Exposure to carcinogens has been implicated in the development of non-Hodgkin lymphoma (NHL), suggesting that an examination of the environment on a population-based level might provide some insights. On that basis, researchers performed a study that found that living in an urban vs. rural area was associated with an increased risk of developing non-Hodgkin lymphoma (NHL) among diverse, urban populations.

The study, published online in Clinical Lymphoma, Myeloma & Leukemia, found an increased incidence of diffuse large B-cell lymphoma (DLBCL) in urban vs. rural Hispanics, and a similar increased incidence of chronic lymphocytic leukemia (CLL) in non-metropolitan urban non-Hispanic blacks.

A total of 482,096 adults aged 20 years and older with incident NHL were reported to 21 Surveillance, Epidemiology,and End Results (SEER) population-based registries for the period 2000 to 2016. Deanna Blansky of the Albert Einstein College of Medicine, Bronx, N.Y., and her colleagues compared patients by NHL subtype and urban-rural status, using rural-urban continuum codes from the U.S. Department of Agriculture.

The researchers found 136,197 DLBCL, 70,882 follicular lymphoma (FL), and 120,319 CLL cases of patients aged ≥ 20 years. The DLBCL patients comprised 73.6% non-Hispanic white, 11.8% Hispanic, and 7.3% non-Hispanic black, with a similar distribution observed for FL and CLL. Patients were adjusted for age, sex, and family poverty.

The study showed that, overall, there was a higher DLBCL incidence rate in metropolitan urban areas, compared with rural areas overall (incidence rate ratio [IRR] = 1.20, 95% confidence interval [CI] 1.11-1.30). Most pronounced was an increased DLBCL incidence among Hispanics in urban areas, compared with rural areas (rural IRR = 1.00; non-metropolitan urban IRR = 1.32, 95% CI 1.16-1.51; metropolitan urban = 1.55, 95% CI 1.36-1.76).

In contrast, metropolitan urban areas had a lower overall incidence of CLL than rural areas (8.4 vs. 9.7 per 100,000; IRR = .87; 95% CI .86-.89).

However, increased CLL incidence rates were found to be associated with non-metropolitan urban areas, compared with rural areas (IRR = 1.19; 95% CI 1.10-1.28), particularly among non-Hispanic Blacks (IRR = 1.49, 95% CI 1.27-1.72).

Unlike DLBCL and CLL, there were no differences observed in FL incidence rates by urban-rural status after adjusting for age, sex, and family poverty rates, the researchers reported.

“Overall, our findings suggest that factors related to urban status may be associated with DLBCL and CLL pathogenesis. Our results may help provide epidemiological clues to understanding the racial disparities seen among hematological malignancies, particularly regarding the risk of DLBCL in Hispanics and CLL in non-Hispanic Blacks,” the researchers concluded.

The study was sponsored by the U.S. National Institutes of Health. The researchers did not report conflict information.
 

mlesney@mdedge.com

SOURCE: Blansky D et al. Clin Lymphoma Myeloma Leuk. 2020 May 15; doi.org/10.1016/j.clml.2020.05.010.

Exposure to carcinogens has been implicated in the development of non-Hodgkin lymphoma (NHL), suggesting that an examination of the environment on a population-based level might provide some insights. On that basis, researchers performed a study that found that living in an urban vs. rural area was associated with an increased risk of developing non-Hodgkin lymphoma (NHL) among diverse, urban populations.

The study, published online in Clinical Lymphoma, Myeloma & Leukemia, found an increased incidence of diffuse large B-cell lymphoma (DLBCL) in urban vs. rural Hispanics, and a similar increased incidence of chronic lymphocytic leukemia (CLL) in non-metropolitan urban non-Hispanic blacks.

A total of 482,096 adults aged 20 years and older with incident NHL were reported to 21 Surveillance, Epidemiology,and End Results (SEER) population-based registries for the period 2000 to 2016. Deanna Blansky of the Albert Einstein College of Medicine, Bronx, N.Y., and her colleagues compared patients by NHL subtype and urban-rural status, using rural-urban continuum codes from the U.S. Department of Agriculture.

The researchers found 136,197 DLBCL, 70,882 follicular lymphoma (FL), and 120,319 CLL cases of patients aged ≥ 20 years. The DLBCL patients comprised 73.6% non-Hispanic white, 11.8% Hispanic, and 7.3% non-Hispanic black, with a similar distribution observed for FL and CLL. Patients were adjusted for age, sex, and family poverty.

The study showed that, overall, there was a higher DLBCL incidence rate in metropolitan urban areas, compared with rural areas overall (incidence rate ratio [IRR] = 1.20, 95% confidence interval [CI] 1.11-1.30). Most pronounced was an increased DLBCL incidence among Hispanics in urban areas, compared with rural areas (rural IRR = 1.00; non-metropolitan urban IRR = 1.32, 95% CI 1.16-1.51; metropolitan urban = 1.55, 95% CI 1.36-1.76).

In contrast, metropolitan urban areas had a lower overall incidence of CLL than rural areas (8.4 vs. 9.7 per 100,000; IRR = .87; 95% CI .86-.89).

However, increased CLL incidence rates were found to be associated with non-metropolitan urban areas, compared with rural areas (IRR = 1.19; 95% CI 1.10-1.28), particularly among non-Hispanic Blacks (IRR = 1.49, 95% CI 1.27-1.72).

Unlike DLBCL and CLL, there were no differences observed in FL incidence rates by urban-rural status after adjusting for age, sex, and family poverty rates, the researchers reported.

“Overall, our findings suggest that factors related to urban status may be associated with DLBCL and CLL pathogenesis. Our results may help provide epidemiological clues to understanding the racial disparities seen among hematological malignancies, particularly regarding the risk of DLBCL in Hispanics and CLL in non-Hispanic Blacks,” the researchers concluded.

The study was sponsored by the U.S. National Institutes of Health. The researchers did not report conflict information.
 

mlesney@mdedge.com

SOURCE: Blansky D et al. Clin Lymphoma Myeloma Leuk. 2020 May 15; doi.org/10.1016/j.clml.2020.05.010.

Exposure to carcinogens has been implicated in the development of non-Hodgkin lymphoma (NHL), suggesting that an examination of the environment on a population-based level might provide some insights. On that basis, researchers performed a study that found that living in an urban vs. rural area was associated with an increased risk of developing non-Hodgkin lymphoma (NHL) among diverse, urban populations.

The study, published online in Clinical Lymphoma, Myeloma & Leukemia, found an increased incidence of diffuse large B-cell lymphoma (DLBCL) in urban vs. rural Hispanics, and a similar increased incidence of chronic lymphocytic leukemia (CLL) in non-metropolitan urban non-Hispanic blacks.

A total of 482,096 adults aged 20 years and older with incident NHL were reported to 21 Surveillance, Epidemiology,and End Results (SEER) population-based registries for the period 2000 to 2016. Deanna Blansky of the Albert Einstein College of Medicine, Bronx, N.Y., and her colleagues compared patients by NHL subtype and urban-rural status, using rural-urban continuum codes from the U.S. Department of Agriculture.

The researchers found 136,197 DLBCL, 70,882 follicular lymphoma (FL), and 120,319 CLL cases of patients aged ≥ 20 years. The DLBCL patients comprised 73.6% non-Hispanic white, 11.8% Hispanic, and 7.3% non-Hispanic black, with a similar distribution observed for FL and CLL. Patients were adjusted for age, sex, and family poverty.

The study showed that, overall, there was a higher DLBCL incidence rate in metropolitan urban areas, compared with rural areas overall (incidence rate ratio [IRR] = 1.20, 95% confidence interval [CI] 1.11-1.30). Most pronounced was an increased DLBCL incidence among Hispanics in urban areas, compared with rural areas (rural IRR = 1.00; non-metropolitan urban IRR = 1.32, 95% CI 1.16-1.51; metropolitan urban = 1.55, 95% CI 1.36-1.76).

In contrast, metropolitan urban areas had a lower overall incidence of CLL than rural areas (8.4 vs. 9.7 per 100,000; IRR = .87; 95% CI .86-.89).

However, increased CLL incidence rates were found to be associated with non-metropolitan urban areas, compared with rural areas (IRR = 1.19; 95% CI 1.10-1.28), particularly among non-Hispanic Blacks (IRR = 1.49, 95% CI 1.27-1.72).

Unlike DLBCL and CLL, there were no differences observed in FL incidence rates by urban-rural status after adjusting for age, sex, and family poverty rates, the researchers reported.

“Overall, our findings suggest that factors related to urban status may be associated with DLBCL and CLL pathogenesis. Our results may help provide epidemiological clues to understanding the racial disparities seen among hematological malignancies, particularly regarding the risk of DLBCL in Hispanics and CLL in non-Hispanic Blacks,” the researchers concluded.

The study was sponsored by the U.S. National Institutes of Health. The researchers did not report conflict information.
 

mlesney@mdedge.com

SOURCE: Blansky D et al. Clin Lymphoma Myeloma Leuk. 2020 May 15; doi.org/10.1016/j.clml.2020.05.010.

To the Editor:

Syringomas are common benign tumors of the eccrine sweat glands that usually manifest clinically as multiple flesh-colored papules. They are most commonly seen on the face, neck, and chest of adolescent girls. Syringomas may appear at any site of the body but are rare in the vulva. We present a case of a 51-year-old woman who was referred to the Division of Gynecologic Oncology at the University of Alabama at Birmingham for further management of a tumor carrying a differential diagnosis of vulvar syringoma vs microcystic adnexal carcinoma (MAC).

A 51-year-old woman presented to dermatology (G.G.) and was referred to the Division of Gynecologic Oncology at the University of Alabama at Birmingham for further management of possible vulvar syringoma vs MAC. The patient previously had been evaluated at an outside community practice due to dyspareunia, vulvar discomfort, and vulvar irregularities of 1 month’s duration. At that time, a small biopsy was performed, and the histologic differential diagnosis included syringoma vs an adnexal carcinoma. Consequently, she was referred to gynecologic oncology for further management.

Pelvic examination revealed multilobular nodular areas overlying the clitoral hood that extended down to the labia majora. The nodular processes did not involve the clitoris, labia minora, or perineum. A mobile isolated lymph node measuring 2.0×1.0 cm in the right inguinal area also was noted. The patient’s clinical history was notable for right breast carcinoma treated with a right mastectomy with axillary lymph node dissection that showed metastatic disease. She also underwent adjuvant chemotherapy with paclitaxel and doxorubicin for breast carcinoma.

After discussing the diagnostic differential and treatment options, the patient elected to undergo a bilateral partial radical vulvectomy with reconstruction and resection of the right inguinal lymph node. Gross examination of the vulvectomy specimen showed multiple flesh-colored papules (Figure 1). Histologic examination revealed a neoplasm with sweat gland differentiation that was broad and poorly circumscribed but confined to the dermis (Figures 2A and 2B). The neoplasm was composed of epithelial cells that formed ductlike structures, lined by 2 layers of cuboidal epithelium within a fibrous stroma (Figure 2C). A toluidine blue special stain was performed and demonstrated an increased amount of mast cells in the tissue (Figure 3). Immunohistochemical stains for gross cystic disease fluid protein, estrogen receptor (ER), and progesterone receptor (PR) were negative in the tumor cells. The lack of cytologic atypia, perineural invasion, and deep infiltration into the subcutis favored a syringoma. One month later, the case was presented at the Tumor Board Conference at the University of Alabama at Birmingham where a final diagnosis of vulvar syringoma was agreed upon and discussed with the patient. At that time, no recurrence was evident and follow-up was recommended.

Syringomas may occur at any site on the body but are prone to occur on the periorbital area, especially the eyelids.¹ Some of the atypical locations for a syringoma include the anterior neck, chest, abdomen, genitals, axillae, groin, and buttocks.² Vulvar syringomas were first reported by Carneiro³ in 1971 as usually affecting adolescent girls and middle-aged women. There have been approximately 40 reported cases affecting women aged 8 to 78 years.^4,5 Vulvar syringomas classically appear as firm or soft, flesh-colored to transparent, papular lesions. The 2 other clinical variants are miliumlike, whitish, cystic papules as well as lichenoid papules.⁶ Pérez-Bustillo et al⁵ reported a case of the lichenoid papule variant on the labia majora of a 78-year-old woman who presented with intermittent vulvar pruritus of 4 years’ duration. Due to this patient’s 9-year history of urinary incontinence, the lesions had been misdiagnosed as irritant dermatitis and associated lichen simplex chronicus (LSC). This case is a reminder to consider vulvar syringoma in patients with LSC who respond poorly to oral antihistamines and topical steroids.⁵ Rarely, multiple clinical variants may coexist. In a case reported by Dereli et al,⁷ a 19-year-old woman presented with concurrent classical and miliumlike forms of vulvar syringoma.

Vulvar syringomas usually present as multiple lesions involving both sides of the labia majora; however, Blasdale and McLelland⁸ reported a single isolated syringoma of the vulva on the anterior right labia minora that measured 1.0×0.5 cm, leading the lesion to be described as a giant syringoma.

Vulvar syringomas usually are asymptomatic and noticed during routine gynecologic examination. Therefore, it is believed that they likely are underdiagnosed.⁵ When symptomatic, they commonly present with constant⁹ or intermittent⁵ pruritus, which may intensify during menstruation, pregnancy, and summertime.^6,10-12 Gerdsen et al¹⁰ documented a 27-year-old woman who presented with a 2-year history of pruritic vulvar skin lesions that became exacerbated during menstruation, which raised the possibility of cyclical hormonal changes being responsible for periodic exacerbation of vulvar pruritus during menstruation. In addition, patients may experience an increase in size and number of the lesions during pregnancy. Bal et al¹¹ reported a 24-year-old primigravida with vulvar papular lesions that intensified during pregnancy. She had experienced intermittent vulvar pruritus for 12 years but had no change in symptoms during menstruation.¹¹ Few studies have attempted to evaluate the presence of ER and PR in the syringomas. A study of 9 nonvulvar syringomas by Wallace and Smoller¹³ showed ER positivity in 1 case and PR positivity in 8 cases, lending support to the hormonal theory; however, in another case series of 15 vulvar syringomas, Huang et al⁶ failed to show ER and PR expression by immunohistochemical staining. A case report published 3 years earlier documented the first case of PR positivity on a vulvar syringoma.¹⁴ Our patient also was negative for ER and PR, which suggested that hormonal status is important in some but not all syringomas.

Patients with vulgar syringomas also might have coexisting extragenital syringomas in the neck,⁴ eyelids,^6,7,10 and periorbital area,⁶ and thorough examination of the body is essential. If an extragenital syringoma is diagnosed, a vulvar syringoma should be considered, especially when the patient presents with unexplained genital symptoms. Although no proven hereditary transmission pattern has been established, family history of syringomas has been established in several cases.¹⁵ In a case series reported by Huang et al,⁶ 4 of 18 patients reported a family history of periorbital syringomas. In our case, the patient did not report a family history of syringomas.

The differential diagnosis of vulvar lesions with pruritus is broad and includes Fox-Fordyce disease, lichen planus, LSC, epidermal cysts, senile angiomas, dystrophic calcinosis, xanthomas, steatocytomas, soft fibromas, condyloma acuminatum, and candidiasis. Vulvar syringomas might have a nonspecific appearance, and histologic examination is essential to confirm the diagnosis and rule out any malignant process such as MAC, vulvar intraepithelial neoplasia, extramammary Paget disease, or other glandular neoplasms of the vulva.

