Original Research

Semaglutide and liraglutide are glucagon-like peptide 1 receptor agonists (GLP-1 RAs) that are approved by the US Food and Drug Administration as subcutaneous injections for patients with type 2 diabetes mellitus (T2DM). Both are recommended by the American Diabetes Association (ADA) as first-line options for patients with concomitant atherosclerotic cardiovascular (CV) disease and exert therapeutic effect via incretin-like mechanisms.¹ These agents lower blood glucose levels by stimulating insulin release, increasing the body’s sensitivity to insulin, and inhibiting inappropriate glucagon secretion.^2,3 They also slow gastric emptying, resulting in decreased appetite and potential weight loss.⁴

The SUSTAIN (1-7) trials concluded that semaglutide presented an equivalent safety profile and greater efficacy compared with other GLP-1 RAs, including exenatide and dulaglutide.² The SUSTAIN-10 open-label, head-to-head trial evaluating 1 mg semaglutide once weekly vs 1.2 mg liraglutide daily concluded that semaglutide was superior in hemoglobin A_1c (HbA_1c) and body weight reduction compared with liraglutide, with slightly increased gastrointestinal (GI) adverse effects (AEs).⁵ Similar to the LEADER trial assessing liraglutide, SUSTAIN-6 evaluated semaglutide in patients at increased CV risk and found that compared with placebo, semaglutide decreased rates of serious CV events, such as CV death, myocardial infarction, and stroke and were similar to the CV outcomes in the LEADER trial.^2,6 Although initial results of the SUSTAIN-6 trial were thought to be nearly equivalent to the LEADER trial, analyses later published comparing both trials noted that semaglutide had more potent HbA_1c lowering and weight loss benefit when compared with liraglutide.^2,6 The cardioprotective outcomes of SUSTAIN-6 qualified semaglutide for inclusion in the current ADA Standards of Medical Care recommendations for CV risk reduction.^6,7 However, despite the CV safety profile and efficacy associated with semaglutide, the SUSTAIN-6 trial noted an increased risk of diabetic retinopathy (DR) complications in 50 of 1648 patients (3%) treated with semaglutide compared with 29 of 1649 (1.8%) who received placebo (P = .02; hazard ratio, 1.76; 95% CI, 1.11-2.78).⁶ Of the 79 total patients who experienced retinopathy complications, 66 had retinopathy at baseline (42 of 50 [84%]) in the semaglutide group; 24 of 29 [83%] in the placebo group).⁶ Worsening of DR became one of the most notable AEs of semaglutide evaluated in clinical trials. This further deemed the effect as a warning in the semaglutide package insert to assist clinicians with treatment decisions.

As part of a US Department of Veterans Affairs (VA) National Lost Opportunity Cost Savings Initiative, which encompasses administrative efforts to promote more cost-effective yet safe and efficacious therapy options for veterans, the Michael E. DeBakey VA Medical Center (MEDVAMC) in Houston, Texas, converted a portion of patients with T2DM established on liraglutide to semaglutide. The 30-day supply cost of the 2-pack liraglutide 6 mg/mL (3 mL) injection pens for the MEDVAMC was $197.64. The 30-day supply cost for the singular multidose semaglutide 0.5 mg/0.375 mL (1.5 mL) injection pen was $115.15. Cost savings for the MEDVAMC facility were initially estimated to reach $642,522.

The subset of patients converted had to have undergone teleretinal imaging and not have a diagnosis of nonproliferative DR (NPDR), proliferative DR (PDR), or PDR with or without diabetic macular edema. These recommendations excluding various forms of retinopathy were implemented per local institution guidance supporting clinical data from the SUSTAIN trials. Patients diagnosed with these conditions were continued on liraglutide due to an increased risk of DR complications associated with semaglutide.

In the fall of 2021, there was also a standing list of patients on liraglutide who were not converted due to a lack of teleretinal imaging. As a result, there was potential for a quality improvement (QI) intervention to target this patient population, which could result in further cost savings for MEDVAMC and improved glycemic control because of increased conversion from liraglutide to semaglutide. The purpose of this project was to perform a QI assessment on this subset of patients both initially converted from liraglutide to semaglutide, and those who were yet to be converted due to a lack of teleretinal imaging to determine the impact on glycemic control and cost savings.

Methods

This QI project was a single-center, prospective cohort study with a retrospective chart review of veterans with T2DM converted from liraglutide to semaglutide at the MEDVAMC. Patient data were collected from the Computerized Patient Record System (CPRS) between March 1, 2021, and November 30, 2021. An initial subset of patients was converted to semaglutide in March and April 2021. Patients initially excluded underwent a second chart review to determine whether they truly met exclusion criteria. Patients who did not have a definitive diagnosis of NPDR or PDR, those due for updated teleretinal imaging, as well as those with updated teleretinal imaging that excluded NPDR or PDR were targeted for clinician education interventions.

Following this intervention, a subset of patients with negative DR findings were converted from liraglutide to semaglutide. Primary care and endocrinology clinicians were notified that patients who met the criteria should be referred for teleretinal imaging if no updated results were present or that patients were eligible for semaglutide conversion based on negative findings. Both patients who were initially converted as well as those converted following education were included for data collection/analysis of glycemic control via HbA_1c and blood glucose levels.

Cost savings were evaluated using outpatient pharmacy procurement pricing data. This project was approved by the MEDVAMC Quality Assurance and Regulatory Affairs Office.

Participants

Patients included in the study were adults aged ≥ 18 years with T2DM, converted from liraglutide 0.6 and 1.2 mg daily to semaglutide 0.25 mg weekly (titrated to 0.5 mg weekly after 4 weeks), and had an active prescription for semaglutide, with or without insulin or other oral antihyperglycemics. Patients with NPDR or PDR, type 1 DM, no HbA_1c data, no filled semaglutide prescriptions, insulin pumps, and those without teleretinal imaging within the postintervention period or who died during the study period were excluded.

Patient baseline characteristics collected included demographic data, CV comorbidities, antihyperglycemic medications, and changes in insulin doses. Parameters analyzed at baseline and 3 to 12 months postconversion included body weight, HbA_1c, and blood glucose levels.

Outcomes

The primary objectives of this QI project were to assess glycemic control (via changes in HbA_1c levels) and cost savings following patient conversion from liraglutide to semaglutide. A second objective was to educate clinicians for referral of T2DM patients without teleretinal imaging in the past 2 years.

The purpose of the latter objective was to encourage conversion from liraglutide to semaglutide in the absence of DR. We predicted that 50% of patients with clinician education would be converted. Secondary objectives included assessing body weight differences, evaluating modifications in diabetes regimen, and documenting AEs. We predicted that glycemic control would either remain stable or improve with conversion to semaglutide.

Statistical Analysis

Patient demographic data were analyzed using descriptive statistics. Quantitative data (HbA_1c, blood glucose, and body weight differences as continuous variables) were analyzed using a paired Student t test, and categorical variables were analyzed using the χ² test.

 

 

Results

During the study period, 692 patients were identified with active liraglutide prescriptions (Figure). Of these, 49 patients who were initially excluded due to outdated teleretinal imaging or negative findings met the criteria for clinician education, and 14 of those 49 patients (28.6%) were converted from liraglutide to semaglutide. Thirty-three patients (67.3%) did not schedule teleretinal imaging or did not convert to semaglutide following negative teleretinal findings. Two patients (4.1%) either scheduled or proceeded with teleretinal imaging, without any further action from the clinician.

Including the 14 patients converted posteducational intervention, 425 patients were converted to semaglutide. Excluded from analysis were 121 patients: 57 for incomplete HbA_1c data or no filled semaglutide prescription; 30 for HbA_1c and weight data outside of the study timeframe; 25 died of causes unrelated to the project; 8 had insulin pumps; and 1 was diagnosed with late-onset type 1 DM. The final sample was 304 patients who underwent analysis.

Two hundred seventy-three patients (89.8%) were male, and 180 (59.2%) were White (Table 1). The mean (SD) age of patients was 65.9 (9.6) years, and 236 (77.6%) were established on insulin therapy (either basal, bolus, or a combination). The 3 most common antihyperglycemic agents (other than insulin) that patients used included 185 metformin (60.9%), 104 empagliflozin (34.2%), and 50 glipizide (16.4%) prescriptions.

Most patients had CV disease. Three hundred patients (98.7%) had comorbid hypertension, 298 (98.0%) had hyperlipidemia, and 114 (37.5%) had coronary artery disease (Table 2). Other diseases that patients were concomitantly diagnosed with included peripheral vascular disease, heart failure, history of stroke or transient ischemic attack, and history of myocardial infarction.

Documented AEs included 83 patients (27.3%) with hypoglycemia at any point within 3 to 12 months of conversion and 25 patients (8.2%) with mainly GI-related events, including nausea, vomiting, diarrhea, decreased appetite, and abdominal pain. Six patients (2.0%) had a new diagnosis of DR 3 to 12 months postconversion.

Glycemic Control and Weight Changes

At baseline, mean (SD) HbA_1c was 8.1% (1.5), blood glucose was 187.4 (44.2) mg/dL, and body weight was 112.9 (23.0) kg (Table 3). In the timeframe evaluated (3 to 12 months postconversion), patients’ mean (SD) HbA_1c was found to have significantly decreased to 7.6% (1.4) (P < .001; 95% CI, -0.7 to -0.3), blood glucose decreased to 172.6 (39.0) mg/dL (P < .001; 95% CI, -19.3 to -10.2), and body weight decreased to 105.2 (32.3) kg (P < .001; 95% CI, -10.6 to -4.8). All parameters evaluated were deemed statistically significant.

Further analyses evaluating specific changes in HbA_1c observed postconversion are as follows: 199 patients (65.5%) experienced a decrease, 92 (30.3%) experienced an increase, and 13 (4.3%) experienced no change in their HbA_1c.

As the timeframe was fairly broad to assess HbA_1c changes, a prespecified subgroup analysis was conducted to determine specific changes in HbA_1c within 3 to 6, 6 to 9, and 9 to 12 months postconversion (Table 4). At 3 to 6 months postconversion, patient mean (SD) HbA_1c levels significantly decreased from 8.2% (1.5) at baseline to 7.6% (1.3) postconversion (P = .002; 95% CI, -1.0 to -0.2). At 6 to 9 months postconversion, the mean (SD) HbA_1c significantly decreased from 8.1% (1.5) at baseline to 7.6% (1.4) postconversion (P = .002; 95% CI, -0.8 to -0.2).

Glucose-Lowering Agent Adjustments

One hundred thirteen patients (37.2%) required no changes to their antihyperglycemic regimen with the conversion, 85 (28.0%) required increased insulin doses, and 77 (25.3%) required decreased insulin doses (Table 5). Forty-five (14.8%) patients underwent discontinuation of either insulin or other antihyperglycemic agents; 44 (14.5%) had other antihyperglycemic agents dose increased, 39 (12.8%) required adding other glucose-lowering agents, 28 (9.2%) discontinued semaglutide, and 10 (3.3%) had other glucose-lowering medication doses decreased.

Cost Savings

Cost savings were evaluated using the MEDVAMC outpatient pharmacy procurement service. The total cost savings per patient per month was $82.49. For the 411 preclinician education patients converted to semaglutide, this resulted in a prospective annual cost savings of $406,840.68. An additional $13,858.32 was saved due to the intervention/clinician education for 14 patients converted to semaglutide. The total annual cost savings was $420,699.00.

Discussion

Overall, glycemic control significantly improved with veterans’ conversion from liraglutide to semaglutide. Not only were significant changes noted with HbA_1c levels and weight, but consistencies were noted with mean HbA_1c decrease and weight loss expected of GLP-1 RAs noted in clinical trials. The typical range for HbA_1c changes expected is -1% to -2% and weight loss of 1 to 6 kg.^4,7 Data from the LEAD-5 and SUSTAIN-4 trials, evaluating glycemic control in liraglutide and semaglutide, respectively, have noted comparable yet slightly more potent HbA_1c decreases (-1.33% for liraglutide 1.8 mg daily vs -1.2% and -1.6% for semaglutide 0.5 mg and 1 mg weekly, respectively).^8,9 However, more robust weight loss has been noted with semaglutide vs liraglutide (-4.62 kg for semaglutide 0.5 mg weekly and -6.33 kg for semaglutide 1 mg weekly vs -3.43 kg for liraglutide 1.8 mg daily).^8,9 Results from the SUSTAIN-10 trial also noted mean changes in HbA_1c of -1.7% for semaglutide 1 mg weekly vs -1.0% for liraglutide 1.2 mg daily; mean body weight differences were -5.8 kg for semaglutide and -1.9 kg for liraglutide at their respective doses.⁵ The mean weight loss noted with this QI project is consistent with prior trials of semaglutide.

Of note, 44 patients (14.5%) required the dosage increase of either one or multiple additional glucose-lowering agents at any time point within the 3- to 12-month period. Of those patients, 38 (86.4%) underwent further semaglutide dose titration to 1 mg weekly. Common reasons for a further dose increase to 1 mg weekly were an indication for more robust HbA_1c lowering, a desire to decrease patients’ either basal or bolus insulin requirements, or a treatment goal of completely titrating patients off insulin.

It is uncertain why 30.3% of patients experienced an increase in HbA_1c and 4.3% experienced no change. However, possibilities for the divergence in HbA_1c outcomes in these subsets of patients may include suboptimal adherence to semaglutide or other antihyperglycemic agents as indicated by clinicians or nonadherence to dietary and lifestyle modifications.

Most patients (65.5%) experienced a decrease in HbA_1c because of conversion to semaglutide, and AEs appeared as follows: 27.3% experienced hypoglycemia, and 8.2% experienced GI intolerance. The semaglutide discontinuation rate neared 10%, a majority due to intolerable AEs as previously described. Overall, patients seemed to tolerate the medication well as their glycemic control and weight loss improved. Adherence was not objectively assessed for this QI project but could be an area of improvement for future studies.

At the MEDVAMC, liraglutide is a nonformulary agent and semaglutide is now the formulary-preferred option. For patients with uncontrolled T2DM, if a GLP-1 RA is desired for therapy, clinicians are to place a prior authorization drug request (PADR) consultation for semaglutide for further evaluation and review of VA Criteria for Use (CFU) by clinical pharmacist practitioners. Liraglutide is the alternative option if patients do not meet the CFU for semaglutide (ie, have a diagnosis of DR among other exclusions). However, the semaglutide CFU was updated in April 2022 to exclude those specifically diagnosed with PDR, severe NPDR, and macular edema unless an ophthalmologist deems semaglutide acceptable. This indicates that patients with mild-to-moderate NPDR (who were originally excluded from this QI project) are now eligible to receive semaglutide. The incidence of new DR diagnoses (2%) observed in this study could indicate an unclear relationship between semaglutide and increased rates of DR; however, no definitive correlation can be established due to the retrospective nature of this project. The implications of the results of this QI project in relation to the updated CFU remain undetermined.

Due to the comparable improvements in HbA_1c and more robust weight loss noted with semaglutide vs liraglutide, we deem it appropriate to select semaglutide as the more cost-efficient GLP-1 RA and formulary preferred option. The data of this QI project supports the overall safety and treatment utility of this option. Although significant cost savings were achieved (> $400,000), the long-term benefit of the liraglutide to semaglutide conversion remains unknown.

Strengths and Limitations

Strengths of this project include the large sample size, its setting in a large VA medical center, and the evaluation of multiple outcomes beyond HbA_1c for assessment of glycemic control (ie, mean blood glucose, insulin titration, and dose adjustment of other glucose-lowering agents).

Limitations of this study include the retrospective chart review used for data collection, limited accuracy of objective data due to the COVID-19 pandemic, and inconsistencies with documentation in patients’ electronic health records. As a protective measure in the height of the pandemic between March 2021 and November 2021, the VA promoted using telephone and virtual-visit clinics to minimize exposure for patients with nonurgent follow-up needs. Patient hesitance to present to the clinic in person due to COVID-19 was also a significant factor in obtaining objective follow-up data. As a result, less accurate and timely baseline and postconversion weight and HbA_1c data resulted, leading to our decision to extend the timeframe evaluated postconversion to 3 to 12 months. We also noted inconsistencies with documentation in CPRS. Unless veterans were closely followed by clinical pharmacist practitioners or endocrine consultation service clinicians, it was more difficult to follow and document trends of insulin titration to assess the impact of semaglutide conversion. The number of AEs, including hypoglycemia and GI intolerance, were also not consistently documented within the CPRS, and the frequency of AEs may be underestimated.

Another possible limitation regarding the interpretation of the results includes the portion of patients titrated up to semaglutide 1 mg weekly. As the focal point of this project was to review changes in glycemic control in the conversion to semaglutide 0.5 mg, this population of patients converted to 1 mg could potentially overestimate the HbA_1c and weight changes described, as it is consistent with the SUSTAIN trials that show more robust decreases in those parameters described earlier.

Conclusions

A subset of patients with T2DM converted from liraglutide to semaglutide experienced significant changes in glycemic control and body weight. Significant differences were noted for a decreased HbA_1c, decreased mean blood glucose, and weight loss. A fair portion of patients’ antihyperglycemic regimens required no changes on conversion to semaglutide. Although the semaglutide discontinuation rate neared 10%, AEs that may have contributed to this discontinuation rate included hypoglycemia and GI intolerance. Clinician education resulted in a substantial number of patients undergoing teleretinal imaging and further conversion to semaglutide; however, due to the low conversion response rate, a more effective method of educating clinicians is warranted. Although the semaglutide cost savings initiative at MEDVAMC resulted in significant savings, a full cost-effective analysis is needed to assess more comprehensive institution savings.

Semaglutide and liraglutide are glucagon-like peptide 1 receptor agonists (GLP-1 RAs) that are approved by the US Food and Drug Administration as subcutaneous injections for patients with type 2 diabetes mellitus (T2DM). Both are recommended by the American Diabetes Association (ADA) as first-line options for patients with concomitant atherosclerotic cardiovascular (CV) disease and exert therapeutic effect via incretin-like mechanisms.¹ These agents lower blood glucose levels by stimulating insulin release, increasing the body’s sensitivity to insulin, and inhibiting inappropriate glucagon secretion.^2,3 They also slow gastric emptying, resulting in decreased appetite and potential weight loss.⁴

The SUSTAIN (1-7) trials concluded that semaglutide presented an equivalent safety profile and greater efficacy compared with other GLP-1 RAs, including exenatide and dulaglutide.² The SUSTAIN-10 open-label, head-to-head trial evaluating 1 mg semaglutide once weekly vs 1.2 mg liraglutide daily concluded that semaglutide was superior in hemoglobin A_1c (HbA_1c) and body weight reduction compared with liraglutide, with slightly increased gastrointestinal (GI) adverse effects (AEs).⁵ Similar to the LEADER trial assessing liraglutide, SUSTAIN-6 evaluated semaglutide in patients at increased CV risk and found that compared with placebo, semaglutide decreased rates of serious CV events, such as CV death, myocardial infarction, and stroke and were similar to the CV outcomes in the LEADER trial.^2,6 Although initial results of the SUSTAIN-6 trial were thought to be nearly equivalent to the LEADER trial, analyses later published comparing both trials noted that semaglutide had more potent HbA_1c lowering and weight loss benefit when compared with liraglutide.^2,6 The cardioprotective outcomes of SUSTAIN-6 qualified semaglutide for inclusion in the current ADA Standards of Medical Care recommendations for CV risk reduction.^6,7 However, despite the CV safety profile and efficacy associated with semaglutide, the SUSTAIN-6 trial noted an increased risk of diabetic retinopathy (DR) complications in 50 of 1648 patients (3%) treated with semaglutide compared with 29 of 1649 (1.8%) who received placebo (P = .02; hazard ratio, 1.76; 95% CI, 1.11-2.78).⁶ Of the 79 total patients who experienced retinopathy complications, 66 had retinopathy at baseline (42 of 50 [84%]) in the semaglutide group; 24 of 29 [83%] in the placebo group).⁶ Worsening of DR became one of the most notable AEs of semaglutide evaluated in clinical trials. This further deemed the effect as a warning in the semaglutide package insert to assist clinicians with treatment decisions.

As part of a US Department of Veterans Affairs (VA) National Lost Opportunity Cost Savings Initiative, which encompasses administrative efforts to promote more cost-effective yet safe and efficacious therapy options for veterans, the Michael E. DeBakey VA Medical Center (MEDVAMC) in Houston, Texas, converted a portion of patients with T2DM established on liraglutide to semaglutide. The 30-day supply cost of the 2-pack liraglutide 6 mg/mL (3 mL) injection pens for the MEDVAMC was $197.64. The 30-day supply cost for the singular multidose semaglutide 0.5 mg/0.375 mL (1.5 mL) injection pen was $115.15. Cost savings for the MEDVAMC facility were initially estimated to reach $642,522.

The subset of patients converted had to have undergone teleretinal imaging and not have a diagnosis of nonproliferative DR (NPDR), proliferative DR (PDR), or PDR with or without diabetic macular edema. These recommendations excluding various forms of retinopathy were implemented per local institution guidance supporting clinical data from the SUSTAIN trials. Patients diagnosed with these conditions were continued on liraglutide due to an increased risk of DR complications associated with semaglutide.

In the fall of 2021, there was also a standing list of patients on liraglutide who were not converted due to a lack of teleretinal imaging. As a result, there was potential for a quality improvement (QI) intervention to target this patient population, which could result in further cost savings for MEDVAMC and improved glycemic control because of increased conversion from liraglutide to semaglutide. The purpose of this project was to perform a QI assessment on this subset of patients both initially converted from liraglutide to semaglutide, and those who were yet to be converted due to a lack of teleretinal imaging to determine the impact on glycemic control and cost savings.

Methods

This QI project was a single-center, prospective cohort study with a retrospective chart review of veterans with T2DM converted from liraglutide to semaglutide at the MEDVAMC. Patient data were collected from the Computerized Patient Record System (CPRS) between March 1, 2021, and November 30, 2021. An initial subset of patients was converted to semaglutide in March and April 2021. Patients initially excluded underwent a second chart review to determine whether they truly met exclusion criteria. Patients who did not have a definitive diagnosis of NPDR or PDR, those due for updated teleretinal imaging, as well as those with updated teleretinal imaging that excluded NPDR or PDR were targeted for clinician education interventions.

Following this intervention, a subset of patients with negative DR findings were converted from liraglutide to semaglutide. Primary care and endocrinology clinicians were notified that patients who met the criteria should be referred for teleretinal imaging if no updated results were present or that patients were eligible for semaglutide conversion based on negative findings. Both patients who were initially converted as well as those converted following education were included for data collection/analysis of glycemic control via HbA_1c and blood glucose levels.

Cost savings were evaluated using outpatient pharmacy procurement pricing data. This project was approved by the MEDVAMC Quality Assurance and Regulatory Affairs Office.

Participants

Patients included in the study were adults aged ≥ 18 years with T2DM, converted from liraglutide 0.6 and 1.2 mg daily to semaglutide 0.25 mg weekly (titrated to 0.5 mg weekly after 4 weeks), and had an active prescription for semaglutide, with or without insulin or other oral antihyperglycemics. Patients with NPDR or PDR, type 1 DM, no HbA_1c data, no filled semaglutide prescriptions, insulin pumps, and those without teleretinal imaging within the postintervention period or who died during the study period were excluded.

Patient baseline characteristics collected included demographic data, CV comorbidities, antihyperglycemic medications, and changes in insulin doses. Parameters analyzed at baseline and 3 to 12 months postconversion included body weight, HbA_1c, and blood glucose levels.

Outcomes

The primary objectives of this QI project were to assess glycemic control (via changes in HbA_1c levels) and cost savings following patient conversion from liraglutide to semaglutide. A second objective was to educate clinicians for referral of T2DM patients without teleretinal imaging in the past 2 years.

The purpose of the latter objective was to encourage conversion from liraglutide to semaglutide in the absence of DR. We predicted that 50% of patients with clinician education would be converted. Secondary objectives included assessing body weight differences, evaluating modifications in diabetes regimen, and documenting AEs. We predicted that glycemic control would either remain stable or improve with conversion to semaglutide.

Statistical Analysis

Patient demographic data were analyzed using descriptive statistics. Quantitative data (HbA_1c, blood glucose, and body weight differences as continuous variables) were analyzed using a paired Student t test, and categorical variables were analyzed using the χ² test.

 

 

Results

During the study period, 692 patients were identified with active liraglutide prescriptions (Figure). Of these, 49 patients who were initially excluded due to outdated teleretinal imaging or negative findings met the criteria for clinician education, and 14 of those 49 patients (28.6%) were converted from liraglutide to semaglutide. Thirty-three patients (67.3%) did not schedule teleretinal imaging or did not convert to semaglutide following negative teleretinal findings. Two patients (4.1%) either scheduled or proceeded with teleretinal imaging, without any further action from the clinician.

Including the 14 patients converted posteducational intervention, 425 patients were converted to semaglutide. Excluded from analysis were 121 patients: 57 for incomplete HbA_1c data or no filled semaglutide prescription; 30 for HbA_1c and weight data outside of the study timeframe; 25 died of causes unrelated to the project; 8 had insulin pumps; and 1 was diagnosed with late-onset type 1 DM. The final sample was 304 patients who underwent analysis.

Two hundred seventy-three patients (89.8%) were male, and 180 (59.2%) were White (Table 1). The mean (SD) age of patients was 65.9 (9.6) years, and 236 (77.6%) were established on insulin therapy (either basal, bolus, or a combination). The 3 most common antihyperglycemic agents (other than insulin) that patients used included 185 metformin (60.9%), 104 empagliflozin (34.2%), and 50 glipizide (16.4%) prescriptions.

Most patients had CV disease. Three hundred patients (98.7%) had comorbid hypertension, 298 (98.0%) had hyperlipidemia, and 114 (37.5%) had coronary artery disease (Table 2). Other diseases that patients were concomitantly diagnosed with included peripheral vascular disease, heart failure, history of stroke or transient ischemic attack, and history of myocardial infarction.

Documented AEs included 83 patients (27.3%) with hypoglycemia at any point within 3 to 12 months of conversion and 25 patients (8.2%) with mainly GI-related events, including nausea, vomiting, diarrhea, decreased appetite, and abdominal pain. Six patients (2.0%) had a new diagnosis of DR 3 to 12 months postconversion.

Glycemic Control and Weight Changes

At baseline, mean (SD) HbA_1c was 8.1% (1.5), blood glucose was 187.4 (44.2) mg/dL, and body weight was 112.9 (23.0) kg (Table 3). In the timeframe evaluated (3 to 12 months postconversion), patients’ mean (SD) HbA_1c was found to have significantly decreased to 7.6% (1.4) (P < .001; 95% CI, -0.7 to -0.3), blood glucose decreased to 172.6 (39.0) mg/dL (P < .001; 95% CI, -19.3 to -10.2), and body weight decreased to 105.2 (32.3) kg (P < .001; 95% CI, -10.6 to -4.8). All parameters evaluated were deemed statistically significant.

Further analyses evaluating specific changes in HbA_1c observed postconversion are as follows: 199 patients (65.5%) experienced a decrease, 92 (30.3%) experienced an increase, and 13 (4.3%) experienced no change in their HbA_1c.

As the timeframe was fairly broad to assess HbA_1c changes, a prespecified subgroup analysis was conducted to determine specific changes in HbA_1c within 3 to 6, 6 to 9, and 9 to 12 months postconversion (Table 4). At 3 to 6 months postconversion, patient mean (SD) HbA_1c levels significantly decreased from 8.2% (1.5) at baseline to 7.6% (1.3) postconversion (P = .002; 95% CI, -1.0 to -0.2). At 6 to 9 months postconversion, the mean (SD) HbA_1c significantly decreased from 8.1% (1.5) at baseline to 7.6% (1.4) postconversion (P = .002; 95% CI, -0.8 to -0.2).

Glucose-Lowering Agent Adjustments

One hundred thirteen patients (37.2%) required no changes to their antihyperglycemic regimen with the conversion, 85 (28.0%) required increased insulin doses, and 77 (25.3%) required decreased insulin doses (Table 5). Forty-five (14.8%) patients underwent discontinuation of either insulin or other antihyperglycemic agents; 44 (14.5%) had other antihyperglycemic agents dose increased, 39 (12.8%) required adding other glucose-lowering agents, 28 (9.2%) discontinued semaglutide, and 10 (3.3%) had other glucose-lowering medication doses decreased.

Cost Savings

Cost savings were evaluated using the MEDVAMC outpatient pharmacy procurement service. The total cost savings per patient per month was $82.49. For the 411 preclinician education patients converted to semaglutide, this resulted in a prospective annual cost savings of $406,840.68. An additional $13,858.32 was saved due to the intervention/clinician education for 14 patients converted to semaglutide. The total annual cost savings was $420,699.00.

