studies have found that simultaneous modulation of multiple pain pathways decreases narcotics consumption and improves patient outcomes.^12,13 Along with other adjuvant therapies, peripheral nerve blocks, such as adductor canal block (ACB) and femoral nerve block (FNB), have been used to decrease postoperative pain.¹⁴ Studies have shown that ACB has fewer complications and shorter functional recovery times compared with FNB.^15,16 The distribution of the ACB excludes the femoral nerve, thus preserving greater quadriceps strength while providing equivalent levels of analgesia compared with FNB.^15,17,18 The ACB has shown decreased near-fall events and improved balance scores in the immediate postoperative period.¹⁹

Our study analyzed opioid consumption patterns of TKA patients from a US Department of Veterans Affairs (VA) medical center before and after the institution of a multimodal analgesic protocol using ACB. The primary purpose of this study was to determine whether a protocol that included intraoperative spinal anesthesia with a postoperative multimodal analgesic regimen and ACB was associated with a decreased postoperative opioid requirement when compared with patients who received intraoperative general anesthesia and a traditional opioid regimen. Secondary outcomes included the effect of opioid consumption on range of motion on postoperative day (POD) 1 and number of opioid prescriptions written at the first postoperative clinic visit.

Methods

Approval for the study was obtained from the institutional review board at the Dayton Veterans Affairs Medical Center (DVAMC) in Ohio. A retrospective chart review was performed to collect data from all patients undergoing TKA at DVAMC from June 1, 2011, through December 31, 2015. Exclusion criteria included multiple surgeries in the study time frame, documented chronic pain, allergy to local anesthetics, daily preoperative use of opioids, and incomplete data in the health record.

All surgeries were performed by 2 staff arthroplasty surgeons at a single VAMC. All patients attended a preoperative visit where a history, physical, and anesthesia evaluation were performed, and watched an educational video detailing surgical indications and postoperative rehabilitation. All surgeries were performed with tourniquets and a periarticular injection was performed at the conclusion of each case. Surgeon 1 treatment of choice was 10 mL 0.5% bupivacaine, whereas surgeon 2 performed a posterior capsular injection of 30 mL 0.25% bupivacaine and a periarticular injection of 30 mg ketorolac in 10 mL 0.25% bupivacaine with epinephrine.

Prior to August 2014, general endotracheal anesthesia was used intraoperatively. A patient-controlled analgesia (PCA) pump of morphine or hydromorphone and additional oral oxycodone or hydrocodone was used for postoperative pain. PCA pumps were patient dependent. In the control group, 245 patients received the morphine PCA while 61 received the hydromorphone PCA. Morphine PCA dosing consisted of 1-mg doses every 10 minutes with potential baseline infusion rates of 0.5 to 1.0 mg/h and a 4-hour limit of 20 mg. Hydromorphone PCA dosing consisted of 0.2 to 0.4-mg doses with a potential continuous dose of 0.2 to 0.4 mg/h and a 4-hour limit of 4 mg.

 

In August 2014, a new analgesic protocol was adopted for TKA consisting of intraoperative spinal anesthesia (0.75% bupivacaine) with IV sedation (propofol), a postoperative multimodal analgesic regimen, an ACB performed in the postanesthesia care unit (PACU), and opioids as needed (protocol group). The ACB catheter was a 0.5% ropivo caine hydrochloride injection. It was attached to a local anesthetic fixed flow rate pump that administers 0.5% ropivacaine without epinephrine at 8 mL/h and was removed on POD 5 by the patient. The multimodal medication regimen included IV ketorolac 15 mg every 6 hours for 3 doses, gabapentin 300 mg every 8 hours, acetaminophen 975 mg every 8 hours, meloxicam 7.5 mg daily, tramadol 50 mg every 6 hours, oxycodone 5 mg 1 to 2 tabs every 4 hours as needed, and IV hydromorphone 0.5 mg every 4 hours as needed for breakthrough pain.