Microcystic adnexal carcinoma was first reported in 1982 by Goldstein et al¹⁶ as a locally aggressive neoplasm that can be confused with benign adnexal neoplasms, particularly desmoplastic trichoepithelioma, trichoadenoma, and syringoma. Microcystic adnexal carcinomas present as slow-growing, flesh-colored papules that may resemble syringomas and appear in similar body sites. Histologic examination is essential to differentiate between these two entities. Syringomas are tumors confined to the dermis and are composed of multiple small ducts lined by 2 layers of cuboidal epithelium within a dense fibrous stroma. Unlike syringomas, MACs usually infiltrate diffusely into the dermis and subcutis and may extend into the underlying muscle. Although bland cytologic features predominate, perineural invasion frequently is present in MACs. A potential pitfall of misdiagnosis can be caused by a superficial biopsy that may reveal benign histologic appearance, particularly in the upper level of the tumor where it may be confused with a syringoma or a benign follicular neoplasm.¹⁷

The initial biopsy performed on our patient was possibly not deep enough to render an unequivocal diagnosis and therefore bilateral partial radical vulvectomy was considered. After surgery, histologic examination of the resection specimen revealed a poorly circumscribed tumor confined to the dermis. The tumor was broad and the lack of deep infiltration into the subcutis and perineural invasion favored a syringoma (Figures 2A and 2B). These findings were consistent with case reports that documented syringomas as being more wide than deep on microscopic examination, whereas the opposite pertained to MAC.¹⁸ Cases of plaque-type syringomas that initially were misdiagnosed as MACs also have been reported.¹⁹ Because misdiagnosis may affect the treatment plan and potentially result in unnecessary surgery, caution should be taken when differentiating between these two entities. When a definitive diagnosis cannot be rendered on a superficial biopsy, a recommendation should be made for a deeper biopsy sampling the subcutis.

For the majority of the patients with vulvar syringomas, treatment is seldom required due to their asymptomatic nature; however, patients who present with symptoms usually report pruritus of variable intensities and patterns. A standardized treatment does not exist for vulvar syringomas, and oral or topical treatment might be used as an initial approach. Commonly prescribed medications with variable results include topical corticosteroids, oral antihistamines, and topical retinoids. In a case reported by Iwao et al,²⁰ vulvar syringomas were successfully treated with tranilast, which has anti-inflammatory and immunomodulatory effects. This medication could have a possible dual action—inhibiting the release of chemical mediators from the mast cells and inhibiting the release of IL-1β from the eccrine duct, which could suppress the proliferation of stromal connective tissue. Our case was stained with toluidine blue and showed an increased number of mast cells in the tissue (Figure 3). Patients who are unresponsive to tranilast or have extensive disease resulting in cosmetic disfigurement might benefit from more invasive treatment methods including a variety of lasers, cryotherapy, electrosurgery, and excision. Excisions should include the entire tumor to avoid recurrence. In a case reported by Garman and Metry,²¹ the lesions were surgically excised using small 2- to 3-mm punches; however, several weeks later the lesions recurred. Our patient presented with a 1-month evolution of dyspareunia, vulvar discomfort, and vulvar irregularities that were probably not treated with oral or topical medications before being referred for surgery.

We report a case of a vulvar syringoma that presented diagnostic challenges in the initial biopsy, which prevented the exclusion of an MAC. After partial radical vulvectomy, histologic examination was more definitive, showing lack of deep infiltration into the subcutis or perineural invasion that are commonly seen in MAC. This case is an example of a notable pitfall in the diagnosis of vulvar syringoma on a limited biopsy leading to overtreatment. Raising awareness of this entity is the only modality to prevent misdiagnosis. We encourage reporting of further cases of syringomas, particularly those with atypical locations or patterns that may cause diagnostic problems.

To the Editor:

Syringomas are common benign tumors of the eccrine sweat glands that usually manifest clinically as multiple flesh-colored papules. They are most commonly seen on the face, neck, and chest of adolescent girls. Syringomas may appear at any site of the body but are rare in the vulva. We present a case of a 51-year-old woman who was referred to the Division of Gynecologic Oncology at the University of Alabama at Birmingham for further management of a tumor carrying a differential diagnosis of vulvar syringoma vs microcystic adnexal carcinoma (MAC).

A 51-year-old woman presented to dermatology (G.G.) and was referred to the Division of Gynecologic Oncology at the University of Alabama at Birmingham for further management of possible vulvar syringoma vs MAC. The patient previously had been evaluated at an outside community practice due to dyspareunia, vulvar discomfort, and vulvar irregularities of 1 month’s duration. At that time, a small biopsy was performed, and the histologic differential diagnosis included syringoma vs an adnexal carcinoma. Consequently, she was referred to gynecologic oncology for further management.

Pelvic examination revealed multilobular nodular areas overlying the clitoral hood that extended down to the labia majora. The nodular processes did not involve the clitoris, labia minora, or perineum. A mobile isolated lymph node measuring 2.0×1.0 cm in the right inguinal area also was noted. The patient’s clinical history was notable for right breast carcinoma treated with a right mastectomy with axillary lymph node dissection that showed metastatic disease. She also underwent adjuvant chemotherapy with paclitaxel and doxorubicin for breast carcinoma.

After discussing the diagnostic differential and treatment options, the patient elected to undergo a bilateral partial radical vulvectomy with reconstruction and resection of the right inguinal lymph node. Gross examination of the vulvectomy specimen showed multiple flesh-colored papules (Figure 1). Histologic examination revealed a neoplasm with sweat gland differentiation that was broad and poorly circumscribed but confined to the dermis (Figures 2A and 2B). The neoplasm was composed of epithelial cells that formed ductlike structures, lined by 2 layers of cuboidal epithelium within a fibrous stroma (Figure 2C). A toluidine blue special stain was performed and demonstrated an increased amount of mast cells in the tissue (Figure 3). Immunohistochemical stains for gross cystic disease fluid protein, estrogen receptor (ER), and progesterone receptor (PR) were negative in the tumor cells. The lack of cytologic atypia, perineural invasion, and deep infiltration into the subcutis favored a syringoma. One month later, the case was presented at the Tumor Board Conference at the University of Alabama at Birmingham where a final diagnosis of vulvar syringoma was agreed upon and discussed with the patient. At that time, no recurrence was evident and follow-up was recommended.

Syringomas may occur at any site on the body but are prone to occur on the periorbital area, especially the eyelids.¹ Some of the atypical locations for a syringoma include the anterior neck, chest, abdomen, genitals, axillae, groin, and buttocks.² Vulvar syringomas were first reported by Carneiro³ in 1971 as usually affecting adolescent girls and middle-aged women. There have been approximately 40 reported cases affecting women aged 8 to 78 years.^4,5 Vulvar syringomas classically appear as firm or soft, flesh-colored to transparent, papular lesions. The 2 other clinical variants are miliumlike, whitish, cystic papules as well as lichenoid papules.⁶ Pérez-Bustillo et al⁵ reported a case of the lichenoid papule variant on the labia majora of a 78-year-old woman who presented with intermittent vulvar pruritus of 4 years’ duration. Due to this patient’s 9-year history of urinary incontinence, the lesions had been misdiagnosed as irritant dermatitis and associated lichen simplex chronicus (LSC). This case is a reminder to consider vulvar syringoma in patients with LSC who respond poorly to oral antihistamines and topical steroids.⁵ Rarely, multiple clinical variants may coexist. In a case reported by Dereli et al,⁷ a 19-year-old woman presented with concurrent classical and miliumlike forms of vulvar syringoma.

Vulvar syringomas usually present as multiple lesions involving both sides of the labia majora; however, Blasdale and McLelland⁸ reported a single isolated syringoma of the vulva on the anterior right labia minora that measured 1.0×0.5 cm, leading the lesion to be described as a giant syringoma.

Vulvar syringomas usually are asymptomatic and noticed during routine gynecologic examination. Therefore, it is believed that they likely are underdiagnosed.⁵ When symptomatic, they commonly present with constant⁹ or intermittent⁵ pruritus, which may intensify during menstruation, pregnancy, and summertime.^6,10-12 Gerdsen et al¹⁰ documented a 27-year-old woman who presented with a 2-year history of pruritic vulvar skin lesions that became exacerbated during menstruation, which raised the possibility of cyclical hormonal changes being responsible for periodic exacerbation of vulvar pruritus during menstruation. In addition, patients may experience an increase in size and number of the lesions during pregnancy. Bal et al¹¹ reported a 24-year-old primigravida with vulvar papular lesions that intensified during pregnancy. She had experienced intermittent vulvar pruritus for 12 years but had no change in symptoms during menstruation.¹¹ Few studies have attempted to evaluate the presence of ER and PR in the syringomas. A study of 9 nonvulvar syringomas by Wallace and Smoller¹³ showed ER positivity in 1 case and PR positivity in 8 cases, lending support to the hormonal theory; however, in another case series of 15 vulvar syringomas, Huang et al⁶ failed to show ER and PR expression by immunohistochemical staining. A case report published 3 years earlier documented the first case of PR positivity on a vulvar syringoma.¹⁴ Our patient also was negative for ER and PR, which suggested that hormonal status is important in some but not all syringomas.

Patients with vulgar syringomas also might have coexisting extragenital syringomas in the neck,⁴ eyelids,^6,7,10 and periorbital area,⁶ and thorough examination of the body is essential. If an extragenital syringoma is diagnosed, a vulvar syringoma should be considered, especially when the patient presents with unexplained genital symptoms. Although no proven hereditary transmission pattern has been established, family history of syringomas has been established in several cases.¹⁵ In a case series reported by Huang et al,⁶ 4 of 18 patients reported a family history of periorbital syringomas. In our case, the patient did not report a family history of syringomas.

The differential diagnosis of vulvar lesions with pruritus is broad and includes Fox-Fordyce disease, lichen planus, LSC, epidermal cysts, senile angiomas, dystrophic calcinosis, xanthomas, steatocytomas, soft fibromas, condyloma acuminatum, and candidiasis. Vulvar syringomas might have a nonspecific appearance, and histologic examination is essential to confirm the diagnosis and rule out any malignant process such as MAC, vulvar intraepithelial neoplasia, extramammary Paget disease, or other glandular neoplasms of the vulva.

Microcystic adnexal carcinoma was first reported in 1982 by Goldstein et al¹⁶ as a locally aggressive neoplasm that can be confused with benign adnexal neoplasms, particularly desmoplastic trichoepithelioma, trichoadenoma, and syringoma. Microcystic adnexal carcinomas present as slow-growing, flesh-colored papules that may resemble syringomas and appear in similar body sites. Histologic examination is essential to differentiate between these two entities. Syringomas are tumors confined to the dermis and are composed of multiple small ducts lined by 2 layers of cuboidal epithelium within a dense fibrous stroma. Unlike syringomas, MACs usually infiltrate diffusely into the dermis and subcutis and may extend into the underlying muscle. Although bland cytologic features predominate, perineural invasion frequently is present in MACs. A potential pitfall of misdiagnosis can be caused by a superficial biopsy that may reveal benign histologic appearance, particularly in the upper level of the tumor where it may be confused with a syringoma or a benign follicular neoplasm.¹⁷

The initial biopsy performed on our patient was possibly not deep enough to render an unequivocal diagnosis and therefore bilateral partial radical vulvectomy was considered. After surgery, histologic examination of the resection specimen revealed a poorly circumscribed tumor confined to the dermis. The tumor was broad and the lack of deep infiltration into the subcutis and perineural invasion favored a syringoma (Figures 2A and 2B). These findings were consistent with case reports that documented syringomas as being more wide than deep on microscopic examination, whereas the opposite pertained to MAC.¹⁸ Cases of plaque-type syringomas that initially were misdiagnosed as MACs also have been reported.¹⁹ Because misdiagnosis may affect the treatment plan and potentially result in unnecessary surgery, caution should be taken when differentiating between these two entities. When a definitive diagnosis cannot be rendered on a superficial biopsy, a recommendation should be made for a deeper biopsy sampling the subcutis.

For the majority of the patients with vulvar syringomas, treatment is seldom required due to their asymptomatic nature; however, patients who present with symptoms usually report pruritus of variable intensities and patterns. A standardized treatment does not exist for vulvar syringomas, and oral or topical treatment might be used as an initial approach. Commonly prescribed medications with variable results include topical corticosteroids, oral antihistamines, and topical retinoids. In a case reported by Iwao et al,²⁰ vulvar syringomas were successfully treated with tranilast, which has anti-inflammatory and immunomodulatory effects. This medication could have a possible dual action—inhibiting the release of chemical mediators from the mast cells and inhibiting the release of IL-1β from the eccrine duct, which could suppress the proliferation of stromal connective tissue. Our case was stained with toluidine blue and showed an increased number of mast cells in the tissue (Figure 3). Patients who are unresponsive to tranilast or have extensive disease resulting in cosmetic disfigurement might benefit from more invasive treatment methods including a variety of lasers, cryotherapy, electrosurgery, and excision. Excisions should include the entire tumor to avoid recurrence. In a case reported by Garman and Metry,²¹ the lesions were surgically excised using small 2- to 3-mm punches; however, several weeks later the lesions recurred. Our patient presented with a 1-month evolution of dyspareunia, vulvar discomfort, and vulvar irregularities that were probably not treated with oral or topical medications before being referred for surgery.

We report a case of a vulvar syringoma that presented diagnostic challenges in the initial biopsy, which prevented the exclusion of an MAC. After partial radical vulvectomy, histologic examination was more definitive, showing lack of deep infiltration into the subcutis or perineural invasion that are commonly seen in MAC. This case is an example of a notable pitfall in the diagnosis of vulvar syringoma on a limited biopsy leading to overtreatment. Raising awareness of this entity is the only modality to prevent misdiagnosis. We encourage reporting of further cases of syringomas, particularly those with atypical locations or patterns that may cause diagnostic problems.

To the Editor:

Syringomas are common benign tumors of the eccrine sweat glands that usually manifest clinically as multiple flesh-colored papules. They are most commonly seen on the face, neck, and chest of adolescent girls. Syringomas may appear at any site of the body but are rare in the vulva. We present a case of a 51-year-old woman who was referred to the Division of Gynecologic Oncology at the University of Alabama at Birmingham for further management of a tumor carrying a differential diagnosis of vulvar syringoma vs microcystic adnexal carcinoma (MAC).

A 51-year-old woman presented to dermatology (G.G.) and was referred to the Division of Gynecologic Oncology at the University of Alabama at Birmingham for further management of possible vulvar syringoma vs MAC. The patient previously had been evaluated at an outside community practice due to dyspareunia, vulvar discomfort, and vulvar irregularities of 1 month’s duration. At that time, a small biopsy was performed, and the histologic differential diagnosis included syringoma vs an adnexal carcinoma. Consequently, she was referred to gynecologic oncology for further management.

Pelvic examination revealed multilobular nodular areas overlying the clitoral hood that extended down to the labia majora. The nodular processes did not involve the clitoris, labia minora, or perineum. A mobile isolated lymph node measuring 2.0×1.0 cm in the right inguinal area also was noted. The patient’s clinical history was notable for right breast carcinoma treated with a right mastectomy with axillary lymph node dissection that showed metastatic disease. She also underwent adjuvant chemotherapy with paclitaxel and doxorubicin for breast carcinoma.

After discussing the diagnostic differential and treatment options, the patient elected to undergo a bilateral partial radical vulvectomy with reconstruction and resection of the right inguinal lymph node. Gross examination of the vulvectomy specimen showed multiple flesh-colored papules (Figure 1). Histologic examination revealed a neoplasm with sweat gland differentiation that was broad and poorly circumscribed but confined to the dermis (Figures 2A and 2B). The neoplasm was composed of epithelial cells that formed ductlike structures, lined by 2 layers of cuboidal epithelium within a fibrous stroma (Figure 2C). A toluidine blue special stain was performed and demonstrated an increased amount of mast cells in the tissue (Figure 3). Immunohistochemical stains for gross cystic disease fluid protein, estrogen receptor (ER), and progesterone receptor (PR) were negative in the tumor cells. The lack of cytologic atypia, perineural invasion, and deep infiltration into the subcutis favored a syringoma. One month later, the case was presented at the Tumor Board Conference at the University of Alabama at Birmingham where a final diagnosis of vulvar syringoma was agreed upon and discussed with the patient. At that time, no recurrence was evident and follow-up was recommended.