Discussion

Overall, glycemic control significantly improved with veterans’ conversion from liraglutide to semaglutide. Not only were significant changes noted with HbA_1c levels and weight, but consistencies were noted with mean HbA_1c decrease and weight loss expected of GLP-1 RAs noted in clinical trials. The typical range for HbA_1c changes expected is -1% to -2% and weight loss of 1 to 6 kg.^4,7 Data from the LEAD-5 and SUSTAIN-4 trials, evaluating glycemic control in liraglutide and semaglutide, respectively, have noted comparable yet slightly more potent HbA_1c decreases (-1.33% for liraglutide 1.8 mg daily vs -1.2% and -1.6% for semaglutide 0.5 mg and 1 mg weekly, respectively).^8,9 However, more robust weight loss has been noted with semaglutide vs liraglutide (-4.62 kg for semaglutide 0.5 mg weekly and -6.33 kg for semaglutide 1 mg weekly vs -3.43 kg for liraglutide 1.8 mg daily).^8,9 Results from the SUSTAIN-10 trial also noted mean changes in HbA_1c of -1.7% for semaglutide 1 mg weekly vs -1.0% for liraglutide 1.2 mg daily; mean body weight differences were -5.8 kg for semaglutide and -1.9 kg for liraglutide at their respective doses.⁵ The mean weight loss noted with this QI project is consistent with prior trials of semaglutide.

Of note, 44 patients (14.5%) required the dosage increase of either one or multiple additional glucose-lowering agents at any time point within the 3- to 12-month period. Of those patients, 38 (86.4%) underwent further semaglutide dose titration to 1 mg weekly. Common reasons for a further dose increase to 1 mg weekly were an indication for more robust HbA_1c lowering, a desire to decrease patients’ either basal or bolus insulin requirements, or a treatment goal of completely titrating patients off insulin.

It is uncertain why 30.3% of patients experienced an increase in HbA_1c and 4.3% experienced no change. However, possibilities for the divergence in HbA_1c outcomes in these subsets of patients may include suboptimal adherence to semaglutide or other antihyperglycemic agents as indicated by clinicians or nonadherence to dietary and lifestyle modifications.

Most patients (65.5%) experienced a decrease in HbA_1c because of conversion to semaglutide, and AEs appeared as follows: 27.3% experienced hypoglycemia, and 8.2% experienced GI intolerance. The semaglutide discontinuation rate neared 10%, a majority due to intolerable AEs as previously described. Overall, patients seemed to tolerate the medication well as their glycemic control and weight loss improved. Adherence was not objectively assessed for this QI project but could be an area of improvement for future studies.

At the MEDVAMC, liraglutide is a nonformulary agent and semaglutide is now the formulary-preferred option. For patients with uncontrolled T2DM, if a GLP-1 RA is desired for therapy, clinicians are to place a prior authorization drug request (PADR) consultation for semaglutide for further evaluation and review of VA Criteria for Use (CFU) by clinical pharmacist practitioners. Liraglutide is the alternative option if patients do not meet the CFU for semaglutide (ie, have a diagnosis of DR among other exclusions). However, the semaglutide CFU was updated in April 2022 to exclude those specifically diagnosed with PDR, severe NPDR, and macular edema unless an ophthalmologist deems semaglutide acceptable. This indicates that patients with mild-to-moderate NPDR (who were originally excluded from this QI project) are now eligible to receive semaglutide. The incidence of new DR diagnoses (2%) observed in this study could indicate an unclear relationship between semaglutide and increased rates of DR; however, no definitive correlation can be established due to the retrospective nature of this project. The implications of the results of this QI project in relation to the updated CFU remain undetermined.

Due to the comparable improvements in HbA_1c and more robust weight loss noted with semaglutide vs liraglutide, we deem it appropriate to select semaglutide as the more cost-efficient GLP-1 RA and formulary preferred option. The data of this QI project supports the overall safety and treatment utility of this option. Although significant cost savings were achieved (> $400,000), the long-term benefit of the liraglutide to semaglutide conversion remains unknown.

Strengths and Limitations

Strengths of this project include the large sample size, its setting in a large VA medical center, and the evaluation of multiple outcomes beyond HbA_1c for assessment of glycemic control (ie, mean blood glucose, insulin titration, and dose adjustment of other glucose-lowering agents).

Limitations of this study include the retrospective chart review used for data collection, limited accuracy of objective data due to the COVID-19 pandemic, and inconsistencies with documentation in patients’ electronic health records. As a protective measure in the height of the pandemic between March 2021 and November 2021, the VA promoted using telephone and virtual-visit clinics to minimize exposure for patients with nonurgent follow-up needs. Patient hesitance to present to the clinic in person due to COVID-19 was also a significant factor in obtaining objective follow-up data. As a result, less accurate and timely baseline and postconversion weight and HbA_1c data resulted, leading to our decision to extend the timeframe evaluated postconversion to 3 to 12 months. We also noted inconsistencies with documentation in CPRS. Unless veterans were closely followed by clinical pharmacist practitioners or endocrine consultation service clinicians, it was more difficult to follow and document trends of insulin titration to assess the impact of semaglutide conversion. The number of AEs, including hypoglycemia and GI intolerance, were also not consistently documented within the CPRS, and the frequency of AEs may be underestimated.

Another possible limitation regarding the interpretation of the results includes the portion of patients titrated up to semaglutide 1 mg weekly. As the focal point of this project was to review changes in glycemic control in the conversion to semaglutide 0.5 mg, this population of patients converted to 1 mg could potentially overestimate the HbA_1c and weight changes described, as it is consistent with the SUSTAIN trials that show more robust decreases in those parameters described earlier.

Conclusions

A subset of patients with T2DM converted from liraglutide to semaglutide experienced significant changes in glycemic control and body weight. Significant differences were noted for a decreased HbA_1c, decreased mean blood glucose, and weight loss. A fair portion of patients’ antihyperglycemic regimens required no changes on conversion to semaglutide. Although the semaglutide discontinuation rate neared 10%, AEs that may have contributed to this discontinuation rate included hypoglycemia and GI intolerance. Clinician education resulted in a substantial number of patients undergoing teleretinal imaging and further conversion to semaglutide; however, due to the low conversion response rate, a more effective method of educating clinicians is warranted. Although the semaglutide cost savings initiative at MEDVAMC resulted in significant savings, a full cost-effective analysis is needed to assess more comprehensive institution savings.

Semaglutide and liraglutide are glucagon-like peptide 1 receptor agonists (GLP-1 RAs) that are approved by the US Food and Drug Administration as subcutaneous injections for patients with type 2 diabetes mellitus (T2DM). Both are recommended by the American Diabetes Association (ADA) as first-line options for patients with concomitant atherosclerotic cardiovascular (CV) disease and exert therapeutic effect via incretin-like mechanisms.¹ These agents lower blood glucose levels by stimulating insulin release, increasing the body’s sensitivity to insulin, and inhibiting inappropriate glucagon secretion.^2,3 They also slow gastric emptying, resulting in decreased appetite and potential weight loss.⁴

The SUSTAIN (1-7) trials concluded that semaglutide presented an equivalent safety profile and greater efficacy compared with other GLP-1 RAs, including exenatide and dulaglutide.² The SUSTAIN-10 open-label, head-to-head trial evaluating 1 mg semaglutide once weekly vs 1.2 mg liraglutide daily concluded that semaglutide was superior in hemoglobin A_1c (HbA_1c) and body weight reduction compared with liraglutide, with slightly increased gastrointestinal (GI) adverse effects (AEs).⁵ Similar to the LEADER trial assessing liraglutide, SUSTAIN-6 evaluated semaglutide in patients at increased CV risk and found that compared with placebo, semaglutide decreased rates of serious CV events, such as CV death, myocardial infarction, and stroke and were similar to the CV outcomes in the LEADER trial.^2,6 Although initial results of the SUSTAIN-6 trial were thought to be nearly equivalent to the LEADER trial, analyses later published comparing both trials noted that semaglutide had more potent HbA_1c lowering and weight loss benefit when compared with liraglutide.^2,6 The cardioprotective outcomes of SUSTAIN-6 qualified semaglutide for inclusion in the current ADA Standards of Medical Care recommendations for CV risk reduction.^6,7 However, despite the CV safety profile and efficacy associated with semaglutide, the SUSTAIN-6 trial noted an increased risk of diabetic retinopathy (DR) complications in 50 of 1648 patients (3%) treated with semaglutide compared with 29 of 1649 (1.8%) who received placebo (P = .02; hazard ratio, 1.76; 95% CI, 1.11-2.78).⁶ Of the 79 total patients who experienced retinopathy complications, 66 had retinopathy at baseline (42 of 50 [84%]) in the semaglutide group; 24 of 29 [83%] in the placebo group).⁶ Worsening of DR became one of the most notable AEs of semaglutide evaluated in clinical trials. This further deemed the effect as a warning in the semaglutide package insert to assist clinicians with treatment decisions.

As part of a US Department of Veterans Affairs (VA) National Lost Opportunity Cost Savings Initiative, which encompasses administrative efforts to promote more cost-effective yet safe and efficacious therapy options for veterans, the Michael E. DeBakey VA Medical Center (MEDVAMC) in Houston, Texas, converted a portion of patients with T2DM established on liraglutide to semaglutide. The 30-day supply cost of the 2-pack liraglutide 6 mg/mL (3 mL) injection pens for the MEDVAMC was $197.64. The 30-day supply cost for the singular multidose semaglutide 0.5 mg/0.375 mL (1.5 mL) injection pen was $115.15. Cost savings for the MEDVAMC facility were initially estimated to reach $642,522.

The subset of patients converted had to have undergone teleretinal imaging and not have a diagnosis of nonproliferative DR (NPDR), proliferative DR (PDR), or PDR with or without diabetic macular edema. These recommendations excluding various forms of retinopathy were implemented per local institution guidance supporting clinical data from the SUSTAIN trials. Patients diagnosed with these conditions were continued on liraglutide due to an increased risk of DR complications associated with semaglutide.

In the fall of 2021, there was also a standing list of patients on liraglutide who were not converted due to a lack of teleretinal imaging. As a result, there was potential for a quality improvement (QI) intervention to target this patient population, which could result in further cost savings for MEDVAMC and improved glycemic control because of increased conversion from liraglutide to semaglutide. The purpose of this project was to perform a QI assessment on this subset of patients both initially converted from liraglutide to semaglutide, and those who were yet to be converted due to a lack of teleretinal imaging to determine the impact on glycemic control and cost savings.

Methods

This QI project was a single-center, prospective cohort study with a retrospective chart review of veterans with T2DM converted from liraglutide to semaglutide at the MEDVAMC. Patient data were collected from the Computerized Patient Record System (CPRS) between March 1, 2021, and November 30, 2021. An initial subset of patients was converted to semaglutide in March and April 2021. Patients initially excluded underwent a second chart review to determine whether they truly met exclusion criteria. Patients who did not have a definitive diagnosis of NPDR or PDR, those due for updated teleretinal imaging, as well as those with updated teleretinal imaging that excluded NPDR or PDR were targeted for clinician education interventions.

Following this intervention, a subset of patients with negative DR findings were converted from liraglutide to semaglutide. Primary care and endocrinology clinicians were notified that patients who met the criteria should be referred for teleretinal imaging if no updated results were present or that patients were eligible for semaglutide conversion based on negative findings. Both patients who were initially converted as well as those converted following education were included for data collection/analysis of glycemic control via HbA_1c and blood glucose levels.

Cost savings were evaluated using outpatient pharmacy procurement pricing data. This project was approved by the MEDVAMC Quality Assurance and Regulatory Affairs Office.

Participants

Patients included in the study were adults aged ≥ 18 years with T2DM, converted from liraglutide 0.6 and 1.2 mg daily to semaglutide 0.25 mg weekly (titrated to 0.5 mg weekly after 4 weeks), and had an active prescription for semaglutide, with or without insulin or other oral antihyperglycemics. Patients with NPDR or PDR, type 1 DM, no HbA_1c data, no filled semaglutide prescriptions, insulin pumps, and those without teleretinal imaging within the postintervention period or who died during the study period were excluded.

Patient baseline characteristics collected included demographic data, CV comorbidities, antihyperglycemic medications, and changes in insulin doses. Parameters analyzed at baseline and 3 to 12 months postconversion included body weight, HbA_1c, and blood glucose levels.

Outcomes

The primary objectives of this QI project were to assess glycemic control (via changes in HbA_1c levels) and cost savings following patient conversion from liraglutide to semaglutide. A second objective was to educate clinicians for referral of T2DM patients without teleretinal imaging in the past 2 years.

The purpose of the latter objective was to encourage conversion from liraglutide to semaglutide in the absence of DR. We predicted that 50% of patients with clinician education would be converted. Secondary objectives included assessing body weight differences, evaluating modifications in diabetes regimen, and documenting AEs. We predicted that glycemic control would either remain stable or improve with conversion to semaglutide.

Statistical Analysis

Patient demographic data were analyzed using descriptive statistics. Quantitative data (HbA_1c, blood glucose, and body weight differences as continuous variables) were analyzed using a paired Student t test, and categorical variables were analyzed using the χ² test.

 

 

Results

During the study period, 692 patients were identified with active liraglutide prescriptions (Figure). Of these, 49 patients who were initially excluded due to outdated teleretinal imaging or negative findings met the criteria for clinician education, and 14 of those 49 patients (28.6%) were converted from liraglutide to semaglutide. Thirty-three patients (67.3%) did not schedule teleretinal imaging or did not convert to semaglutide following negative teleretinal findings. Two patients (4.1%) either scheduled or proceeded with teleretinal imaging, without any further action from the clinician.

Including the 14 patients converted posteducational intervention, 425 patients were converted to semaglutide. Excluded from analysis were 121 patients: 57 for incomplete HbA_1c data or no filled semaglutide prescription; 30 for HbA_1c and weight data outside of the study timeframe; 25 died of causes unrelated to the project; 8 had insulin pumps; and 1 was diagnosed with late-onset type 1 DM. The final sample was 304 patients who underwent analysis.

Two hundred seventy-three patients (89.8%) were male, and 180 (59.2%) were White (Table 1). The mean (SD) age of patients was 65.9 (9.6) years, and 236 (77.6%) were established on insulin therapy (either basal, bolus, or a combination). The 3 most common antihyperglycemic agents (other than insulin) that patients used included 185 metformin (60.9%), 104 empagliflozin (34.2%), and 50 glipizide (16.4%) prescriptions.

Most patients had CV disease. Three hundred patients (98.7%) had comorbid hypertension, 298 (98.0%) had hyperlipidemia, and 114 (37.5%) had coronary artery disease (Table 2). Other diseases that patients were concomitantly diagnosed with included peripheral vascular disease, heart failure, history of stroke or transient ischemic attack, and history of myocardial infarction.

Documented AEs included 83 patients (27.3%) with hypoglycemia at any point within 3 to 12 months of conversion and 25 patients (8.2%) with mainly GI-related events, including nausea, vomiting, diarrhea, decreased appetite, and abdominal pain. Six patients (2.0%) had a new diagnosis of DR 3 to 12 months postconversion.

Glycemic Control and Weight Changes

At baseline, mean (SD) HbA_1c was 8.1% (1.5), blood glucose was 187.4 (44.2) mg/dL, and body weight was 112.9 (23.0) kg (Table 3). In the timeframe evaluated (3 to 12 months postconversion), patients’ mean (SD) HbA_1c was found to have significantly decreased to 7.6% (1.4) (P < .001; 95% CI, -0.7 to -0.3), blood glucose decreased to 172.6 (39.0) mg/dL (P < .001; 95% CI, -19.3 to -10.2), and body weight decreased to 105.2 (32.3) kg (P < .001; 95% CI, -10.6 to -4.8). All parameters evaluated were deemed statistically significant.

Further analyses evaluating specific changes in HbA_1c observed postconversion are as follows: 199 patients (65.5%) experienced a decrease, 92 (30.3%) experienced an increase, and 13 (4.3%) experienced no change in their HbA_1c.

As the timeframe was fairly broad to assess HbA_1c changes, a prespecified subgroup analysis was conducted to determine specific changes in HbA_1c within 3 to 6, 6 to 9, and 9 to 12 months postconversion (Table 4). At 3 to 6 months postconversion, patient mean (SD) HbA_1c levels significantly decreased from 8.2% (1.5) at baseline to 7.6% (1.3) postconversion (P = .002; 95% CI, -1.0 to -0.2). At 6 to 9 months postconversion, the mean (SD) HbA_1c significantly decreased from 8.1% (1.5) at baseline to 7.6% (1.4) postconversion (P = .002; 95% CI, -0.8 to -0.2).

Glucose-Lowering Agent Adjustments

One hundred thirteen patients (37.2%) required no changes to their antihyperglycemic regimen with the conversion, 85 (28.0%) required increased insulin doses, and 77 (25.3%) required decreased insulin doses (Table 5). Forty-five (14.8%) patients underwent discontinuation of either insulin or other antihyperglycemic agents; 44 (14.5%) had other antihyperglycemic agents dose increased, 39 (12.8%) required adding other glucose-lowering agents, 28 (9.2%) discontinued semaglutide, and 10 (3.3%) had other glucose-lowering medication doses decreased.

Cost Savings

Cost savings were evaluated using the MEDVAMC outpatient pharmacy procurement service. The total cost savings per patient per month was $82.49. For the 411 preclinician education patients converted to semaglutide, this resulted in a prospective annual cost savings of $406,840.68. An additional $13,858.32 was saved due to the intervention/clinician education for 14 patients converted to semaglutide. The total annual cost savings was $420,699.00.

Discussion

Overall, glycemic control significantly improved with veterans’ conversion from liraglutide to semaglutide. Not only were significant changes noted with HbA_1c levels and weight, but consistencies were noted with mean HbA_1c decrease and weight loss expected of GLP-1 RAs noted in clinical trials. The typical range for HbA_1c changes expected is -1% to -2% and weight loss of 1 to 6 kg.^4,7 Data from the LEAD-5 and SUSTAIN-4 trials, evaluating glycemic control in liraglutide and semaglutide, respectively, have noted comparable yet slightly more potent HbA_1c decreases (-1.33% for liraglutide 1.8 mg daily vs -1.2% and -1.6% for semaglutide 0.5 mg and 1 mg weekly, respectively).^8,9 However, more robust weight loss has been noted with semaglutide vs liraglutide (-4.62 kg for semaglutide 0.5 mg weekly and -6.33 kg for semaglutide 1 mg weekly vs -3.43 kg for liraglutide 1.8 mg daily).^8,9 Results from the SUSTAIN-10 trial also noted mean changes in HbA_1c of -1.7% for semaglutide 1 mg weekly vs -1.0% for liraglutide 1.2 mg daily; mean body weight differences were -5.8 kg for semaglutide and -1.9 kg for liraglutide at their respective doses.⁵ The mean weight loss noted with this QI project is consistent with prior trials of semaglutide.

Of note, 44 patients (14.5%) required the dosage increase of either one or multiple additional glucose-lowering agents at any time point within the 3- to 12-month period. Of those patients, 38 (86.4%) underwent further semaglutide dose titration to 1 mg weekly. Common reasons for a further dose increase to 1 mg weekly were an indication for more robust HbA_1c lowering, a desire to decrease patients’ either basal or bolus insulin requirements, or a treatment goal of completely titrating patients off insulin.

It is uncertain why 30.3% of patients experienced an increase in HbA_1c and 4.3% experienced no change. However, possibilities for the divergence in HbA_1c outcomes in these subsets of patients may include suboptimal adherence to semaglutide or other antihyperglycemic agents as indicated by clinicians or nonadherence to dietary and lifestyle modifications.

Most patients (65.5%) experienced a decrease in HbA_1c because of conversion to semaglutide, and AEs appeared as follows: 27.3% experienced hypoglycemia, and 8.2% experienced GI intolerance. The semaglutide discontinuation rate neared 10%, a majority due to intolerable AEs as previously described. Overall, patients seemed to tolerate the medication well as their glycemic control and weight loss improved. Adherence was not objectively assessed for this QI project but could be an area of improvement for future studies.

At the MEDVAMC, liraglutide is a nonformulary agent and semaglutide is now the formulary-preferred option. For patients with uncontrolled T2DM, if a GLP-1 RA is desired for therapy, clinicians are to place a prior authorization drug request (PADR) consultation for semaglutide for further evaluation and review of VA Criteria for Use (CFU) by clinical pharmacist practitioners. Liraglutide is the alternative option if patients do not meet the CFU for semaglutide (ie, have a diagnosis of DR among other exclusions). However, the semaglutide CFU was updated in April 2022 to exclude those specifically diagnosed with PDR, severe NPDR, and macular edema unless an ophthalmologist deems semaglutide acceptable. This indicates that patients with mild-to-moderate NPDR (who were originally excluded from this QI project) are now eligible to receive semaglutide. The incidence of new DR diagnoses (2%) observed in this study could indicate an unclear relationship between semaglutide and increased rates of DR; however, no definitive correlation can be established due to the retrospective nature of this project. The implications of the results of this QI project in relation to the updated CFU remain undetermined.

Due to the comparable improvements in HbA_1c and more robust weight loss noted with semaglutide vs liraglutide, we deem it appropriate to select semaglutide as the more cost-efficient GLP-1 RA and formulary preferred option. The data of this QI project supports the overall safety and treatment utility of this option. Although significant cost savings were achieved (> $400,000), the long-term benefit of the liraglutide to semaglutide conversion remains unknown.

Strengths and Limitations

Strengths of this project include the large sample size, its setting in a large VA medical center, and the evaluation of multiple outcomes beyond HbA_1c for assessment of glycemic control (ie, mean blood glucose, insulin titration, and dose adjustment of other glucose-lowering agents).

Limitations of this study include the retrospective chart review used for data collection, limited accuracy of objective data due to the COVID-19 pandemic, and inconsistencies with documentation in patients’ electronic health records. As a protective measure in the height of the pandemic between March 2021 and November 2021, the VA promoted using telephone and virtual-visit clinics to minimize exposure for patients with nonurgent follow-up needs. Patient hesitance to present to the clinic in person due to COVID-19 was also a significant factor in obtaining objective follow-up data. As a result, less accurate and timely baseline and postconversion weight and HbA_1c data resulted, leading to our decision to extend the timeframe evaluated postconversion to 3 to 12 months. We also noted inconsistencies with documentation in CPRS. Unless veterans were closely followed by clinical pharmacist practitioners or endocrine consultation service clinicians, it was more difficult to follow and document trends of insulin titration to assess the impact of semaglutide conversion. The number of AEs, including hypoglycemia and GI intolerance, were also not consistently documented within the CPRS, and the frequency of AEs may be underestimated.

Another possible limitation regarding the interpretation of the results includes the portion of patients titrated up to semaglutide 1 mg weekly. As the focal point of this project was to review changes in glycemic control in the conversion to semaglutide 0.5 mg, this population of patients converted to 1 mg could potentially overestimate the HbA_1c and weight changes described, as it is consistent with the SUSTAIN trials that show more robust decreases in those parameters described earlier.

Conclusions

A subset of patients with T2DM converted from liraglutide to semaglutide experienced significant changes in glycemic control and body weight. Significant differences were noted for a decreased HbA_1c, decreased mean blood glucose, and weight loss. A fair portion of patients’ antihyperglycemic regimens required no changes on conversion to semaglutide. Although the semaglutide discontinuation rate neared 10%, AEs that may have contributed to this discontinuation rate included hypoglycemia and GI intolerance. Clinician education resulted in a substantial number of patients undergoing teleretinal imaging and further conversion to semaglutide; however, due to the low conversion response rate, a more effective method of educating clinicians is warranted. Although the semaglutide cost savings initiative at MEDVAMC resulted in significant savings, a full cost-effective analysis is needed to assess more comprehensive institution savings.

To the Editor:

Video-based coaching (VBC) involves a surgeon recording a surgery and then reviewing the video with a surgical coach; it is a form of education that is gaining popularity among surgical specialties.¹ Video-based education is underutilized in dermatology residency training.² We conducted a pilot study at our dermatology residency program to evaluate the efficacy and feasibility of VBC.

The University of Texas at Austin Dell Medical School institutional review board approved this study. All 4 first-year dermatology residents were recruited to participate in this study. Participants filled out a prestudy survey assessing their surgical experience, confidence in performing surgery, and attitudes on VBC. Participants used a head-mounted point-of-view camera to record themselves performing a wide local excision on the trunk or extremities of a live human patient. Participants then reviewed the recording on their own and scored themselves using the Objective Structured Assessment of Technical Skills (OSATS) scoring table (scored from 1 to 5, with 5 being the highest possible score for each element), which is a validated tool for assessing surgical skills (eTable 1).³ Given that there were no assistants participating in the surgery, this element of the OSATS scoring table was excluded, making a maximum possible score of 30 and a minimum possible score of 6. After scoring themselves, participants then had a 1-on-1 coaching session with a fellowship-trained dermatologic surgeon (M.F. or T.H.) via online teleconferencing.

During the coaching session, participants and coaches reviewed the video. The surgical coaches also scored the residents using the OSATS, then residents and coaches discussed how the resident could improve using the OSATS scores as a guide. The residents then completed a poststudy survey assessing their surgical experience, confidence in performing surgery, and attitudes on VBC. Descriptive statistics were reported.

On average, residents spent 31.3 minutes reviewing their own surgeries and scoring themselves. The average time for a coaching session, which included time spent scoring, was 13.8 minutes. Residents scored themselves lower than the surgical coaches did by an average of 5.25 points (eTable 2). Residents gave themselves an average total score of 10.5, while their respective surgical coaches gave the residents an average score of 15.75. There was a trend of residents with greater surgical experience having higher OSATS scores (Figure). After the coaching session, 3 of 4 residents reported that they felt more confident in their surgical skills. All residents felt more confident in assessing their surgical skills and felt that VBC was an effective teaching measure. All residents agreed that VBC should be continued as part of their residency training.

Video-based coaching has the potential to provide several benefits for dermatology trainees. Because receiving feedback intraoperatively often can be distracting and incomplete, video review can instead allow the surgeon to focus on performing the surgery and then later focus on learning while reviewing the video.^1,4 Feedback also can be more comprehensive and delivered without concern for time constraints or disturbing clinic flow as well as without the additional concern of the patient overhearing comments and feedback.³ Although independent video review in the absence of coaching can lead to improvement in surgical skills, the addition of VBC provides even greater potential educational benefit.⁴ During the COVID-19 pandemic, VBC allowed coaches to provide feedback without additional exposures. We utilized dermatologic surgery faculty as coaches, but this format of training also would apply to general dermatology faculty.

Another goal of VBC is to enhance a trainee’s ability to perform self-directed learning, which requires accurate self-assessment.⁴ Accurately assessing one’s own strengths empowers a trainee to act with appropriate confidence, while understanding one’s own weaknesses allows a trainee to effectively balance confidence and caution in daily practice.⁵ Interestingly, in our study all residents scored themselves lower than surgical coaches, but with 1 coaching session, the residents subsequently reported greater surgical confidence.

Time constraints can be a potential barrier to surgical coaching.⁴ Our study demonstrates that VBC requires minimal time investment. Increasing the speed of video playback allowed for efficient evaluation of resident surgeries without compromising the coach’s ability to provide comprehensive feedback. Our feedback sessions were performed virtually, which allowed for ease of scheduling between trainees and coaches.

Our pilot study demonstrated that VBC is relatively easy to implement in a dermatology residency training setting, leveraging relatively low-cost technologies and allowing for a means of learning that residents felt was effective. Video-based coaching requires minimal time investment from both trainees and coaches and has the potential to enhance surgical confidence. Our current study is limited by its small sample size. Future studies should include follow-up recordings and assess the efficacy of VBC in enhancing surgical skills.

To the Editor:

Video-based coaching (VBC) involves a surgeon recording a surgery and then reviewing the video with a surgical coach; it is a form of education that is gaining popularity among surgical specialties.¹ Video-based education is underutilized in dermatology residency training.² We conducted a pilot study at our dermatology residency program to evaluate the efficacy and feasibility of VBC.

The University of Texas at Austin Dell Medical School institutional review board approved this study. All 4 first-year dermatology residents were recruited to participate in this study. Participants filled out a prestudy survey assessing their surgical experience, confidence in performing surgery, and attitudes on VBC. Participants used a head-mounted point-of-view camera to record themselves performing a wide local excision on the trunk or extremities of a live human patient. Participants then reviewed the recording on their own and scored themselves using the Objective Structured Assessment of Technical Skills (OSATS) scoring table (scored from 1 to 5, with 5 being the highest possible score for each element), which is a validated tool for assessing surgical skills (eTable 1).³ Given that there were no assistants participating in the surgery, this element of the OSATS scoring table was excluded, making a maximum possible score of 30 and a minimum possible score of 6. After scoring themselves, participants then had a 1-on-1 coaching session with a fellowship-trained dermatologic surgeon (M.F. or T.H.) via online teleconferencing.