Preoperative demographic characteristics were collected (Table 1). Data on all IV and oral opioid requirements were collected for both groups, converted to morphine milligram equivalents (MME), and a total morphine equivalent dose (MED) was calculated.^20,21

In April 2015, a separate protocol change occurred at the DVAMC with the goal of discharge on POD 1. To standardize outcomes before and after this change, data collection regarding opioid requirements was concluded at midnight on POD 1. If a patient was discharged before midnight on POD 1, opioid requirement through the time of discharge was collected. All surgeries were performed in the morning to early afternoon; however, specific surgical times were not collected. Patients were also evaluated by a physical therapist on POD 0, and maximal knee flexion and extension were measured on POD 1. Patients were discharged with prescriptions for oxycodone/acetaminophen and tramadol and were seen 3 weeks later for their first postoperative visit. Opioid refills at the first postoperative visit were recorded. All statistical analyses were performed in SAS 9.4 with significance set to α = 0.05. Between-groups differences in preoperative and perioperative characteristics as well as postoperative outcomes were analyzed using independent samples t tests for continuous variables and Fisher exact tests for dichotomous discrete variables. Where groups differed for a pre- or perioperative variable, linear mixed models analysis was used to determine whether IV, oral, and total MEDs were significantly affected by the interaction between the pre- or perioperative variable with analgesia group. For refills at the postoperative visit, the effects of pre- or perioperative differences were tested using χ² tests. Effect sizes for outcome variables were estimated using Cohen d and probability of superiority (Δ) for continuous variables, and relative risk (RR) in the case of discrete variables.²²

Results

During the study period from June 1, 2011, through December 31, 2015, 533 eligible TKAs were performed, 306 in the control group and 227 in the protocol group. The groups had similar sex distribution; body mass index; knee range of motion; diagnoses of diabetes mellitus, coronary artery disease, and chronic kidney disease; and history of deep vein thrombosis (DVT) or pulmonary embolism (P ≥ .05). The protocol group was significantly older (P = .04) and had a significantly higher rate of chronic obstructive pulmonary disease (COPD) (P = .002). There were no significant differences between number of procedures performed by surgeon (P = .48) or total tourniquet time (P = .13) (Table 2). Mean (SD) length of stay was significantly greater in the control group compared with the protocol group (2.5 [1.3] vs 1.4 [0.7] days, P < .001).

Figure 1 shows the distributions of each type of opioid used. Compared with the control group, the protocol group had a significantly lower mean (SD) IV opioid use: 178.2 (98.0) MED vs 12.0 (24.6) MED (P < .001; d = 2.19; Δ = 0.94) and mean (SD) total opioid use: 241.7 (120.1) MED vs 74.8 (42.7) MED (P < .001; d = 1.76; Δ = 0.89). Mean (SD) oral opioid use did not differ between groups (control, 63.6 [45.4] MED; protocol, 62.9 [31.4] MED; P = .85; d = 0.02; Δ = 0.51). A significantly lower percentage of patients in the protocol group received additional opioids at the 3-week follow-up when compared to the control group: 46.7% vs 61.3%, respectively (P < .001; RR, 0.76; 95% CI, 0.65-0.90).

There were no significant differences in postoperative mean (SD) maximum knee flexion (control, 67.2 [15.7]°; protocol, 67.8 [19.2]°; P = .72; d = 0.03; Δ = 0.51) or mean (SD) total flexion/extension arc (control, 66.2 [15.9]°; protocol, 67.9 [19.4]°; P = .32; d = 0.10; Δ = 0.53). Mean (SD) postoperative maximum knee extension was significantly higher in the protocol group compared with the control group (-0.1 [2.1]° vs 1.0 [3.7]°; P < .001; d = 0.35; Δ = 0.60). More patients in the protocol group (92.5%) were discharged to home compared with the control group (86.6%) (P = .02; RR, 1.07; 95% CI, 1.01-1.13).