Syringomas may occur at any site on the body but are prone to occur on the periorbital area, especially the eyelids.¹ Some of the atypical locations for a syringoma include the anterior neck, chest, abdomen, genitals, axillae, groin, and buttocks.² Vulvar syringomas were first reported by Carneiro³ in 1971 as usually affecting adolescent girls and middle-aged women. There have been approximately 40 reported cases affecting women aged 8 to 78 years.^4,5 Vulvar syringomas classically appear as firm or soft, flesh-colored to transparent, papular lesions. The 2 other clinical variants are miliumlike, whitish, cystic papules as well as lichenoid papules.⁶ Pérez-Bustillo et al⁵ reported a case of the lichenoid papule variant on the labia majora of a 78-year-old woman who presented with intermittent vulvar pruritus of 4 years’ duration. Due to this patient’s 9-year history of urinary incontinence, the lesions had been misdiagnosed as irritant dermatitis and associated lichen simplex chronicus (LSC). This case is a reminder to consider vulvar syringoma in patients with LSC who respond poorly to oral antihistamines and topical steroids.⁵ Rarely, multiple clinical variants may coexist. In a case reported by Dereli et al,⁷ a 19-year-old woman presented with concurrent classical and miliumlike forms of vulvar syringoma.

Vulvar syringomas usually present as multiple lesions involving both sides of the labia majora; however, Blasdale and McLelland⁸ reported a single isolated syringoma of the vulva on the anterior right labia minora that measured 1.0×0.5 cm, leading the lesion to be described as a giant syringoma.

Vulvar syringomas usually are asymptomatic and noticed during routine gynecologic examination. Therefore, it is believed that they likely are underdiagnosed.⁵ When symptomatic, they commonly present with constant⁹ or intermittent⁵ pruritus, which may intensify during menstruation, pregnancy, and summertime.^6,10-12 Gerdsen et al¹⁰ documented a 27-year-old woman who presented with a 2-year history of pruritic vulvar skin lesions that became exacerbated during menstruation, which raised the possibility of cyclical hormonal changes being responsible for periodic exacerbation of vulvar pruritus during menstruation. In addition, patients may experience an increase in size and number of the lesions during pregnancy. Bal et al¹¹ reported a 24-year-old primigravida with vulvar papular lesions that intensified during pregnancy. She had experienced intermittent vulvar pruritus for 12 years but had no change in symptoms during menstruation.¹¹ Few studies have attempted to evaluate the presence of ER and PR in the syringomas. A study of 9 nonvulvar syringomas by Wallace and Smoller¹³ showed ER positivity in 1 case and PR positivity in 8 cases, lending support to the hormonal theory; however, in another case series of 15 vulvar syringomas, Huang et al⁶ failed to show ER and PR expression by immunohistochemical staining. A case report published 3 years earlier documented the first case of PR positivity on a vulvar syringoma.¹⁴ Our patient also was negative for ER and PR, which suggested that hormonal status is important in some but not all syringomas.

Patients with vulgar syringomas also might have coexisting extragenital syringomas in the neck,⁴ eyelids,^6,7,10 and periorbital area,⁶ and thorough examination of the body is essential. If an extragenital syringoma is diagnosed, a vulvar syringoma should be considered, especially when the patient presents with unexplained genital symptoms. Although no proven hereditary transmission pattern has been established, family history of syringomas has been established in several cases.¹⁵ In a case series reported by Huang et al,⁶ 4 of 18 patients reported a family history of periorbital syringomas. In our case, the patient did not report a family history of syringomas.

The differential diagnosis of vulvar lesions with pruritus is broad and includes Fox-Fordyce disease, lichen planus, LSC, epidermal cysts, senile angiomas, dystrophic calcinosis, xanthomas, steatocytomas, soft fibromas, condyloma acuminatum, and candidiasis. Vulvar syringomas might have a nonspecific appearance, and histologic examination is essential to confirm the diagnosis and rule out any malignant process such as MAC, vulvar intraepithelial neoplasia, extramammary Paget disease, or other glandular neoplasms of the vulva.

Microcystic adnexal carcinoma was first reported in 1982 by Goldstein et al¹⁶ as a locally aggressive neoplasm that can be confused with benign adnexal neoplasms, particularly desmoplastic trichoepithelioma, trichoadenoma, and syringoma. Microcystic adnexal carcinomas present as slow-growing, flesh-colored papules that may resemble syringomas and appear in similar body sites. Histologic examination is essential to differentiate between these two entities. Syringomas are tumors confined to the dermis and are composed of multiple small ducts lined by 2 layers of cuboidal epithelium within a dense fibrous stroma. Unlike syringomas, MACs usually infiltrate diffusely into the dermis and subcutis and may extend into the underlying muscle. Although bland cytologic features predominate, perineural invasion frequently is present in MACs. A potential pitfall of misdiagnosis can be caused by a superficial biopsy that may reveal benign histologic appearance, particularly in the upper level of the tumor where it may be confused with a syringoma or a benign follicular neoplasm.¹⁷

The initial biopsy performed on our patient was possibly not deep enough to render an unequivocal diagnosis and therefore bilateral partial radical vulvectomy was considered. After surgery, histologic examination of the resection specimen revealed a poorly circumscribed tumor confined to the dermis. The tumor was broad and the lack of deep infiltration into the subcutis and perineural invasion favored a syringoma (Figures 2A and 2B). These findings were consistent with case reports that documented syringomas as being more wide than deep on microscopic examination, whereas the opposite pertained to MAC.¹⁸ Cases of plaque-type syringomas that initially were misdiagnosed as MACs also have been reported.¹⁹ Because misdiagnosis may affect the treatment plan and potentially result in unnecessary surgery, caution should be taken when differentiating between these two entities. When a definitive diagnosis cannot be rendered on a superficial biopsy, a recommendation should be made for a deeper biopsy sampling the subcutis.

For the majority of the patients with vulvar syringomas, treatment is seldom required due to their asymptomatic nature; however, patients who present with symptoms usually report pruritus of variable intensities and patterns. A standardized treatment does not exist for vulvar syringomas, and oral or topical treatment might be used as an initial approach. Commonly prescribed medications with variable results include topical corticosteroids, oral antihistamines, and topical retinoids. In a case reported by Iwao et al,²⁰ vulvar syringomas were successfully treated with tranilast, which has anti-inflammatory and immunomodulatory effects. This medication could have a possible dual action—inhibiting the release of chemical mediators from the mast cells and inhibiting the release of IL-1β from the eccrine duct, which could suppress the proliferation of stromal connective tissue. Our case was stained with toluidine blue and showed an increased number of mast cells in the tissue (Figure 3). Patients who are unresponsive to tranilast or have extensive disease resulting in cosmetic disfigurement might benefit from more invasive treatment methods including a variety of lasers, cryotherapy, electrosurgery, and excision. Excisions should include the entire tumor to avoid recurrence. In a case reported by Garman and Metry,²¹ the lesions were surgically excised using small 2- to 3-mm punches; however, several weeks later the lesions recurred. Our patient presented with a 1-month evolution of dyspareunia, vulvar discomfort, and vulvar irregularities that were probably not treated with oral or topical medications before being referred for surgery.

We report a case of a vulvar syringoma that presented diagnostic challenges in the initial biopsy, which prevented the exclusion of an MAC. After partial radical vulvectomy, histologic examination was more definitive, showing lack of deep infiltration into the subcutis or perineural invasion that are commonly seen in MAC. This case is an example of a notable pitfall in the diagnosis of vulvar syringoma on a limited biopsy leading to overtreatment. Raising awareness of this entity is the only modality to prevent misdiagnosis. We encourage reporting of further cases of syringomas, particularly those with atypical locations or patterns that may cause diagnostic problems.

To the Editor:

Sweet syndrome (SS), also known as acute febrile neutrophilic dermatosis, is an uncommon inflammatory skin disorder characterized by sudden onset of fever, leukocytosis, neutrophilia, and tender erythematous papules or plaques or both. Skin biopsy usually reveals extensive infiltration of neutrophils into the epidermis and dermis.^1-3 Although rare, cases of eosinophil-rich SS have been reported in patients with drug-induced and malignancy-associated SS.^4,5 We report a case of a patient with classical SS with dermal eosinophilic infiltration.

An 80-year-old Hispanic man presented with abrupt onset of a rash on the posterior scalp, left ear, back, and hands of 5 days’ duration. The lesions were painful and had progressed to the point of impairing hand grip. The patient’s medical history included a reported common cold the week prior, hyperlipidemia, and hypertension, for which he took metoprolol, simvastatin, aspirin, and clopidogrel. He denied oral lesions and medication changes. He was afebrile and did not experience dietary changes, weight loss, or fatigue. He recently returned from travel to the Dominican Republic.

Physical examination revealed tender, well demarcated, pink to violaceous, pseudovesicular papules and plaques on the palms and dorsal hands (Figure 1), the posterior scalp, left ear, proximal left arm, and back. Pink, juicy, targetoid papules were also found on the scalp, back, and left arm. There was no evidence of lymphadenopathy. Laboratory test results revealed an elevated white blood cell count (11,500/µL [reference range, 3800-10,800/µL]), absolute neutrophil count (8073/µL [reference range, 1500–7800/µL]), and eosinophil count (610/µL [reference range, 15–500/µL]). These results indicated leukocytosis with neutrophilia and mild eosinophilia. The patient also was anemic (hemoglobin, 11.5 g/dL [reference range, 13.2–17.1 g/dL]; hematocrit, 35.1% [reference range, 38.5%–50%]). Urine testing revealed altered renal function (serum creatinine, 2.42 mg/dL [reference range, 0.7–1.1 mg/dL]; blood urea nitrogen, 34 mg/dL [reference range, 7–25 mg/dL]; glomerular filtration rate, 4 mL/min/1.73 m² (reference range, ≥60 mL/min/1.73 m²]), suggesting stage 4 chronic kidney disease. Urinalysis showed mild hematuria and proteinuria.

In our patient, the observed clinical and histological findings were consistent with a diagnosis of SS,^2,10 specifically tender erythematous plaques of sudden onset, fast response to systemic corticosteroid therapy, a dermal neutrophilic infiltrate without evidence of leukocytoclastic vasculitis, and leukocytosis greater than 8000/µL with more than 70% neutrophils. He also exhibited targetoid lesions, which have been reported in 7% to 12% of SS patients.^10,11

The predominant cells involved in the dermis of SS lesions are mature neutrophils; however, eosinophils have been observed in small numbers within dermal infiltrates in skin lesions of patients with either classical SS or drug-induced dermatosis.² In 2 studies of cases of SS (N=73 and N=31), eosinophils were reported in 35% and 41% of skin biopsies, respectively.^4,5 Nevertheless, cases with dense eosinophilic infiltrates are rare. Furthermore, Masuda et al¹² reported a case of eosinophil-rich SS in a 29-year-old woman after treatment of an upper respiratory tract infection with an antibiotic, and Soon et al¹³ described an eosinophil-rich case of SS in the setting of new-onset enteropathy-associated T-cell lymphoma.

Our patient was considered to have classical SS because he had an episode of an upper respiratory tract infection 1 week prior to onset of clinical manifestations. The histologic finding of numerous eosinophils in our case was unusual for idiopathic SS. This finding might suggest a drug hypersensitivity reaction, but the lack of any change in the patient’s long-term medication list and the lack of any other episodes made a diagnosis of drug-induced SS less likely in our patient.

Eosinophilic dermatosis of hematologic malignancy is a rare cutaneous condition in which nodules, pruritic papules, and vesicles arise in patients with a hematologic malignancy, such as chronic lymphocytic leukemia and mantle cell lymphoma,¹³ in which a deep perivascular lymphocytic infiltrate and numerous eosinophils are observed. Malignancy was ruled out in our patient because of the lack of characteristic abnormalities in blood testing, the fast response to corticosteroid therapy, and the lack of recurrence posttreatment or additional systemic concerns.

The typical pathology findings of SS consist of mature neutrophils found in the dermis without evidence of leukocytoclastic vasculitis. Eosinophil-rich infiltration, however rare, has been reported in SS. This report highlights a case of classical SS with a particularly dense eosinophilic infiltrate, which could be mistaken for other eosinophilic dermatoses. Dermatologists should be aware of the possibility of marked eosinophilic infiltration in all subtypes of this disorder.

To the Editor:

Sweet syndrome (SS), also known as acute febrile neutrophilic dermatosis, is an uncommon inflammatory skin disorder characterized by sudden onset of fever, leukocytosis, neutrophilia, and tender erythematous papules or plaques or both. Skin biopsy usually reveals extensive infiltration of neutrophils into the epidermis and dermis.^1-3 Although rare, cases of eosinophil-rich SS have been reported in patients with drug-induced and malignancy-associated SS.^4,5 We report a case of a patient with classical SS with dermal eosinophilic infiltration.

An 80-year-old Hispanic man presented with abrupt onset of a rash on the posterior scalp, left ear, back, and hands of 5 days’ duration. The lesions were painful and had progressed to the point of impairing hand grip. The patient’s medical history included a reported common cold the week prior, hyperlipidemia, and hypertension, for which he took metoprolol, simvastatin, aspirin, and clopidogrel. He denied oral lesions and medication changes. He was afebrile and did not experience dietary changes, weight loss, or fatigue. He recently returned from travel to the Dominican Republic.

Physical examination revealed tender, well demarcated, pink to violaceous, pseudovesicular papules and plaques on the palms and dorsal hands (Figure 1), the posterior scalp, left ear, proximal left arm, and back. Pink, juicy, targetoid papules were also found on the scalp, back, and left arm. There was no evidence of lymphadenopathy. Laboratory test results revealed an elevated white blood cell count (11,500/µL [reference range, 3800-10,800/µL]), absolute neutrophil count (8073/µL [reference range, 1500–7800/µL]), and eosinophil count (610/µL [reference range, 15–500/µL]). These results indicated leukocytosis with neutrophilia and mild eosinophilia. The patient also was anemic (hemoglobin, 11.5 g/dL [reference range, 13.2–17.1 g/dL]; hematocrit, 35.1% [reference range, 38.5%–50%]). Urine testing revealed altered renal function (serum creatinine, 2.42 mg/dL [reference range, 0.7–1.1 mg/dL]; blood urea nitrogen, 34 mg/dL [reference range, 7–25 mg/dL]; glomerular filtration rate, 4 mL/min/1.73 m² (reference range, ≥60 mL/min/1.73 m²]), suggesting stage 4 chronic kidney disease. Urinalysis showed mild hematuria and proteinuria.

In our patient, the observed clinical and histological findings were consistent with a diagnosis of SS,^2,10 specifically tender erythematous plaques of sudden onset, fast response to systemic corticosteroid therapy, a dermal neutrophilic infiltrate without evidence of leukocytoclastic vasculitis, and leukocytosis greater than 8000/µL with more than 70% neutrophils. He also exhibited targetoid lesions, which have been reported in 7% to 12% of SS patients.^10,11

The predominant cells involved in the dermis of SS lesions are mature neutrophils; however, eosinophils have been observed in small numbers within dermal infiltrates in skin lesions of patients with either classical SS or drug-induced dermatosis.² In 2 studies of cases of SS (N=73 and N=31), eosinophils were reported in 35% and 41% of skin biopsies, respectively.^4,5 Nevertheless, cases with dense eosinophilic infiltrates are rare. Furthermore, Masuda et al¹² reported a case of eosinophil-rich SS in a 29-year-old woman after treatment of an upper respiratory tract infection with an antibiotic, and Soon et al¹³ described an eosinophil-rich case of SS in the setting of new-onset enteropathy-associated T-cell lymphoma.