During the coaching session, participants and coaches reviewed the video. The surgical coaches also scored the residents using the OSATS, then residents and coaches discussed how the resident could improve using the OSATS scores as a guide. The residents then completed a poststudy survey assessing their surgical experience, confidence in performing surgery, and attitudes on VBC. Descriptive statistics were reported.

On average, residents spent 31.3 minutes reviewing their own surgeries and scoring themselves. The average time for a coaching session, which included time spent scoring, was 13.8 minutes. Residents scored themselves lower than the surgical coaches did by an average of 5.25 points (eTable 2). Residents gave themselves an average total score of 10.5, while their respective surgical coaches gave the residents an average score of 15.75. There was a trend of residents with greater surgical experience having higher OSATS scores (Figure). After the coaching session, 3 of 4 residents reported that they felt more confident in their surgical skills. All residents felt more confident in assessing their surgical skills and felt that VBC was an effective teaching measure. All residents agreed that VBC should be continued as part of their residency training.

Video-based coaching has the potential to provide several benefits for dermatology trainees. Because receiving feedback intraoperatively often can be distracting and incomplete, video review can instead allow the surgeon to focus on performing the surgery and then later focus on learning while reviewing the video.^1,4 Feedback also can be more comprehensive and delivered without concern for time constraints or disturbing clinic flow as well as without the additional concern of the patient overhearing comments and feedback.³ Although independent video review in the absence of coaching can lead to improvement in surgical skills, the addition of VBC provides even greater potential educational benefit.⁴ During the COVID-19 pandemic, VBC allowed coaches to provide feedback without additional exposures. We utilized dermatologic surgery faculty as coaches, but this format of training also would apply to general dermatology faculty.

Another goal of VBC is to enhance a trainee’s ability to perform self-directed learning, which requires accurate self-assessment.⁴ Accurately assessing one’s own strengths empowers a trainee to act with appropriate confidence, while understanding one’s own weaknesses allows a trainee to effectively balance confidence and caution in daily practice.⁵ Interestingly, in our study all residents scored themselves lower than surgical coaches, but with 1 coaching session, the residents subsequently reported greater surgical confidence.

Time constraints can be a potential barrier to surgical coaching.⁴ Our study demonstrates that VBC requires minimal time investment. Increasing the speed of video playback allowed for efficient evaluation of resident surgeries without compromising the coach’s ability to provide comprehensive feedback. Our feedback sessions were performed virtually, which allowed for ease of scheduling between trainees and coaches.

Our pilot study demonstrated that VBC is relatively easy to implement in a dermatology residency training setting, leveraging relatively low-cost technologies and allowing for a means of learning that residents felt was effective. Video-based coaching requires minimal time investment from both trainees and coaches and has the potential to enhance surgical confidence. Our current study is limited by its small sample size. Future studies should include follow-up recordings and assess the efficacy of VBC in enhancing surgical skills.

To the Editor:

Video-based coaching (VBC) involves a surgeon recording a surgery and then reviewing the video with a surgical coach; it is a form of education that is gaining popularity among surgical specialties.¹ Video-based education is underutilized in dermatology residency training.² We conducted a pilot study at our dermatology residency program to evaluate the efficacy and feasibility of VBC.

The University of Texas at Austin Dell Medical School institutional review board approved this study. All 4 first-year dermatology residents were recruited to participate in this study. Participants filled out a prestudy survey assessing their surgical experience, confidence in performing surgery, and attitudes on VBC. Participants used a head-mounted point-of-view camera to record themselves performing a wide local excision on the trunk or extremities of a live human patient. Participants then reviewed the recording on their own and scored themselves using the Objective Structured Assessment of Technical Skills (OSATS) scoring table (scored from 1 to 5, with 5 being the highest possible score for each element), which is a validated tool for assessing surgical skills (eTable 1).³ Given that there were no assistants participating in the surgery, this element of the OSATS scoring table was excluded, making a maximum possible score of 30 and a minimum possible score of 6. After scoring themselves, participants then had a 1-on-1 coaching session with a fellowship-trained dermatologic surgeon (M.F. or T.H.) via online teleconferencing.

During the coaching session, participants and coaches reviewed the video. The surgical coaches also scored the residents using the OSATS, then residents and coaches discussed how the resident could improve using the OSATS scores as a guide. The residents then completed a poststudy survey assessing their surgical experience, confidence in performing surgery, and attitudes on VBC. Descriptive statistics were reported.

On average, residents spent 31.3 minutes reviewing their own surgeries and scoring themselves. The average time for a coaching session, which included time spent scoring, was 13.8 minutes. Residents scored themselves lower than the surgical coaches did by an average of 5.25 points (eTable 2). Residents gave themselves an average total score of 10.5, while their respective surgical coaches gave the residents an average score of 15.75. There was a trend of residents with greater surgical experience having higher OSATS scores (Figure). After the coaching session, 3 of 4 residents reported that they felt more confident in their surgical skills. All residents felt more confident in assessing their surgical skills and felt that VBC was an effective teaching measure. All residents agreed that VBC should be continued as part of their residency training.

Video-based coaching has the potential to provide several benefits for dermatology trainees. Because receiving feedback intraoperatively often can be distracting and incomplete, video review can instead allow the surgeon to focus on performing the surgery and then later focus on learning while reviewing the video.^1,4 Feedback also can be more comprehensive and delivered without concern for time constraints or disturbing clinic flow as well as without the additional concern of the patient overhearing comments and feedback.³ Although independent video review in the absence of coaching can lead to improvement in surgical skills, the addition of VBC provides even greater potential educational benefit.⁴ During the COVID-19 pandemic, VBC allowed coaches to provide feedback without additional exposures. We utilized dermatologic surgery faculty as coaches, but this format of training also would apply to general dermatology faculty.

Another goal of VBC is to enhance a trainee’s ability to perform self-directed learning, which requires accurate self-assessment.⁴ Accurately assessing one’s own strengths empowers a trainee to act with appropriate confidence, while understanding one’s own weaknesses allows a trainee to effectively balance confidence and caution in daily practice.⁵ Interestingly, in our study all residents scored themselves lower than surgical coaches, but with 1 coaching session, the residents subsequently reported greater surgical confidence.

Time constraints can be a potential barrier to surgical coaching.⁴ Our study demonstrates that VBC requires minimal time investment. Increasing the speed of video playback allowed for efficient evaluation of resident surgeries without compromising the coach’s ability to provide comprehensive feedback. Our feedback sessions were performed virtually, which allowed for ease of scheduling between trainees and coaches.

Our pilot study demonstrated that VBC is relatively easy to implement in a dermatology residency training setting, leveraging relatively low-cost technologies and allowing for a means of learning that residents felt was effective. Video-based coaching requires minimal time investment from both trainees and coaches and has the potential to enhance surgical confidence. Our current study is limited by its small sample size. Future studies should include follow-up recordings and assess the efficacy of VBC in enhancing surgical skills.

Dermatology residency positions continue to be highly coveted among applicants in the match. In 2019, dermatology proved to be the most competitive specialty, with 36.3% of US medical school seniors and independent applicants going unmatched.¹ Prior to the transition to a pass/fail system, the mean US Medical Licensing Examination (USMLE) Step 1 score for matched applicants increased from 247 in 2014 to 251 in 2019. The growing number of scholarly activities reported by applicants has contributed to the competitiveness of the specialty. In 2018, the mean number of abstracts, presentations, and publications reported by matched applicants was 14.71, which was higher than other competitive specialties, including orthopedic surgery and otolaryngology (11.5 and 10.4, respectively). Dermatology applicants who did not match in 2018 reported a mean of 8.6 abstracts, presentations, and publications, which was on par with successful applicants in many other specialties.¹ In 2011, Stratman and Ness² found that publishing manuscripts and listing research experience were factors strongly associated with matching into dermatology for reapplicants. These trends in reported research have added pressure for applicants to increase their publications.

Given that many students do not choose a career in dermatology until later in medical school, some students choose to take a gap year between their third and fourth years of medical school to pursue a research fellowship (RF) and produce publications, in theory to increase the chances of matching in dermatology. A survey of dermatology applicants conducted by Costello et al³ in 2021 found that, of the students who completed a gap year (n=90; 31.25%), 78.7% (n=71) of them completed an RF, and those who completed RFs were more likely to match at top dermatology residency programs (P<.01). The authors also reported that there was no significant difference in overall match rates between gap-year and non–gap-year applicants.³ Another survey of 328 medical students found that the most common reason students take years off for research during medical school is to increase competitiveness for residency application.⁴ Although it is clear that students completing an RF often find success in the match, there are limited published data on how those involved in selecting dermatology residents view this additional year. We surveyed faculty members participating in the resident selection process to assess their viewpoints on how RFs factored into an applicant’s odds of matching into dermatology residency and performance as a resident.

Materials and Methods

An institutional review board application was submitted through the Geisinger Health System (Danville, Pennsylvania), and an exemption to complete the survey was granted. The survey consisted of 16 questions via REDCap electronic data capture and was sent to a listserve of dermatology program directors who were asked to distribute the survey to program chairs and faculty members within their department. Survey questions evaluated the participants’ involvement in medical student advising and the residency selection process. Questions relating to the respondents’ opinions were based on a 5-point Likert scale on level of agreement (1=strongly agree; 5=strongly disagree) or importance (1=a great deal; 5=not at all). All responses were collected anonymously. Data points were compiled and analyzed using REDCap. Statistical analysis via χ² tests were conducted when appropriate.

Results

The survey was sent to 142 individuals and distributed to faculty members within those departments between August 16, 2019, and September 24, 2019. The survey elicited a total of 110 respondents. Demographic information is shown in eTable 1. Of these respondents, 35.5% were program directors, 23.6% were program chairs, 3.6% were both program director and program chair, and 37.3% were core faculty members. Although respondents’ roles were varied, 96.4% indicated that they were involved in both advising medical students and in selecting residents.

None of the respondents indicated that they always recommend that students complete an RF, and only 4.5% indicated that they usually recommend it; 40% of respondents rarely or never recommend an RF, while 55.5% sometimes recommend it. Although there was a variety of responses to how frequently faculty members recommend an RF, almost all respondents (98.2%) agreed that the reason medical students pursued an RF prior to residency application was to increase the competitiveness of their residency application. However, 20% of respondents believed that students in this cohort were seeking to gain a deeper understanding of the specialty, and 27.3% thought that this cohort had genuine interest in research. Interestingly, despite the medical students’ intentions of choosing an RF, most respondents (67.3%) agreed or strongly agreed that the publications produced by fellows make an impact on the dermatologic scientific community.

Although some respondents indicated that completion of an RF positively impacts resident performance with regard to patient care, most indicated that the impact was a little (26.4%) or not at all (50%). Additionally, a minority of respondents (11.8%) believed that RFs positively impact resident performance on in-service and board examinations at least a moderate amount, with 62.7% indicating no positive impact at all. Only 12.7% of participants agreed or strongly agreed that completion of an RF led to increased applicant involvement in research throughout their career, and most (73.6%) believed there were downsides to completing an RF. Finally, only 20% agreed or strongly agreed that students who completed an RF were more dedicated to the field of dermatology (eTable 2).

Further evaluation of the data indicated that the perceived utility of RFs did not affect respondents’ recommendation on whether to pursue an RF or not. For example, of the 4.5% of respondents who indicated that they always or usually recommended RFs, only 1 respondent believed that students who completed an RF were more dedicated to the field of dermatology than those who did not. Although 55.5% of respondents answered that they sometimes recommended completion of an RF, less than a quarter of this group believed that students who completed an RF were more likely to be heavily involved in research throughout their career (P=.99).

Overall, 11.8% of respondents indicated that completion of a dermatology RF influenced the evaluation of an applicant a great deal or a lot, while 53.6% of respondents indicated a little or no influence at all. Most respondents (62.8%) agreed or strongly agreed that completion of an RF can compensate for flaws in a residency application. Furthermore, when asked if completion of an RF could set 2 otherwise equivocal applicants apart from one another, 46.4% of respondents agreed or strongly agreed with the statement, while only 17.3% disagreed or strongly disagreed (eTable 2).

Comment

This study characterized how completion of an RF is viewed by those involved in advising medical students and selecting dermatology residents. The growing pressure for applicants to increase the number of publications combined with the competitiveness of applying for a dermatology residency position has led to increased participation in RFs. However, studies have found that students who completed an RF often did so despite a lack of interest.⁴ Nonetheless, little is known about how this is perceived by those involved in choosing residents.

We found that few respondents always or usually advised applicants to complete an RF, but the majority sometimes recommended them, demonstrating the complexity of this issue. Completion of an RF impacted 11.8% of respondents’ overall opinion of an applicant a lot or a great deal, while most respondents (53.6%) were influenced a little or not at all. However, 46.4% of respondents indicated that completion of a dermatology RF would set apart 2 applicants of otherwise equal standing, and 62.8% agreed or strongly agreed that completion of an RF would compensate for flaws in an application. These responses align with the findings of a study conducted by Kaffenberger et al,⁵ who surveyed members of the Association of Professors of Dermatology and found that 74.5% (73/98) of mentors almost always or sometimes recommended a research gap year for reasons that included low grades, low USMLE Step scores, and little research. These data suggest that completion of an RF can give a competitive advantage to applicants despite most advisors acknowledging that these applicants are not likely to be involved in research throughout their careers, perform better on standardized examinations, or provide better patient care.

Given the complexity of this issue, respondents may not have been able to accurately answer the question about how much an RF influenced their overall opinion of an applicant because of subconscious bias. Furthermore, respondents likely tailored their recommendations to complete an RF based on individual applicant strengths and weaknesses, and the specific reasons why one may recommend an RF need to be further investigated.

Although there may be other perceived advantages to RFs that were not captured by our survey, completion of a dermatology RF is not without disadvantages. Fellowships often are unfunded and offered in cities with high costs of living. Additionally, students are forced to delay graduation from medical school by a year at minimum and continue to accrue interest on medical school loans during this time. The financial burdens of completing an RF may exclude students of lower socioeconomic status and contribute to a decrease in diversity within the field. Dermatology has been found to be the second least diverse specialty, behind orthopedics.⁶ Soliman et al⁷ found that racial minorities and low-income students were more likely to cite socioeconomic barriers as factors involved in their decision not to pursue a career in dermatology. This notion was supported by Rinderknecht et al,⁸ who found that Black and Latinx dermatology applicants were more likely to come from disadvantaged backgrounds, and Black applicants were more likely to indicate financial concerns as their primary reason for not pursuing an RF. The impact of accumulated student debt and decreased access should be carefully weighed against the potential benefits of an RF. However, as the USMLE transitions their Step 1 score reporting from numerical to a pass/fail system, it also is possible that dermatology programs will place more emphasis on research productivity when evaluating applications for residency. Overall, the decision to recommend an RF represents an extremely complex topic, as indicated by the results of this study.

Limitations—Our survey-based study is limited by response rate and response bias. Despite the large number of responses, the overall response rate cannot be determined because it is unknown how many total faculty members actually received the survey. Moreover, data collected from current dermatology residents who have completed RFs vs those who have not as they pertain to resident performance and preparedness for the rigors of a dermatology residency would be useful.

Dermatology residency positions continue to be highly coveted among applicants in the match. In 2019, dermatology proved to be the most competitive specialty, with 36.3% of US medical school seniors and independent applicants going unmatched.¹ Prior to the transition to a pass/fail system, the mean US Medical Licensing Examination (USMLE) Step 1 score for matched applicants increased from 247 in 2014 to 251 in 2019. The growing number of scholarly activities reported by applicants has contributed to the competitiveness of the specialty. In 2018, the mean number of abstracts, presentations, and publications reported by matched applicants was 14.71, which was higher than other competitive specialties, including orthopedic surgery and otolaryngology (11.5 and 10.4, respectively). Dermatology applicants who did not match in 2018 reported a mean of 8.6 abstracts, presentations, and publications, which was on par with successful applicants in many other specialties.¹ In 2011, Stratman and Ness² found that publishing manuscripts and listing research experience were factors strongly associated with matching into dermatology for reapplicants. These trends in reported research have added pressure for applicants to increase their publications.

Given that many students do not choose a career in dermatology until later in medical school, some students choose to take a gap year between their third and fourth years of medical school to pursue a research fellowship (RF) and produce publications, in theory to increase the chances of matching in dermatology. A survey of dermatology applicants conducted by Costello et al³ in 2021 found that, of the students who completed a gap year (n=90; 31.25%), 78.7% (n=71) of them completed an RF, and those who completed RFs were more likely to match at top dermatology residency programs (P<.01). The authors also reported that there was no significant difference in overall match rates between gap-year and non–gap-year applicants.³ Another survey of 328 medical students found that the most common reason students take years off for research during medical school is to increase competitiveness for residency application.⁴ Although it is clear that students completing an RF often find success in the match, there are limited published data on how those involved in selecting dermatology residents view this additional year. We surveyed faculty members participating in the resident selection process to assess their viewpoints on how RFs factored into an applicant’s odds of matching into dermatology residency and performance as a resident.

Materials and Methods

An institutional review board application was submitted through the Geisinger Health System (Danville, Pennsylvania), and an exemption to complete the survey was granted. The survey consisted of 16 questions via REDCap electronic data capture and was sent to a listserve of dermatology program directors who were asked to distribute the survey to program chairs and faculty members within their department. Survey questions evaluated the participants’ involvement in medical student advising and the residency selection process. Questions relating to the respondents’ opinions were based on a 5-point Likert scale on level of agreement (1=strongly agree; 5=strongly disagree) or importance (1=a great deal; 5=not at all). All responses were collected anonymously. Data points were compiled and analyzed using REDCap. Statistical analysis via χ² tests were conducted when appropriate.

Results

The survey was sent to 142 individuals and distributed to faculty members within those departments between August 16, 2019, and September 24, 2019. The survey elicited a total of 110 respondents. Demographic information is shown in eTable 1. Of these respondents, 35.5% were program directors, 23.6% were program chairs, 3.6% were both program director and program chair, and 37.3% were core faculty members. Although respondents’ roles were varied, 96.4% indicated that they were involved in both advising medical students and in selecting residents.

None of the respondents indicated that they always recommend that students complete an RF, and only 4.5% indicated that they usually recommend it; 40% of respondents rarely or never recommend an RF, while 55.5% sometimes recommend it. Although there was a variety of responses to how frequently faculty members recommend an RF, almost all respondents (98.2%) agreed that the reason medical students pursued an RF prior to residency application was to increase the competitiveness of their residency application. However, 20% of respondents believed that students in this cohort were seeking to gain a deeper understanding of the specialty, and 27.3% thought that this cohort had genuine interest in research. Interestingly, despite the medical students’ intentions of choosing an RF, most respondents (67.3%) agreed or strongly agreed that the publications produced by fellows make an impact on the dermatologic scientific community.

Although some respondents indicated that completion of an RF positively impacts resident performance with regard to patient care, most indicated that the impact was a little (26.4%) or not at all (50%). Additionally, a minority of respondents (11.8%) believed that RFs positively impact resident performance on in-service and board examinations at least a moderate amount, with 62.7% indicating no positive impact at all. Only 12.7% of participants agreed or strongly agreed that completion of an RF led to increased applicant involvement in research throughout their career, and most (73.6%) believed there were downsides to completing an RF. Finally, only 20% agreed or strongly agreed that students who completed an RF were more dedicated to the field of dermatology (eTable 2).

Further evaluation of the data indicated that the perceived utility of RFs did not affect respondents’ recommendation on whether to pursue an RF or not. For example, of the 4.5% of respondents who indicated that they always or usually recommended RFs, only 1 respondent believed that students who completed an RF were more dedicated to the field of dermatology than those who did not. Although 55.5% of respondents answered that they sometimes recommended completion of an RF, less than a quarter of this group believed that students who completed an RF were more likely to be heavily involved in research throughout their career (P=.99).

Overall, 11.8% of respondents indicated that completion of a dermatology RF influenced the evaluation of an applicant a great deal or a lot, while 53.6% of respondents indicated a little or no influence at all. Most respondents (62.8%) agreed or strongly agreed that completion of an RF can compensate for flaws in a residency application. Furthermore, when asked if completion of an RF could set 2 otherwise equivocal applicants apart from one another, 46.4% of respondents agreed or strongly agreed with the statement, while only 17.3% disagreed or strongly disagreed (eTable 2).

Comment

This study characterized how completion of an RF is viewed by those involved in advising medical students and selecting dermatology residents. The growing pressure for applicants to increase the number of publications combined with the competitiveness of applying for a dermatology residency position has led to increased participation in RFs. However, studies have found that students who completed an RF often did so despite a lack of interest.⁴ Nonetheless, little is known about how this is perceived by those involved in choosing residents.

We found that few respondents always or usually advised applicants to complete an RF, but the majority sometimes recommended them, demonstrating the complexity of this issue. Completion of an RF impacted 11.8% of respondents’ overall opinion of an applicant a lot or a great deal, while most respondents (53.6%) were influenced a little or not at all. However, 46.4% of respondents indicated that completion of a dermatology RF would set apart 2 applicants of otherwise equal standing, and 62.8% agreed or strongly agreed that completion of an RF would compensate for flaws in an application. These responses align with the findings of a study conducted by Kaffenberger et al,⁵ who surveyed members of the Association of Professors of Dermatology and found that 74.5% (73/98) of mentors almost always or sometimes recommended a research gap year for reasons that included low grades, low USMLE Step scores, and little research. These data suggest that completion of an RF can give a competitive advantage to applicants despite most advisors acknowledging that these applicants are not likely to be involved in research throughout their careers, perform better on standardized examinations, or provide better patient care.

Given the complexity of this issue, respondents may not have been able to accurately answer the question about how much an RF influenced their overall opinion of an applicant because of subconscious bias. Furthermore, respondents likely tailored their recommendations to complete an RF based on individual applicant strengths and weaknesses, and the specific reasons why one may recommend an RF need to be further investigated.

Although there may be other perceived advantages to RFs that were not captured by our survey, completion of a dermatology RF is not without disadvantages. Fellowships often are unfunded and offered in cities with high costs of living. Additionally, students are forced to delay graduation from medical school by a year at minimum and continue to accrue interest on medical school loans during this time. The financial burdens of completing an RF may exclude students of lower socioeconomic status and contribute to a decrease in diversity within the field. Dermatology has been found to be the second least diverse specialty, behind orthopedics.⁶ Soliman et al⁷ found that racial minorities and low-income students were more likely to cite socioeconomic barriers as factors involved in their decision not to pursue a career in dermatology. This notion was supported by Rinderknecht et al,⁸ who found that Black and Latinx dermatology applicants were more likely to come from disadvantaged backgrounds, and Black applicants were more likely to indicate financial concerns as their primary reason for not pursuing an RF. The impact of accumulated student debt and decreased access should be carefully weighed against the potential benefits of an RF. However, as the USMLE transitions their Step 1 score reporting from numerical to a pass/fail system, it also is possible that dermatology programs will place more emphasis on research productivity when evaluating applications for residency. Overall, the decision to recommend an RF represents an extremely complex topic, as indicated by the results of this study.

Limitations—Our survey-based study is limited by response rate and response bias. Despite the large number of responses, the overall response rate cannot be determined because it is unknown how many total faculty members actually received the survey. Moreover, data collected from current dermatology residents who have completed RFs vs those who have not as they pertain to resident performance and preparedness for the rigors of a dermatology residency would be useful.

Dermatology residency positions continue to be highly coveted among applicants in the match. In 2019, dermatology proved to be the most competitive specialty, with 36.3% of US medical school seniors and independent applicants going unmatched.¹ Prior to the transition to a pass/fail system, the mean US Medical Licensing Examination (USMLE) Step 1 score for matched applicants increased from 247 in 2014 to 251 in 2019. The growing number of scholarly activities reported by applicants has contributed to the competitiveness of the specialty. In 2018, the mean number of abstracts, presentations, and publications reported by matched applicants was 14.71, which was higher than other competitive specialties, including orthopedic surgery and otolaryngology (11.5 and 10.4, respectively). Dermatology applicants who did not match in 2018 reported a mean of 8.6 abstracts, presentations, and publications, which was on par with successful applicants in many other specialties.¹ In 2011, Stratman and Ness² found that publishing manuscripts and listing research experience were factors strongly associated with matching into dermatology for reapplicants. These trends in reported research have added pressure for applicants to increase their publications.

Given that many students do not choose a career in dermatology until later in medical school, some students choose to take a gap year between their third and fourth years of medical school to pursue a research fellowship (RF) and produce publications, in theory to increase the chances of matching in dermatology. A survey of dermatology applicants conducted by Costello et al³ in 2021 found that, of the students who completed a gap year (n=90; 31.25%), 78.7% (n=71) of them completed an RF, and those who completed RFs were more likely to match at top dermatology residency programs (P<.01). The authors also reported that there was no significant difference in overall match rates between gap-year and non–gap-year applicants.³ Another survey of 328 medical students found that the most common reason students take years off for research during medical school is to increase competitiveness for residency application.⁴ Although it is clear that students completing an RF often find success in the match, there are limited published data on how those involved in selecting dermatology residents view this additional year. We surveyed faculty members participating in the resident selection process to assess their viewpoints on how RFs factored into an applicant’s odds of matching into dermatology residency and performance as a resident.

Materials and Methods

An institutional review board application was submitted through the Geisinger Health System (Danville, Pennsylvania), and an exemption to complete the survey was granted. The survey consisted of 16 questions via REDCap electronic data capture and was sent to a listserve of dermatology program directors who were asked to distribute the survey to program chairs and faculty members within their department. Survey questions evaluated the participants’ involvement in medical student advising and the residency selection process. Questions relating to the respondents’ opinions were based on a 5-point Likert scale on level of agreement (1=strongly agree; 5=strongly disagree) or importance (1=a great deal; 5=not at all). All responses were collected anonymously. Data points were compiled and analyzed using REDCap. Statistical analysis via χ² tests were conducted when appropriate.

Results

The survey was sent to 142 individuals and distributed to faculty members within those departments between August 16, 2019, and September 24, 2019. The survey elicited a total of 110 respondents. Demographic information is shown in eTable 1. Of these respondents, 35.5% were program directors, 23.6% were program chairs, 3.6% were both program director and program chair, and 37.3% were core faculty members. Although respondents’ roles were varied, 96.4% indicated that they were involved in both advising medical students and in selecting residents.

None of the respondents indicated that they always recommend that students complete an RF, and only 4.5% indicated that they usually recommend it; 40% of respondents rarely or never recommend an RF, while 55.5% sometimes recommend it. Although there was a variety of responses to how frequently faculty members recommend an RF, almost all respondents (98.2%) agreed that the reason medical students pursued an RF prior to residency application was to increase the competitiveness of their residency application. However, 20% of respondents believed that students in this cohort were seeking to gain a deeper understanding of the specialty, and 27.3% thought that this cohort had genuine interest in research. Interestingly, despite the medical students’ intentions of choosing an RF, most respondents (67.3%) agreed or strongly agreed that the publications produced by fellows make an impact on the dermatologic scientific community.

Although some respondents indicated that completion of an RF positively impacts resident performance with regard to patient care, most indicated that the impact was a little (26.4%) or not at all (50%). Additionally, a minority of respondents (11.8%) believed that RFs positively impact resident performance on in-service and board examinations at least a moderate amount, with 62.7% indicating no positive impact at all. Only 12.7% of participants agreed or strongly agreed that completion of an RF led to increased applicant involvement in research throughout their career, and most (73.6%) believed there were downsides to completing an RF. Finally, only 20% agreed or strongly agreed that students who completed an RF were more dedicated to the field of dermatology (eTable 2).

Further evaluation of the data indicated that the perceived utility of RFs did not affect respondents’ recommendation on whether to pursue an RF or not. For example, of the 4.5% of respondents who indicated that they always or usually recommended RFs, only 1 respondent believed that students who completed an RF were more dedicated to the field of dermatology than those who did not. Although 55.5% of respondents answered that they sometimes recommended completion of an RF, less than a quarter of this group believed that students who completed an RF were more likely to be heavily involved in research throughout their career (P=.99).

Overall, 11.8% of respondents indicated that completion of a dermatology RF influenced the evaluation of an applicant a great deal or a lot, while 53.6% of respondents indicated a little or no influence at all. Most respondents (62.8%) agreed or strongly agreed that completion of an RF can compensate for flaws in a residency application. Furthermore, when asked if completion of an RF could set 2 otherwise equivocal applicants apart from one another, 46.4% of respondents agreed or strongly agreed with the statement, while only 17.3% disagreed or strongly disagreed (eTable 2).