 

Because age and rates of COPD differed between groups, sensitivity analyses were conducted to determine whether these variables influenced postoperative opioid use. The relationship between age and group was significant for IV (P < .001) and total opioid use (P < .001). Younger patients received higher MED doses than older patients within the control group, while dosages were fairly consistent regardless of age in the protocol group (Figure 2). There was no significance in age interaction effect with regard to oral opioids (P = .83) nor opioid refills at 3-week follow-up (P = .24).

The sensitivity analysis for COPD found that a diagnosis of COPD did not significantly influence utilization of IV opioids (P = .10), or total opioids (P = .68). There was a significant interaction effect for oral opioids (Figure 3). Patients in the control group with COPD required significantly higher mean (SD) oral opioids than patients without COPD (91.5 [123.9] MED and 62.0 [36.0] MED, respectively; P = .03). In the control group, the χ² test was significant regarding opioid prescription refills at the 3-week visit (P = .004) with 62.4% of patients with COPD requiring refills vs 44.4% without COPD (P = .004). There was no difference in refills in the protocol group (46.4% vs 48.4%).

Finally, 2-sided independent samples t test evaluated total MED use between the 2 surgeons. There was no difference in total MED per patient for the surgeons. In the control group, mean (SD) total MED for surgeon 1 was 232.9 (118.7) MED vs 252.8 (121.5) MED for surgeon 2 (P = .18). In the protocol group, the mean (SD) total MED was 72.5 (43.2) and 77.4 (42.1) for surgeon 1 and surgeon 2, respectively (P = .39).

Discussion

Coordinated efforts with major medical organizations are being made to decrease opioid prescriptions and exposure.^5,6 To our knowledge, no study has quantified a decrease in opioid requirement in a VA population after implementation of a protocol that includes intraoperative spinal anesthesia and a postoperative multimodal analgesic regimen including ACB after TKA. The analgesic protocol described in this study aligns with recommendations from both the CDC and the AAOS to decrease opioid use and misuse by maximizing nonopioid medications and limiting the size and number of opioid prescriptions. However, public and medical opinion of opioids as well as prescribing practices have changed over time with a trend toward lower opioid use. The interventions, as part of the described protocol, are a result of these changes and attempt to minimize opioid use while maximizing postoperative analgesia.

Our data showed a significant decrease in total opioid use through POD 1, IV opioid use, and opioid prescriptions provided at the first postoperative visit. The protocol group used only 6.7% of the IV opioids and 30.9% of the total opioids that were used by the control group. The substantial difference in IV opioid requirement, 166.2 MED, is equivalent to 8 mg of IV hydromorphone or 55 mg of IV morphine. The difference in total opioid requirement was similar at 166.9 MED, equivalent to 111 mg of oral oxycodone.

Decreasing opioid use has the additional benefit of improving outcomes, as higher doses of opioids have been associated with increased length of stay, greater rates of DVT, and postoperative infection.²³ These complications occurred in a stepwise manner, suggesting a dose-response gradient that makes the sizable decrease noted in our data of greater relevance.²³ While the adverse effects (AEs) of opioids are well known, there are limited data on opioid dosing and its effect on perioperative outcomes.²³

A significant decrease in the percentage of patients receiving an opioid prescription at the first postoperative visit suggests a decrease in the number of patients on prolonged opioids after TKA with implementation of modern analgesic modalities. The duration of postoperative opioid use has been found to be the strongest predictor of misuse, and each postoperative refill increases the probability of misuse by 44%.²⁴ In addition, opioid use for > 3 months after TKA is associated with increased risk of periprosthetic infection, increased overall revision rate, and stiffness at 1 year postoperatively.⁹ While not entirely under the control of the surgeon, measures to decrease the number of postoperative opioid refills may lead to a decrease in opioid misuse.

 

In the control group, older patients tended to receive less opioids. This is likely due to physiologic changes in opioid metabolism associated with aging, including decreased renal and hepatic opioid metabolism and alterations in overall body composition that increase relative potency and duration of action of opioids in a geriatric population.^25,26 No difference in opioid use by age was found for the protocol group.