Our patient was considered to have classical SS because he had an episode of an upper respiratory tract infection 1 week prior to onset of clinical manifestations. The histologic finding of numerous eosinophils in our case was unusual for idiopathic SS. This finding might suggest a drug hypersensitivity reaction, but the lack of any change in the patient’s long-term medication list and the lack of any other episodes made a diagnosis of drug-induced SS less likely in our patient.

Eosinophilic dermatosis of hematologic malignancy is a rare cutaneous condition in which nodules, pruritic papules, and vesicles arise in patients with a hematologic malignancy, such as chronic lymphocytic leukemia and mantle cell lymphoma,¹³ in which a deep perivascular lymphocytic infiltrate and numerous eosinophils are observed. Malignancy was ruled out in our patient because of the lack of characteristic abnormalities in blood testing, the fast response to corticosteroid therapy, and the lack of recurrence posttreatment or additional systemic concerns.

The typical pathology findings of SS consist of mature neutrophils found in the dermis without evidence of leukocytoclastic vasculitis. Eosinophil-rich infiltration, however rare, has been reported in SS. This report highlights a case of classical SS with a particularly dense eosinophilic infiltrate, which could be mistaken for other eosinophilic dermatoses. Dermatologists should be aware of the possibility of marked eosinophilic infiltration in all subtypes of this disorder.

To the Editor:

Sweet syndrome (SS), also known as acute febrile neutrophilic dermatosis, is an uncommon inflammatory skin disorder characterized by sudden onset of fever, leukocytosis, neutrophilia, and tender erythematous papules or plaques or both. Skin biopsy usually reveals extensive infiltration of neutrophils into the epidermis and dermis.^1-3 Although rare, cases of eosinophil-rich SS have been reported in patients with drug-induced and malignancy-associated SS.^4,5 We report a case of a patient with classical SS with dermal eosinophilic infiltration.

An 80-year-old Hispanic man presented with abrupt onset of a rash on the posterior scalp, left ear, back, and hands of 5 days’ duration. The lesions were painful and had progressed to the point of impairing hand grip. The patient’s medical history included a reported common cold the week prior, hyperlipidemia, and hypertension, for which he took metoprolol, simvastatin, aspirin, and clopidogrel. He denied oral lesions and medication changes. He was afebrile and did not experience dietary changes, weight loss, or fatigue. He recently returned from travel to the Dominican Republic.

Physical examination revealed tender, well demarcated, pink to violaceous, pseudovesicular papules and plaques on the palms and dorsal hands (Figure 1), the posterior scalp, left ear, proximal left arm, and back. Pink, juicy, targetoid papules were also found on the scalp, back, and left arm. There was no evidence of lymphadenopathy. Laboratory test results revealed an elevated white blood cell count (11,500/µL [reference range, 3800-10,800/µL]), absolute neutrophil count (8073/µL [reference range, 1500–7800/µL]), and eosinophil count (610/µL [reference range, 15–500/µL]). These results indicated leukocytosis with neutrophilia and mild eosinophilia. The patient also was anemic (hemoglobin, 11.5 g/dL [reference range, 13.2–17.1 g/dL]; hematocrit, 35.1% [reference range, 38.5%–50%]). Urine testing revealed altered renal function (serum creatinine, 2.42 mg/dL [reference range, 0.7–1.1 mg/dL]; blood urea nitrogen, 34 mg/dL [reference range, 7–25 mg/dL]; glomerular filtration rate, 4 mL/min/1.73 m² (reference range, ≥60 mL/min/1.73 m²]), suggesting stage 4 chronic kidney disease. Urinalysis showed mild hematuria and proteinuria.

In our patient, the observed clinical and histological findings were consistent with a diagnosis of SS,^2,10 specifically tender erythematous plaques of sudden onset, fast response to systemic corticosteroid therapy, a dermal neutrophilic infiltrate without evidence of leukocytoclastic vasculitis, and leukocytosis greater than 8000/µL with more than 70% neutrophils. He also exhibited targetoid lesions, which have been reported in 7% to 12% of SS patients.^10,11

The predominant cells involved in the dermis of SS lesions are mature neutrophils; however, eosinophils have been observed in small numbers within dermal infiltrates in skin lesions of patients with either classical SS or drug-induced dermatosis.² In 2 studies of cases of SS (N=73 and N=31), eosinophils were reported in 35% and 41% of skin biopsies, respectively.^4,5 Nevertheless, cases with dense eosinophilic infiltrates are rare. Furthermore, Masuda et al¹² reported a case of eosinophil-rich SS in a 29-year-old woman after treatment of an upper respiratory tract infection with an antibiotic, and Soon et al¹³ described an eosinophil-rich case of SS in the setting of new-onset enteropathy-associated T-cell lymphoma.

Our patient was considered to have classical SS because he had an episode of an upper respiratory tract infection 1 week prior to onset of clinical manifestations. The histologic finding of numerous eosinophils in our case was unusual for idiopathic SS. This finding might suggest a drug hypersensitivity reaction, but the lack of any change in the patient’s long-term medication list and the lack of any other episodes made a diagnosis of drug-induced SS less likely in our patient.

Eosinophilic dermatosis of hematologic malignancy is a rare cutaneous condition in which nodules, pruritic papules, and vesicles arise in patients with a hematologic malignancy, such as chronic lymphocytic leukemia and mantle cell lymphoma,¹³ in which a deep perivascular lymphocytic infiltrate and numerous eosinophils are observed. Malignancy was ruled out in our patient because of the lack of characteristic abnormalities in blood testing, the fast response to corticosteroid therapy, and the lack of recurrence posttreatment or additional systemic concerns.

The typical pathology findings of SS consist of mature neutrophils found in the dermis without evidence of leukocytoclastic vasculitis. Eosinophil-rich infiltration, however rare, has been reported in SS. This report highlights a case of classical SS with a particularly dense eosinophilic infiltrate, which could be mistaken for other eosinophilic dermatoses. Dermatologists should be aware of the possibility of marked eosinophilic infiltration in all subtypes of this disorder.

A new CT scan pulmonary angiography model may help optimize the diagnostic work-up process for patients with suspected pulmonary hypertension (PH), according to a recent study.

The diagnostic and prognostic utility of the model was validated in a tertiary referral population of treatment-naive patients who had a high pretest probability of PH.

“The aim of this study was to (a) build a diagnostic CT model in patients with suspected PH using the current guideline definition of PH (mPAP [mean pulmonary arterial pressure] ≥25 mm Hg) and the recent proposed definition of >20 mm Hg and (b) test its prognostic significance,” wrote Andrew J. Swift, MBChB, PhD, of the University of Sheffield (England) and colleagues in European Radiology.

The study cohort included 491 patients with suspected PH who underwent routine CT pulmonary angiography and right-heart catheterization between April 2012 and March 2016. CT metrics for patients with PH were developed using axial and reconstructed images.

The researchers identified the derivation (n = 247) and validation (n = 244) cohorts using random patient selection. In the derivation cohort, multivariate regression analysis was conducted to develop a model with the ability to predict mPAP ≥25 mm Hg and >20 mm Hg.

In the validation cohort, receiver operating characteristic analysis was performed to establish compromise CT thresholds, as well as sensitivity and specificity. The prognostic utility of the model was evaluated using Kaplan-Meier analysis.
 

Derivation cohort

Among the 247 patients in the derivation cohort, a CT regression model was identified, which included right-ventricle outflow tract thickness, main pulmonary artery diameter, and left ventricular area and interventricular septal angle; the area under the curve (AUC) in this cohort was 0.92.

Validation cohort

Among the 244 patients in the validation cohort, the model demonstrated strong diagnostic utility for the detection of PH, with an AUC of 0.91 and 0.94 for mPAP >20 mm Hg and ≥25 mm Hg, respectively.

With respect to the prognostic utility of the model, the researchers found that the diagnostic thresholds were prognostic in the CT model (all P < .01).

“The diagnostic CT thresholds are also of prognostic value; patients found not to have PH on CT have an excellent outcome,” they explained.

Dr. Swift and colleagues acknowledged that positive and negative predictive values will change based on the diagnostic setting. As a result, the findings from the current study may only be applicable to tertiary referral patient populations.

“This data may be particularly helpful when triaging patients with suspected severe PH for consideration of targeted pulmonary vascular therapies,” they concluded.

The study was supported by Wellcome Trust, the National Institute for Health Research, MRC POLARIS, and Bayer. The authors reported having no conflicts of interest with any companies related to the publication.

SOURCE: Swift AJ et al. Eur Radiol. 2020 Apr 27. doi: 10.1007/s00330-020-06846-1.

A new CT scan pulmonary angiography model may help optimize the diagnostic work-up process for patients with suspected pulmonary hypertension (PH), according to a recent study.

The diagnostic and prognostic utility of the model was validated in a tertiary referral population of treatment-naive patients who had a high pretest probability of PH.

“The aim of this study was to (a) build a diagnostic CT model in patients with suspected PH using the current guideline definition of PH (mPAP [mean pulmonary arterial pressure] ≥25 mm Hg) and the recent proposed definition of >20 mm Hg and (b) test its prognostic significance,” wrote Andrew J. Swift, MBChB, PhD, of the University of Sheffield (England) and colleagues in European Radiology.

The study cohort included 491 patients with suspected PH who underwent routine CT pulmonary angiography and right-heart catheterization between April 2012 and March 2016. CT metrics for patients with PH were developed using axial and reconstructed images.

The researchers identified the derivation (n = 247) and validation (n = 244) cohorts using random patient selection. In the derivation cohort, multivariate regression analysis was conducted to develop a model with the ability to predict mPAP ≥25 mm Hg and >20 mm Hg.

In the validation cohort, receiver operating characteristic analysis was performed to establish compromise CT thresholds, as well as sensitivity and specificity. The prognostic utility of the model was evaluated using Kaplan-Meier analysis.
 

Derivation cohort

Among the 247 patients in the derivation cohort, a CT regression model was identified, which included right-ventricle outflow tract thickness, main pulmonary artery diameter, and left ventricular area and interventricular septal angle; the area under the curve (AUC) in this cohort was 0.92.

Validation cohort

Among the 244 patients in the validation cohort, the model demonstrated strong diagnostic utility for the detection of PH, with an AUC of 0.91 and 0.94 for mPAP >20 mm Hg and ≥25 mm Hg, respectively.

With respect to the prognostic utility of the model, the researchers found that the diagnostic thresholds were prognostic in the CT model (all P < .01).

“The diagnostic CT thresholds are also of prognostic value; patients found not to have PH on CT have an excellent outcome,” they explained.

Dr. Swift and colleagues acknowledged that positive and negative predictive values will change based on the diagnostic setting. As a result, the findings from the current study may only be applicable to tertiary referral patient populations.

“This data may be particularly helpful when triaging patients with suspected severe PH for consideration of targeted pulmonary vascular therapies,” they concluded.

The study was supported by Wellcome Trust, the National Institute for Health Research, MRC POLARIS, and Bayer. The authors reported having no conflicts of interest with any companies related to the publication.

SOURCE: Swift AJ et al. Eur Radiol. 2020 Apr 27. doi: 10.1007/s00330-020-06846-1.

A new CT scan pulmonary angiography model may help optimize the diagnostic work-up process for patients with suspected pulmonary hypertension (PH), according to a recent study.

The diagnostic and prognostic utility of the model was validated in a tertiary referral population of treatment-naive patients who had a high pretest probability of PH.

“The aim of this study was to (a) build a diagnostic CT model in patients with suspected PH using the current guideline definition of PH (mPAP [mean pulmonary arterial pressure] ≥25 mm Hg) and the recent proposed definition of >20 mm Hg and (b) test its prognostic significance,” wrote Andrew J. Swift, MBChB, PhD, of the University of Sheffield (England) and colleagues in European Radiology.

The study cohort included 491 patients with suspected PH who underwent routine CT pulmonary angiography and right-heart catheterization between April 2012 and March 2016. CT metrics for patients with PH were developed using axial and reconstructed images.

The researchers identified the derivation (n = 247) and validation (n = 244) cohorts using random patient selection. In the derivation cohort, multivariate regression analysis was conducted to develop a model with the ability to predict mPAP ≥25 mm Hg and >20 mm Hg.

In the validation cohort, receiver operating characteristic analysis was performed to establish compromise CT thresholds, as well as sensitivity and specificity. The prognostic utility of the model was evaluated using Kaplan-Meier analysis.
 

Derivation cohort

Among the 247 patients in the derivation cohort, a CT regression model was identified, which included right-ventricle outflow tract thickness, main pulmonary artery diameter, and left ventricular area and interventricular septal angle; the area under the curve (AUC) in this cohort was 0.92.

Validation cohort

Among the 244 patients in the validation cohort, the model demonstrated strong diagnostic utility for the detection of PH, with an AUC of 0.91 and 0.94 for mPAP >20 mm Hg and ≥25 mm Hg, respectively.

With respect to the prognostic utility of the model, the researchers found that the diagnostic thresholds were prognostic in the CT model (all P < .01).

“The diagnostic CT thresholds are also of prognostic value; patients found not to have PH on CT have an excellent outcome,” they explained.

Dr. Swift and colleagues acknowledged that positive and negative predictive values will change based on the diagnostic setting. As a result, the findings from the current study may only be applicable to tertiary referral patient populations.

“This data may be particularly helpful when triaging patients with suspected severe PH for consideration of targeted pulmonary vascular therapies,” they concluded.

The study was supported by Wellcome Trust, the National Institute for Health Research, MRC POLARIS, and Bayer. The authors reported having no conflicts of interest with any companies related to the publication.

SOURCE: Swift AJ et al. Eur Radiol. 2020 Apr 27. doi: 10.1007/s00330-020-06846-1.

Biomarkers for inflammation and thrombosis may predict deaths from COVID-19 among critically ill patients, researchers say.

Courtesy NIAID

Their prospective cohort study of 1150 patients hospitalized with the disease in New York City also revealed a high proportion of racial and ethnic minorities, and confirmed high rates of critical illness and mortality.

“Of particular interest is the finding that over three quarters of critically ill patients required a ventilator and almost one third required renal dialysis support,” Max O’Donnell, MD, MPH, assistant professor of medicine and epidemiology at Columbia University in New York City, said in a press release.

O’Donnell and colleagues published the results of their study online today in The Lancet. It is the largest prospective cohort study published in the United States, they said.

“Although the clinical spectrum of disease has been characterised in reports from China and Italy, until now, detailed understanding of how the virus is affecting critically ill patients in the US has been limited to reports from a small number of cases,” said Natalie Yip, MD, assistant professor of medicine at Columbia University.

In the cohort, drawn from two NewYork-Presbyterian hospitals, the researchers focused on the 257 (22%) patients who required intensive care. When they estimated inflammation through interleukin-6 (IL-6) concentrations and thrombosis through D-dimer concentrations, they found a 10% increased risk for death with every 10% increase of IL-6 (adjusted hazard ratio [aHR], 1.11; 95% confidence interval [CI], 1.02–1.20) or D-dimer concentration (aHR, 1.10; 95% CI, 1.01–1.19).

“The association of mortality with higher concentrations of IL-6 and d-dimer is particularly relevant for two reasons,” write Giacomo Grasselli, from the Fondazione IRCCS Ca’ Granda Ospediale Maggiore Policlinico, and Alberto Zanella, from the University of Milan, Italy, in an accompanying commentary.

“First, it confirms the key pathogenic role played by the activation of systemic inflammation and endothelial-vascular damage in the development of organ dysfunction,” they write. “Second, it provides the rationale for the design of clinical trials for measuring the efficacy of treatment with immunomodulating and anticoagulant drugs.”

Seventeen percent of patients received interleukin receptor antagonists and 26% received corticosteroids, but the authors did not report any data on the effects of these treatments, or any data about anticoagulant therapies administered.
 