Comment

This study characterized how completion of an RF is viewed by those involved in advising medical students and selecting dermatology residents. The growing pressure for applicants to increase the number of publications combined with the competitiveness of applying for a dermatology residency position has led to increased participation in RFs. However, studies have found that students who completed an RF often did so despite a lack of interest.⁴ Nonetheless, little is known about how this is perceived by those involved in choosing residents.

We found that few respondents always or usually advised applicants to complete an RF, but the majority sometimes recommended them, demonstrating the complexity of this issue. Completion of an RF impacted 11.8% of respondents’ overall opinion of an applicant a lot or a great deal, while most respondents (53.6%) were influenced a little or not at all. However, 46.4% of respondents indicated that completion of a dermatology RF would set apart 2 applicants of otherwise equal standing, and 62.8% agreed or strongly agreed that completion of an RF would compensate for flaws in an application. These responses align with the findings of a study conducted by Kaffenberger et al,⁵ who surveyed members of the Association of Professors of Dermatology and found that 74.5% (73/98) of mentors almost always or sometimes recommended a research gap year for reasons that included low grades, low USMLE Step scores, and little research. These data suggest that completion of an RF can give a competitive advantage to applicants despite most advisors acknowledging that these applicants are not likely to be involved in research throughout their careers, perform better on standardized examinations, or provide better patient care.

Given the complexity of this issue, respondents may not have been able to accurately answer the question about how much an RF influenced their overall opinion of an applicant because of subconscious bias. Furthermore, respondents likely tailored their recommendations to complete an RF based on individual applicant strengths and weaknesses, and the specific reasons why one may recommend an RF need to be further investigated.

Although there may be other perceived advantages to RFs that were not captured by our survey, completion of a dermatology RF is not without disadvantages. Fellowships often are unfunded and offered in cities with high costs of living. Additionally, students are forced to delay graduation from medical school by a year at minimum and continue to accrue interest on medical school loans during this time. The financial burdens of completing an RF may exclude students of lower socioeconomic status and contribute to a decrease in diversity within the field. Dermatology has been found to be the second least diverse specialty, behind orthopedics.⁶ Soliman et al⁷ found that racial minorities and low-income students were more likely to cite socioeconomic barriers as factors involved in their decision not to pursue a career in dermatology. This notion was supported by Rinderknecht et al,⁸ who found that Black and Latinx dermatology applicants were more likely to come from disadvantaged backgrounds, and Black applicants were more likely to indicate financial concerns as their primary reason for not pursuing an RF. The impact of accumulated student debt and decreased access should be carefully weighed against the potential benefits of an RF. However, as the USMLE transitions their Step 1 score reporting from numerical to a pass/fail system, it also is possible that dermatology programs will place more emphasis on research productivity when evaluating applications for residency. Overall, the decision to recommend an RF represents an extremely complex topic, as indicated by the results of this study.

Limitations—Our survey-based study is limited by response rate and response bias. Despite the large number of responses, the overall response rate cannot be determined because it is unknown how many total faculty members actually received the survey. Moreover, data collected from current dermatology residents who have completed RFs vs those who have not as they pertain to resident performance and preparedness for the rigors of a dermatology residency would be useful.

Melasma is a complex, long-lasting, acquired dermatologic pigmentation disorder resulting in grey-brown patches that last for more than 3 months. Sun-exposed areas including the nose, cheeks, forehead, and forearms are most likely to be affected.¹ In Southeast Asia, 0.25% to 4% of the population affected by melasma is aged 30 to 40 years.² In particular, melasma is a concern among pregnant women due to increased levels of melanocyte-stimulating hormones (MSHs) and is impacted by genetics, hormonal influence, and exposure to UV light.^3,4 In Pakistan, approximately 46% of women are affected by melasma during pregnancy.^2,5 Although few studies have focused on the clinical approaches to melasma in darker skin types, it continues to disproportionately affect the skin of color population.⁴

The areas of hyperpigmentation seen in melasma exhibit increased deposition of melanin in the epidermis and dermis, but melanocytes are not elevated. However, in areas of hyperpigmentation, the melanocytes are larger and more dendritic and demonstrate an increased level of melanogenesis.⁶ During pregnancy, especially in the third trimester, elevated levels of estrogen, progesterone, and MSH often are found in association with melasma.⁷ Tyrosinase (TYR) activity increases and cellular proliferation is reduced after treatment of melanocytes in culture with β-estradiol.⁸ Sex steroids increase transcription of genes encoding melanogenic enzymes in normal human melanocytes, especially TYR.⁹ These results are consistent with the notable increases in melanin synthesis and TYR activity reported for normal human melanocytes under similar conditions in culture.¹⁰ Because melanocytes contain both cytosolic and nuclear estrogen receptors, melanocytes in patients with melasma may be inherently more sensitive to the stimulatory effects of estrogens and possibly other steroid hormones.¹¹

The current treatment options for melasma have varying levels of success and include topical depigmenting agents such as hydroquinone, tretinoin, azelaic acid, kojic acid, and corticosteroids; dermabrasion; and chemical peels.^12-14 Chemical peels with glycolic acid, salicylic acid, lactic acid, trichloroacetic acid, and phenol, as well as laser therapy, are reliable management options.^13,14 Traditionally, melasma has been treated with a combination of modalities along with photoprotection and trigger avoidance.¹²

The efficacy and safety of the available therapies for melasma are still controversial and require further exploration. In recent years, off-label tranexamic acid (TA) has emerged as a potential therapy for melasma. Although the mechanism of action remains unclear, TA may inhibit melanin synthesis by blocking the interaction between melanocytes and keratinocytes.¹⁵ Tranexamic acid also may reverse the abnormal dermal changes associated with melasma by inhibiting melanogenesis and angiogenesis.¹⁶

Although various therapeutic options exist for melasma, the search for a reliable option in patients with darker skin types continues.¹³ We sought to evaluate the efficacy of TA solution 5% in reducing the severity of melasma in South Asian patients, thereby improving patient outcomes and maximizing patient satisfaction. Topical TA is inexpensive and readily accessible and does not cause systemic side effects. These qualities make it a promising treatment compared to traditional therapies.

Methods

We conducted a randomized controlled trial at Rawalpindi Medical Institute (Punjab, Pakistan). The researchers obtained informed consent for all enrolled patients. Cases were sampled from the original patient population seen at the office using nonprobability consecutive sampling. The sample size was calculated with a 95% CI, margin of error of 9%, and expected percentage of efficacy of 86.1% by using TA solution 5%. South Asian male and female patients aged 20 to 45 years with melasma were included in the analysis. Patients were excluded if they were already taking TA, oral contraceptive pills, or photosensitizing drugs (eg, nonsteroidal anti-inflammatory drugs, tetracyclines, phenytoin, carbamazepine); were pregnant; had chronic kidney disease (creatinine >2.0 mg/dL); had cardiac abnormalities (abnormal electrocardiogram); had hematologic disorders (international normalized ratio >2); or had received another melasma treatment within the last 3 to 6 months.

All enrolled patients underwent a detailed history and physical examination. Patient demographics were subsequently noted, including age, sex, history of diabetes mellitus or hypertension, and duration of melasma. The melasma area and severity index (MASI) score of each patient was calculated at baseline, and a corresponding photograph was taken.

The topical solution was prepared with 5 g of TA dissolved in 10 cc of ethanol at 96 °F, 10 cc of 1,3-butanediol, and distilled water up to 100 cc. The TA solution was applied to the affected areas once daily by the patient for 12 weeks. Each application covered the affected areas completely. Patients were instructed to apply sunscreen with sun protection factor 60 to those same areas for UV protection after 15 minutes of TA application. Biweekly follow-ups were scheduled during the trial, and the MASI score was recorded at these visits. If the mean MASI score was reduced by half after 12 weeks of treatment, then the treatment was considered efficacious with a 95% CI.

The percentage reduction from baseline was calculated as follows: percentage reduction=(baseline score– follow-up score)/baseline score×100.

Statistical Analysis—Data were analyzed in SPSS Statistics 25 (IBM). The quantitative variables of age, duration of melasma, and body mass index were presented as mean (SD). Qualitative variables such as sex, history of diabetes mellitus or hypertension, site of melasma, and efficacy were presented as frequencies and percentages. Mean MASI scores at baseline and 12 weeks posttreatment were compared using a paired t test (P≤.05). Data were stratified for age, sex, history of diabetes mellitus or hypertension, site of melasma, and duration of melasma, and a χ² test was applied to compare efficacy in stratified groups (P≤.05).

Results

Sixty patients were enrolled in the study. Of them, 17 (28.33%) were male, and 43 (71.67%) were female (2:5 ratio). They ranged in age from 20 to 45 years (mean [SD], 31.93 [6.26] years). Thirty-seven patients (61.67%) were aged 31 to 45 years of age (Table 1). The mean (SD) duration of disease was 10.18 (2.10) months. The response to TA was recorded based on patient distribution according to the site of melasma as well as history of diabetes mellitus and hypertension.

Topical TA was found to be efficacious for melasma in 50 (83.33%) patients. The mean (SD) baseline and week 12 MASI scores were 23.15 (5.02) and 12.71 (4.10)(P<.0001), respectively (Table 2). The stratification of efficacy with respect to age, sex, duration of melasma, site of melasma, and history of diabetes mellitus or hypertension is shown in the eTable. The site of melasma was significant with respect to stratification of efficacy. On the forehead, TA was found to be efficacious in 11 patients and nonefficacious in 0 patients (P=.036). In the malar region, it was efficacious in 16 patients and nonefficacious in 1 patient (P=.036). Finally, on the chin, it was efficacious in 23 patients and nonefficacious in 9 patients (P=.036).

Comment

Melasma Presentation and Development—Melasma is a chronic skin condition that more often affects patients with darker skin types. This condition is characterized by hyperpigmentation of skin that is directly exposed to the sun, such as the cheek, nose, forehead, and above the upper lip.¹⁷ Although the mechanism behind how melasma develops is unknown, one theory suggests that UV light can lead to increased plasmin in keratinocytes.¹⁸ This increased plasmin will thereby increase the arachidonic acid and α-MSH, leading to the observed uneven hyperpigmentation that is notable in melasma. Melasma is common in patients using oral contraceptives or expired cosmetic drugs; in those who are pregnant; and in those with liver dysfunction.¹⁸ Melasma has a negative impact on patients’ quality of life because of substantial psychological and social distress. Thus, finding an accessible treatment is imperative.¹⁹

Melasma Management—The most common treatments for melasma have been topical bleaching agents and photoprotection. Combination therapy options include chemical peels, dermabrasion, and laser treatments, though they present with limited efficacy.^17,20 Because melasma focuses on pigmentation correction, topical treatments work to disturb melanocyte pigment production at the enzymatic level.²¹ Tyrosinase is rate limiting in melanin production, as it converts L-tyrosinase to L-3,4-dihydroxyphenylalanine, using copper to interact with L-3,4-dihydroxyphenylalanine as a cofactor in the active site.²² Therefore, tyrosine is a major target for many drugs that have been developed for melasma to decrease melaninization.²¹

Recently, research has focused on the effects of topical, intradermal, and oral TA for melasma.¹⁷ Tranexamic acid most commonly has been used in medicine as a fibrinolytic agent because of its antiplasmin properties. It has been hypothesized that TA can inhibit the release of paracrine melanogenic factors that normally act to stimulate melanocytes.¹⁷ Although studies have supported the safety and efficacy of TA, there remains a lack of clinical studies that are sufficiently powered. No definitive consensus on the use of TA for melasma currently exists, which indicates the need for large-scale, randomized, controlled trials.²³

One trial (N=25) found that TA solution 5% achieved efficacy (>50% reduction in MASI score from baseline) in 86.1% of patients with melasma.²⁴ In another study (N=18), topical TA 5% achieved efficacy (>50% reduction in MASI score) in 86% of patients with melasma.²⁵

Melasma Comorbidities—To determine if certain comorbidities, such as diabetes mellitus or hypertension, influenced the progression of melasma, we stratified the efficacy results for patients with these 2 comorbidities, which showed no significant difference (P=.794 and P=.101, respectively). Thus, the relatively higher prevalence of diabetes mellitus (16 patients) and hypertension (11 patients) did not contribute to the efficacy of TA in lowering MASI scores over the 12-week period, which supports the findings of Doolan and Gupta,²⁶ who investigated the endocrinologic conditions associated with melasma and found no such association with diabetes mellitus or hypertension.

TA Formulations for Melasma—The efficacy of topical TA has been explored in several studies. Six studies with sample sizes of 13 to 50 patients each showed statistically significant differences in MASI scores between baseline and following TA treatment (P<.001).^27-32 Several formulations and regimens were utilized, including TA cream 3% for 12 weeks, TA gel 5% for 12 weeks, TA solution 3% for 12 weeks, TA liposome 5% for 12 weeks, and TA solution 2% for 12 weeks.¹⁸ Additionally, these studies found TA to be effective in limiting dyschromia and decreasing MASI scores. There were no statistically significant differences between formulations and method of application. Topical TA has been found to be just as effective as other treatments for melasma, including intradermal TA injections, topical hydroquinone, and a combination of topical hydroquinone and dexamethasone.¹⁸

Further study of the efficacy of intradermal TA is necessary because many human trials have lacked statistical significance or a control group. Lee et al³² conducted a trial of 100 female patients who received weekly intradermal TA microinjections for 12 weeks. After 8 and 12 weeks, MASI scores decreased significantly (P<.01).³² Similarly, Badran et al³³ observed 60 female patients in 3 trial groups: group A received TA (4 mg/mL) intradermal injections every 2 weeks, group B received TA (10 mg/mL) intradermal injections every 2 weeks, and group C received TA cream 10% twice daily. Although all groups showed improvement in MASI, group B, which had the highest intradermal TA concentration, exhibited the most improvement. Thus, it was determined that intradermal application led to better results, but the cream was still effective.³³

Saki et al³⁴ conducted a randomized, split-face trial of 37 patients comparing the efficacy of intradermal TA and topical hydroquinone. Each group was treated with either monthly intradermal TA injections or nightly hydroquinone for 3 months. After 4 weeks of treatment, TA initially had a greater improvement. However, after 20 weeks, the overall changes were not significant between the 2 groups.³⁴ Pazyar et al³⁵ conducted a randomized, split-face trial of 49 patients comparing the efficacy of intradermal TA and hydroquinone cream. After 24 weeks of biweekly TA injections or twice-daily hydroquinone, there were no statistically significant differences in the decreased MASI scores between treatments.³⁵ Additional large, double-blind, controlled trials are needed to thoroughly assess the role of intradermal TA in comparison to its treatment counterpart of hydroquinone.

Ebrahimi and Naeini²⁹ conducted a 12-week, double-blind, split-phase trial of 50 Iranian melasma patients, which showed that 27.3% of patients rated the improvement in melasma as excellent, 42.4% as good, and 30.3% as fair after using TA solution 3%. Wu et al³⁶ also showed a total melasma improvement rate of 80.9% in 256 patients with long-term oral use of TA. In a study by Kim et al³¹ (N=245), the mean MASI score considerably decreased after topical TA use, with a total response rate of 95.6%. In another study, Atefi et al³⁷ presented significantly increased levels of satisfaction in patients treated with topical TA 5% vs hydroquinone (P=.015).

Melasma in Patients With Darker Skin Types—Special attention must be given to choosing the appropriate medication in melasma patients with darker skin types, as there is an increased risk for postinflammatory hyperpigmentation. Currently, few randomized controlled trials exist that fulfill the criteria of evaluating pharmacologic options for patients with melasma, and even fewer studies solely focus on patients with darker skin types.³⁸ In addition to treatment advances, patients must be educated on the need to avoid sun exposure when possible or to use photoprotection, especially in the South Asian region, where these practices rarely are taught. Our study provided a unique analysis regarding the efficacy of TA solution 5% for the treatment of melasma in patients of South Asian descent. Clinicians can use these findings as a foundation for treating all patients with melasma but particularly those with darker skin types.

Study Limitations—Our study consisted of 60 patients; although our study had more patients than similar trials, larger studies are needed. Additionally, other variables were excluded from our analysis, such as comorbidities beyond diabetes mellitus and hypertension.

Conclusion

This study contributes to the growing field of melasma therapeutics by evaluating the efficacy of using TA solution 5% for the treatment of melasma in South Asian patients with darker skin types. Clinicians may use our study to broaden their treatment options for a common condition while also addressing the lack of clinical options for patients with darker skin types. Further studies investigating the effectiveness of TA in large clinical trials in humans are warranted to understand the efficacy and the risk for any complications.

Melasma is a complex, long-lasting, acquired dermatologic pigmentation disorder resulting in grey-brown patches that last for more than 3 months. Sun-exposed areas including the nose, cheeks, forehead, and forearms are most likely to be affected.¹ In Southeast Asia, 0.25% to 4% of the population affected by melasma is aged 30 to 40 years.² In particular, melasma is a concern among pregnant women due to increased levels of melanocyte-stimulating hormones (MSHs) and is impacted by genetics, hormonal influence, and exposure to UV light.^3,4 In Pakistan, approximately 46% of women are affected by melasma during pregnancy.^2,5 Although few studies have focused on the clinical approaches to melasma in darker skin types, it continues to disproportionately affect the skin of color population.⁴

The areas of hyperpigmentation seen in melasma exhibit increased deposition of melanin in the epidermis and dermis, but melanocytes are not elevated. However, in areas of hyperpigmentation, the melanocytes are larger and more dendritic and demonstrate an increased level of melanogenesis.⁶ During pregnancy, especially in the third trimester, elevated levels of estrogen, progesterone, and MSH often are found in association with melasma.⁷ Tyrosinase (TYR) activity increases and cellular proliferation is reduced after treatment of melanocytes in culture with β-estradiol.⁸ Sex steroids increase transcription of genes encoding melanogenic enzymes in normal human melanocytes, especially TYR.⁹ These results are consistent with the notable increases in melanin synthesis and TYR activity reported for normal human melanocytes under similar conditions in culture.¹⁰ Because melanocytes contain both cytosolic and nuclear estrogen receptors, melanocytes in patients with melasma may be inherently more sensitive to the stimulatory effects of estrogens and possibly other steroid hormones.¹¹

The current treatment options for melasma have varying levels of success and include topical depigmenting agents such as hydroquinone, tretinoin, azelaic acid, kojic acid, and corticosteroids; dermabrasion; and chemical peels.^12-14 Chemical peels with glycolic acid, salicylic acid, lactic acid, trichloroacetic acid, and phenol, as well as laser therapy, are reliable management options.^13,14 Traditionally, melasma has been treated with a combination of modalities along with photoprotection and trigger avoidance.¹²

The efficacy and safety of the available therapies for melasma are still controversial and require further exploration. In recent years, off-label tranexamic acid (TA) has emerged as a potential therapy for melasma. Although the mechanism of action remains unclear, TA may inhibit melanin synthesis by blocking the interaction between melanocytes and keratinocytes.¹⁵ Tranexamic acid also may reverse the abnormal dermal changes associated with melasma by inhibiting melanogenesis and angiogenesis.¹⁶

Although various therapeutic options exist for melasma, the search for a reliable option in patients with darker skin types continues.¹³ We sought to evaluate the efficacy of TA solution 5% in reducing the severity of melasma in South Asian patients, thereby improving patient outcomes and maximizing patient satisfaction. Topical TA is inexpensive and readily accessible and does not cause systemic side effects. These qualities make it a promising treatment compared to traditional therapies.

Methods

We conducted a randomized controlled trial at Rawalpindi Medical Institute (Punjab, Pakistan). The researchers obtained informed consent for all enrolled patients. Cases were sampled from the original patient population seen at the office using nonprobability consecutive sampling. The sample size was calculated with a 95% CI, margin of error of 9%, and expected percentage of efficacy of 86.1% by using TA solution 5%. South Asian male and female patients aged 20 to 45 years with melasma were included in the analysis. Patients were excluded if they were already taking TA, oral contraceptive pills, or photosensitizing drugs (eg, nonsteroidal anti-inflammatory drugs, tetracyclines, phenytoin, carbamazepine); were pregnant; had chronic kidney disease (creatinine >2.0 mg/dL); had cardiac abnormalities (abnormal electrocardiogram); had hematologic disorders (international normalized ratio >2); or had received another melasma treatment within the last 3 to 6 months.

All enrolled patients underwent a detailed history and physical examination. Patient demographics were subsequently noted, including age, sex, history of diabetes mellitus or hypertension, and duration of melasma. The melasma area and severity index (MASI) score of each patient was calculated at baseline, and a corresponding photograph was taken.

The topical solution was prepared with 5 g of TA dissolved in 10 cc of ethanol at 96 °F, 10 cc of 1,3-butanediol, and distilled water up to 100 cc. The TA solution was applied to the affected areas once daily by the patient for 12 weeks. Each application covered the affected areas completely. Patients were instructed to apply sunscreen with sun protection factor 60 to those same areas for UV protection after 15 minutes of TA application. Biweekly follow-ups were scheduled during the trial, and the MASI score was recorded at these visits. If the mean MASI score was reduced by half after 12 weeks of treatment, then the treatment was considered efficacious with a 95% CI.

The percentage reduction from baseline was calculated as follows: percentage reduction=(baseline score– follow-up score)/baseline score×100.

Statistical Analysis—Data were analyzed in SPSS Statistics 25 (IBM). The quantitative variables of age, duration of melasma, and body mass index were presented as mean (SD). Qualitative variables such as sex, history of diabetes mellitus or hypertension, site of melasma, and efficacy were presented as frequencies and percentages. Mean MASI scores at baseline and 12 weeks posttreatment were compared using a paired t test (P≤.05). Data were stratified for age, sex, history of diabetes mellitus or hypertension, site of melasma, and duration of melasma, and a χ² test was applied to compare efficacy in stratified groups (P≤.05).

Results

Sixty patients were enrolled in the study. Of them, 17 (28.33%) were male, and 43 (71.67%) were female (2:5 ratio). They ranged in age from 20 to 45 years (mean [SD], 31.93 [6.26] years). Thirty-seven patients (61.67%) were aged 31 to 45 years of age (Table 1). The mean (SD) duration of disease was 10.18 (2.10) months. The response to TA was recorded based on patient distribution according to the site of melasma as well as history of diabetes mellitus and hypertension.

Topical TA was found to be efficacious for melasma in 50 (83.33%) patients. The mean (SD) baseline and week 12 MASI scores were 23.15 (5.02) and 12.71 (4.10)(P<.0001), respectively (Table 2). The stratification of efficacy with respect to age, sex, duration of melasma, site of melasma, and history of diabetes mellitus or hypertension is shown in the eTable. The site of melasma was significant with respect to stratification of efficacy. On the forehead, TA was found to be efficacious in 11 patients and nonefficacious in 0 patients (P=.036). In the malar region, it was efficacious in 16 patients and nonefficacious in 1 patient (P=.036). Finally, on the chin, it was efficacious in 23 patients and nonefficacious in 9 patients (P=.036).

Comment

Melasma Presentation and Development—Melasma is a chronic skin condition that more often affects patients with darker skin types. This condition is characterized by hyperpigmentation of skin that is directly exposed to the sun, such as the cheek, nose, forehead, and above the upper lip.¹⁷ Although the mechanism behind how melasma develops is unknown, one theory suggests that UV light can lead to increased plasmin in keratinocytes.¹⁸ This increased plasmin will thereby increase the arachidonic acid and α-MSH, leading to the observed uneven hyperpigmentation that is notable in melasma. Melasma is common in patients using oral contraceptives or expired cosmetic drugs; in those who are pregnant; and in those with liver dysfunction.¹⁸ Melasma has a negative impact on patients’ quality of life because of substantial psychological and social distress. Thus, finding an accessible treatment is imperative.¹⁹

Melasma Management—The most common treatments for melasma have been topical bleaching agents and photoprotection. Combination therapy options include chemical peels, dermabrasion, and laser treatments, though they present with limited efficacy.^17,20 Because melasma focuses on pigmentation correction, topical treatments work to disturb melanocyte pigment production at the enzymatic level.²¹ Tyrosinase is rate limiting in melanin production, as it converts L-tyrosinase to L-3,4-dihydroxyphenylalanine, using copper to interact with L-3,4-dihydroxyphenylalanine as a cofactor in the active site.²² Therefore, tyrosine is a major target for many drugs that have been developed for melasma to decrease melaninization.²¹

Recently, research has focused on the effects of topical, intradermal, and oral TA for melasma.¹⁷ Tranexamic acid most commonly has been used in medicine as a fibrinolytic agent because of its antiplasmin properties. It has been hypothesized that TA can inhibit the release of paracrine melanogenic factors that normally act to stimulate melanocytes.¹⁷ Although studies have supported the safety and efficacy of TA, there remains a lack of clinical studies that are sufficiently powered. No definitive consensus on the use of TA for melasma currently exists, which indicates the need for large-scale, randomized, controlled trials.²³

One trial (N=25) found that TA solution 5% achieved efficacy (>50% reduction in MASI score from baseline) in 86.1% of patients with melasma.²⁴ In another study (N=18), topical TA 5% achieved efficacy (>50% reduction in MASI score) in 86% of patients with melasma.²⁵

Melasma Comorbidities—To determine if certain comorbidities, such as diabetes mellitus or hypertension, influenced the progression of melasma, we stratified the efficacy results for patients with these 2 comorbidities, which showed no significant difference (P=.794 and P=.101, respectively). Thus, the relatively higher prevalence of diabetes mellitus (16 patients) and hypertension (11 patients) did not contribute to the efficacy of TA in lowering MASI scores over the 12-week period, which supports the findings of Doolan and Gupta,²⁶ who investigated the endocrinologic conditions associated with melasma and found no such association with diabetes mellitus or hypertension.

TA Formulations for Melasma—The efficacy of topical TA has been explored in several studies. Six studies with sample sizes of 13 to 50 patients each showed statistically significant differences in MASI scores between baseline and following TA treatment (P<.001).^27-32 Several formulations and regimens were utilized, including TA cream 3% for 12 weeks, TA gel 5% for 12 weeks, TA solution 3% for 12 weeks, TA liposome 5% for 12 weeks, and TA solution 2% for 12 weeks.¹⁸ Additionally, these studies found TA to be effective in limiting dyschromia and decreasing MASI scores. There were no statistically significant differences between formulations and method of application. Topical TA has been found to be just as effective as other treatments for melasma, including intradermal TA injections, topical hydroquinone, and a combination of topical hydroquinone and dexamethasone.¹⁸

Further study of the efficacy of intradermal TA is necessary because many human trials have lacked statistical significance or a control group. Lee et al³² conducted a trial of 100 female patients who received weekly intradermal TA microinjections for 12 weeks. After 8 and 12 weeks, MASI scores decreased significantly (P<.01).³² Similarly, Badran et al³³ observed 60 female patients in 3 trial groups: group A received TA (4 mg/mL) intradermal injections every 2 weeks, group B received TA (10 mg/mL) intradermal injections every 2 weeks, and group C received TA cream 10% twice daily. Although all groups showed improvement in MASI, group B, which had the highest intradermal TA concentration, exhibited the most improvement. Thus, it was determined that intradermal application led to better results, but the cream was still effective.³³

Saki et al³⁴ conducted a randomized, split-face trial of 37 patients comparing the efficacy of intradermal TA and topical hydroquinone. Each group was treated with either monthly intradermal TA injections or nightly hydroquinone for 3 months. After 4 weeks of treatment, TA initially had a greater improvement. However, after 20 weeks, the overall changes were not significant between the 2 groups.³⁴ Pazyar et al³⁵ conducted a randomized, split-face trial of 49 patients comparing the efficacy of intradermal TA and hydroquinone cream. After 24 weeks of biweekly TA injections or twice-daily hydroquinone, there were no statistically significant differences in the decreased MASI scores between treatments.³⁵ Additional large, double-blind, controlled trials are needed to thoroughly assess the role of intradermal TA in comparison to its treatment counterpart of hydroquinone.

Ebrahimi and Naeini²⁹ conducted a 12-week, double-blind, split-phase trial of 50 Iranian melasma patients, which showed that 27.3% of patients rated the improvement in melasma as excellent, 42.4% as good, and 30.3% as fair after using TA solution 3%. Wu et al³⁶ also showed a total melasma improvement rate of 80.9% in 256 patients with long-term oral use of TA. In a study by Kim et al³¹ (N=245), the mean MASI score considerably decreased after topical TA use, with a total response rate of 95.6%. In another study, Atefi et al³⁷ presented significantly increased levels of satisfaction in patients treated with topical TA 5% vs hydroquinone (P=.015).