Patients in the protocol group demonstrated significantly greater maximal knee extension on POD 1 compared with the control group. No difference in maximal flexion was found. This difference in extension may partially be explained by the use of an ACB. One benefit of ACB is greater quadriceps strength and fewer near-fall events when compared with FNB.^15,19

Our results corroborate the findings of similar studies. A randomized controlled trial comparing a multimodal analgesic regimen with a periarticular injection without a postoperative ACB to a hydromorphone PCA revealed a significant decrease in opioid use in the multimodal analgesic group.²⁷ Along with lower opioid requirements, the multimodal analgesic group had lower visual analog scale pain scores, fewer AEs, faster progression to physical therapy milestones, and higher satisfaction.²⁷ Recent guidelines from the French Society of Anaesthesia and Intensive Care Medicine recommend against the use of gabapentin as a method of postoperative pain control. However, this specifically refers to the preoperative administration of gabapentin. This same set of guidelines later cites a high level of evidence suggesting patients undergoing arthroplasty benefit more from gabapentinoids.²⁸ Multiple analgesic protocols that include gabapentin as a part of a multimodal approach have been shown to have positive results.^13,29

In our study, patients receiving the multimodal analgesic regimen were significantly more likely to be discharged home rather than to postacute care facilities, which have been associated with increased rates of major complications, 30-day readmission, and 30-day reoperation.^30,31 In addition, discharge to an inpatient rehabilitation or skilled nursing facility has not been found to result in higher functional outcomes, despite $3.2 billion spent yearly on rehabilitation services after primary TKA.^32,33

A component of our described analgesic protocol included spinal anesthesia intraoperatively. The differences between groups regarding anesthesia type can be attributed to this protocol change. A significantly greater percentage of patients in the protocol group received spinal anesthesia, while more patients in the control group received general anesthesia. While patients who received spinal anesthesia may have enhanced analgesia in the immediate postoperative period, no differences in opioid outcomes were seen based on anesthesia type. Known benefits of intraoperative spinal anesthesia include decreased perioperative blood loss and a smaller decrease in hemoglobin postoperatively, as well as lower rates of in-hospital complications, including pulmonary embolism, pneumonia, cerebrovascular events, and acute renal failure.³⁴

Limitations

A number of limitations of this study should be noted. One was a protocol change regarding length of stay, which occurred during the study period and resulted in a significantly shorter length of stay in the protocol group. As a result, opioid use data were analyzed only through midnight at the end of POD 1. Patients who were discharged on POD 1 did not have opioid use data available for the full duration of the first POD, which may exaggerate the decrease in opioid requirements, as opioids used after discharge but prior to midnight on POD 1 were not recorded. However, opioids taken at home are oral with a low MME compared with IV opioids received by hospitalized patients in the control group. In addition, if taken as prescribed, patients at home would only have enough time to take a few doses of opioids prior to the midnight cutoff. We do not believe this difference in time of opioid use meaningfully affected the data. An additional limitation includes the variability between periarticular injections between surgeons. While the percentage of patients that received injections from surgeon 1 vs surgeon 2 were similar, it cannot be ruled out as a potential confounding factor. Other limitations include a lack of pain scores to compare subjective pain ratings, the retrospective nature of the study, and a largely homogenous male VA population.

Conclusions

Ease of access to opioids is a risk factor for opioid abuse, which itself is a risk factor for subsequent heroin use.^1,2 The CDC and AAOS have thus published recommendations regarding opioid prescribing practices to decrease opioid use and abuse.^5,6 Our described protocol, which aligns with these recommendations, resulted in a significant decrease in IV opioid requirement, total opioid requirement, and lower rates of opioid prescriptions provided at the first postoperative visit. These promising findings demonstrate a lower percentage of patients on long-term opioids after TKA and a significantly decreased cumulative opioid exposure.