Severe disease common

The study also highlighted a high proportion of ethnic and racial minorities. Sixty-two percent of the critically ill patients were Hispanic or Latinx, 19% Black, 32% White, and 3% Asian.

Their median age was 62 years and 67% were men. Eighty-two percent had at least one chronic illness, most commonly hypertension (63%), followed by diabetes (36%). Forty-six percent were obese.

As of April 28, 2020, 101 (39%) of the critically ill patients had died following a median of 9 days (interquartile range (IQR), 5–15) in the hospital and 94 (37%) remained hospitalized. Of the 203 patients who received invasive mechanical ventilation, 84 (41%) had died.

The poor prognosis of patients requiring ventilation is consistent with data from a report on patients treated in National Health Service intensive care units in England, Wales, and Northern Ireland through May 15. Overall, 11,292 patients with COVID-19 required critical care, and 4855 needed advanced respiratory support. Approximately half of the patients receiving mechanical ventilation had died 30 days after starting critical care.

In the New York study, patients spent an average of 18 days on a ventilator (IQR, 9–28 days). This is a longer period than reported in smaller studies of cases from Washington state, but corresponds with a recent report from Italy, the researchers said.

Remarkably, O’Donnell and colleagues report that almost a third (31%) of critically ill patients developed severe kidney damage and required dialysis.

Mortality was associated with several baseline factors, including older age (aHR, 1.31 [95% CI, 1.09–1.57] per 10-year increase), chronic cardiac disease (aHR, 1.76; 95% CI, 1.08–2·86), and chronic pulmonary disease (aHR, 2.94; 95% CI, 1.48–5.84).

Authors of the New York study reported financial relationships to ICE Neurosystems, ALung Technologies, Baxter, BREETHE, Xenios, Hemovent, Gilead Sciences, Amazon, and Karyopharm Therapeutics. Grasselli reports personal fees from Biotest, Draeger, Fisher & Paykel, Maquet, Merck Sharp & Dohme, and Pfizer, all outside the area of work commented on here. Zanella has disclosed no relevant financial relationships.

This article first appeared on Medscape.com.

Biomarkers for inflammation and thrombosis may predict deaths from COVID-19 among critically ill patients, researchers say.

Courtesy NIAID

Their prospective cohort study of 1150 patients hospitalized with the disease in New York City also revealed a high proportion of racial and ethnic minorities, and confirmed high rates of critical illness and mortality.

“Of particular interest is the finding that over three quarters of critically ill patients required a ventilator and almost one third required renal dialysis support,” Max O’Donnell, MD, MPH, assistant professor of medicine and epidemiology at Columbia University in New York City, said in a press release.

O’Donnell and colleagues published the results of their study online today in The Lancet. It is the largest prospective cohort study published in the United States, they said.

“Although the clinical spectrum of disease has been characterised in reports from China and Italy, until now, detailed understanding of how the virus is affecting critically ill patients in the US has been limited to reports from a small number of cases,” said Natalie Yip, MD, assistant professor of medicine at Columbia University.

In the cohort, drawn from two NewYork-Presbyterian hospitals, the researchers focused on the 257 (22%) patients who required intensive care. When they estimated inflammation through interleukin-6 (IL-6) concentrations and thrombosis through D-dimer concentrations, they found a 10% increased risk for death with every 10% increase of IL-6 (adjusted hazard ratio [aHR], 1.11; 95% confidence interval [CI], 1.02–1.20) or D-dimer concentration (aHR, 1.10; 95% CI, 1.01–1.19).

“The association of mortality with higher concentrations of IL-6 and d-dimer is particularly relevant for two reasons,” write Giacomo Grasselli, from the Fondazione IRCCS Ca’ Granda Ospediale Maggiore Policlinico, and Alberto Zanella, from the University of Milan, Italy, in an accompanying commentary.

“First, it confirms the key pathogenic role played by the activation of systemic inflammation and endothelial-vascular damage in the development of organ dysfunction,” they write. “Second, it provides the rationale for the design of clinical trials for measuring the efficacy of treatment with immunomodulating and anticoagulant drugs.”

Seventeen percent of patients received interleukin receptor antagonists and 26% received corticosteroids, but the authors did not report any data on the effects of these treatments, or any data about anticoagulant therapies administered.
 

Severe disease common

The study also highlighted a high proportion of ethnic and racial minorities. Sixty-two percent of the critically ill patients were Hispanic or Latinx, 19% Black, 32% White, and 3% Asian.

Their median age was 62 years and 67% were men. Eighty-two percent had at least one chronic illness, most commonly hypertension (63%), followed by diabetes (36%). Forty-six percent were obese.

As of April 28, 2020, 101 (39%) of the critically ill patients had died following a median of 9 days (interquartile range (IQR), 5–15) in the hospital and 94 (37%) remained hospitalized. Of the 203 patients who received invasive mechanical ventilation, 84 (41%) had died.

The poor prognosis of patients requiring ventilation is consistent with data from a report on patients treated in National Health Service intensive care units in England, Wales, and Northern Ireland through May 15. Overall, 11,292 patients with COVID-19 required critical care, and 4855 needed advanced respiratory support. Approximately half of the patients receiving mechanical ventilation had died 30 days after starting critical care.

In the New York study, patients spent an average of 18 days on a ventilator (IQR, 9–28 days). This is a longer period than reported in smaller studies of cases from Washington state, but corresponds with a recent report from Italy, the researchers said.

Remarkably, O’Donnell and colleagues report that almost a third (31%) of critically ill patients developed severe kidney damage and required dialysis.

Mortality was associated with several baseline factors, including older age (aHR, 1.31 [95% CI, 1.09–1.57] per 10-year increase), chronic cardiac disease (aHR, 1.76; 95% CI, 1.08–2·86), and chronic pulmonary disease (aHR, 2.94; 95% CI, 1.48–5.84).

Authors of the New York study reported financial relationships to ICE Neurosystems, ALung Technologies, Baxter, BREETHE, Xenios, Hemovent, Gilead Sciences, Amazon, and Karyopharm Therapeutics. Grasselli reports personal fees from Biotest, Draeger, Fisher & Paykel, Maquet, Merck Sharp & Dohme, and Pfizer, all outside the area of work commented on here. Zanella has disclosed no relevant financial relationships.

This article first appeared on Medscape.com.

Biomarkers for inflammation and thrombosis may predict deaths from COVID-19 among critically ill patients, researchers say.

Courtesy NIAID

Their prospective cohort study of 1150 patients hospitalized with the disease in New York City also revealed a high proportion of racial and ethnic minorities, and confirmed high rates of critical illness and mortality.

“Of particular interest is the finding that over three quarters of critically ill patients required a ventilator and almost one third required renal dialysis support,” Max O’Donnell, MD, MPH, assistant professor of medicine and epidemiology at Columbia University in New York City, said in a press release.

O’Donnell and colleagues published the results of their study online today in The Lancet. It is the largest prospective cohort study published in the United States, they said.

“Although the clinical spectrum of disease has been characterised in reports from China and Italy, until now, detailed understanding of how the virus is affecting critically ill patients in the US has been limited to reports from a small number of cases,” said Natalie Yip, MD, assistant professor of medicine at Columbia University.

In the cohort, drawn from two NewYork-Presbyterian hospitals, the researchers focused on the 257 (22%) patients who required intensive care. When they estimated inflammation through interleukin-6 (IL-6) concentrations and thrombosis through D-dimer concentrations, they found a 10% increased risk for death with every 10% increase of IL-6 (adjusted hazard ratio [aHR], 1.11; 95% confidence interval [CI], 1.02–1.20) or D-dimer concentration (aHR, 1.10; 95% CI, 1.01–1.19).

“The association of mortality with higher concentrations of IL-6 and d-dimer is particularly relevant for two reasons,” write Giacomo Grasselli, from the Fondazione IRCCS Ca’ Granda Ospediale Maggiore Policlinico, and Alberto Zanella, from the University of Milan, Italy, in an accompanying commentary.

“First, it confirms the key pathogenic role played by the activation of systemic inflammation and endothelial-vascular damage in the development of organ dysfunction,” they write. “Second, it provides the rationale for the design of clinical trials for measuring the efficacy of treatment with immunomodulating and anticoagulant drugs.”

Seventeen percent of patients received interleukin receptor antagonists and 26% received corticosteroids, but the authors did not report any data on the effects of these treatments, or any data about anticoagulant therapies administered.
 

Severe disease common

The study also highlighted a high proportion of ethnic and racial minorities. Sixty-two percent of the critically ill patients were Hispanic or Latinx, 19% Black, 32% White, and 3% Asian.

Their median age was 62 years and 67% were men. Eighty-two percent had at least one chronic illness, most commonly hypertension (63%), followed by diabetes (36%). Forty-six percent were obese.

As of April 28, 2020, 101 (39%) of the critically ill patients had died following a median of 9 days (interquartile range (IQR), 5–15) in the hospital and 94 (37%) remained hospitalized. Of the 203 patients who received invasive mechanical ventilation, 84 (41%) had died.

The poor prognosis of patients requiring ventilation is consistent with data from a report on patients treated in National Health Service intensive care units in England, Wales, and Northern Ireland through May 15. Overall, 11,292 patients with COVID-19 required critical care, and 4855 needed advanced respiratory support. Approximately half of the patients receiving mechanical ventilation had died 30 days after starting critical care.

In the New York study, patients spent an average of 18 days on a ventilator (IQR, 9–28 days). This is a longer period than reported in smaller studies of cases from Washington state, but corresponds with a recent report from Italy, the researchers said.

Remarkably, O’Donnell and colleagues report that almost a third (31%) of critically ill patients developed severe kidney damage and required dialysis.

Mortality was associated with several baseline factors, including older age (aHR, 1.31 [95% CI, 1.09–1.57] per 10-year increase), chronic cardiac disease (aHR, 1.76; 95% CI, 1.08–2·86), and chronic pulmonary disease (aHR, 2.94; 95% CI, 1.48–5.84).

Authors of the New York study reported financial relationships to ICE Neurosystems, ALung Technologies, Baxter, BREETHE, Xenios, Hemovent, Gilead Sciences, Amazon, and Karyopharm Therapeutics. Grasselli reports personal fees from Biotest, Draeger, Fisher & Paykel, Maquet, Merck Sharp & Dohme, and Pfizer, all outside the area of work commented on here. Zanella has disclosed no relevant financial relationships.

This article first appeared on Medscape.com.

Among patients with atrial fibrillation, use of direct oral anticoagulants is linked to lower osteoporotic fracture risk versus warfarin, results of a recent population-based cohort study show.

iStock/Thinkstock

The choice of direct oral anticoagulant (DOAC) didn’t appear to have an impact, as each individual agent yielded a substantially lower risk of fracture versus the vitamin K antagonist, with risk reductions ranging from 38% to 48%, according to the study authors.

This is one of the latest reports to suggest DOACs could have an edge over warfarin for preventing fractures, providing new evidence that “may help inform the benefit risk assessment” when it comes to choosing an anticoagulant for a patient with atrial fibrillation (AFib) in the clinic, wrote the authors, led by Wallis C.Y. Lau, PhD, with the University College London.

“There exists a compelling case for evaluating whether the risk for osteoporotic fractures should be considered at the point of prescribing an oral anticoagulant to minimize fracture risk,” Dr. Lau and coauthors wrote in a report on the study that appears in Annals of Internal Medicine.

The case is especially compelling since fracture risk is “often neglected” when choosing an anticoagulant, the authors wrote. Surgeries to treat fracture are difficult because of the need for perioperative management of anticoagulation as “a balance between the risk for stroke and excessive bleeding must be achieved,” they added.

Based on these data, physicians should strongly consider DOACs as an alternative to vitamin K antagonists to reduce the risk of osteoporosis over the long term in patients with AFib, according to Victor Lawrence Roberts, MD, a Florida endocrinologist.

“Osteoporosis takes years, sometimes decades to develop, and if you then overlay warfarin on top of a readily evolving metabolic bone disease, you probably accelerate that process, said Dr. Roberts, professor of internal medicine at the University of Central Florida, Orlando, and editorial advisory board member of Internal Medicine News.

There’s a considerable amount of concerning preclinical data that warfarin could increase osteoporotic fracture risk. Of note, vitamin K antagonists modulate osteocalcin, a calcium-binding bone matrix protein, Dr. Roberts said.

“Osteocalcin is important for bone metabolism and health, and inhibiting osteocalcin will inhibit the ability to have a healthy bone matrix,” he explained.

The impact of anticoagulants on fracture risk is particularly relevant to patients with AFib, according to Dr. Lau and colleagues, who referenced one 2017 report showing a higher incidence of hip fracture among AFib patients versus those without AFib.

In their more recent study, Dr. Lau and colleagues reviewed electronic health records in a Hong Kong database for 23,515 older adults with a new diagnosis of AFib who received a new prescription of warfarin or DOACs including apixaban, dabigatran, or rivaroxaban.

DOAC use was consistently associated with a lower risk of osteoporotic fractures versus warfarin, regardless of the DOAC considered. The hazard ratios were 0.62 (95% confidence interval, 0.41-0.94) for apixaban, 0.65 (95% CI, 0.49-0.86) for dabigatran, and 0.52 (95% CI, 0.37-0.73) for rivaroxaban versus warfarin, the report showed.

Head-to-head comparisons between DOACS didn’t yield any statistically significant differences, though the analyses were underpowered in this respect, according to the investigators.

“This study can only rule out more than a twofold higher or a 50% lower relative risk for osteoporotic fractures between individual DOACs,” they wrote. “However, any absolute risk differences were small and would likely be of minor clinical significance.”

The reduced risk of fracture for DOACs versus warfarin was consistent in men and women with AFib, suggesting that women may particularly benefit from DOACs, given that they have a higher risk of fracture than men, the investigators added.

The results of this study suggest yet another benefit of DOACs over warfarin in patients with AFib, according to internist Noel Deep, MD, who is the chief medical officer of Aspirus Langlade Hospital in Antigo, Wisconsin.

“The lower risk of osteoporotic fractures with DOACS, in addition to other advantages such as lower risk of intracranial bleeding, once- or twice-daily consistent dosing, no dietary restrictions, and no blood tests to regulate the dose might be another reason that physicians may favor them over warfarin in older individuals requiring anticoagulation,” Dr. Deep said in an interview.

Results of this and several other recent studies may help in recommending DOACs to internal medicine patients who have a diagnosis of AFib requiring anticoagulation, according to Dr. Deep, who is also a physician at Aspirus Antigo Clinic and a member of Internal Medicine News’ editorial advisory board. These include a 2019 U.S.-based study of more than 167,000 patients with AFib (JAMA Intern Med. 2019;180[2]:245‐253) showing that use of DOACs, particularly apixaban, were linked to lower fracture risk versus warfarin use. Similarly, a Danish national registry study also published in 2019 showed that the absolute risk of osteoporotic fractures was low overall and significantly lower in patients who received DOACs (J Am Coll Cardiol. 2019;74[17]:2150-2158).

Funding for the study came from the University of Hong Kong and University College London Strategic Planning Fund. The study authors reported disclosures related to Bayer, Bristol-Myers Squibb, Pfizer, Janssen, Amgen, Takeda, IQVIA, and others.

SOURCE: Lau WCY et al. Ann Intern Med. 2020 May 18. doi: 10.7326/M19-3671.

Among patients with atrial fibrillation, use of direct oral anticoagulants is linked to lower osteoporotic fracture risk versus warfarin, results of a recent population-based cohort study show.

iStock/Thinkstock

The choice of direct oral anticoagulant (DOAC) didn’t appear to have an impact, as each individual agent yielded a substantially lower risk of fracture versus the vitamin K antagonist, with risk reductions ranging from 38% to 48%, according to the study authors.