Melasma in Patients With Darker Skin Types—Special attention must be given to choosing the appropriate medication in melasma patients with darker skin types, as there is an increased risk for postinflammatory hyperpigmentation. Currently, few randomized controlled trials exist that fulfill the criteria of evaluating pharmacologic options for patients with melasma, and even fewer studies solely focus on patients with darker skin types.³⁸ In addition to treatment advances, patients must be educated on the need to avoid sun exposure when possible or to use photoprotection, especially in the South Asian region, where these practices rarely are taught. Our study provided a unique analysis regarding the efficacy of TA solution 5% for the treatment of melasma in patients of South Asian descent. Clinicians can use these findings as a foundation for treating all patients with melasma but particularly those with darker skin types.

Study Limitations—Our study consisted of 60 patients; although our study had more patients than similar trials, larger studies are needed. Additionally, other variables were excluded from our analysis, such as comorbidities beyond diabetes mellitus and hypertension.

Conclusion

This study contributes to the growing field of melasma therapeutics by evaluating the efficacy of using TA solution 5% for the treatment of melasma in South Asian patients with darker skin types. Clinicians may use our study to broaden their treatment options for a common condition while also addressing the lack of clinical options for patients with darker skin types. Further studies investigating the effectiveness of TA in large clinical trials in humans are warranted to understand the efficacy and the risk for any complications.

Melasma is a complex, long-lasting, acquired dermatologic pigmentation disorder resulting in grey-brown patches that last for more than 3 months. Sun-exposed areas including the nose, cheeks, forehead, and forearms are most likely to be affected.¹ In Southeast Asia, 0.25% to 4% of the population affected by melasma is aged 30 to 40 years.² In particular, melasma is a concern among pregnant women due to increased levels of melanocyte-stimulating hormones (MSHs) and is impacted by genetics, hormonal influence, and exposure to UV light.^3,4 In Pakistan, approximately 46% of women are affected by melasma during pregnancy.^2,5 Although few studies have focused on the clinical approaches to melasma in darker skin types, it continues to disproportionately affect the skin of color population.⁴

The areas of hyperpigmentation seen in melasma exhibit increased deposition of melanin in the epidermis and dermis, but melanocytes are not elevated. However, in areas of hyperpigmentation, the melanocytes are larger and more dendritic and demonstrate an increased level of melanogenesis.⁶ During pregnancy, especially in the third trimester, elevated levels of estrogen, progesterone, and MSH often are found in association with melasma.⁷ Tyrosinase (TYR) activity increases and cellular proliferation is reduced after treatment of melanocytes in culture with β-estradiol.⁸ Sex steroids increase transcription of genes encoding melanogenic enzymes in normal human melanocytes, especially TYR.⁹ These results are consistent with the notable increases in melanin synthesis and TYR activity reported for normal human melanocytes under similar conditions in culture.¹⁰ Because melanocytes contain both cytosolic and nuclear estrogen receptors, melanocytes in patients with melasma may be inherently more sensitive to the stimulatory effects of estrogens and possibly other steroid hormones.¹¹

The current treatment options for melasma have varying levels of success and include topical depigmenting agents such as hydroquinone, tretinoin, azelaic acid, kojic acid, and corticosteroids; dermabrasion; and chemical peels.^12-14 Chemical peels with glycolic acid, salicylic acid, lactic acid, trichloroacetic acid, and phenol, as well as laser therapy, are reliable management options.^13,14 Traditionally, melasma has been treated with a combination of modalities along with photoprotection and trigger avoidance.¹²

The efficacy and safety of the available therapies for melasma are still controversial and require further exploration. In recent years, off-label tranexamic acid (TA) has emerged as a potential therapy for melasma. Although the mechanism of action remains unclear, TA may inhibit melanin synthesis by blocking the interaction between melanocytes and keratinocytes.¹⁵ Tranexamic acid also may reverse the abnormal dermal changes associated with melasma by inhibiting melanogenesis and angiogenesis.¹⁶

Although various therapeutic options exist for melasma, the search for a reliable option in patients with darker skin types continues.¹³ We sought to evaluate the efficacy of TA solution 5% in reducing the severity of melasma in South Asian patients, thereby improving patient outcomes and maximizing patient satisfaction. Topical TA is inexpensive and readily accessible and does not cause systemic side effects. These qualities make it a promising treatment compared to traditional therapies.

Methods

We conducted a randomized controlled trial at Rawalpindi Medical Institute (Punjab, Pakistan). The researchers obtained informed consent for all enrolled patients. Cases were sampled from the original patient population seen at the office using nonprobability consecutive sampling. The sample size was calculated with a 95% CI, margin of error of 9%, and expected percentage of efficacy of 86.1% by using TA solution 5%. South Asian male and female patients aged 20 to 45 years with melasma were included in the analysis. Patients were excluded if they were already taking TA, oral contraceptive pills, or photosensitizing drugs (eg, nonsteroidal anti-inflammatory drugs, tetracyclines, phenytoin, carbamazepine); were pregnant; had chronic kidney disease (creatinine >2.0 mg/dL); had cardiac abnormalities (abnormal electrocardiogram); had hematologic disorders (international normalized ratio >2); or had received another melasma treatment within the last 3 to 6 months.

All enrolled patients underwent a detailed history and physical examination. Patient demographics were subsequently noted, including age, sex, history of diabetes mellitus or hypertension, and duration of melasma. The melasma area and severity index (MASI) score of each patient was calculated at baseline, and a corresponding photograph was taken.

The topical solution was prepared with 5 g of TA dissolved in 10 cc of ethanol at 96 °F, 10 cc of 1,3-butanediol, and distilled water up to 100 cc. The TA solution was applied to the affected areas once daily by the patient for 12 weeks. Each application covered the affected areas completely. Patients were instructed to apply sunscreen with sun protection factor 60 to those same areas for UV protection after 15 minutes of TA application. Biweekly follow-ups were scheduled during the trial, and the MASI score was recorded at these visits. If the mean MASI score was reduced by half after 12 weeks of treatment, then the treatment was considered efficacious with a 95% CI.

The percentage reduction from baseline was calculated as follows: percentage reduction=(baseline score– follow-up score)/baseline score×100.

Statistical Analysis—Data were analyzed in SPSS Statistics 25 (IBM). The quantitative variables of age, duration of melasma, and body mass index were presented as mean (SD). Qualitative variables such as sex, history of diabetes mellitus or hypertension, site of melasma, and efficacy were presented as frequencies and percentages. Mean MASI scores at baseline and 12 weeks posttreatment were compared using a paired t test (P≤.05). Data were stratified for age, sex, history of diabetes mellitus or hypertension, site of melasma, and duration of melasma, and a χ² test was applied to compare efficacy in stratified groups (P≤.05).

Results

Sixty patients were enrolled in the study. Of them, 17 (28.33%) were male, and 43 (71.67%) were female (2:5 ratio). They ranged in age from 20 to 45 years (mean [SD], 31.93 [6.26] years). Thirty-seven patients (61.67%) were aged 31 to 45 years of age (Table 1). The mean (SD) duration of disease was 10.18 (2.10) months. The response to TA was recorded based on patient distribution according to the site of melasma as well as history of diabetes mellitus and hypertension.

Topical TA was found to be efficacious for melasma in 50 (83.33%) patients. The mean (SD) baseline and week 12 MASI scores were 23.15 (5.02) and 12.71 (4.10)(P<.0001), respectively (Table 2). The stratification of efficacy with respect to age, sex, duration of melasma, site of melasma, and history of diabetes mellitus or hypertension is shown in the eTable. The site of melasma was significant with respect to stratification of efficacy. On the forehead, TA was found to be efficacious in 11 patients and nonefficacious in 0 patients (P=.036). In the malar region, it was efficacious in 16 patients and nonefficacious in 1 patient (P=.036). Finally, on the chin, it was efficacious in 23 patients and nonefficacious in 9 patients (P=.036).

Comment

Melasma Presentation and Development—Melasma is a chronic skin condition that more often affects patients with darker skin types. This condition is characterized by hyperpigmentation of skin that is directly exposed to the sun, such as the cheek, nose, forehead, and above the upper lip.¹⁷ Although the mechanism behind how melasma develops is unknown, one theory suggests that UV light can lead to increased plasmin in keratinocytes.¹⁸ This increased plasmin will thereby increase the arachidonic acid and α-MSH, leading to the observed uneven hyperpigmentation that is notable in melasma. Melasma is common in patients using oral contraceptives or expired cosmetic drugs; in those who are pregnant; and in those with liver dysfunction.¹⁸ Melasma has a negative impact on patients’ quality of life because of substantial psychological and social distress. Thus, finding an accessible treatment is imperative.¹⁹

Melasma Management—The most common treatments for melasma have been topical bleaching agents and photoprotection. Combination therapy options include chemical peels, dermabrasion, and laser treatments, though they present with limited efficacy.^17,20 Because melasma focuses on pigmentation correction, topical treatments work to disturb melanocyte pigment production at the enzymatic level.²¹ Tyrosinase is rate limiting in melanin production, as it converts L-tyrosinase to L-3,4-dihydroxyphenylalanine, using copper to interact with L-3,4-dihydroxyphenylalanine as a cofactor in the active site.²² Therefore, tyrosine is a major target for many drugs that have been developed for melasma to decrease melaninization.²¹

Recently, research has focused on the effects of topical, intradermal, and oral TA for melasma.¹⁷ Tranexamic acid most commonly has been used in medicine as a fibrinolytic agent because of its antiplasmin properties. It has been hypothesized that TA can inhibit the release of paracrine melanogenic factors that normally act to stimulate melanocytes.¹⁷ Although studies have supported the safety and efficacy of TA, there remains a lack of clinical studies that are sufficiently powered. No definitive consensus on the use of TA for melasma currently exists, which indicates the need for large-scale, randomized, controlled trials.²³

One trial (N=25) found that TA solution 5% achieved efficacy (>50% reduction in MASI score from baseline) in 86.1% of patients with melasma.²⁴ In another study (N=18), topical TA 5% achieved efficacy (>50% reduction in MASI score) in 86% of patients with melasma.²⁵

Melasma Comorbidities—To determine if certain comorbidities, such as diabetes mellitus or hypertension, influenced the progression of melasma, we stratified the efficacy results for patients with these 2 comorbidities, which showed no significant difference (P=.794 and P=.101, respectively). Thus, the relatively higher prevalence of diabetes mellitus (16 patients) and hypertension (11 patients) did not contribute to the efficacy of TA in lowering MASI scores over the 12-week period, which supports the findings of Doolan and Gupta,²⁶ who investigated the endocrinologic conditions associated with melasma and found no such association with diabetes mellitus or hypertension.

TA Formulations for Melasma—The efficacy of topical TA has been explored in several studies. Six studies with sample sizes of 13 to 50 patients each showed statistically significant differences in MASI scores between baseline and following TA treatment (P<.001).^27-32 Several formulations and regimens were utilized, including TA cream 3% for 12 weeks, TA gel 5% for 12 weeks, TA solution 3% for 12 weeks, TA liposome 5% for 12 weeks, and TA solution 2% for 12 weeks.¹⁸ Additionally, these studies found TA to be effective in limiting dyschromia and decreasing MASI scores. There were no statistically significant differences between formulations and method of application. Topical TA has been found to be just as effective as other treatments for melasma, including intradermal TA injections, topical hydroquinone, and a combination of topical hydroquinone and dexamethasone.¹⁸

Further study of the efficacy of intradermal TA is necessary because many human trials have lacked statistical significance or a control group. Lee et al³² conducted a trial of 100 female patients who received weekly intradermal TA microinjections for 12 weeks. After 8 and 12 weeks, MASI scores decreased significantly (P<.01).³² Similarly, Badran et al³³ observed 60 female patients in 3 trial groups: group A received TA (4 mg/mL) intradermal injections every 2 weeks, group B received TA (10 mg/mL) intradermal injections every 2 weeks, and group C received TA cream 10% twice daily. Although all groups showed improvement in MASI, group B, which had the highest intradermal TA concentration, exhibited the most improvement. Thus, it was determined that intradermal application led to better results, but the cream was still effective.³³

Saki et al³⁴ conducted a randomized, split-face trial of 37 patients comparing the efficacy of intradermal TA and topical hydroquinone. Each group was treated with either monthly intradermal TA injections or nightly hydroquinone for 3 months. After 4 weeks of treatment, TA initially had a greater improvement. However, after 20 weeks, the overall changes were not significant between the 2 groups.³⁴ Pazyar et al³⁵ conducted a randomized, split-face trial of 49 patients comparing the efficacy of intradermal TA and hydroquinone cream. After 24 weeks of biweekly TA injections or twice-daily hydroquinone, there were no statistically significant differences in the decreased MASI scores between treatments.³⁵ Additional large, double-blind, controlled trials are needed to thoroughly assess the role of intradermal TA in comparison to its treatment counterpart of hydroquinone.

Ebrahimi and Naeini²⁹ conducted a 12-week, double-blind, split-phase trial of 50 Iranian melasma patients, which showed that 27.3% of patients rated the improvement in melasma as excellent, 42.4% as good, and 30.3% as fair after using TA solution 3%. Wu et al³⁶ also showed a total melasma improvement rate of 80.9% in 256 patients with long-term oral use of TA. In a study by Kim et al³¹ (N=245), the mean MASI score considerably decreased after topical TA use, with a total response rate of 95.6%. In another study, Atefi et al³⁷ presented significantly increased levels of satisfaction in patients treated with topical TA 5% vs hydroquinone (P=.015).

Melasma in Patients With Darker Skin Types—Special attention must be given to choosing the appropriate medication in melasma patients with darker skin types, as there is an increased risk for postinflammatory hyperpigmentation. Currently, few randomized controlled trials exist that fulfill the criteria of evaluating pharmacologic options for patients with melasma, and even fewer studies solely focus on patients with darker skin types.³⁸ In addition to treatment advances, patients must be educated on the need to avoid sun exposure when possible or to use photoprotection, especially in the South Asian region, where these practices rarely are taught. Our study provided a unique analysis regarding the efficacy of TA solution 5% for the treatment of melasma in patients of South Asian descent. Clinicians can use these findings as a foundation for treating all patients with melasma but particularly those with darker skin types.

Study Limitations—Our study consisted of 60 patients; although our study had more patients than similar trials, larger studies are needed. Additionally, other variables were excluded from our analysis, such as comorbidities beyond diabetes mellitus and hypertension.

Conclusion

This study contributes to the growing field of melasma therapeutics by evaluating the efficacy of using TA solution 5% for the treatment of melasma in South Asian patients with darker skin types. Clinicians may use our study to broaden their treatment options for a common condition while also addressing the lack of clinical options for patients with darker skin types. Further studies investigating the effectiveness of TA in large clinical trials in humans are warranted to understand the efficacy and the risk for any complications.

Endocrine mucin-producing sweat gland carcinoma (EMPSGC) and primary cutaneous mucinous carcinoma (PCMC) are rare low-grade neoplasms thought to arise from apocrine glands. These neoplasms share many histologic features and are proposed to be on a single histopathologic continuum, with EMPSGC as the in situ form that may progress to the invasive PCMC,¹ which is analogous to ductal carcinoma in situ and mucinous carcinoma of the breast, respectively.^2-5 Management involves a metastatic workup and either wide local excision (WLE) with margins greater than 5 mm or Mohs micrographic surgery (MMS) in anatomically sensitive areas.² We present 2 cases of EMPSGC and 3 cases of PCMC. We also review the clinical and histopathological features, differential diagnoses, and treatments.

Methods

Following institutional review board approval, we conducted a retrospective, single-institution case series. We searched electronic medical records dating from 2000 to 2019 for tumors diagnosed as PCMC or extramammary Paget disease treated with MMS. We gathered demographic, clinical, pathologic, and follow-up information from the electronic medical records for each case (Tables 1 and 2). Two dermatopathologists (B.P. and B.F.K.) reviewed the hematoxylin and eosin–stained slides of each tumor as well as all available immunohistochemical stains. One of the reviewers (B.F.K.) is a board-certified dermatologist, dermatopathologist, and fellowship-trained Mohs surgeon.

Results

Demographic and Clinical Information—We identified 2 cases of EMPSGC and 3 cases of PCMC diagnosed and treated at our institution; 4 of these cases had been treated within the last 2 years. One had been treated 18 years prior; case information was limited due to planned institutional record destruction. Three of the patients were female and 2 were male. The mean age at presentation was 71 years (range, 62–87 years). None had experienced recurrence or metastases after a mean follow-up of 30 months.

Case 1—A 68-year-old woman noted a slow-growing, flesh-colored papule measuring 12×10 mm on the right lower eyelid. An excisional biopsy was completed with 2-mm clinical margins, and the defect was closed in a linear fashion. Histologic sections demonstrated EMPSGC with uninvolved margins. The patient desired no further intervention and was clinically followed. Magnetic resonance imaging (MRI) of the head and neck found no evidence of metastasis. She has had no recurrence after 15 months.

Case 2—A 62-year-old man presented with a 7×5-mm, flesh-colored papule on the left lower eyelid margin (Figure 1). It was previously treated conservatively as a hordeolum but was biopsied after it failed to resolve with 3-mm margins. Histopathology demonstrated an EMPSGC (Figure 2). The lesion was treated with modified MMS with permanent en face section processing and cleared after 1 stage. Computed tomography of the head and neck showed no abnormalities. He has had no recurrence after 9 months.

Case 3—A 72-year-old man presented with a nontender papule near the right lateral canthus. A punch biopsy demonstrated PCMC. He was treated via modified MMS with permanent en face section processing. The tumor was cleared in 1 stage. He showed no evidence of recurrence after 112 months and died of unrelated causes. The rest of his clinical information was limited because of planned institutional destruction of records.

Case 4—An 87-year-old woman presented with a 25×25-mm, slow-growing mass of 12 months’ duration on the left lower abdomen (Figure 3). A biopsy demonstrated PCMC (Figure 4). Because of the size of the lesion, she underwent WLE with 20- to 30-mm margins by a general surgeon under general anesthesia. Positron emission tomography/computed tomography was unremarkable. She has remained disease free for 11 months.

Case 5—A 66-year-old woman presented for evaluation of a posterior scalp mass measuring 23×18 mm that had grown over the last 24 months. Biopsy showed mucinous carcinoma with lymphovascular invasion consistent with PCMC (Figure 5) confirmed on multiple tissue levels and with the aid of immunohistochemistry. She was sent for an MRI of the head, neck, chest, abdomen, and pelvis, which demonstrated 2 enlarged postauricular lymph nodes and raised suspicion for metastatic disease vs reactive lymphadenopathy. Mohs micrographic surgery with frozen sections was performed with 1- to 3-mm margins; the final layer was sent for permanent processing and confirmed negative margins. Sentinel lymph node biopsy and lymphadenectomy of the 2 nodes present on imaging showed no evidence of metastasis. The patient had no recurrence in 1 month.

Comment

Endocrine mucin-producing sweat gland carcinoma and PCMC are sweat gland malignancies that carry low metastatic potential but are locally aggressive. Endocrine mucin-producing sweat gland carcinoma has a strong predilection for the periorbital region, especially the lower eyelids of older women.³ Primary cutaneous mucinous carcinoma may arise on the eyelids, scalp, axillae, and trunk and has been reported more often in older men. These slow-growing tumors appear as nonspecific nodules.³ Lesions frequently are asymptomatic but rarely may cause pruritus and bleeding. Histologically, EMPSGC appears as solid or cystic nodules of cells with a papillary, cribriform, or pseudopapillary appearance. Intracellular or extracellular mucin as well as malignant spread of tumor cells along pre-existing ductlike structures make it difficult to histologically distinguish EMPSGC from ductal carcinoma in situ.³

A key histopathologic feature of PCMC is basophilic epithelioid cell nests in mucinous lakes.⁴ Rosettelike structures are seen within solid areas of the tumor. Fibrous septae separate individual collections of mucin, creating a lobulated appearance. The histopathologic differential diagnosis of EMPSGC and PCMC is broad, including basal cell carcinoma, hidradenoma, hidradenocarcinoma, apocrine adenoma, and dermal duct tumor. Positive expression of at least 1 neuroendocrine marker (ie, synaptophysin, neuron-specific enolase, chromogranin) and low-molecular cytokeratin (cytokeratin 7, CAM5.2, Ber-EP4) can aid in the diagnosis of both EMPSGC and PCMC.⁴ The use of p63 immunostaining is beneficial in delineating adnexal neoplasms. Adnexal tumors that stain positively with p63 are more likely to be of primary cutaneous origin, whereas lack of p63 staining usually denotes a secondary metastatic process. However, p63 staining is less reliable when distinguishing primary and metastatic mucinous neoplasms. Metastatic mucinous carcinomas often stain positive with p63, while PCMC usually stains negative despite its primary cutaneous origin, decreasing the clinical utility of p63. The tumor may be identical to metastatic mucinous adenocarcinoma of the breast, gastrointestinal tract, lung, ovary, and pancreas. Tumor islands floating in mucin are identified in both primary cutaneous and metastatic disease to the skin.^3,6 Areas of tumor necrosis, notable atypia, and perineural or lymphovascular invasion are infrequently reported in EMPSGC or PCMC, though lymphatic invasion was identified in case 5 presented herein.

A metastatic workup is warranted in all cases of PCMC, including a thorough history, review of systems, breast examination, and imaging. A workup may be considered in cases of EMPSGC depending on histologic features or clinical history.

There is uncertainty regarding the optimal management of these slow-growing yet locally destructive tumors.⁵ The incidence of local recurrence of PCMC after WLE with narrow margins of at least 1 cm can be as high as 30% to 40%, especially on the eyelid.⁴ There is no consensus on surgical care for either of these tumors.⁵ Because of the high recurrence rate and the predilection for the eyelid and face, MMS provides an excellent alternative to WLE for tissue preservation and meticulous margin control. We advocate for the use of the Mohs technique with permanent sectioning, which may delay the repair, but reviewing tissue with permanent fixation improves the quality and accuracy of the margin evaluation because these tumors often are infiltrative and difficult to delineate under frozen section processing. Permanent en face sectioning allows the laboratory to utilize the full array of immunohistochemical stains for these tumors, providing accurate and timely results.

Limitations to our retrospective uncontrolled study include missing or incomplete data points and short follow-up time. Additionally, there was no standardization to the margins removed with MMS or WLE because of the limited available data that comment on appropriate margins.

Endocrine mucin-producing sweat gland carcinoma (EMPSGC) and primary cutaneous mucinous carcinoma (PCMC) are rare low-grade neoplasms thought to arise from apocrine glands. These neoplasms share many histologic features and are proposed to be on a single histopathologic continuum, with EMPSGC as the in situ form that may progress to the invasive PCMC,¹ which is analogous to ductal carcinoma in situ and mucinous carcinoma of the breast, respectively.^2-5 Management involves a metastatic workup and either wide local excision (WLE) with margins greater than 5 mm or Mohs micrographic surgery (MMS) in anatomically sensitive areas.² We present 2 cases of EMPSGC and 3 cases of PCMC. We also review the clinical and histopathological features, differential diagnoses, and treatments.

Methods

Following institutional review board approval, we conducted a retrospective, single-institution case series. We searched electronic medical records dating from 2000 to 2019 for tumors diagnosed as PCMC or extramammary Paget disease treated with MMS. We gathered demographic, clinical, pathologic, and follow-up information from the electronic medical records for each case (Tables 1 and 2). Two dermatopathologists (B.P. and B.F.K.) reviewed the hematoxylin and eosin–stained slides of each tumor as well as all available immunohistochemical stains. One of the reviewers (B.F.K.) is a board-certified dermatologist, dermatopathologist, and fellowship-trained Mohs surgeon.

Results

Demographic and Clinical Information—We identified 2 cases of EMPSGC and 3 cases of PCMC diagnosed and treated at our institution; 4 of these cases had been treated within the last 2 years. One had been treated 18 years prior; case information was limited due to planned institutional record destruction. Three of the patients were female and 2 were male. The mean age at presentation was 71 years (range, 62–87 years). None had experienced recurrence or metastases after a mean follow-up of 30 months.

Case 1—A 68-year-old woman noted a slow-growing, flesh-colored papule measuring 12×10 mm on the right lower eyelid. An excisional biopsy was completed with 2-mm clinical margins, and the defect was closed in a linear fashion. Histologic sections demonstrated EMPSGC with uninvolved margins. The patient desired no further intervention and was clinically followed. Magnetic resonance imaging (MRI) of the head and neck found no evidence of metastasis. She has had no recurrence after 15 months.

Case 2—A 62-year-old man presented with a 7×5-mm, flesh-colored papule on the left lower eyelid margin (Figure 1). It was previously treated conservatively as a hordeolum but was biopsied after it failed to resolve with 3-mm margins. Histopathology demonstrated an EMPSGC (Figure 2). The lesion was treated with modified MMS with permanent en face section processing and cleared after 1 stage. Computed tomography of the head and neck showed no abnormalities. He has had no recurrence after 9 months.

Case 3—A 72-year-old man presented with a nontender papule near the right lateral canthus. A punch biopsy demonstrated PCMC. He was treated via modified MMS with permanent en face section processing. The tumor was cleared in 1 stage. He showed no evidence of recurrence after 112 months and died of unrelated causes. The rest of his clinical information was limited because of planned institutional destruction of records.

Case 4—An 87-year-old woman presented with a 25×25-mm, slow-growing mass of 12 months’ duration on the left lower abdomen (Figure 3). A biopsy demonstrated PCMC (Figure 4). Because of the size of the lesion, she underwent WLE with 20- to 30-mm margins by a general surgeon under general anesthesia. Positron emission tomography/computed tomography was unremarkable. She has remained disease free for 11 months.

Case 5—A 66-year-old woman presented for evaluation of a posterior scalp mass measuring 23×18 mm that had grown over the last 24 months. Biopsy showed mucinous carcinoma with lymphovascular invasion consistent with PCMC (Figure 5) confirmed on multiple tissue levels and with the aid of immunohistochemistry. She was sent for an MRI of the head, neck, chest, abdomen, and pelvis, which demonstrated 2 enlarged postauricular lymph nodes and raised suspicion for metastatic disease vs reactive lymphadenopathy. Mohs micrographic surgery with frozen sections was performed with 1- to 3-mm margins; the final layer was sent for permanent processing and confirmed negative margins. Sentinel lymph node biopsy and lymphadenectomy of the 2 nodes present on imaging showed no evidence of metastasis. The patient had no recurrence in 1 month.

Comment

Endocrine mucin-producing sweat gland carcinoma and PCMC are sweat gland malignancies that carry low metastatic potential but are locally aggressive. Endocrine mucin-producing sweat gland carcinoma has a strong predilection for the periorbital region, especially the lower eyelids of older women.³ Primary cutaneous mucinous carcinoma may arise on the eyelids, scalp, axillae, and trunk and has been reported more often in older men. These slow-growing tumors appear as nonspecific nodules.³ Lesions frequently are asymptomatic but rarely may cause pruritus and bleeding. Histologically, EMPSGC appears as solid or cystic nodules of cells with a papillary, cribriform, or pseudopapillary appearance. Intracellular or extracellular mucin as well as malignant spread of tumor cells along pre-existing ductlike structures make it difficult to histologically distinguish EMPSGC from ductal carcinoma in situ.³

A key histopathologic feature of PCMC is basophilic epithelioid cell nests in mucinous lakes.⁴ Rosettelike structures are seen within solid areas of the tumor. Fibrous septae separate individual collections of mucin, creating a lobulated appearance. The histopathologic differential diagnosis of EMPSGC and PCMC is broad, including basal cell carcinoma, hidradenoma, hidradenocarcinoma, apocrine adenoma, and dermal duct tumor. Positive expression of at least 1 neuroendocrine marker (ie, synaptophysin, neuron-specific enolase, chromogranin) and low-molecular cytokeratin (cytokeratin 7, CAM5.2, Ber-EP4) can aid in the diagnosis of both EMPSGC and PCMC.⁴ The use of p63 immunostaining is beneficial in delineating adnexal neoplasms. Adnexal tumors that stain positively with p63 are more likely to be of primary cutaneous origin, whereas lack of p63 staining usually denotes a secondary metastatic process. However, p63 staining is less reliable when distinguishing primary and metastatic mucinous neoplasms. Metastatic mucinous carcinomas often stain positive with p63, while PCMC usually stains negative despite its primary cutaneous origin, decreasing the clinical utility of p63. The tumor may be identical to metastatic mucinous adenocarcinoma of the breast, gastrointestinal tract, lung, ovary, and pancreas. Tumor islands floating in mucin are identified in both primary cutaneous and metastatic disease to the skin.^3,6 Areas of tumor necrosis, notable atypia, and perineural or lymphovascular invasion are infrequently reported in EMPSGC or PCMC, though lymphatic invasion was identified in case 5 presented herein.

A metastatic workup is warranted in all cases of PCMC, including a thorough history, review of systems, breast examination, and imaging. A workup may be considered in cases of EMPSGC depending on histologic features or clinical history.

There is uncertainty regarding the optimal management of these slow-growing yet locally destructive tumors.⁵ The incidence of local recurrence of PCMC after WLE with narrow margins of at least 1 cm can be as high as 30% to 40%, especially on the eyelid.⁴ There is no consensus on surgical care for either of these tumors.⁵ Because of the high recurrence rate and the predilection for the eyelid and face, MMS provides an excellent alternative to WLE for tissue preservation and meticulous margin control. We advocate for the use of the Mohs technique with permanent sectioning, which may delay the repair, but reviewing tissue with permanent fixation improves the quality and accuracy of the margin evaluation because these tumors often are infiltrative and difficult to delineate under frozen section processing. Permanent en face sectioning allows the laboratory to utilize the full array of immunohistochemical stains for these tumors, providing accurate and timely results.

Limitations to our retrospective uncontrolled study include missing or incomplete data points and short follow-up time. Additionally, there was no standardization to the margins removed with MMS or WLE because of the limited available data that comment on appropriate margins.

Endocrine mucin-producing sweat gland carcinoma (EMPSGC) and primary cutaneous mucinous carcinoma (PCMC) are rare low-grade neoplasms thought to arise from apocrine glands. These neoplasms share many histologic features and are proposed to be on a single histopathologic continuum, with EMPSGC as the in situ form that may progress to the invasive PCMC,¹ which is analogous to ductal carcinoma in situ and mucinous carcinoma of the breast, respectively.^2-5 Management involves a metastatic workup and either wide local excision (WLE) with margins greater than 5 mm or Mohs micrographic surgery (MMS) in anatomically sensitive areas.² We present 2 cases of EMPSGC and 3 cases of PCMC. We also review the clinical and histopathological features, differential diagnoses, and treatments.

Methods

Following institutional review board approval, we conducted a retrospective, single-institution case series. We searched electronic medical records dating from 2000 to 2019 for tumors diagnosed as PCMC or extramammary Paget disease treated with MMS. We gathered demographic, clinical, pathologic, and follow-up information from the electronic medical records for each case (Tables 1 and 2). Two dermatopathologists (B.P. and B.F.K.) reviewed the hematoxylin and eosin–stained slides of each tumor as well as all available immunohistochemical stains. One of the reviewers (B.F.K.) is a board-certified dermatologist, dermatopathologist, and fellowship-trained Mohs surgeon.

Results

Demographic and Clinical Information—We identified 2 cases of EMPSGC and 3 cases of PCMC diagnosed and treated at our institution; 4 of these cases had been treated within the last 2 years. One had been treated 18 years prior; case information was limited due to planned institutional record destruction. Three of the patients were female and 2 were male. The mean age at presentation was 71 years (range, 62–87 years). None had experienced recurrence or metastases after a mean follow-up of 30 months.

Case 1—A 68-year-old woman noted a slow-growing, flesh-colored papule measuring 12×10 mm on the right lower eyelid. An excisional biopsy was completed with 2-mm clinical margins, and the defect was closed in a linear fashion. Histologic sections demonstrated EMPSGC with uninvolved margins. The patient desired no further intervention and was clinically followed. Magnetic resonance imaging (MRI) of the head and neck found no evidence of metastasis. She has had no recurrence after 15 months.

Case 2—A 62-year-old man presented with a 7×5-mm, flesh-colored papule on the left lower eyelid margin (Figure 1). It was previously treated conservatively as a hordeolum but was biopsied after it failed to resolve with 3-mm margins. Histopathology demonstrated an EMPSGC (Figure 2). The lesion was treated with modified MMS with permanent en face section processing and cleared after 1 stage. Computed tomography of the head and neck showed no abnormalities. He has had no recurrence after 9 months.

Case 3—A 72-year-old man presented with a nontender papule near the right lateral canthus. A punch biopsy demonstrated PCMC. He was treated via modified MMS with permanent en face section processing. The tumor was cleared in 1 stage. He showed no evidence of recurrence after 112 months and died of unrelated causes. The rest of his clinical information was limited because of planned institutional destruction of records.

Case 4—An 87-year-old woman presented with a 25×25-mm, slow-growing mass of 12 months’ duration on the left lower abdomen (Figure 3). A biopsy demonstrated PCMC (Figure 4). Because of the size of the lesion, she underwent WLE with 20- to 30-mm margins by a general surgeon under general anesthesia. Positron emission tomography/computed tomography was unremarkable. She has remained disease free for 11 months.

Case 5—A 66-year-old woman presented for evaluation of a posterior scalp mass measuring 23×18 mm that had grown over the last 24 months. Biopsy showed mucinous carcinoma with lymphovascular invasion consistent with PCMC (Figure 5) confirmed on multiple tissue levels and with the aid of immunohistochemistry. She was sent for an MRI of the head, neck, chest, abdomen, and pelvis, which demonstrated 2 enlarged postauricular lymph nodes and raised suspicion for metastatic disease vs reactive lymphadenopathy. Mohs micrographic surgery with frozen sections was performed with 1- to 3-mm margins; the final layer was sent for permanent processing and confirmed negative margins. Sentinel lymph node biopsy and lymphadenectomy of the 2 nodes present on imaging showed no evidence of metastasis. The patient had no recurrence in 1 month.

Comment

Endocrine mucin-producing sweat gland carcinoma and PCMC are sweat gland malignancies that carry low metastatic potential but are locally aggressive. Endocrine mucin-producing sweat gland carcinoma has a strong predilection for the periorbital region, especially the lower eyelids of older women.³ Primary cutaneous mucinous carcinoma may arise on the eyelids, scalp, axillae, and trunk and has been reported more often in older men. These slow-growing tumors appear as nonspecific nodules.³ Lesions frequently are asymptomatic but rarely may cause pruritus and bleeding. Histologically, EMPSGC appears as solid or cystic nodules of cells with a papillary, cribriform, or pseudopapillary appearance. Intracellular or extracellular mucin as well as malignant spread of tumor cells along pre-existing ductlike structures make it difficult to histologically distinguish EMPSGC from ductal carcinoma in situ.³

A key histopathologic feature of PCMC is basophilic epithelioid cell nests in mucinous lakes.⁴ Rosettelike structures are seen within solid areas of the tumor. Fibrous septae separate individual collections of mucin, creating a lobulated appearance. The histopathologic differential diagnosis of EMPSGC and PCMC is broad, including basal cell carcinoma, hidradenoma, hidradenocarcinoma, apocrine adenoma, and dermal duct tumor. Positive expression of at least 1 neuroendocrine marker (ie, synaptophysin, neuron-specific enolase, chromogranin) and low-molecular cytokeratin (cytokeratin 7, CAM5.2, Ber-EP4) can aid in the diagnosis of both EMPSGC and PCMC.⁴ The use of p63 immunostaining is beneficial in delineating adnexal neoplasms. Adnexal tumors that stain positively with p63 are more likely to be of primary cutaneous origin, whereas lack of p63 staining usually denotes a secondary metastatic process. However, p63 staining is less reliable when distinguishing primary and metastatic mucinous neoplasms. Metastatic mucinous carcinomas often stain positive with p63, while PCMC usually stains negative despite its primary cutaneous origin, decreasing the clinical utility of p63. The tumor may be identical to metastatic mucinous adenocarcinoma of the breast, gastrointestinal tract, lung, ovary, and pancreas. Tumor islands floating in mucin are identified in both primary cutaneous and metastatic disease to the skin.^3,6 Areas of tumor necrosis, notable atypia, and perineural or lymphovascular invasion are infrequently reported in EMPSGC or PCMC, though lymphatic invasion was identified in case 5 presented herein.

A metastatic workup is warranted in all cases of PCMC, including a thorough history, review of systems, breast examination, and imaging. A workup may be considered in cases of EMPSGC depending on histologic features or clinical history.

There is uncertainty regarding the optimal management of these slow-growing yet locally destructive tumors.⁵ The incidence of local recurrence of PCMC after WLE with narrow margins of at least 1 cm can be as high as 30% to 40%, especially on the eyelid.⁴ There is no consensus on surgical care for either of these tumors.⁵ Because of the high recurrence rate and the predilection for the eyelid and face, MMS provides an excellent alternative to WLE for tissue preservation and meticulous margin control. We advocate for the use of the Mohs technique with permanent sectioning, which may delay the repair, but reviewing tissue with permanent fixation improves the quality and accuracy of the margin evaluation because these tumors often are infiltrative and difficult to delineate under frozen section processing. Permanent en face sectioning allows the laboratory to utilize the full array of immunohistochemical stains for these tumors, providing accurate and timely results.

Limitations to our retrospective uncontrolled study include missing or incomplete data points and short follow-up time. Additionally, there was no standardization to the margins removed with MMS or WLE because of the limited available data that comment on appropriate margins.

ABSTRACT

Background: We sought to examine whether electrical impedance spectroscopy (EIS), a diagnostic tool approved by the US Food and Drug Administration for the evaluation of pigmented skin lesions (PSLs), is beneficial to primary care providers (PCPs) by comparing the accuracy of PCPs’ management decisions for PSLs based on visual examination alone with those based on concurrent visual and EIS evaluation.

Methods: Physicians and nurse practitioners (NPs) participated in an anonymous online survey in which they viewed clinical images of PSLs and were asked to make 2 clinical decisions before and after being provided an EIS score that indicated the likelihood that the lesion was a melanoma. They were asked (1) if they would biopsy the lesion/refer the patient out and (2) what they expected the pathology results would show.

Results: Forty-four physicians and 17 NPs participated, making clinical decisions for 1354 presented lesions. Overall, with the addition of EIS to visual inspection of clinical images, the sensitivity of biopsy/referral decisions for melanomas and severely dysplastic nevi (SDN) increased from 69.2% to 90.0% (P < .001), while specificity increased from 44.0% to 72.6% (P < .001). Physicians and NPs, regardless of years of experience, each saw significant improvements in sensitivity, specificity, and diagnostic accuracy with the addition of EIS scores.

Conclusions: The incorporation of EIS data into clinical decision-making by PCPs significantly increased the sensitivity and specificity of biopsy/referral decisions for melanomas and SDN and overall diagnostic accuracy compared with visual inspection alone. The results of this study suggest that diagnostic accuracy for PSLs by PCPs may be improved with adjunctive use of EIS with visual inspection.

Primary care providers (PCPs) are often the first line of defense in detecting skin cancers. For patients with concerning skin lesions, PCPs may choose to perform a biopsy or facilitate access to specialty services (eg, Dermatology). Consequently, PCPs play a critical role in the timely detection of skin cancers, and it is paramount to employ continually improving detection methods, such as the application of technologic advances.¹

Differentiating benign nevi from melanoma and severely dysplastic nevi (SDN), both of which warrant excision, poses a unique challenge to clinicians examining pigmented skin lesions (PSLs). PCPs often rely on visual inspection to differentiate benign skin lesions from malignant skin cancers. In some primary care practices, dermoscopy, which involves using a handheld device to evaluate lesions with polarized light and magnification, is used to improve melanoma detection. However, while visual inspection and dermoscopy are valid, effective techniques for the diagnosis of melanocytic lesions, in many instances they still can lead to missed cancers or unnecessary biopsies and specialty referrals. Adjunctive use of dermoscopy with visual inspection has been shown to increase the probability of skin cancer detection, but it fails to achieve a near-100% success rate.² Furthermore, dermoscopy is heavily user-dependent, requiring significant training and experience for appropriate use.³

Another option is an electrical impedance spectroscopy (EIS) device (Nevisense, Scibase, Stockholm, Sweden), which has been approved by the US Food and Drug Administration (FDA) to assist in the detection of melanoma and differentiation from benign PSLs.⁴ EIS is a noninvasive, rapidly applied technology designed to accompany the visual examination of melanocytic lesions in office, with or without dermoscopy. Still relatively new, the technology is employed today by many dermatologists, increasing diagnostic accuracy for PSLs.⁵ The lightweight and portable instrument features a handheld probe, which is held against a lesion to obtain a reading. EIS uses a low-voltage electrode to apply a harmless electrical current to the skin at various frequencies.⁶ As benign and malignant tissues vary in cell shape, size, and composition, EIS distinguishes differential electrical resistance of the tissue to aid in diagnosis.⁷

Continue to: EIS provides high-sensitivity...

EIS provides high-sensitivity melanoma diagnosis vs histopathologic confirmation from biopsies, with 1 study showing a 96.6% sensitivity rating, detecting 256 of 265 melanomas.⁴ The EIS device, by measuring differences in electrical resistance between benign and cancerous cells, outputs a simple integer score ranging from 0 to 10 associated with the likelihood of the lesion being a melanoma.⁸ Based on data from the Nevisense pivotal trial,⁴ Nevisense reports that scores of 0 to 3 carry a negative predictive value of 99% for melanoma, whereas scores of 4 to 10 signify increasingly greater positive predictive values from 7% to 61%.

Findings suggest that the use of electrical impedance spectroscopy is particularly advantageous to clinicians who are less proficient in assessing melanocytic lesions.

We aimed to assess whether EIS may be beneficial to PCPs by comparing the accuracy of clinical decision-making for PSLs based on visual examination alone with that based on concurrent visual and EIS evaluation.

METHODS

A questionnaire was distributed via email to 142 clinicians at clinics affiliated with either of 2 organizations delivering care to the New York City area through a network of community health centers: the Institute for Family Health (IFH) and the Community Healthcare Network (CHN). Of these recipients, 72 were affiliated with IFH across 27 community health centers and 70 were affiliated with CHN across 14 community health centers. Recipients were physicians and nurse practitioners (NPs) practicing at primary health care facilities.

Survey instrument. The first section of the survey instrument (APPENDIX) solicited demographic information and explained how to apply the EIS scores for diagnostic ­decision-making. The second featured images of 12 randomly selected, histologically confirmed, and EIS-evaluated PSLs from a previously published prospective blinded trial of 2416 lesions.⁴ The Institutional Review Board of the Icahn School of Medicine at Mount Sinai reviewed and approved the study and survey instrument.

Clinical images of these lesions, comprising 4 melanocytic nevi, 4 dysplastic nevi (including 3 mild-moderately dysplastic and 1 severely dysplastic nevus), and 4 melanomas, were first presented to respondents with 2 tasks: (1) rate on a scale of 1 to 5 their likelihood to biopsy or refer this lesion to a dermatologist (1: not likely; 5: extremely likely); and (2) select what they expect the pathology results to be: melanocytic nevus, dysplastic nevus, or malignant melanoma. Subsequently, respondents repeated the assessments after being presented with the EIS score for the same lesion in conjunction with the clinical image.

Continue to: Analysis

Analysis. A biopsy or referral rating of 4 or 5 was considered a decision to biopsy or refer (ie, a diagnostic decision consistent with melanoma or SDN warranting excision), whereas a selection of 1 to 3 was considered a decision not to biopsy or refer (ie, a diagnostic decision consistent with a benign PSL). The sensitivity and specificity of biopsy/­referral decisions for melanomas and SDN, the proportion of missed melanomas and SDN, and the proportion of biopsy/referral decisions for benign lesions were separately determined for visual inspection alone and visual inspection with EIS score. Similarly, diagnostic accuracy was calculated for these clinical scenarios. These metrics were further stratified among different subsets of the respondent population. Differences in sensitivity, specificity, biopsy/referral decision proportions, and diagnostic accuracy were calculated using McNemar’s test for paired proportions.

RESULTS

Sixty-one respondents, comprising 44 physicians and 17 NPs, completed the survey, yielding a response rate of 43% (TABLE 1). In total, 1354 clinical decisions (677 based on visual inspection alone and 677 based on visual inspection plus EIS) were made. A biopsy/­referral decision was made after assessing 416 of 677 cases (61%) with visual inspection alone and 360 of 677 cases (53%) when relying on visual inspection plus EIS. None of the respondents reported any prior experience with EIS.

When incorporating EIS scores, respondents’ mean sensitivity for melanomas and SDN increased from 69.2% to 90.0% (P < .001) and specificity from 44.0% to 72.6% (P < .001; TABLE 2). At baseline, physicians demonstrated a sensitivity and specificity of 74.6% and 46.5%, respectively, while NPs demonstrated a sensitivity and specificity of 56.1% and 37.9%, respectively.

All respondent subgroups stratified by occupation and years of experience saw significant increases in both sensitivity and specificity upon the incorporation of EIS scores, with NPs seeing a greater increase in sensitivity (56.1% vs 85.4%; P < .001) and specificity (37.9% vs 69.0%; P < .001) than physicians (sensitivity: 74.6% vs 91.9%; P < .001; specificity: 46.5% vs 74.1%; P < .001). The only difference in diagnostic performance based on years of experience was a greater pre-EIS sensitivity by clinicians who had been in practice for ≥ 15 years, compared with those in practice for shorter periods (TABLE 2).

The improvements, seen in clinicians of varying training and experience, suggest that the learning curve of EIS may not be as steep as that of dermoscopy.

Diagnostic accuracy increased significantly from 48% when based on visual inspection alone to 73% with the addition of EIS scores (P < .001; TABLE 3). Physicians and NPs each significantly increased their diagnostic accuracy upon the incorporation of EIS, with NPs exhibiting the greatest increase (from 36.9% to 65.7%; P < .001). PCPs with 6 to 14 years of experience saw the greatest increase in diagnostic accuracy when adding EIS (45.9% vs 76.4%; P < .001). Overall, the addition of EIS scores resulted in 58 fewer missed melanomas and SDN and 114 fewer benign referrals or biopsies (TABLE 4).

Continue to: DISCUSSION

DISCUSSION

Primary care evaluation plays a significant role in the diagnosis and management of PSLs, ultimately shaping outcomes for patients with melanoma. Improved accuracy of PSL classification could yield greater sensitivity for the diagnosis of melanomas and high-risk melanocytic lesions at earlier stages, while also reducing the number of unnecessary biopsies and referrals—leading to decreased patient morbidity and mortality and reduced health care spending.⁹

Diagnostic tools are valuable insofar as they can improve accuracy and positively impact clinical management and patient outcomes.¹⁰ In this case, increased sensitivity reduced missed melanoma diagnoses, while increased specificity avoided the additional costs and patient toll associated with a biopsy or referral for a benign lesion.

Dermoscopy has been shown to improve the sensitivity and specificity of PSL diagnosis compared with visual inspection alone; however, without substantial training and experience, accuracy with dermoscopy can be no better than examination with the naked eye.^3,11,12 The dropout rates are high for training PCPs in its use, given that several months of training may be needed for competent use.^13,14 To improve the clinical management of PSLs broadly in primary care, a need exists for easy-to-use adjunctive tools that increase diagnostic accuracy.¹⁵

In this study, with only a brief explanation of how to interpret EIS scores, clinicians without any prior experience using EIS demonstrated significantly improved accuracy in deciding appropriate management and classifying melanocytic lesions with the addition of EIS to visual inspection. These improvements, seen in clinicians of varying training and experience, suggest that the learning curve of EIS may not be as steep as that of dermoscopy.

The greater baseline sensitivity, specificity, and diagnostic accuracy of physicians’ clinical decision-making compared with NPs before the incorporation of EIS in the study may be a product of comparatively more extensive medical training. In addition, EIS yielded a greater benefit to NPs than to physicians, with greater increases in sensitivity and specificity noted. This suggests that the use of EIS is particularly advantageous to clinicians who are less proficient in assessing melanocytic lesions. Using visual inspection alone, more experienced respondents made biopsy/referral decisions with greater sensitivity but similar specificity to those with less experience. With the incorporation of EIS scores, the sensitivity and specificity of respondents’ clinical decision-making rose to comparable levels across all experience groups, providing further indication of EIS’s particular value to clinicians who are less proficient in PSL evaluation.

Continue to: This technology holds the potential...

This technology holds the potential to be seamlessly implemented into primary care practice, given that dermatology expertise training is not required to use the EIS device; this could allow for EIS measurement of lesions to be delegated to office staff (eg, nurses, medical assistants).¹⁶ Future studies are needed to assess EIS use among PCPs in a real-world setting, where factors such as its application on nonmelanocytic lesions (eg, seborrheic keratoses) and its pairing with patient historical data could produce varying results.

Limitations. While revealing, this study had its limitations. Respondents did not have access to additional pertinent clinical information, such as patients’ histories and risk factors. Clinical decisions in this survey were made based on digital images rather than in vivo examination. This may not represent a real-life evaluation; there is the potential for minimization of the true consequences of a missed melanoma or unnecessary biopsy in the minds of participants, and this does not factor in the operation of the actual EIS device. The Hawthorne effect may also have influenced PCPs’ diagnostic selections. Also, the limited sample size constitutes another limitation.

The results of this preliminary study suggest that diagnostic accuracy for pigmented skin lesions by PCPs may be improved with the adjunctive use of electrical impedance spectroscopy with visual inspection.

Of note, in this survey format, respondents rated their inclination to biopsy or refer each lesion from 1 to 5. For statistical analyses, lesions rated 1 to 3 were considered as not biopsied/referred and those rated 4 to 5 as biopsied/referred. The sensitivity and specificity values observed, for both visual examination and concurrent visual and EIS evaluation, are therefore based on this classification system of participants’ provided ratings. It is conceivable that differing sensitivity and specificity values might have been detected if clinicians were instead given a binary choice for referral/biopsy decisions.

CONCLUSIONS

Among PCPs tasked with evaluating melanocytic lesions, the incorporation of EIS data into clinical decision-making in this study significantly increased the sensitivity, specificity, and overall diagnostic accuracy of biopsy or referral decisions for melanomas and SDN compared with visual inspection alone. Overall, the results of this preliminary study suggest that diagnostic accuracy for PSLs by PCPs may be improved with the adjunctive use of EIS with visual inspection. This would ultimately improve patient care and reduce the morbidity and mortality of a melanoma diagnosis.

CORRESPONDENCE
Jonathan Ungar, MD, Kimberly and Eric J. Waldman Department of Dermatology, Icahn School of Medicine at Mount Sinai, 5 East 98th Street, 5th Floor, New York, NY 10029; jonathan.ungar@mountsinai.org

ABSTRACT

Background: We sought to examine whether electrical impedance spectroscopy (EIS), a diagnostic tool approved by the US Food and Drug Administration for the evaluation of pigmented skin lesions (PSLs), is beneficial to primary care providers (PCPs) by comparing the accuracy of PCPs’ management decisions for PSLs based on visual examination alone with those based on concurrent visual and EIS evaluation.

Methods: Physicians and nurse practitioners (NPs) participated in an anonymous online survey in which they viewed clinical images of PSLs and were asked to make 2 clinical decisions before and after being provided an EIS score that indicated the likelihood that the lesion was a melanoma. They were asked (1) if they would biopsy the lesion/refer the patient out and (2) what they expected the pathology results would show.

Results: Forty-four physicians and 17 NPs participated, making clinical decisions for 1354 presented lesions. Overall, with the addition of EIS to visual inspection of clinical images, the sensitivity of biopsy/referral decisions for melanomas and severely dysplastic nevi (SDN) increased from 69.2% to 90.0% (P < .001), while specificity increased from 44.0% to 72.6% (P < .001). Physicians and NPs, regardless of years of experience, each saw significant improvements in sensitivity, specificity, and diagnostic accuracy with the addition of EIS scores.

Conclusions: The incorporation of EIS data into clinical decision-making by PCPs significantly increased the sensitivity and specificity of biopsy/referral decisions for melanomas and SDN and overall diagnostic accuracy compared with visual inspection alone. The results of this study suggest that diagnostic accuracy for PSLs by PCPs may be improved with adjunctive use of EIS with visual inspection.

Primary care providers (PCPs) are often the first line of defense in detecting skin cancers. For patients with concerning skin lesions, PCPs may choose to perform a biopsy or facilitate access to specialty services (eg, Dermatology). Consequently, PCPs play a critical role in the timely detection of skin cancers, and it is paramount to employ continually improving detection methods, such as the application of technologic advances.¹

Differentiating benign nevi from melanoma and severely dysplastic nevi (SDN), both of which warrant excision, poses a unique challenge to clinicians examining pigmented skin lesions (PSLs). PCPs often rely on visual inspection to differentiate benign skin lesions from malignant skin cancers. In some primary care practices, dermoscopy, which involves using a handheld device to evaluate lesions with polarized light and magnification, is used to improve melanoma detection. However, while visual inspection and dermoscopy are valid, effective techniques for the diagnosis of melanocytic lesions, in many instances they still can lead to missed cancers or unnecessary biopsies and specialty referrals. Adjunctive use of dermoscopy with visual inspection has been shown to increase the probability of skin cancer detection, but it fails to achieve a near-100% success rate.² Furthermore, dermoscopy is heavily user-dependent, requiring significant training and experience for appropriate use.³

Another option is an electrical impedance spectroscopy (EIS) device (Nevisense, Scibase, Stockholm, Sweden), which has been approved by the US Food and Drug Administration (FDA) to assist in the detection of melanoma and differentiation from benign PSLs.⁴ EIS is a noninvasive, rapidly applied technology designed to accompany the visual examination of melanocytic lesions in office, with or without dermoscopy. Still relatively new, the technology is employed today by many dermatologists, increasing diagnostic accuracy for PSLs.⁵ The lightweight and portable instrument features a handheld probe, which is held against a lesion to obtain a reading. EIS uses a low-voltage electrode to apply a harmless electrical current to the skin at various frequencies.⁶ As benign and malignant tissues vary in cell shape, size, and composition, EIS distinguishes differential electrical resistance of the tissue to aid in diagnosis.⁷

Continue to: EIS provides high-sensitivity...

EIS provides high-sensitivity melanoma diagnosis vs histopathologic confirmation from biopsies, with 1 study showing a 96.6% sensitivity rating, detecting 256 of 265 melanomas.⁴ The EIS device, by measuring differences in electrical resistance between benign and cancerous cells, outputs a simple integer score ranging from 0 to 10 associated with the likelihood of the lesion being a melanoma.⁸ Based on data from the Nevisense pivotal trial,⁴ Nevisense reports that scores of 0 to 3 carry a negative predictive value of 99% for melanoma, whereas scores of 4 to 10 signify increasingly greater positive predictive values from 7% to 61%.

Findings suggest that the use of electrical impedance spectroscopy is particularly advantageous to clinicians who are less proficient in assessing melanocytic lesions.

We aimed to assess whether EIS may be beneficial to PCPs by comparing the accuracy of clinical decision-making for PSLs based on visual examination alone with that based on concurrent visual and EIS evaluation.

METHODS

A questionnaire was distributed via email to 142 clinicians at clinics affiliated with either of 2 organizations delivering care to the New York City area through a network of community health centers: the Institute for Family Health (IFH) and the Community Healthcare Network (CHN). Of these recipients, 72 were affiliated with IFH across 27 community health centers and 70 were affiliated with CHN across 14 community health centers. Recipients were physicians and nurse practitioners (NPs) practicing at primary health care facilities.

Survey instrument. The first section of the survey instrument (APPENDIX) solicited demographic information and explained how to apply the EIS scores for diagnostic ­decision-making. The second featured images of 12 randomly selected, histologically confirmed, and EIS-evaluated PSLs from a previously published prospective blinded trial of 2416 lesions.⁴ The Institutional Review Board of the Icahn School of Medicine at Mount Sinai reviewed and approved the study and survey instrument.

Clinical images of these lesions, comprising 4 melanocytic nevi, 4 dysplastic nevi (including 3 mild-moderately dysplastic and 1 severely dysplastic nevus), and 4 melanomas, were first presented to respondents with 2 tasks: (1) rate on a scale of 1 to 5 their likelihood to biopsy or refer this lesion to a dermatologist (1: not likely; 5: extremely likely); and (2) select what they expect the pathology results to be: melanocytic nevus, dysplastic nevus, or malignant melanoma. Subsequently, respondents repeated the assessments after being presented with the EIS score for the same lesion in conjunction with the clinical image.

Continue to: Analysis

Analysis. A biopsy or referral rating of 4 or 5 was considered a decision to biopsy or refer (ie, a diagnostic decision consistent with melanoma or SDN warranting excision), whereas a selection of 1 to 3 was considered a decision not to biopsy or refer (ie, a diagnostic decision consistent with a benign PSL). The sensitivity and specificity of biopsy/­referral decisions for melanomas and SDN, the proportion of missed melanomas and SDN, and the proportion of biopsy/referral decisions for benign lesions were separately determined for visual inspection alone and visual inspection with EIS score. Similarly, diagnostic accuracy was calculated for these clinical scenarios. These metrics were further stratified among different subsets of the respondent population. Differences in sensitivity, specificity, biopsy/referral decision proportions, and diagnostic accuracy were calculated using McNemar’s test for paired proportions.