This is one of the latest reports to suggest DOACs could have an edge over warfarin for preventing fractures, providing new evidence that “may help inform the benefit risk assessment” when it comes to choosing an anticoagulant for a patient with atrial fibrillation (AFib) in the clinic, wrote the authors, led by Wallis C.Y. Lau, PhD, with the University College London.

“There exists a compelling case for evaluating whether the risk for osteoporotic fractures should be considered at the point of prescribing an oral anticoagulant to minimize fracture risk,” Dr. Lau and coauthors wrote in a report on the study that appears in Annals of Internal Medicine.

The case is especially compelling since fracture risk is “often neglected” when choosing an anticoagulant, the authors wrote. Surgeries to treat fracture are difficult because of the need for perioperative management of anticoagulation as “a balance between the risk for stroke and excessive bleeding must be achieved,” they added.

Based on these data, physicians should strongly consider DOACs as an alternative to vitamin K antagonists to reduce the risk of osteoporosis over the long term in patients with AFib, according to Victor Lawrence Roberts, MD, a Florida endocrinologist.

“Osteoporosis takes years, sometimes decades to develop, and if you then overlay warfarin on top of a readily evolving metabolic bone disease, you probably accelerate that process, said Dr. Roberts, professor of internal medicine at the University of Central Florida, Orlando, and editorial advisory board member of Internal Medicine News.

There’s a considerable amount of concerning preclinical data that warfarin could increase osteoporotic fracture risk. Of note, vitamin K antagonists modulate osteocalcin, a calcium-binding bone matrix protein, Dr. Roberts said.

“Osteocalcin is important for bone metabolism and health, and inhibiting osteocalcin will inhibit the ability to have a healthy bone matrix,” he explained.

The impact of anticoagulants on fracture risk is particularly relevant to patients with AFib, according to Dr. Lau and colleagues, who referenced one 2017 report showing a higher incidence of hip fracture among AFib patients versus those without AFib.

In their more recent study, Dr. Lau and colleagues reviewed electronic health records in a Hong Kong database for 23,515 older adults with a new diagnosis of AFib who received a new prescription of warfarin or DOACs including apixaban, dabigatran, or rivaroxaban.

DOAC use was consistently associated with a lower risk of osteoporotic fractures versus warfarin, regardless of the DOAC considered. The hazard ratios were 0.62 (95% confidence interval, 0.41-0.94) for apixaban, 0.65 (95% CI, 0.49-0.86) for dabigatran, and 0.52 (95% CI, 0.37-0.73) for rivaroxaban versus warfarin, the report showed.

Head-to-head comparisons between DOACS didn’t yield any statistically significant differences, though the analyses were underpowered in this respect, according to the investigators.

“This study can only rule out more than a twofold higher or a 50% lower relative risk for osteoporotic fractures between individual DOACs,” they wrote. “However, any absolute risk differences were small and would likely be of minor clinical significance.”

The reduced risk of fracture for DOACs versus warfarin was consistent in men and women with AFib, suggesting that women may particularly benefit from DOACs, given that they have a higher risk of fracture than men, the investigators added.

The results of this study suggest yet another benefit of DOACs over warfarin in patients with AFib, according to internist Noel Deep, MD, who is the chief medical officer of Aspirus Langlade Hospital in Antigo, Wisconsin.

“The lower risk of osteoporotic fractures with DOACS, in addition to other advantages such as lower risk of intracranial bleeding, once- or twice-daily consistent dosing, no dietary restrictions, and no blood tests to regulate the dose might be another reason that physicians may favor them over warfarin in older individuals requiring anticoagulation,” Dr. Deep said in an interview.

Results of this and several other recent studies may help in recommending DOACs to internal medicine patients who have a diagnosis of AFib requiring anticoagulation, according to Dr. Deep, who is also a physician at Aspirus Antigo Clinic and a member of Internal Medicine News’ editorial advisory board. These include a 2019 U.S.-based study of more than 167,000 patients with AFib (JAMA Intern Med. 2019;180[2]:245‐253) showing that use of DOACs, particularly apixaban, were linked to lower fracture risk versus warfarin use. Similarly, a Danish national registry study also published in 2019 showed that the absolute risk of osteoporotic fractures was low overall and significantly lower in patients who received DOACs (J Am Coll Cardiol. 2019;74[17]:2150-2158).

Funding for the study came from the University of Hong Kong and University College London Strategic Planning Fund. The study authors reported disclosures related to Bayer, Bristol-Myers Squibb, Pfizer, Janssen, Amgen, Takeda, IQVIA, and others.

SOURCE: Lau WCY et al. Ann Intern Med. 2020 May 18. doi: 10.7326/M19-3671.

Among patients with atrial fibrillation, use of direct oral anticoagulants is linked to lower osteoporotic fracture risk versus warfarin, results of a recent population-based cohort study show.

iStock/Thinkstock

The choice of direct oral anticoagulant (DOAC) didn’t appear to have an impact, as each individual agent yielded a substantially lower risk of fracture versus the vitamin K antagonist, with risk reductions ranging from 38% to 48%, according to the study authors.

This is one of the latest reports to suggest DOACs could have an edge over warfarin for preventing fractures, providing new evidence that “may help inform the benefit risk assessment” when it comes to choosing an anticoagulant for a patient with atrial fibrillation (AFib) in the clinic, wrote the authors, led by Wallis C.Y. Lau, PhD, with the University College London.

“There exists a compelling case for evaluating whether the risk for osteoporotic fractures should be considered at the point of prescribing an oral anticoagulant to minimize fracture risk,” Dr. Lau and coauthors wrote in a report on the study that appears in Annals of Internal Medicine.

The case is especially compelling since fracture risk is “often neglected” when choosing an anticoagulant, the authors wrote. Surgeries to treat fracture are difficult because of the need for perioperative management of anticoagulation as “a balance between the risk for stroke and excessive bleeding must be achieved,” they added.

Based on these data, physicians should strongly consider DOACs as an alternative to vitamin K antagonists to reduce the risk of osteoporosis over the long term in patients with AFib, according to Victor Lawrence Roberts, MD, a Florida endocrinologist.

“Osteoporosis takes years, sometimes decades to develop, and if you then overlay warfarin on top of a readily evolving metabolic bone disease, you probably accelerate that process, said Dr. Roberts, professor of internal medicine at the University of Central Florida, Orlando, and editorial advisory board member of Internal Medicine News.

There’s a considerable amount of concerning preclinical data that warfarin could increase osteoporotic fracture risk. Of note, vitamin K antagonists modulate osteocalcin, a calcium-binding bone matrix protein, Dr. Roberts said.

“Osteocalcin is important for bone metabolism and health, and inhibiting osteocalcin will inhibit the ability to have a healthy bone matrix,” he explained.

The impact of anticoagulants on fracture risk is particularly relevant to patients with AFib, according to Dr. Lau and colleagues, who referenced one 2017 report showing a higher incidence of hip fracture among AFib patients versus those without AFib.

In their more recent study, Dr. Lau and colleagues reviewed electronic health records in a Hong Kong database for 23,515 older adults with a new diagnosis of AFib who received a new prescription of warfarin or DOACs including apixaban, dabigatran, or rivaroxaban.

DOAC use was consistently associated with a lower risk of osteoporotic fractures versus warfarin, regardless of the DOAC considered. The hazard ratios were 0.62 (95% confidence interval, 0.41-0.94) for apixaban, 0.65 (95% CI, 0.49-0.86) for dabigatran, and 0.52 (95% CI, 0.37-0.73) for rivaroxaban versus warfarin, the report showed.

Head-to-head comparisons between DOACS didn’t yield any statistically significant differences, though the analyses were underpowered in this respect, according to the investigators.

“This study can only rule out more than a twofold higher or a 50% lower relative risk for osteoporotic fractures between individual DOACs,” they wrote. “However, any absolute risk differences were small and would likely be of minor clinical significance.”

The reduced risk of fracture for DOACs versus warfarin was consistent in men and women with AFib, suggesting that women may particularly benefit from DOACs, given that they have a higher risk of fracture than men, the investigators added.

The results of this study suggest yet another benefit of DOACs over warfarin in patients with AFib, according to internist Noel Deep, MD, who is the chief medical officer of Aspirus Langlade Hospital in Antigo, Wisconsin.

“The lower risk of osteoporotic fractures with DOACS, in addition to other advantages such as lower risk of intracranial bleeding, once- or twice-daily consistent dosing, no dietary restrictions, and no blood tests to regulate the dose might be another reason that physicians may favor them over warfarin in older individuals requiring anticoagulation,” Dr. Deep said in an interview.

Results of this and several other recent studies may help in recommending DOACs to internal medicine patients who have a diagnosis of AFib requiring anticoagulation, according to Dr. Deep, who is also a physician at Aspirus Antigo Clinic and a member of Internal Medicine News’ editorial advisory board. These include a 2019 U.S.-based study of more than 167,000 patients with AFib (JAMA Intern Med. 2019;180[2]:245‐253) showing that use of DOACs, particularly apixaban, were linked to lower fracture risk versus warfarin use. Similarly, a Danish national registry study also published in 2019 showed that the absolute risk of osteoporotic fractures was low overall and significantly lower in patients who received DOACs (J Am Coll Cardiol. 2019;74[17]:2150-2158).

Funding for the study came from the University of Hong Kong and University College London Strategic Planning Fund. The study authors reported disclosures related to Bayer, Bristol-Myers Squibb, Pfizer, Janssen, Amgen, Takeda, IQVIA, and others.

SOURCE: Lau WCY et al. Ann Intern Med. 2020 May 18. doi: 10.7326/M19-3671.

A saliva test for detecting oropharyngeal cancer caused by human papillomavirus–16 (HPV-16) has scored a world first: It detected such a cancer in an asymptomatic adult.

If the finding can be replicated in a larger cohort of healthy asymptomatic individuals, widespread screening for HPV-16 – the main driver behind skyrocketing rates of oropharyngeal cancer – ­­­­­could be one step closer, the researchers suggested.

“Oropharyngeal squamous cell carcinomas often presents at a late stage with patients suffering huge morbidity as a result of treatment, [so] we must find strategies to detect these cancers earlier,” senior author Chamindie Punyadeera, PhD, Queensland University of Technology in Brisbane, Australia, told Medscape Medical News in an email.

“This study, for the first time, provides a solid scientific foundation to initiate a screening trial in high-risk individuals to detect HPV-driven oropharyngeal cancer, she added. “Saliva testing could be broadly implemented and [used] in a screening trial in the future,” she said.

The case report was published online March 31 in Frontiers in Oncology.

The saliva test was developed by Dr. Punyadeera and first author Kai Dun Tang, PhD, also from Queensland University of Technology. It is administered as an oral rinse: the individual swishes a saline solution around in his or her mouth for a minute or 2, and then spits the sample into a tube.
 

Prevalence Study

The saliva test was being scrutinized in an ongoing HPV-16 DNA prevalence study, which involved 650 healthy participants being tested for oral HPV-16 DNA.

“Of these, 3 have been identified to have persistent oral HPV-16 DNA infection,” the investigators reported.

After having approached these three participants, one middle-aged male who had been consistently HPV-16 DNA positive for a period of 36 months – and whose HPV-16 viral load had been steadily rising over time – was invited to attend an ear, nose, and throat clinic for assessment.

“Initial clinical examination of the oropharynx including palpation and white light revealed no significant abnormalities,” the researchers emphasized.

As Dr. Punyadeera explained, standard clinical assessment for oropharyngeal malignancy includes white light examination for masses, detection of irregularities or asymmetry of the underlying structures and palpation of the tonsil and tongue base.

Cross-sectional imaging with CT or MRI can be helpful as well, she said, but these imaging studies are unable to detect lesions smaller than a few millimeters in size.

In the case of this individual, salivary oral rinse samples had been collected at baseline, and again at 6, 12, and 36 months after study enrollment as well as 2 weeks after the patient decided to undergo a bilateral tonsillectomy.

DNA was extracted from the salivary oral rinse samples, as well as from the tonsillar tissue obtained after resection. HPV-16 DNA genotyping and viral loads were analyzed with a PCR assay.

Results from the salivary samples indicated that the patient’s HPV-16 DNA viral load had increased exponentially across the 36 months of follow-up, from 3.43 copies/50 ng at baseline to 1281.69 copies/50 ng at 36 months.

On surgery, the patient was found to have a 2 mm squamous cell carcinoma in the left tonsil, but all other oropharyngeal tissues were normal and HPV-16 DNA negative.

Two weeks after undergoing the tonsillectomy, the patient’s HPV-16 DNA viral load in the saliva samples became undetectable.

This case report demonstrates that salivary HPV can both detect smaller lesions than either clinical examination or even radiological investigation, and that the same salivary test can likely also be used to monitor treatment response, Dr. Punyadeera commented.
 

Long-term persistence

As researchers explained in their paper, long-term persistence of HPV-16 infection is most likely a prerequisite for the development of subsequent malignancy.

Unlike cervical cancer caused by HPV-16 infection, the natural history of HPV infection in the oropharyngeal cavity is not known.

However, clinical assessment of patients with either persistent HPV infection or microscopic carcinoma has failed to detect any identifiable abnormalities.

Thus, this is the first report of a histologically confirmed diagnosis of an asymptomatic occult oropharyngeal cancer detected by a screening test through serial measurements of HPV-16 DNA, the investigators emphasized.

The report also demonstrated that very early lesions can be eradicated with minimal morbidity. Unfortunately, most oropharyngeal cancer is currently diagnosed at much later stages, and surgical removal of these is often associated with significant disabilities including difficulties with swallowing and even communicating.

“It’s amazing to think that this man was cured of his disease with a 15-minute procedure which left him with no lasting issues at all,” Dr. Punyadeera commented. “We need to try and make this the norm, not the exception.

“So we must have a well-designed screening study using all the insights we have gained from this case. We owe it to patients to explore these findings to their fullest,” Dr. Punyadeera emphasized.

The authors have disclosed no relevant financial relationships.

This article first appeared on Medscape.com.

A saliva test for detecting oropharyngeal cancer caused by human papillomavirus–16 (HPV-16) has scored a world first: It detected such a cancer in an asymptomatic adult.

If the finding can be replicated in a larger cohort of healthy asymptomatic individuals, widespread screening for HPV-16 – the main driver behind skyrocketing rates of oropharyngeal cancer – ­­­­­could be one step closer, the researchers suggested.

“Oropharyngeal squamous cell carcinomas often presents at a late stage with patients suffering huge morbidity as a result of treatment, [so] we must find strategies to detect these cancers earlier,” senior author Chamindie Punyadeera, PhD, Queensland University of Technology in Brisbane, Australia, told Medscape Medical News in an email.

“This study, for the first time, provides a solid scientific foundation to initiate a screening trial in high-risk individuals to detect HPV-driven oropharyngeal cancer, she added. “Saliva testing could be broadly implemented and [used] in a screening trial in the future,” she said.

The case report was published online March 31 in Frontiers in Oncology.

The saliva test was developed by Dr. Punyadeera and first author Kai Dun Tang, PhD, also from Queensland University of Technology. It is administered as an oral rinse: the individual swishes a saline solution around in his or her mouth for a minute or 2, and then spits the sample into a tube.
 

Prevalence Study

The saliva test was being scrutinized in an ongoing HPV-16 DNA prevalence study, which involved 650 healthy participants being tested for oral HPV-16 DNA.

“Of these, 3 have been identified to have persistent oral HPV-16 DNA infection,” the investigators reported.

After having approached these three participants, one middle-aged male who had been consistently HPV-16 DNA positive for a period of 36 months – and whose HPV-16 viral load had been steadily rising over time – was invited to attend an ear, nose, and throat clinic for assessment.

“Initial clinical examination of the oropharynx including palpation and white light revealed no significant abnormalities,” the researchers emphasized.

As Dr. Punyadeera explained, standard clinical assessment for oropharyngeal malignancy includes white light examination for masses, detection of irregularities or asymmetry of the underlying structures and palpation of the tonsil and tongue base.

Cross-sectional imaging with CT or MRI can be helpful as well, she said, but these imaging studies are unable to detect lesions smaller than a few millimeters in size.

In the case of this individual, salivary oral rinse samples had been collected at baseline, and again at 6, 12, and 36 months after study enrollment as well as 2 weeks after the patient decided to undergo a bilateral tonsillectomy.

DNA was extracted from the salivary oral rinse samples, as well as from the tonsillar tissue obtained after resection. HPV-16 DNA genotyping and viral loads were analyzed with a PCR assay.

Results from the salivary samples indicated that the patient’s HPV-16 DNA viral load had increased exponentially across the 36 months of follow-up, from 3.43 copies/50 ng at baseline to 1281.69 copies/50 ng at 36 months.

On surgery, the patient was found to have a 2 mm squamous cell carcinoma in the left tonsil, but all other oropharyngeal tissues were normal and HPV-16 DNA negative.

Two weeks after undergoing the tonsillectomy, the patient’s HPV-16 DNA viral load in the saliva samples became undetectable.

This case report demonstrates that salivary HPV can both detect smaller lesions than either clinical examination or even radiological investigation, and that the same salivary test can likely also be used to monitor treatment response, Dr. Punyadeera commented.
 

Long-term persistence

As researchers explained in their paper, long-term persistence of HPV-16 infection is most likely a prerequisite for the development of subsequent malignancy.

Unlike cervical cancer caused by HPV-16 infection, the natural history of HPV infection in the oropharyngeal cavity is not known.

However, clinical assessment of patients with either persistent HPV infection or microscopic carcinoma has failed to detect any identifiable abnormalities.

Thus, this is the first report of a histologically confirmed diagnosis of an asymptomatic occult oropharyngeal cancer detected by a screening test through serial measurements of HPV-16 DNA, the investigators emphasized.

The report also demonstrated that very early lesions can be eradicated with minimal morbidity. Unfortunately, most oropharyngeal cancer is currently diagnosed at much later stages, and surgical removal of these is often associated with significant disabilities including difficulties with swallowing and even communicating.

“It’s amazing to think that this man was cured of his disease with a 15-minute procedure which left him with no lasting issues at all,” Dr. Punyadeera commented. “We need to try and make this the norm, not the exception.

“So we must have a well-designed screening study using all the insights we have gained from this case. We owe it to patients to explore these findings to their fullest,” Dr. Punyadeera emphasized.

The authors have disclosed no relevant financial relationships.

This article first appeared on Medscape.com.

A saliva test for detecting oropharyngeal cancer caused by human papillomavirus–16 (HPV-16) has scored a world first: It detected such a cancer in an asymptomatic adult.

If the finding can be replicated in a larger cohort of healthy asymptomatic individuals, widespread screening for HPV-16 – the main driver behind skyrocketing rates of oropharyngeal cancer – ­­­­­could be one step closer, the researchers suggested.

“Oropharyngeal squamous cell carcinomas often presents at a late stage with patients suffering huge morbidity as a result of treatment, [so] we must find strategies to detect these cancers earlier,” senior author Chamindie Punyadeera, PhD, Queensland University of Technology in Brisbane, Australia, told Medscape Medical News in an email.

“This study, for the first time, provides a solid scientific foundation to initiate a screening trial in high-risk individuals to detect HPV-driven oropharyngeal cancer, she added. “Saliva testing could be broadly implemented and [used] in a screening trial in the future,” she said.

The case report was published online March 31 in Frontiers in Oncology.

The saliva test was developed by Dr. Punyadeera and first author Kai Dun Tang, PhD, also from Queensland University of Technology. It is administered as an oral rinse: the individual swishes a saline solution around in his or her mouth for a minute or 2, and then spits the sample into a tube.
 

Prevalence Study

The saliva test was being scrutinized in an ongoing HPV-16 DNA prevalence study, which involved 650 healthy participants being tested for oral HPV-16 DNA.

“Of these, 3 have been identified to have persistent oral HPV-16 DNA infection,” the investigators reported.

After having approached these three participants, one middle-aged male who had been consistently HPV-16 DNA positive for a period of 36 months – and whose HPV-16 viral load had been steadily rising over time – was invited to attend an ear, nose, and throat clinic for assessment.

“Initial clinical examination of the oropharynx including palpation and white light revealed no significant abnormalities,” the researchers emphasized.

As Dr. Punyadeera explained, standard clinical assessment for oropharyngeal malignancy includes white light examination for masses, detection of irregularities or asymmetry of the underlying structures and palpation of the tonsil and tongue base.

Cross-sectional imaging with CT or MRI can be helpful as well, she said, but these imaging studies are unable to detect lesions smaller than a few millimeters in size.

In the case of this individual, salivary oral rinse samples had been collected at baseline, and again at 6, 12, and 36 months after study enrollment as well as 2 weeks after the patient decided to undergo a bilateral tonsillectomy.

DNA was extracted from the salivary oral rinse samples, as well as from the tonsillar tissue obtained after resection. HPV-16 DNA genotyping and viral loads were analyzed with a PCR assay.

Results from the salivary samples indicated that the patient’s HPV-16 DNA viral load had increased exponentially across the 36 months of follow-up, from 3.43 copies/50 ng at baseline to 1281.69 copies/50 ng at 36 months.

On surgery, the patient was found to have a 2 mm squamous cell carcinoma in the left tonsil, but all other oropharyngeal tissues were normal and HPV-16 DNA negative.

Two weeks after undergoing the tonsillectomy, the patient’s HPV-16 DNA viral load in the saliva samples became undetectable.

This case report demonstrates that salivary HPV can both detect smaller lesions than either clinical examination or even radiological investigation, and that the same salivary test can likely also be used to monitor treatment response, Dr. Punyadeera commented.
 

Long-term persistence

As researchers explained in their paper, long-term persistence of HPV-16 infection is most likely a prerequisite for the development of subsequent malignancy.

Unlike cervical cancer caused by HPV-16 infection, the natural history of HPV infection in the oropharyngeal cavity is not known.

However, clinical assessment of patients with either persistent HPV infection or microscopic carcinoma has failed to detect any identifiable abnormalities.

Thus, this is the first report of a histologically confirmed diagnosis of an asymptomatic occult oropharyngeal cancer detected by a screening test through serial measurements of HPV-16 DNA, the investigators emphasized.

The report also demonstrated that very early lesions can be eradicated with minimal morbidity. Unfortunately, most oropharyngeal cancer is currently diagnosed at much later stages, and surgical removal of these is often associated with significant disabilities including difficulties with swallowing and even communicating.

“It’s amazing to think that this man was cured of his disease with a 15-minute procedure which left him with no lasting issues at all,” Dr. Punyadeera commented. “We need to try and make this the norm, not the exception.

“So we must have a well-designed screening study using all the insights we have gained from this case. We owe it to patients to explore these findings to their fullest,” Dr. Punyadeera emphasized.

The authors have disclosed no relevant financial relationships.

This article first appeared on Medscape.com.

Children with neurologic impairment (NI; eg, hypoxic-­ischemic encephalopathy, muscular dystrophy) are characterized by functional and/or intellectual impairments resulting from a variety of neurologic diseases.¹ These children commonly have respiratory comorbidities, including central hypoventilation, impaired cough, and oromotor dysfunction, that may lead to chronic respiratory insufficiency and a need for chronic respiratory support at baseline.^2,3 Baseline respiratory support modalities can include supplemental oxygen, noninvasive positive pressure ventilation, or invasive mechanical ventilation.

Acute respiratory infections (ARI; eg, pneumonia, bronchiolitis) are the most common cause of hospitalization, intensive care unit (ICU) admission, and death for children with NI.^1,3 Discharge criteria for otherwise healthy children admitted to the hospital with ARI often include return to respiratory baseline.⁴ Children with complex chronic conditions have longer hospitalizations when hospitalized with respiratory infections,^5-7 because, in part, comorbidities and complications prolong the time to return to baseline. This prolonged return to respiratory baseline in combination with family knowledge, comfort, and skill in managing respiratory support and other complexities at home may alter discharge practices in the population of children with NI. In our clinical experience, discharge before return to baseline respiratory support occurs more frequently in children with NI than in otherwise healthy children when hospitalized with ARI. However, the consequences of discharging children with NI prior to return to respiratory baseline are unknown.

In this study, we sought to determine if discharge prior to return to baseline respiratory support is associated with reutilization among children with NI hospitalized with ARI. We hypothesized that patients discharged prior to return to respiratory baseline would have higher rates of 30-day hospital reutilization.

Study Design and Data Source

This single-center, retrospective cohort study of children hospitalized at Cincinnati Children’s Hospital Medical Center (CCHMC) used data from the Pediatric Health Information System (PHIS) and the electronic medical record (EMR). PHIS, an administrative database of 45 not-for-profit, tertiary care, US pediatric hospitals managed by Children’s Hospital Association (Lenexa, Kansas), was used to identify eligible children, examine demographic and clinical variables, and define outcomes. PHIS contains data regarding patient demographics, inpatient resource utilization, and diagnoses. Encrypted medical record numbers in PHIS allowed for local identification of patients’ medical records to complete EMR review to confirm eligibility and obtain detailed patient-level clinical information (eg, respiratory support needs) not available in PHIS.

Pilot medical record reviews allowed for standardized study definitions and procedures. All study staff underwent training with the primary investigator, including detailed review of 10 initial abstractions. Two investigators (K.M. and S.C.) performed repeat abstractions from 40 randomly selected records to enable assessment of interrater reliability. Average reliability, indicated by the κ statistic, indicated substantial to near-perfect reliability⁸ (κ = 0.97, 95% CI 0.90-1.00) for the primary exposure. EMR data were managed using Research Electronic Data Capture (REDCap, Nashville, Tennessee)⁹ and subsequently merged with PHIS data.

Study Population

Hospitalizations of children with NI aged 1 to 18 years at CCHMC between January 2010 and September 2015 were eligible for inclusion if they had a principal discharge diagnosis indicative of ARI and required increased respiratory support from baseline during hospitalization. NI was defined as a high-intensity, chronic neurological diagnosis with substantial functional impairments according to previously defined diagnosis codes.^1,10 ARI was identified using codes in the Clinical Classification Software (Agency for Healthcare Research and Quality, Rockville, MD) respiratory group indicative of ARI (eg, pneumonia, bronchiolitis, influenza; Appendix Table).

Children transferred to CCHMC were excluded because records from their initial illness presentation and management were not available. Because of expected differences in management and outcomes, children with a known diagnosis of tuberculosis or human immunodeficiency virus were excluded. Because exposure criteria were dependent on hospital discharge status, hospitalizations for children who died during admission (4 of 632 hospitalizations, 0.63%) were excluded from the final cohort (Appendix Figure).

Study Definitions

Baseline respiratory support (ie, “respiratory baseline”) was defined as the child’s highest level of respiratory support needed prior to admission when well (ie, no support, supplemental oxygen, continuous positive airway pressure [CPAP] or bilevel positive airway pressure [BiPAP], or ventilator support), and further characterized by night or day/night requirement. Respiratory baseline was identified using EMR documentation of home respiratory support use at the time of index admission. Return to respiratory baseline was defined as the date on which the child achieved documented home respiratory support settings, regardless of clinical symptoms.

Children may have required increased respiratory support from baseline at any time during hospitalization. Maximum respiratory support required was categorized as one of the following: (1) initiation of supplemental oxygen or increase in oxygen flow or duration; (2) initiation of CPAP or BiPAP; (3) increase in pressure settings or duration of pressure support for those with baseline CPAP, BiPAP, or ventilator use; and (4) initiation of full mechanical ventilation. Respiratory support categories were mutually exclusive: children requiring multiple types of increased respiratory support were classified for analysis by the most invasive form of respiratory support used (eg, a child requiring increase in both oxygen flow and pressure settings was categorized as an increase in pressure settings). Children who received heated high-flow nasal cannula therapy as maximum support were categorized as initiation or increase in oxygen support.

Time to return to respiratory baseline was defined as the difference in days between date of return to respiratory baseline and date of admission. Time to return to respiratory baseline was determined only for children who were discharged at respiratory baseline.

Primary Exposure and Outcome Measures

The primary exposure was hospital discharge before return to respiratory baseline (ie, discharge on higher respiratory support than at baseline settings). At our institution, standardized discharge criteria for children with NI do not exist. The primary outcome was all-cause, 30-day hospital reutilization, including hospital readmissions and emergency department (ED) revisits. Secondary outcomes included 30-day reutilization for ARI and hospital length of stay (LOS) in days.

Patient Demographics and Clinical Characteristics

Demographic and patient characteristics that might influence hospital discharge before return to respiratory baseline or readmission were obtained from PHIS (eg, demographic information, age, insurance type, measures of clinical complexity, illness severity) and by EMR review (eg, baseline respiratory support needs, maximum respiratory support during hospitalization). Measures of clinical complexity included comorbid complex chronic conditions (CCCs)^11-14 and technology dependence^14-16 using previously defined diagnostic codes. Measures of illness severity included sepsis¹⁷ and ICU-level care. At our institution, children with baseline ventilator use do not require admission to the ICU unless they are clinically unstable.

Statistical Analysis

Continuous variables were described using medians and interquartile ranges (IQR). Categorical variables were described using counts and percentages. Patient characteristics and outcomes were stratified by primary exposure and compared using chi-square test or Fisher exact test for categorical variables and Wilcoxon rank sum test for continuous variables.

To examine the independent association between discharge before return to respiratory baseline and hospital reutilization, a generalized estimating equation was used that included potential confounders while accounting for within-patient clustering. Patient demographics included age, race, ethnicity, and insurance type; measures of clinical complexity included number of CCCs, technology dependence, and baseline respiratory support; and measures of acute illness severity included ARI diagnosis, degree of increase in respiratory support during hospitalization, and ICU-level care. LOS was also included in the model as a covariate because of its expected association with both exposure and outcome.

Secondary analyses were conducted using the outcome of 30-day reutilization for ARI. Subgroup analysis excluding hospitalizations of children lost to follow-up (ie, no encounters in the 6 months after hospital discharge) was also conducted. All analyses were performed with SAS v9.3 (SAS Institute, Cary, North Carolina). P values < .05 were considered statistically significant. This study was approved by the Institutional Review Board.

Study Cohort

A total of 632 hospitalizations experienced by 366 children with NI who were admitted with ARI were included (Appendix Figure). Most children (66.4%) in the cohort experienced only one hospitalization, 17.5% had two hospitalizations, 7.9% had three hospitalizations, and 8.2% had four or more hospitalizations. The median age at hospitalization was 5.0 years (IQR 2.8-10.5) and most hospitalizations were for children who were male (56.6%), white (78.3%), non-Hispanic (96.0%), and publicly insured (51.7%; Table 1). More than one-quarter (28.6%) of hospitalizations were for children with four or more CCCs, and in 73.4% of hospitalizations, children were technology dependent (Table 1). Baseline respiratory support was common (46.8%), including home mechanical ventilation in 11.1% of hospitalizations (Table 1). Bacterial pneumonia, including aspiration pneumonia, was the most common discharge diagnosis (50.5%, Table 1).

Demographic and Clinical Characteristics

Children were discharged before return to respiratory baseline in 30.4% of hospitalizations (Appendix Figure). Children discharged before return to respiratory baseline were older (median age 5.7 years, IQR 3.1-11.0, vs 4.9 years, IQR 2.6-9.7; P = .04) and more likely to be privately insured (54.2% vs 43.4%; P = .04), compared with children discharged at respiratory baseline (Table 1). Children discharged before return to respiratory baseline were also more likely to have a respiratory CCC (59.9% vs 30.9%; P < .001), have a respiratory technology dependence diagnosis code (44.8% vs 24.1%; P < .001), and have baseline respiratory support needs on EMR review (67.7% vs 37.7%; P < .001), compared with children discharged at baseline (Table 1).