RESULTS

Sixty-one respondents, comprising 44 physicians and 17 NPs, completed the survey, yielding a response rate of 43% (TABLE 1). In total, 1354 clinical decisions (677 based on visual inspection alone and 677 based on visual inspection plus EIS) were made. A biopsy/­referral decision was made after assessing 416 of 677 cases (61%) with visual inspection alone and 360 of 677 cases (53%) when relying on visual inspection plus EIS. None of the respondents reported any prior experience with EIS.

When incorporating EIS scores, respondents’ mean sensitivity for melanomas and SDN increased from 69.2% to 90.0% (P < .001) and specificity from 44.0% to 72.6% (P < .001; TABLE 2). At baseline, physicians demonstrated a sensitivity and specificity of 74.6% and 46.5%, respectively, while NPs demonstrated a sensitivity and specificity of 56.1% and 37.9%, respectively.

All respondent subgroups stratified by occupation and years of experience saw significant increases in both sensitivity and specificity upon the incorporation of EIS scores, with NPs seeing a greater increase in sensitivity (56.1% vs 85.4%; P < .001) and specificity (37.9% vs 69.0%; P < .001) than physicians (sensitivity: 74.6% vs 91.9%; P < .001; specificity: 46.5% vs 74.1%; P < .001). The only difference in diagnostic performance based on years of experience was a greater pre-EIS sensitivity by clinicians who had been in practice for ≥ 15 years, compared with those in practice for shorter periods (TABLE 2).

The improvements, seen in clinicians of varying training and experience, suggest that the learning curve of EIS may not be as steep as that of dermoscopy.

Diagnostic accuracy increased significantly from 48% when based on visual inspection alone to 73% with the addition of EIS scores (P < .001; TABLE 3). Physicians and NPs each significantly increased their diagnostic accuracy upon the incorporation of EIS, with NPs exhibiting the greatest increase (from 36.9% to 65.7%; P < .001). PCPs with 6 to 14 years of experience saw the greatest increase in diagnostic accuracy when adding EIS (45.9% vs 76.4%; P < .001). Overall, the addition of EIS scores resulted in 58 fewer missed melanomas and SDN and 114 fewer benign referrals or biopsies (TABLE 4).

Continue to: DISCUSSION

DISCUSSION

Primary care evaluation plays a significant role in the diagnosis and management of PSLs, ultimately shaping outcomes for patients with melanoma. Improved accuracy of PSL classification could yield greater sensitivity for the diagnosis of melanomas and high-risk melanocytic lesions at earlier stages, while also reducing the number of unnecessary biopsies and referrals—leading to decreased patient morbidity and mortality and reduced health care spending.⁹

Diagnostic tools are valuable insofar as they can improve accuracy and positively impact clinical management and patient outcomes.¹⁰ In this case, increased sensitivity reduced missed melanoma diagnoses, while increased specificity avoided the additional costs and patient toll associated with a biopsy or referral for a benign lesion.

Dermoscopy has been shown to improve the sensitivity and specificity of PSL diagnosis compared with visual inspection alone; however, without substantial training and experience, accuracy with dermoscopy can be no better than examination with the naked eye.^3,11,12 The dropout rates are high for training PCPs in its use, given that several months of training may be needed for competent use.^13,14 To improve the clinical management of PSLs broadly in primary care, a need exists for easy-to-use adjunctive tools that increase diagnostic accuracy.¹⁵

In this study, with only a brief explanation of how to interpret EIS scores, clinicians without any prior experience using EIS demonstrated significantly improved accuracy in deciding appropriate management and classifying melanocytic lesions with the addition of EIS to visual inspection. These improvements, seen in clinicians of varying training and experience, suggest that the learning curve of EIS may not be as steep as that of dermoscopy.

The greater baseline sensitivity, specificity, and diagnostic accuracy of physicians’ clinical decision-making compared with NPs before the incorporation of EIS in the study may be a product of comparatively more extensive medical training. In addition, EIS yielded a greater benefit to NPs than to physicians, with greater increases in sensitivity and specificity noted. This suggests that the use of EIS is particularly advantageous to clinicians who are less proficient in assessing melanocytic lesions. Using visual inspection alone, more experienced respondents made biopsy/referral decisions with greater sensitivity but similar specificity to those with less experience. With the incorporation of EIS scores, the sensitivity and specificity of respondents’ clinical decision-making rose to comparable levels across all experience groups, providing further indication of EIS’s particular value to clinicians who are less proficient in PSL evaluation.

Continue to: This technology holds the potential...

This technology holds the potential to be seamlessly implemented into primary care practice, given that dermatology expertise training is not required to use the EIS device; this could allow for EIS measurement of lesions to be delegated to office staff (eg, nurses, medical assistants).¹⁶ Future studies are needed to assess EIS use among PCPs in a real-world setting, where factors such as its application on nonmelanocytic lesions (eg, seborrheic keratoses) and its pairing with patient historical data could produce varying results.

Limitations. While revealing, this study had its limitations. Respondents did not have access to additional pertinent clinical information, such as patients’ histories and risk factors. Clinical decisions in this survey were made based on digital images rather than in vivo examination. This may not represent a real-life evaluation; there is the potential for minimization of the true consequences of a missed melanoma or unnecessary biopsy in the minds of participants, and this does not factor in the operation of the actual EIS device. The Hawthorne effect may also have influenced PCPs’ diagnostic selections. Also, the limited sample size constitutes another limitation.

The results of this preliminary study suggest that diagnostic accuracy for pigmented skin lesions by PCPs may be improved with the adjunctive use of electrical impedance spectroscopy with visual inspection.

Of note, in this survey format, respondents rated their inclination to biopsy or refer each lesion from 1 to 5. For statistical analyses, lesions rated 1 to 3 were considered as not biopsied/referred and those rated 4 to 5 as biopsied/referred. The sensitivity and specificity values observed, for both visual examination and concurrent visual and EIS evaluation, are therefore based on this classification system of participants’ provided ratings. It is conceivable that differing sensitivity and specificity values might have been detected if clinicians were instead given a binary choice for referral/biopsy decisions.

CONCLUSIONS

Among PCPs tasked with evaluating melanocytic lesions, the incorporation of EIS data into clinical decision-making in this study significantly increased the sensitivity, specificity, and overall diagnostic accuracy of biopsy or referral decisions for melanomas and SDN compared with visual inspection alone. Overall, the results of this preliminary study suggest that diagnostic accuracy for PSLs by PCPs may be improved with the adjunctive use of EIS with visual inspection. This would ultimately improve patient care and reduce the morbidity and mortality of a melanoma diagnosis.

CORRESPONDENCE
Jonathan Ungar, MD, Kimberly and Eric J. Waldman Department of Dermatology, Icahn School of Medicine at Mount Sinai, 5 East 98th Street, 5th Floor, New York, NY 10029; jonathan.ungar@mountsinai.org

ABSTRACT

Background: We sought to examine whether electrical impedance spectroscopy (EIS), a diagnostic tool approved by the US Food and Drug Administration for the evaluation of pigmented skin lesions (PSLs), is beneficial to primary care providers (PCPs) by comparing the accuracy of PCPs’ management decisions for PSLs based on visual examination alone with those based on concurrent visual and EIS evaluation.

Methods: Physicians and nurse practitioners (NPs) participated in an anonymous online survey in which they viewed clinical images of PSLs and were asked to make 2 clinical decisions before and after being provided an EIS score that indicated the likelihood that the lesion was a melanoma. They were asked (1) if they would biopsy the lesion/refer the patient out and (2) what they expected the pathology results would show.

Results: Forty-four physicians and 17 NPs participated, making clinical decisions for 1354 presented lesions. Overall, with the addition of EIS to visual inspection of clinical images, the sensitivity of biopsy/referral decisions for melanomas and severely dysplastic nevi (SDN) increased from 69.2% to 90.0% (P < .001), while specificity increased from 44.0% to 72.6% (P < .001). Physicians and NPs, regardless of years of experience, each saw significant improvements in sensitivity, specificity, and diagnostic accuracy with the addition of EIS scores.

Conclusions: The incorporation of EIS data into clinical decision-making by PCPs significantly increased the sensitivity and specificity of biopsy/referral decisions for melanomas and SDN and overall diagnostic accuracy compared with visual inspection alone. The results of this study suggest that diagnostic accuracy for PSLs by PCPs may be improved with adjunctive use of EIS with visual inspection.

Primary care providers (PCPs) are often the first line of defense in detecting skin cancers. For patients with concerning skin lesions, PCPs may choose to perform a biopsy or facilitate access to specialty services (eg, Dermatology). Consequently, PCPs play a critical role in the timely detection of skin cancers, and it is paramount to employ continually improving detection methods, such as the application of technologic advances.¹

Differentiating benign nevi from melanoma and severely dysplastic nevi (SDN), both of which warrant excision, poses a unique challenge to clinicians examining pigmented skin lesions (PSLs). PCPs often rely on visual inspection to differentiate benign skin lesions from malignant skin cancers. In some primary care practices, dermoscopy, which involves using a handheld device to evaluate lesions with polarized light and magnification, is used to improve melanoma detection. However, while visual inspection and dermoscopy are valid, effective techniques for the diagnosis of melanocytic lesions, in many instances they still can lead to missed cancers or unnecessary biopsies and specialty referrals. Adjunctive use of dermoscopy with visual inspection has been shown to increase the probability of skin cancer detection, but it fails to achieve a near-100% success rate.² Furthermore, dermoscopy is heavily user-dependent, requiring significant training and experience for appropriate use.³

Another option is an electrical impedance spectroscopy (EIS) device (Nevisense, Scibase, Stockholm, Sweden), which has been approved by the US Food and Drug Administration (FDA) to assist in the detection of melanoma and differentiation from benign PSLs.⁴ EIS is a noninvasive, rapidly applied technology designed to accompany the visual examination of melanocytic lesions in office, with or without dermoscopy. Still relatively new, the technology is employed today by many dermatologists, increasing diagnostic accuracy for PSLs.⁵ The lightweight and portable instrument features a handheld probe, which is held against a lesion to obtain a reading. EIS uses a low-voltage electrode to apply a harmless electrical current to the skin at various frequencies.⁶ As benign and malignant tissues vary in cell shape, size, and composition, EIS distinguishes differential electrical resistance of the tissue to aid in diagnosis.⁷

Continue to: EIS provides high-sensitivity...

EIS provides high-sensitivity melanoma diagnosis vs histopathologic confirmation from biopsies, with 1 study showing a 96.6% sensitivity rating, detecting 256 of 265 melanomas.⁴ The EIS device, by measuring differences in electrical resistance between benign and cancerous cells, outputs a simple integer score ranging from 0 to 10 associated with the likelihood of the lesion being a melanoma.⁸ Based on data from the Nevisense pivotal trial,⁴ Nevisense reports that scores of 0 to 3 carry a negative predictive value of 99% for melanoma, whereas scores of 4 to 10 signify increasingly greater positive predictive values from 7% to 61%.

Findings suggest that the use of electrical impedance spectroscopy is particularly advantageous to clinicians who are less proficient in assessing melanocytic lesions.

We aimed to assess whether EIS may be beneficial to PCPs by comparing the accuracy of clinical decision-making for PSLs based on visual examination alone with that based on concurrent visual and EIS evaluation.

METHODS

A questionnaire was distributed via email to 142 clinicians at clinics affiliated with either of 2 organizations delivering care to the New York City area through a network of community health centers: the Institute for Family Health (IFH) and the Community Healthcare Network (CHN). Of these recipients, 72 were affiliated with IFH across 27 community health centers and 70 were affiliated with CHN across 14 community health centers. Recipients were physicians and nurse practitioners (NPs) practicing at primary health care facilities.

Survey instrument. The first section of the survey instrument (APPENDIX) solicited demographic information and explained how to apply the EIS scores for diagnostic ­decision-making. The second featured images of 12 randomly selected, histologically confirmed, and EIS-evaluated PSLs from a previously published prospective blinded trial of 2416 lesions.⁴ The Institutional Review Board of the Icahn School of Medicine at Mount Sinai reviewed and approved the study and survey instrument.

Clinical images of these lesions, comprising 4 melanocytic nevi, 4 dysplastic nevi (including 3 mild-moderately dysplastic and 1 severely dysplastic nevus), and 4 melanomas, were first presented to respondents with 2 tasks: (1) rate on a scale of 1 to 5 their likelihood to biopsy or refer this lesion to a dermatologist (1: not likely; 5: extremely likely); and (2) select what they expect the pathology results to be: melanocytic nevus, dysplastic nevus, or malignant melanoma. Subsequently, respondents repeated the assessments after being presented with the EIS score for the same lesion in conjunction with the clinical image.

Continue to: Analysis

Analysis. A biopsy or referral rating of 4 or 5 was considered a decision to biopsy or refer (ie, a diagnostic decision consistent with melanoma or SDN warranting excision), whereas a selection of 1 to 3 was considered a decision not to biopsy or refer (ie, a diagnostic decision consistent with a benign PSL). The sensitivity and specificity of biopsy/­referral decisions for melanomas and SDN, the proportion of missed melanomas and SDN, and the proportion of biopsy/referral decisions for benign lesions were separately determined for visual inspection alone and visual inspection with EIS score. Similarly, diagnostic accuracy was calculated for these clinical scenarios. These metrics were further stratified among different subsets of the respondent population. Differences in sensitivity, specificity, biopsy/referral decision proportions, and diagnostic accuracy were calculated using McNemar’s test for paired proportions.

RESULTS

Sixty-one respondents, comprising 44 physicians and 17 NPs, completed the survey, yielding a response rate of 43% (TABLE 1). In total, 1354 clinical decisions (677 based on visual inspection alone and 677 based on visual inspection plus EIS) were made. A biopsy/­referral decision was made after assessing 416 of 677 cases (61%) with visual inspection alone and 360 of 677 cases (53%) when relying on visual inspection plus EIS. None of the respondents reported any prior experience with EIS.

When incorporating EIS scores, respondents’ mean sensitivity for melanomas and SDN increased from 69.2% to 90.0% (P < .001) and specificity from 44.0% to 72.6% (P < .001; TABLE 2). At baseline, physicians demonstrated a sensitivity and specificity of 74.6% and 46.5%, respectively, while NPs demonstrated a sensitivity and specificity of 56.1% and 37.9%, respectively.

All respondent subgroups stratified by occupation and years of experience saw significant increases in both sensitivity and specificity upon the incorporation of EIS scores, with NPs seeing a greater increase in sensitivity (56.1% vs 85.4%; P < .001) and specificity (37.9% vs 69.0%; P < .001) than physicians (sensitivity: 74.6% vs 91.9%; P < .001; specificity: 46.5% vs 74.1%; P < .001). The only difference in diagnostic performance based on years of experience was a greater pre-EIS sensitivity by clinicians who had been in practice for ≥ 15 years, compared with those in practice for shorter periods (TABLE 2).

The improvements, seen in clinicians of varying training and experience, suggest that the learning curve of EIS may not be as steep as that of dermoscopy.

Diagnostic accuracy increased significantly from 48% when based on visual inspection alone to 73% with the addition of EIS scores (P < .001; TABLE 3). Physicians and NPs each significantly increased their diagnostic accuracy upon the incorporation of EIS, with NPs exhibiting the greatest increase (from 36.9% to 65.7%; P < .001). PCPs with 6 to 14 years of experience saw the greatest increase in diagnostic accuracy when adding EIS (45.9% vs 76.4%; P < .001). Overall, the addition of EIS scores resulted in 58 fewer missed melanomas and SDN and 114 fewer benign referrals or biopsies (TABLE 4).

Continue to: DISCUSSION

DISCUSSION

Primary care evaluation plays a significant role in the diagnosis and management of PSLs, ultimately shaping outcomes for patients with melanoma. Improved accuracy of PSL classification could yield greater sensitivity for the diagnosis of melanomas and high-risk melanocytic lesions at earlier stages, while also reducing the number of unnecessary biopsies and referrals—leading to decreased patient morbidity and mortality and reduced health care spending.⁹

Diagnostic tools are valuable insofar as they can improve accuracy and positively impact clinical management and patient outcomes.¹⁰ In this case, increased sensitivity reduced missed melanoma diagnoses, while increased specificity avoided the additional costs and patient toll associated with a biopsy or referral for a benign lesion.

Dermoscopy has been shown to improve the sensitivity and specificity of PSL diagnosis compared with visual inspection alone; however, without substantial training and experience, accuracy with dermoscopy can be no better than examination with the naked eye.^3,11,12 The dropout rates are high for training PCPs in its use, given that several months of training may be needed for competent use.^13,14 To improve the clinical management of PSLs broadly in primary care, a need exists for easy-to-use adjunctive tools that increase diagnostic accuracy.¹⁵

In this study, with only a brief explanation of how to interpret EIS scores, clinicians without any prior experience using EIS demonstrated significantly improved accuracy in deciding appropriate management and classifying melanocytic lesions with the addition of EIS to visual inspection. These improvements, seen in clinicians of varying training and experience, suggest that the learning curve of EIS may not be as steep as that of dermoscopy.

The greater baseline sensitivity, specificity, and diagnostic accuracy of physicians’ clinical decision-making compared with NPs before the incorporation of EIS in the study may be a product of comparatively more extensive medical training. In addition, EIS yielded a greater benefit to NPs than to physicians, with greater increases in sensitivity and specificity noted. This suggests that the use of EIS is particularly advantageous to clinicians who are less proficient in assessing melanocytic lesions. Using visual inspection alone, more experienced respondents made biopsy/referral decisions with greater sensitivity but similar specificity to those with less experience. With the incorporation of EIS scores, the sensitivity and specificity of respondents’ clinical decision-making rose to comparable levels across all experience groups, providing further indication of EIS’s particular value to clinicians who are less proficient in PSL evaluation.

Continue to: This technology holds the potential...

This technology holds the potential to be seamlessly implemented into primary care practice, given that dermatology expertise training is not required to use the EIS device; this could allow for EIS measurement of lesions to be delegated to office staff (eg, nurses, medical assistants).¹⁶ Future studies are needed to assess EIS use among PCPs in a real-world setting, where factors such as its application on nonmelanocytic lesions (eg, seborrheic keratoses) and its pairing with patient historical data could produce varying results.

Limitations. While revealing, this study had its limitations. Respondents did not have access to additional pertinent clinical information, such as patients’ histories and risk factors. Clinical decisions in this survey were made based on digital images rather than in vivo examination. This may not represent a real-life evaluation; there is the potential for minimization of the true consequences of a missed melanoma or unnecessary biopsy in the minds of participants, and this does not factor in the operation of the actual EIS device. The Hawthorne effect may also have influenced PCPs’ diagnostic selections. Also, the limited sample size constitutes another limitation.

The results of this preliminary study suggest that diagnostic accuracy for pigmented skin lesions by PCPs may be improved with the adjunctive use of electrical impedance spectroscopy with visual inspection.

Of note, in this survey format, respondents rated their inclination to biopsy or refer each lesion from 1 to 5. For statistical analyses, lesions rated 1 to 3 were considered as not biopsied/referred and those rated 4 to 5 as biopsied/referred. The sensitivity and specificity values observed, for both visual examination and concurrent visual and EIS evaluation, are therefore based on this classification system of participants’ provided ratings. It is conceivable that differing sensitivity and specificity values might have been detected if clinicians were instead given a binary choice for referral/biopsy decisions.

CONCLUSIONS

Among PCPs tasked with evaluating melanocytic lesions, the incorporation of EIS data into clinical decision-making in this study significantly increased the sensitivity, specificity, and overall diagnostic accuracy of biopsy or referral decisions for melanomas and SDN compared with visual inspection alone. Overall, the results of this preliminary study suggest that diagnostic accuracy for PSLs by PCPs may be improved with the adjunctive use of EIS with visual inspection. This would ultimately improve patient care and reduce the morbidity and mortality of a melanoma diagnosis.

CORRESPONDENCE
Jonathan Ungar, MD, Kimberly and Eric J. Waldman Department of Dermatology, Icahn School of Medicine at Mount Sinai, 5 East 98th Street, 5th Floor, New York, NY 10029; jonathan.ungar@mountsinai.org

ABSTRACT

Background: The Altman Rule, a simple tool for consumers seeking to make healthier packaged food choices at the point of sale, applies to packaged carbohydrates. According to the Altman Rule, a food is a healthier option if it has at least 3 g of fiber per serving and the grams of fiber plus the grams of protein exceed the grams of sugar per serving. This study sought to evaluate whether the Altman Rule is a valid proxy for glycemic load (GL).

Methods: We compared the binary outcome of whether a food item meets the Altman Rule with the GL of all foods categorized as cereals, chips, crackers, and granola bars in the Nutrition Data System for Research Database (University of Minnesota, Version 2010). We examined the percentage of foods in low-, medium-, and high-GL categories that met the Altman Rule.

Results: There were 1235 foods (342 cereals, 305 chips, 379 crackers, and 209 granola bars) in this analysis. There was a significant relationship between the GL of foods and the Altman Rule (P < .001) in that most low-GL (68%), almost half of medium-GL (48%), and very few high-GL (7%) foods met the criteria of the rule.

Conclusions: The Altman Rule is a reasonable proxy for GL and can be a useful and accessible tool for consumers interested in buying healthier packaged carbohydrate foods.

Nutrition can be complicated for consumers interested in making healthier choices at the grocery store. Consumers may have difficulty identifying more nutritious options, especially when food labels are adorned with claims such as “Good Source of Fiber” or “Heart Healthy.”¹ In addition, when reading food labels, consumers may find it difficult to decipher which data to prioritize when carbohydrates, total sugars, added sugars, total dietary fiber, soluble fiber, and insoluble fiber are all listed.

The concept of glycemic load (GL) is an important consideration, especially for people with diabetes. GL approximates the blood sugar response to different foods. A food with a high GL is digested quickly, and its carbohydrates are taken into the bloodstream rapidly. This leads to a spike and subsequent drop in blood sugars, which can cause symptoms of hyperglycemia and hypoglycemia in a person with diabetes.^2,3 Despite its usefulness, GL may be too complicated for a consumer to understand, and it does not appear anywhere on the food label. Since GL is calculated using pooled blood sugar response from individuals after the ingestion of the particular food, estimation of the GL is not intuitable.⁴

Point-of-sale tools. People seeking to lose weight, control diabetes, improve dyslipidemia and/or blood pressure, and/or decrease their risk for heart disease may benefit from point-of-sale tools such as the Altman Rule, which simplifies and encourages the selection of more nutritious foods.¹ Other tools—such as Guiding Stars (https://guidingstars.com), NuVal (www.nuval.com), and different variations of traffic lights—have been created to help consumers make more informed and healthier food choices.^5-8 However, Guiding Stars and NuVal are based on complicated algorithms that are not entirely transparent and not accessible to the average consumer.^6,7 Evaluations of these nutrition tools indicate that consumers tend to underrate the healthiness of some foods, such as raw almonds and salmon, and overrate the healthiness of others, such as fruit punch and diet soda, when using traffic light systems.⁶ Furthermore, these nutrition tools are not available in many supermarkets. Previous research suggests that the use of point-of-sale nutrition apps decreases with the time and effort involved in using an app.⁹

Continue to: The Altman Rule

The Altman Rule was developed by a family physician (author WA) to provide a more accessible tool for people interested in choosing healthier prepackaged carbohydrate foods while shopping. Since the user does not need to have a smartphone, and they are not required to download or understand an app for each purchase, the Altman Rule may be more usable compared with more complicated alternatives.

The Altman Rule can be used with nutrition labels that feature serving information and calories in enlarged and bold type, in compliance with the most recent US Food and Drug Administration (FDA) guideline from 2016. Many foods with high fiber also have high amounts of sugar, so the criteria of the Altman Rule includes a 2-step process requiring (1) a minimum of 3 g of total dietary fiber per serving and (2) the sum of the grams of fiber plus the grams of protein per serving to be greater than the total grams of sugar (not grams of added sugar or grams of carbohydrate) per serving (FIGURE 1A). Unlike the relatively complicated formula related to GL, this 2-part rule can be applied in seconds while shopping (FIGURE 1B).

The rule is intended only to be used for packaged carbohydrate products, such as bread, muffins, bagels, pasta, rice, oatmeal, cereals, snack bars, chips, and crackers. It does not apply to whole foods, such as meat, dairy, fruits, or vegetables. These foods are excluded to prevent any consumer confusion related to the nutritional content of whole foods (eg, an apple may have more sugar than fiber and protein combined, but it is still a nutritious option).

Since the user does not need to have a smartphone, the Altman Rule may be more usable compared with more complicated alternatives.

This study aimed to determine if the Altman Rule is a reasonable proxy for the more complicated concept of GL. We calculated the relationship between the GL of commercially available packaged carbohydrate foods and whether those foods met the Altman Rule.

METHODS

The Altman Rule was tested by comparing the binary outcome of the rule (meets/does not meet) with data on all foods categorized as cereals, chips, crackers, and granola bars in the Nutrition Data System for Research (NDSR) Database (University of Minnesota, Version 2010).

Continue to: To account for differences...

To account for differences in serving size, we used the standard of 50 g for each product as 1 serving. We used 50 g (about 1.7 oz) to help compare the different foods and between foods within the same group. Additionally, 50 g is close to 1 serving for most foods in these groups; it is about the size of a typical granola bar, three-quarters to 2 cups of cereal, 10 to 12 crackers, and 15 to 25 chips. We determined the GL for each product by multiplying the number of available carbohydrates (total carbohydrate – dietary fiber) by the product’s glycemic index/100. In general, GL is categorized as low (≤ 10), medium (11-19), or high (≥ 20).

We applied the Altman Rule to categorize each product as meeting or not meeting the rule. We compared the proportion of foods meeting the Altman Rule, stratified by GL and by specific foods, and used chi-square to determine if differences were statistically significant. These data were collected and analyzed in the summer of 2019.

RESULTS

There were 1235 foods (342 breakfast cereals, 305 chips, 379 crackers, and 209 granola bars) used for this analysis. There is a significant relationship between the GL of foods and the Altman Rule in that most low-GL (68%), almost half of medium-GL (48%), and only a few high-GL foods (7%) met the rule (P < .001) (TABLE 1). There was also a significant relationship between “meeting the ­Altman Rule” and GL within each food type (P < .001) (TABLE 2).

The medium-GL foods were the second largest category of foods we calculated; thus we further broke them into binary categories of low-medium GL (values 11-14) and high-medium GL (values 15-19) to explore the results of the Altman Rule. About half of the foods in medium-GL category met the Altman Rule. About eighty-five percent of the foods with low-medium GL passed the Altman Rule, while only 39% of the foods with high-medium GL did.

Foods that met the rule were more likely to be low GL and foods that did not pass the rule were more likely high GL. Within the medium-GL category, foods that met the rule were more likely to be low-medium GL. 

Continue to: The findings within food categories...

The findings within food categories showed that very few cereals, chips, crackers, and granola bars were low GL. For every food category, except granola bars, far more low-GL foods met the Altman Rule than those that did not. At the same time, very few high-GL foods met the Altman Rule. The category with the most individual high-GL food items meeting the Altman Rule was cereal. This was also the subcategory with the largest percentage of high-GL food items meeting the Altman Rule. Thirty-nine cereals that were high GL met the rule, but more than 4 times as many high-GL cereals did not (n = 190).

DISCUSSION

Marketing and nutrition messaging create consumer confusion that makes it challenging to identify packaged food items that are more nutrient dense. The Altman Rule simplifies food choices that have become unnecessarily complex. Our findings suggest this 2-step rule is a reasonable proxy for the more complicated and less accessible GL for packaged carbohydrates, such as cereals, chips, crackers, and snack bars. Foods that meet the rule are likely low or low-medium GL and thus are foods that are likely to be healthier choices.

Our findings suggest this 2-step rule is a reasonable proxy for the more complicated and less accessible glycemic load for packaged carbohydrates.

Of note, only 9% of chips (n = 27) passed the Altman Rule, likely due to their low dietary fiber content, which was typical of chips. If a food item does not have at least 3 grams of total dietary fiber per serving, it does not pass the Altman Rule, regardless of how much protein or sugar is in the product. This may be considered a strength or a weakness of the Altman Rule. Few nutrition-dense foods are low in fiber, but some foods could be nutritious but do not meet the Altman Rule due to having < 3 g of fiber.

With the high prevalence of chronic diseases such as hypertension, diabetes, hyperlipidemia, and cardiovascular disease, it is essential to help consumers prevent chronic disease altogether or manage their chronic disease by providing tools to identify healthier food choices. The tool also has a place in clinical medicine for use by physicians and other health care professionals. Research shows that physicians find both time and lack of ­knowledge/resources to be a barrier to providing nutritional counseling to patients.¹⁰ Since the Altman Rule can be shared and explained with very little time and without extensive nutritional knowledge, it meets these needs.

Limitations

Glycemic load. We acknowledge that the Altman Rule is not foolproof and that assessing this rule based on GL has some limitations. GL is not a perfect or comprehensive way to measure the nutritional value of a food. For example, fruits such as watermelon and grapes are nutritionally dense. However, they contain high amounts of natural sugars—and as such, their GL is relatively high, which could lead a consumer to perceive them as unhealthy. Nevertheless, GL is both a useful and accepted tool and a reasonable way to assess the validity of the rule, specifically when assessing packaged carbohydrates. The simplicity of the Altman Rule and its relationship with GL makes it such that consumers are more likely to make a healthier food choice using it.⁹

Continue to: Specificity and sensitivity

Specificity and sensitivity. There are other limitations to the Altman Rule, given that a small number of high-GL foods meet the rule. For example, some granola bars had high dietary protein, which offset a high sugar content just enough to pass the rule despite a higher GL. As such, concluding that a snack bar is a healthier choice because it meets the Altman Rule when it has high amounts of sugar may not be appropriate. This limitation could be considered a lack of specificity (the rule includes food it ought not to include). Another limitation to consider would be a lack of sensitivity, given that only 68% of low-GL foods passed the Altman Rule. Since GL is associated with carbohydrate content, foods with a low carbohydrate count often have little to no fiber and thus would fall into the category of foods that did not meet the Altman Rule but had low GL. In this case, however, the low amount of fiber may render the Altman Rule a better indicator of a healthier food choice than the GL.

Hidden sugars. Foods with sugar alcohols and artificial sweeteners may be as deleterious as caloric alternatives while not being accounted for when reporting the grams of sugar per serving on the nutrition label.⁷ This may represent an exception to the Altman Rule, as foods that are not healthier choices may pass the rule because the sugar content on the nutrition label is, in a sense, artificially lowered. Future research may investigate the hypothesis that these foods are nutritionally inferior despite meeting the Altman Rule.

The sample. Our study also was limited to working only with foods that were included in the NDSR database up to 2010. This limitation is mitigated by the fact that the sample size was large (> 1000 packaged food items were included in our analyses). The study also could be limited by the food categories that were analyzed; food categories such as bread, rice, pasta, and bagels were not included.

The objective of this research was to investigate the relationship between GL and the Altman Rule, rather than to conduct an exhaustive analysis of the Altman Rule for every possible food category. Studying the relationship between the Altman Rule and GL in other categories of food is an objective for future research. The data so far support a relationship between these entities. The likelihood of the nutrition facts of foods changing without the GL changing (or vice versa) is very low. As such, the Altman Rule still seems to be a reasonable proxy of GL.

CONCLUSIONS

Research indicates that point-of-sale tools, such as Guiding Stars, NuVal, and other stoplight tools, can successfully alter consumers’ behaviors.⁹ These tools can be helpful but are not available in many supermarkets. Despite the limitations, the Altman Rule is a useful decision aid that is accessible to all consumers no matter where they live or shop and is easy to use and remember.

The Altman rule can be used in clinical practice by health care professionals, such as physicians, nurse practitioners, physician assistants, dietitians, and health coaches. It also has the potential to be used in commercial settings, such as grocery stores, to help consumers easily identify healthier convenience foods. This has public health implications, as the rule can both empower consumers and potentially incentivize food manufacturers to upgrade their products nutritionally.

Additional research would be useful to evaluate consumers’ preferences and perceptions about how user-friendly the Altman Rule is at the point of sale with packaged carbohydrate foods. This would help to further understand how the use of information on food packaging can motivate healthier decisions—thereby helping to alleviate the burden of chronic disease.

CORRESPONDENCE
Kimberly R. Dong, DrPH, MS, RDN, Tufts University School of Medicine, Department of Public Health and Community Medicine, 136 Harrison Avenue, MV Building, Boston, MA 02111; kimberly.dong@tufts.edu

ABSTRACT

Background: The Altman Rule, a simple tool for consumers seeking to make healthier packaged food choices at the point of sale, applies to packaged carbohydrates. According to the Altman Rule, a food is a healthier option if it has at least 3 g of fiber per serving and the grams of fiber plus the grams of protein exceed the grams of sugar per serving. This study sought to evaluate whether the Altman Rule is a valid proxy for glycemic load (GL).

Methods: We compared the binary outcome of whether a food item meets the Altman Rule with the GL of all foods categorized as cereals, chips, crackers, and granola bars in the Nutrition Data System for Research Database (University of Minnesota, Version 2010). We examined the percentage of foods in low-, medium-, and high-GL categories that met the Altman Rule.

Results: There were 1235 foods (342 cereals, 305 chips, 379 crackers, and 209 granola bars) in this analysis. There was a significant relationship between the GL of foods and the Altman Rule (P < .001) in that most low-GL (68%), almost half of medium-GL (48%), and very few high-GL (7%) foods met the criteria of the rule.

Conclusions: The Altman Rule is a reasonable proxy for GL and can be a useful and accessible tool for consumers interested in buying healthier packaged carbohydrate foods.

Nutrition can be complicated for consumers interested in making healthier choices at the grocery store. Consumers may have difficulty identifying more nutritious options, especially when food labels are adorned with claims such as “Good Source of Fiber” or “Heart Healthy.”¹ In addition, when reading food labels, consumers may find it difficult to decipher which data to prioritize when carbohydrates, total sugars, added sugars, total dietary fiber, soluble fiber, and insoluble fiber are all listed.

The concept of glycemic load (GL) is an important consideration, especially for people with diabetes. GL approximates the blood sugar response to different foods. A food with a high GL is digested quickly, and its carbohydrates are taken into the bloodstream rapidly. This leads to a spike and subsequent drop in blood sugars, which can cause symptoms of hyperglycemia and hypoglycemia in a person with diabetes.^2,3 Despite its usefulness, GL may be too complicated for a consumer to understand, and it does not appear anywhere on the food label. Since GL is calculated using pooled blood sugar response from individuals after the ingestion of the particular food, estimation of the GL is not intuitable.⁴

Point-of-sale tools. People seeking to lose weight, control diabetes, improve dyslipidemia and/or blood pressure, and/or decrease their risk for heart disease may benefit from point-of-sale tools such as the Altman Rule, which simplifies and encourages the selection of more nutritious foods.¹ Other tools—such as Guiding Stars (https://guidingstars.com), NuVal (www.nuval.com), and different variations of traffic lights—have been created to help consumers make more informed and healthier food choices.^5-8 However, Guiding Stars and NuVal are based on complicated algorithms that are not entirely transparent and not accessible to the average consumer.^6,7 Evaluations of these nutrition tools indicate that consumers tend to underrate the healthiness of some foods, such as raw almonds and salmon, and overrate the healthiness of others, such as fruit punch and diet soda, when using traffic light systems.⁶ Furthermore, these nutrition tools are not available in many supermarkets. Previous research suggests that the use of point-of-sale nutrition apps decreases with the time and effort involved in using an app.⁹

Continue to: The Altman Rule

The Altman Rule was developed by a family physician (author WA) to provide a more accessible tool for people interested in choosing healthier prepackaged carbohydrate foods while shopping. Since the user does not need to have a smartphone, and they are not required to download or understand an app for each purchase, the Altman Rule may be more usable compared with more complicated alternatives.

The Altman Rule can be used with nutrition labels that feature serving information and calories in enlarged and bold type, in compliance with the most recent US Food and Drug Administration (FDA) guideline from 2016. Many foods with high fiber also have high amounts of sugar, so the criteria of the Altman Rule includes a 2-step process requiring (1) a minimum of 3 g of total dietary fiber per serving and (2) the sum of the grams of fiber plus the grams of protein per serving to be greater than the total grams of sugar (not grams of added sugar or grams of carbohydrate) per serving (FIGURE 1A). Unlike the relatively complicated formula related to GL, this 2-part rule can be applied in seconds while shopping (FIGURE 1B).

The rule is intended only to be used for packaged carbohydrate products, such as bread, muffins, bagels, pasta, rice, oatmeal, cereals, snack bars, chips, and crackers. It does not apply to whole foods, such as meat, dairy, fruits, or vegetables. These foods are excluded to prevent any consumer confusion related to the nutritional content of whole foods (eg, an apple may have more sugar than fiber and protein combined, but it is still a nutritious option).

Since the user does not need to have a smartphone, the Altman Rule may be more usable compared with more complicated alternatives.

This study aimed to determine if the Altman Rule is a reasonable proxy for the more complicated concept of GL. We calculated the relationship between the GL of commercially available packaged carbohydrate foods and whether those foods met the Altman Rule.

METHODS

The Altman Rule was tested by comparing the binary outcome of the rule (meets/does not meet) with data on all foods categorized as cereals, chips, crackers, and granola bars in the Nutrition Data System for Research (NDSR) Database (University of Minnesota, Version 2010).

Continue to: To account for differences...

To account for differences in serving size, we used the standard of 50 g for each product as 1 serving. We used 50 g (about 1.7 oz) to help compare the different foods and between foods within the same group. Additionally, 50 g is close to 1 serving for most foods in these groups; it is about the size of a typical granola bar, three-quarters to 2 cups of cereal, 10 to 12 crackers, and 15 to 25 chips. We determined the GL for each product by multiplying the number of available carbohydrates (total carbohydrate – dietary fiber) by the product’s glycemic index/100. In general, GL is categorized as low (≤ 10), medium (11-19), or high (≥ 20).

We applied the Altman Rule to categorize each product as meeting or not meeting the rule. We compared the proportion of foods meeting the Altman Rule, stratified by GL and by specific foods, and used chi-square to determine if differences were statistically significant. These data were collected and analyzed in the summer of 2019.

RESULTS

There were 1235 foods (342 breakfast cereals, 305 chips, 379 crackers, and 209 granola bars) used for this analysis. There is a significant relationship between the GL of foods and the Altman Rule in that most low-GL (68%), almost half of medium-GL (48%), and only a few high-GL foods (7%) met the rule (P < .001) (TABLE 1). There was also a significant relationship between “meeting the ­Altman Rule” and GL within each food type (P < .001) (TABLE 2).

The medium-GL foods were the second largest category of foods we calculated; thus we further broke them into binary categories of low-medium GL (values 11-14) and high-medium GL (values 15-19) to explore the results of the Altman Rule. About half of the foods in medium-GL category met the Altman Rule. About eighty-five percent of the foods with low-medium GL passed the Altman Rule, while only 39% of the foods with high-medium GL did.

Foods that met the rule were more likely to be low GL and foods that did not pass the rule were more likely high GL. Within the medium-GL category, foods that met the rule were more likely to be low-medium GL. 

Continue to: The findings within food categories...

The findings within food categories showed that very few cereals, chips, crackers, and granola bars were low GL. For every food category, except granola bars, far more low-GL foods met the Altman Rule than those that did not. At the same time, very few high-GL foods met the Altman Rule. The category with the most individual high-GL food items meeting the Altman Rule was cereal. This was also the subcategory with the largest percentage of high-GL food items meeting the Altman Rule. Thirty-nine cereals that were high GL met the rule, but more than 4 times as many high-GL cereals did not (n = 190).

DISCUSSION

Marketing and nutrition messaging create consumer confusion that makes it challenging to identify packaged food items that are more nutrient dense. The Altman Rule simplifies food choices that have become unnecessarily complex. Our findings suggest this 2-step rule is a reasonable proxy for the more complicated and less accessible GL for packaged carbohydrates, such as cereals, chips, crackers, and snack bars. Foods that meet the rule are likely low or low-medium GL and thus are foods that are likely to be healthier choices.

Our findings suggest this 2-step rule is a reasonable proxy for the more complicated and less accessible glycemic load for packaged carbohydrates.

Of note, only 9% of chips (n = 27) passed the Altman Rule, likely due to their low dietary fiber content, which was typical of chips. If a food item does not have at least 3 grams of total dietary fiber per serving, it does not pass the Altman Rule, regardless of how much protein or sugar is in the product. This may be considered a strength or a weakness of the Altman Rule. Few nutrition-dense foods are low in fiber, but some foods could be nutritious but do not meet the Altman Rule due to having < 3 g of fiber.

With the high prevalence of chronic diseases such as hypertension, diabetes, hyperlipidemia, and cardiovascular disease, it is essential to help consumers prevent chronic disease altogether or manage their chronic disease by providing tools to identify healthier food choices. The tool also has a place in clinical medicine for use by physicians and other health care professionals. Research shows that physicians find both time and lack of ­knowledge/resources to be a barrier to providing nutritional counseling to patients.¹⁰ Since the Altman Rule can be shared and explained with very little time and without extensive nutritional knowledge, it meets these needs.

Limitations

Glycemic load. We acknowledge that the Altman Rule is not foolproof and that assessing this rule based on GL has some limitations. GL is not a perfect or comprehensive way to measure the nutritional value of a food. For example, fruits such as watermelon and grapes are nutritionally dense. However, they contain high amounts of natural sugars—and as such, their GL is relatively high, which could lead a consumer to perceive them as unhealthy. Nevertheless, GL is both a useful and accepted tool and a reasonable way to assess the validity of the rule, specifically when assessing packaged carbohydrates. The simplicity of the Altman Rule and its relationship with GL makes it such that consumers are more likely to make a healthier food choice using it.⁹

Continue to: Specificity and sensitivity

Specificity and sensitivity. There are other limitations to the Altman Rule, given that a small number of high-GL foods meet the rule. For example, some granola bars had high dietary protein, which offset a high sugar content just enough to pass the rule despite a higher GL. As such, concluding that a snack bar is a healthier choice because it meets the Altman Rule when it has high amounts of sugar may not be appropriate. This limitation could be considered a lack of specificity (the rule includes food it ought not to include). Another limitation to consider would be a lack of sensitivity, given that only 68% of low-GL foods passed the Altman Rule. Since GL is associated with carbohydrate content, foods with a low carbohydrate count often have little to no fiber and thus would fall into the category of foods that did not meet the Altman Rule but had low GL. In this case, however, the low amount of fiber may render the Altman Rule a better indicator of a healthier food choice than the GL.

Hidden sugars. Foods with sugar alcohols and artificial sweeteners may be as deleterious as caloric alternatives while not being accounted for when reporting the grams of sugar per serving on the nutrition label.⁷ This may represent an exception to the Altman Rule, as foods that are not healthier choices may pass the rule because the sugar content on the nutrition label is, in a sense, artificially lowered. Future research may investigate the hypothesis that these foods are nutritionally inferior despite meeting the Altman Rule.

The sample. Our study also was limited to working only with foods that were included in the NDSR database up to 2010. This limitation is mitigated by the fact that the sample size was large (> 1000 packaged food items were included in our analyses). The study also could be limited by the food categories that were analyzed; food categories such as bread, rice, pasta, and bagels were not included.

The objective of this research was to investigate the relationship between GL and the Altman Rule, rather than to conduct an exhaustive analysis of the Altman Rule for every possible food category. Studying the relationship between the Altman Rule and GL in other categories of food is an objective for future research. The data so far support a relationship between these entities. The likelihood of the nutrition facts of foods changing without the GL changing (or vice versa) is very low. As such, the Altman Rule still seems to be a reasonable proxy of GL.

CONCLUSIONS

Research indicates that point-of-sale tools, such as Guiding Stars, NuVal, and other stoplight tools, can successfully alter consumers’ behaviors.⁹ These tools can be helpful but are not available in many supermarkets. Despite the limitations, the Altman Rule is a useful decision aid that is accessible to all consumers no matter where they live or shop and is easy to use and remember.

The Altman rule can be used in clinical practice by health care professionals, such as physicians, nurse practitioners, physician assistants, dietitians, and health coaches. It also has the potential to be used in commercial settings, such as grocery stores, to help consumers easily identify healthier convenience foods. This has public health implications, as the rule can both empower consumers and potentially incentivize food manufacturers to upgrade their products nutritionally.

Additional research would be useful to evaluate consumers’ preferences and perceptions about how user-friendly the Altman Rule is at the point of sale with packaged carbohydrate foods. This would help to further understand how the use of information on food packaging can motivate healthier decisions—thereby helping to alleviate the burden of chronic disease.

CORRESPONDENCE
Kimberly R. Dong, DrPH, MS, RDN, Tufts University School of Medicine, Department of Public Health and Community Medicine, 136 Harrison Avenue, MV Building, Boston, MA 02111; kimberly.dong@tufts.edu

ABSTRACT

Background: The Altman Rule, a simple tool for consumers seeking to make healthier packaged food choices at the point of sale, applies to packaged carbohydrates. According to the Altman Rule, a food is a healthier option if it has at least 3 g of fiber per serving and the grams of fiber plus the grams of protein exceed the grams of sugar per serving. This study sought to evaluate whether the Altman Rule is a valid proxy for glycemic load (GL).

Methods: We compared the binary outcome of whether a food item meets the Altman Rule with the GL of all foods categorized as cereals, chips, crackers, and granola bars in the Nutrition Data System for Research Database (University of Minnesota, Version 2010). We examined the percentage of foods in low-, medium-, and high-GL categories that met the Altman Rule.

Results: There were 1235 foods (342 cereals, 305 chips, 379 crackers, and 209 granola bars) in this analysis. There was a significant relationship between the GL of foods and the Altman Rule (P < .001) in that most low-GL (68%), almost half of medium-GL (48%), and very few high-GL (7%) foods met the criteria of the rule.

Conclusions: The Altman Rule is a reasonable proxy for GL and can be a useful and accessible tool for consumers interested in buying healthier packaged carbohydrate foods.

Nutrition can be complicated for consumers interested in making healthier choices at the grocery store. Consumers may have difficulty identifying more nutritious options, especially when food labels are adorned with claims such as “Good Source of Fiber” or “Heart Healthy.”¹ In addition, when reading food labels, consumers may find it difficult to decipher which data to prioritize when carbohydrates, total sugars, added sugars, total dietary fiber, soluble fiber, and insoluble fiber are all listed.

The concept of glycemic load (GL) is an important consideration, especially for people with diabetes. GL approximates the blood sugar response to different foods. A food with a high GL is digested quickly, and its carbohydrates are taken into the bloodstream rapidly. This leads to a spike and subsequent drop in blood sugars, which can cause symptoms of hyperglycemia and hypoglycemia in a person with diabetes.^2,3 Despite its usefulness, GL may be too complicated for a consumer to understand, and it does not appear anywhere on the food label. Since GL is calculated using pooled blood sugar response from individuals after the ingestion of the particular food, estimation of the GL is not intuitable.⁴

Point-of-sale tools. People seeking to lose weight, control diabetes, improve dyslipidemia and/or blood pressure, and/or decrease their risk for heart disease may benefit from point-of-sale tools such as the Altman Rule, which simplifies and encourages the selection of more nutritious foods.¹ Other tools—such as Guiding Stars (https://guidingstars.com), NuVal (www.nuval.com), and different variations of traffic lights—have been created to help consumers make more informed and healthier food choices.^5-8 However, Guiding Stars and NuVal are based on complicated algorithms that are not entirely transparent and not accessible to the average consumer.^6,7 Evaluations of these nutrition tools indicate that consumers tend to underrate the healthiness of some foods, such as raw almonds and salmon, and overrate the healthiness of others, such as fruit punch and diet soda, when using traffic light systems.⁶ Furthermore, these nutrition tools are not available in many supermarkets. Previous research suggests that the use of point-of-sale nutrition apps decreases with the time and effort involved in using an app.⁹

Continue to: The Altman Rule

The Altman Rule was developed by a family physician (author WA) to provide a more accessible tool for people interested in choosing healthier prepackaged carbohydrate foods while shopping. Since the user does not need to have a smartphone, and they are not required to download or understand an app for each purchase, the Altman Rule may be more usable compared with more complicated alternatives.

The Altman Rule can be used with nutrition labels that feature serving information and calories in enlarged and bold type, in compliance with the most recent US Food and Drug Administration (FDA) guideline from 2016. Many foods with high fiber also have high amounts of sugar, so the criteria of the Altman Rule includes a 2-step process requiring (1) a minimum of 3 g of total dietary fiber per serving and (2) the sum of the grams of fiber plus the grams of protein per serving to be greater than the total grams of sugar (not grams of added sugar or grams of carbohydrate) per serving (FIGURE 1A). Unlike the relatively complicated formula related to GL, this 2-part rule can be applied in seconds while shopping (FIGURE 1B).

The rule is intended only to be used for packaged carbohydrate products, such as bread, muffins, bagels, pasta, rice, oatmeal, cereals, snack bars, chips, and crackers. It does not apply to whole foods, such as meat, dairy, fruits, or vegetables. These foods are excluded to prevent any consumer confusion related to the nutritional content of whole foods (eg, an apple may have more sugar than fiber and protein combined, but it is still a nutritious option).

Since the user does not need to have a smartphone, the Altman Rule may be more usable compared with more complicated alternatives.

This study aimed to determine if the Altman Rule is a reasonable proxy for the more complicated concept of GL. We calculated the relationship between the GL of commercially available packaged carbohydrate foods and whether those foods met the Altman Rule.

METHODS

The Altman Rule was tested by comparing the binary outcome of the rule (meets/does not meet) with data on all foods categorized as cereals, chips, crackers, and granola bars in the Nutrition Data System for Research (NDSR) Database (University of Minnesota, Version 2010).

Continue to: To account for differences...

To account for differences in serving size, we used the standard of 50 g for each product as 1 serving. We used 50 g (about 1.7 oz) to help compare the different foods and between foods within the same group. Additionally, 50 g is close to 1 serving for most foods in these groups; it is about the size of a typical granola bar, three-quarters to 2 cups of cereal, 10 to 12 crackers, and 15 to 25 chips. We determined the GL for each product by multiplying the number of available carbohydrates (total carbohydrate – dietary fiber) by the product’s glycemic index/100. In general, GL is categorized as low (≤ 10), medium (11-19), or high (≥ 20).

We applied the Altman Rule to categorize each product as meeting or not meeting the rule. We compared the proportion of foods meeting the Altman Rule, stratified by GL and by specific foods, and used chi-square to determine if differences were statistically significant. These data were collected and analyzed in the summer of 2019.

RESULTS

There were 1235 foods (342 breakfast cereals, 305 chips, 379 crackers, and 209 granola bars) used for this analysis. There is a significant relationship between the GL of foods and the Altman Rule in that most low-GL (68%), almost half of medium-GL (48%), and only a few high-GL foods (7%) met the rule (P < .001) (TABLE 1). There was also a significant relationship between “meeting the ­Altman Rule” and GL within each food type (P < .001) (TABLE 2).

The medium-GL foods were the second largest category of foods we calculated; thus we further broke them into binary categories of low-medium GL (values 11-14) and high-medium GL (values 15-19) to explore the results of the Altman Rule. About half of the foods in medium-GL category met the Altman Rule. About eighty-five percent of the foods with low-medium GL passed the Altman Rule, while only 39% of the foods with high-medium GL did.

Foods that met the rule were more likely to be low GL and foods that did not pass the rule were more likely high GL. Within the medium-GL category, foods that met the rule were more likely to be low-medium GL. 

Continue to: The findings within food categories...

The findings within food categories showed that very few cereals, chips, crackers, and granola bars were low GL. For every food category, except granola bars, far more low-GL foods met the Altman Rule than those that did not. At the same time, very few high-GL foods met the Altman Rule. The category with the most individual high-GL food items meeting the Altman Rule was cereal. This was also the subcategory with the largest percentage of high-GL food items meeting the Altman Rule. Thirty-nine cereals that were high GL met the rule, but more than 4 times as many high-GL cereals did not (n = 190).

DISCUSSION

Marketing and nutrition messaging create consumer confusion that makes it challenging to identify packaged food items that are more nutrient dense. The Altman Rule simplifies food choices that have become unnecessarily complex. Our findings suggest this 2-step rule is a reasonable proxy for the more complicated and less accessible GL for packaged carbohydrates, such as cereals, chips, crackers, and snack bars. Foods that meet the rule are likely low or low-medium GL and thus are foods that are likely to be healthier choices.

Our findings suggest this 2-step rule is a reasonable proxy for the more complicated and less accessible glycemic load for packaged carbohydrates.

Of note, only 9% of chips (n = 27) passed the Altman Rule, likely due to their low dietary fiber content, which was typical of chips. If a food item does not have at least 3 grams of total dietary fiber per serving, it does not pass the Altman Rule, regardless of how much protein or sugar is in the product. This may be considered a strength or a weakness of the Altman Rule. Few nutrition-dense foods are low in fiber, but some foods could be nutritious but do not meet the Altman Rule due to having < 3 g of fiber.

With the high prevalence of chronic diseases such as hypertension, diabetes, hyperlipidemia, and cardiovascular disease, it is essential to help consumers prevent chronic disease altogether or manage their chronic disease by providing tools to identify healthier food choices. The tool also has a place in clinical medicine for use by physicians and other health care professionals. Research shows that physicians find both time and lack of ­knowledge/resources to be a barrier to providing nutritional counseling to patients.¹⁰ Since the Altman Rule can be shared and explained with very little time and without extensive nutritional knowledge, it meets these needs.

Limitations

Glycemic load. We acknowledge that the Altman Rule is not foolproof and that assessing this rule based on GL has some limitations. GL is not a perfect or comprehensive way to measure the nutritional value of a food. For example, fruits such as watermelon and grapes are nutritionally dense. However, they contain high amounts of natural sugars—and as such, their GL is relatively high, which could lead a consumer to perceive them as unhealthy. Nevertheless, GL is both a useful and accepted tool and a reasonable way to assess the validity of the rule, specifically when assessing packaged carbohydrates. The simplicity of the Altman Rule and its relationship with GL makes it such that consumers are more likely to make a healthier food choice using it.⁹

Continue to: Specificity and sensitivity

Specificity and sensitivity. There are other limitations to the Altman Rule, given that a small number of high-GL foods meet the rule. For example, some granola bars had high dietary protein, which offset a high sugar content just enough to pass the rule despite a higher GL. As such, concluding that a snack bar is a healthier choice because it meets the Altman Rule when it has high amounts of sugar may not be appropriate. This limitation could be considered a lack of specificity (the rule includes food it ought not to include). Another limitation to consider would be a lack of sensitivity, given that only 68% of low-GL foods passed the Altman Rule. Since GL is associated with carbohydrate content, foods with a low carbohydrate count often have little to no fiber and thus would fall into the category of foods that did not meet the Altman Rule but had low GL. In this case, however, the low amount of fiber may render the Altman Rule a better indicator of a healthier food choice than the GL.

Hidden sugars. Foods with sugar alcohols and artificial sweeteners may be as deleterious as caloric alternatives while not being accounted for when reporting the grams of sugar per serving on the nutrition label.⁷ This may represent an exception to the Altman Rule, as foods that are not healthier choices may pass the rule because the sugar content on the nutrition label is, in a sense, artificially lowered. Future research may investigate the hypothesis that these foods are nutritionally inferior despite meeting the Altman Rule.

The sample. Our study also was limited to working only with foods that were included in the NDSR database up to 2010. This limitation is mitigated by the fact that the sample size was large (> 1000 packaged food items were included in our analyses). The study also could be limited by the food categories that were analyzed; food categories such as bread, rice, pasta, and bagels were not included.

The objective of this research was to investigate the relationship between GL and the Altman Rule, rather than to conduct an exhaustive analysis of the Altman Rule for every possible food category. Studying the relationship between the Altman Rule and GL in other categories of food is an objective for future research. The data so far support a relationship between these entities. The likelihood of the nutrition facts of foods changing without the GL changing (or vice versa) is very low. As such, the Altman Rule still seems to be a reasonable proxy of GL.

CONCLUSIONS

Research indicates that point-of-sale tools, such as Guiding Stars, NuVal, and other stoplight tools, can successfully alter consumers’ behaviors.⁹ These tools can be helpful but are not available in many supermarkets. Despite the limitations, the Altman Rule is a useful decision aid that is accessible to all consumers no matter where they live or shop and is easy to use and remember.

The Altman rule can be used in clinical practice by health care professionals, such as physicians, nurse practitioners, physician assistants, dietitians, and health coaches. It also has the potential to be used in commercial settings, such as grocery stores, to help consumers easily identify healthier convenience foods. This has public health implications, as the rule can both empower consumers and potentially incentivize food manufacturers to upgrade their products nutritionally.

Additional research would be useful to evaluate consumers’ preferences and perceptions about how user-friendly the Altman Rule is at the point of sale with packaged carbohydrate foods. This would help to further understand how the use of information on food packaging can motivate healthier decisions—thereby helping to alleviate the burden of chronic disease.

CORRESPONDENCE
Kimberly R. Dong, DrPH, MS, RDN, Tufts University School of Medicine, Department of Public Health and Community Medicine, 136 Harrison Avenue, MV Building, Boston, MA 02111; kimberly.dong@tufts.edu