Children discharged before return to respiratory baseline required significantly greater escalation in respiratory support during hospitalization, compared with children discharged at respiratory baseline, including higher rates of initiation of CPAP or BiPAP, increased pressure settings from baseline (for home CPAP, BiPAP, or ventilator users), and initiation of full mechanical ventilation (Table 1). Hospitalizations in which children were discharged before return to respiratory baseline were also more likely to include ICU care than were those for children discharged at baseline (52.1% vs 35.2%; P < .001; Table 1).

Clinical Outcomes and Utilization

Reutilization within 30 days occurred after 32.1% of hospitalizations, with 26.1% requiring hospital readmission and 6.0% requiring ED revisit (Table 2). There was no statistical association in either unadjusted (Table 2) or adjusted (Table 3) analysis between children discharged before return to respiratory baseline and 30-day all-cause hospital reutilizations, hospital readmissions, or ED revisits.

In analysis of secondary outcomes, 30-day reutilization because of ARI occurred after 21.5% of hospitalizations, with 19.0% requiring hospital readmission and 2.5% requiring ED revisit. Median hospital LOS for the cohort was 4 days (IQR 2-8; Table 2). Hospitalizations in which children were discharged before return to respiratory baseline were longer than in those discharged at baseline (median 6 days, IQR 3-11, vs 4 days, IQR 2-7; P < .001; Table 2).

For hospitalizations of children discharged at respiratory baseline, the median time to return to respiratory baseline was 3 days (IQR 1-5, complete range 0-80). In these encounters, discharge occurred soon after return to respiratory baseline (median 1 day, IQR 0-1.5, complete range 0-54).

In subgroup analysis excluding the 18 hospitalizations in which children were lost to follow-up (2.8% of the total cohort), discharge before return to respiratory baseline was not associated with 30-day all-cause hospital reutilization (Table 4).

In this retrospective cohort study, children with NI hospitalized with ARI were frequently discharged using increased respiratory support from baseline. However, those discharged before return to respiratory baseline, despite their greater clinical complexity and acute illness severity, did not have increased hospital reutilization, compared with children discharged at respiratory baseline. Our findings suggest that discharge before return to baseline respiratory support after ARI may be clinically appropriate in some children with NI.

With the growing emphasis on decreasing hospital costs, concern exists that patients are being discharged from hospitals “quicker and sicker,”^18,19 with shortening lengths of stay and higher patient instability at discharge. Clinical instability at discharge has been associated with adverse postdischarge outcomes in adults with pneumonia^20-23; however, studies evaluating discharge readiness have not examined the population of children with NI. Our findings of no difference in reutilization for children with NI discharged before return to respiratory baseline, which would be expected to approximate one or more clinical instabilities, contrast these concerns.

Clinicians caring for children with NI hospitalized with ARI may find it difficult to determine a child’s discharge readiness, in part because many children with NI have longer times to return to respiratory baseline and some never return to their pre-­illness baseline.²⁴ In otherwise healthy children hospitalized with respiratory infections such as pneumonia, discharge criteria typically include complete wean from respiratory support prior to discharge.^4,25 In our study’s more complex children, whose parents already manage respiratory support at home, we hypothesize that discharging providers may be comfortable with discharge when the child has certain types of increased respiratory support compatible with home equipment, a parent skilled with monitoring the child’s respiratory status, and the support of an experienced outpatient provider and home nursing providers. At our institution, outpatient respiratory support weans are primarily performed by pediatric pulmonologists and, for isolated weaning of supplemental oxygen or time using support, by parents and outpatient pediatricians.

Another important factor in determining a child’s discharge readiness is the perspective of the child’s parent. Berry et al found that children whose parents believe they are not healthy enough for discharge are more likely to experience unplanned hospital readmissions,²⁴ signaling the role of child- and family-­specific factors in safe discharge decisions. Therefore, parents of children with NI should be proactively involved throughout the multidisciplinary discharge process,^26,27 including the decision to discharge before return to respiratory baseline. Parents have identified ongoing provider support, opportunities to practice home care skills, and written instructions with contingency plans as important components of discharge readiness.²⁸ Further work to create partnerships with these highly skilled caregivers in discharge decision making and transition planning are needed to promote safe discharge practices in this complex population.

In our study, children discharged before return to respiratory baseline were more likely to be older and privately insured compared with children discharged at respiratory baseline. Prior studies have found that social factors including low socioeconomic status influence ED provider admissions decisions for children with pneumonia.^29,30 However, the role of socioeconomic factors in provider discharge decisions for children with NI has not been assessed. These traits may also be proxy markers of other sociodemographic factors, such as parent education level, financial hardship influencing ability to participate in a child’s care at the bedside, access to comprehensive outpatient primary care, and availability of private home nursing. We hypothesize that these related characteristics directly and indirectly influence provider discharge decisions.

Discharging providers are likely more comfortable with discharge prior to return to respiratory baseline when the family has private duty nursing in the home. Home nurses can assist families in providing increased respiratory support and recognizing respiratory problems that may arise following discharge. However, home nursing shortages are common nationwide.^31,32 Low-income children, children with respiratory technology use, and children without Medicaid have been found to have larger gaps in home nursing availability.^31,32 Further studies are needed to understand the role of home nursing availability in provider discharge decisions in this population.

This study has several limitations. The retrospective design of this study creates the potential for residual confounding; there may be other clinical or demographic factors influencing hospital discharge decisions that we are unable to capture using EMR review, including parental knowledge and comfort managing illness, quality of discharge instructions, frequency of follow-up visits, and presence of skilled home nursing services. Categorization of children based on respiratory support status at discharge lends potential for misclassification of exposure; however, our substantial interrater reliability suggests that misclassification bias is small. This study’s primary finding indicated no difference between exposure groups; although we may be unable to detect small differences, we had sufficient power with our sample size to detect meaningful differences in reutilization outcomes.

This study was not designed to capture outpatient time to return to respiratory baseline; prospective studies are needed to identify rates of return to respiratory baseline following ARI in children with NI. We did not measure the level of respiratory support used by children at the time of discharge and, therefore, are unable to estimate the amount of respiratory support weaning needed following discharge or the compatibility of support with home equipment using our data. In addition, this study focused on respiratory support modalities and, thus, did not measure inpatient utilization of mucociliary clearance technologies that might be hypothesized to decrease the time to return to baseline respiratory support. Next steps in evaluating treatment of ARI include investigating the effect of mucociliary clearance on both exposure and outcome in this population.

There may be other clinically meaningful outcomes for this population apart from reutilization that we have not assessed in this study, including increased respiratory support required following discharge, primary care reutilization, healthcare costs, or parent satisfaction with timing of and outcomes after discharge. Finally, although our hospital has reutilization rates for children with NI that are similar to other institutions in the United States,³³ our results may not be generalizable to children with NI hospitalized at other institutions because local discharge processes and systems of care may be different. Prospective, multicenter investigation is needed to evaluate the clinical consequences of discharge before return to respiratory baseline more broadly.

At our institution, approximately one-quarter of children with NI hospitalized with ARI were discharged before return to respiratory baseline, but these children were not at increased risk of reutilization, compared with children discharged at respiratory baseline. Our findings suggest that return to baseline respiratory support might not be a necessary component of hospital discharge criteria. In otherwise clinically stable children with NI, discharge before return to respiratory baseline may be reasonable if their parents are comfortable managing respiratory support at home.

The authors thank Jonathan Rodean of the Children’s Hospital Association for his assistance with abstraction of PHIS data.

Children with neurologic impairment (NI; eg, hypoxic-­ischemic encephalopathy, muscular dystrophy) are characterized by functional and/or intellectual impairments resulting from a variety of neurologic diseases.¹ These children commonly have respiratory comorbidities, including central hypoventilation, impaired cough, and oromotor dysfunction, that may lead to chronic respiratory insufficiency and a need for chronic respiratory support at baseline.^2,3 Baseline respiratory support modalities can include supplemental oxygen, noninvasive positive pressure ventilation, or invasive mechanical ventilation.

Acute respiratory infections (ARI; eg, pneumonia, bronchiolitis) are the most common cause of hospitalization, intensive care unit (ICU) admission, and death for children with NI.^1,3 Discharge criteria for otherwise healthy children admitted to the hospital with ARI often include return to respiratory baseline.⁴ Children with complex chronic conditions have longer hospitalizations when hospitalized with respiratory infections,^5-7 because, in part, comorbidities and complications prolong the time to return to baseline. This prolonged return to respiratory baseline in combination with family knowledge, comfort, and skill in managing respiratory support and other complexities at home may alter discharge practices in the population of children with NI. In our clinical experience, discharge before return to baseline respiratory support occurs more frequently in children with NI than in otherwise healthy children when hospitalized with ARI. However, the consequences of discharging children with NI prior to return to respiratory baseline are unknown.

In this study, we sought to determine if discharge prior to return to baseline respiratory support is associated with reutilization among children with NI hospitalized with ARI. We hypothesized that patients discharged prior to return to respiratory baseline would have higher rates of 30-day hospital reutilization.

Study Design and Data Source

This single-center, retrospective cohort study of children hospitalized at Cincinnati Children’s Hospital Medical Center (CCHMC) used data from the Pediatric Health Information System (PHIS) and the electronic medical record (EMR). PHIS, an administrative database of 45 not-for-profit, tertiary care, US pediatric hospitals managed by Children’s Hospital Association (Lenexa, Kansas), was used to identify eligible children, examine demographic and clinical variables, and define outcomes. PHIS contains data regarding patient demographics, inpatient resource utilization, and diagnoses. Encrypted medical record numbers in PHIS allowed for local identification of patients’ medical records to complete EMR review to confirm eligibility and obtain detailed patient-level clinical information (eg, respiratory support needs) not available in PHIS.

Pilot medical record reviews allowed for standardized study definitions and procedures. All study staff underwent training with the primary investigator, including detailed review of 10 initial abstractions. Two investigators (K.M. and S.C.) performed repeat abstractions from 40 randomly selected records to enable assessment of interrater reliability. Average reliability, indicated by the κ statistic, indicated substantial to near-perfect reliability⁸ (κ = 0.97, 95% CI 0.90-1.00) for the primary exposure. EMR data were managed using Research Electronic Data Capture (REDCap, Nashville, Tennessee)⁹ and subsequently merged with PHIS data.

Study Population

Hospitalizations of children with NI aged 1 to 18 years at CCHMC between January 2010 and September 2015 were eligible for inclusion if they had a principal discharge diagnosis indicative of ARI and required increased respiratory support from baseline during hospitalization. NI was defined as a high-intensity, chronic neurological diagnosis with substantial functional impairments according to previously defined diagnosis codes.^1,10 ARI was identified using codes in the Clinical Classification Software (Agency for Healthcare Research and Quality, Rockville, MD) respiratory group indicative of ARI (eg, pneumonia, bronchiolitis, influenza; Appendix Table).

Children transferred to CCHMC were excluded because records from their initial illness presentation and management were not available. Because of expected differences in management and outcomes, children with a known diagnosis of tuberculosis or human immunodeficiency virus were excluded. Because exposure criteria were dependent on hospital discharge status, hospitalizations for children who died during admission (4 of 632 hospitalizations, 0.63%) were excluded from the final cohort (Appendix Figure).

Study Definitions

Baseline respiratory support (ie, “respiratory baseline”) was defined as the child’s highest level of respiratory support needed prior to admission when well (ie, no support, supplemental oxygen, continuous positive airway pressure [CPAP] or bilevel positive airway pressure [BiPAP], or ventilator support), and further characterized by night or day/night requirement. Respiratory baseline was identified using EMR documentation of home respiratory support use at the time of index admission. Return to respiratory baseline was defined as the date on which the child achieved documented home respiratory support settings, regardless of clinical symptoms.

Children may have required increased respiratory support from baseline at any time during hospitalization. Maximum respiratory support required was categorized as one of the following: (1) initiation of supplemental oxygen or increase in oxygen flow or duration; (2) initiation of CPAP or BiPAP; (3) increase in pressure settings or duration of pressure support for those with baseline CPAP, BiPAP, or ventilator use; and (4) initiation of full mechanical ventilation. Respiratory support categories were mutually exclusive: children requiring multiple types of increased respiratory support were classified for analysis by the most invasive form of respiratory support used (eg, a child requiring increase in both oxygen flow and pressure settings was categorized as an increase in pressure settings). Children who received heated high-flow nasal cannula therapy as maximum support were categorized as initiation or increase in oxygen support.

Time to return to respiratory baseline was defined as the difference in days between date of return to respiratory baseline and date of admission. Time to return to respiratory baseline was determined only for children who were discharged at respiratory baseline.

Primary Exposure and Outcome Measures

The primary exposure was hospital discharge before return to respiratory baseline (ie, discharge on higher respiratory support than at baseline settings). At our institution, standardized discharge criteria for children with NI do not exist. The primary outcome was all-cause, 30-day hospital reutilization, including hospital readmissions and emergency department (ED) revisits. Secondary outcomes included 30-day reutilization for ARI and hospital length of stay (LOS) in days.

Patient Demographics and Clinical Characteristics

Demographic and patient characteristics that might influence hospital discharge before return to respiratory baseline or readmission were obtained from PHIS (eg, demographic information, age, insurance type, measures of clinical complexity, illness severity) and by EMR review (eg, baseline respiratory support needs, maximum respiratory support during hospitalization). Measures of clinical complexity included comorbid complex chronic conditions (CCCs)^11-14 and technology dependence^14-16 using previously defined diagnostic codes. Measures of illness severity included sepsis¹⁷ and ICU-level care. At our institution, children with baseline ventilator use do not require admission to the ICU unless they are clinically unstable.

Statistical Analysis

Continuous variables were described using medians and interquartile ranges (IQR). Categorical variables were described using counts and percentages. Patient characteristics and outcomes were stratified by primary exposure and compared using chi-square test or Fisher exact test for categorical variables and Wilcoxon rank sum test for continuous variables.

To examine the independent association between discharge before return to respiratory baseline and hospital reutilization, a generalized estimating equation was used that included potential confounders while accounting for within-patient clustering. Patient demographics included age, race, ethnicity, and insurance type; measures of clinical complexity included number of CCCs, technology dependence, and baseline respiratory support; and measures of acute illness severity included ARI diagnosis, degree of increase in respiratory support during hospitalization, and ICU-level care. LOS was also included in the model as a covariate because of its expected association with both exposure and outcome.

Secondary analyses were conducted using the outcome of 30-day reutilization for ARI. Subgroup analysis excluding hospitalizations of children lost to follow-up (ie, no encounters in the 6 months after hospital discharge) was also conducted. All analyses were performed with SAS v9.3 (SAS Institute, Cary, North Carolina). P values < .05 were considered statistically significant. This study was approved by the Institutional Review Board.

Study Cohort

A total of 632 hospitalizations experienced by 366 children with NI who were admitted with ARI were included (Appendix Figure). Most children (66.4%) in the cohort experienced only one hospitalization, 17.5% had two hospitalizations, 7.9% had three hospitalizations, and 8.2% had four or more hospitalizations. The median age at hospitalization was 5.0 years (IQR 2.8-10.5) and most hospitalizations were for children who were male (56.6%), white (78.3%), non-Hispanic (96.0%), and publicly insured (51.7%; Table 1). More than one-quarter (28.6%) of hospitalizations were for children with four or more CCCs, and in 73.4% of hospitalizations, children were technology dependent (Table 1). Baseline respiratory support was common (46.8%), including home mechanical ventilation in 11.1% of hospitalizations (Table 1). Bacterial pneumonia, including aspiration pneumonia, was the most common discharge diagnosis (50.5%, Table 1).

Demographic and Clinical Characteristics

Children were discharged before return to respiratory baseline in 30.4% of hospitalizations (Appendix Figure). Children discharged before return to respiratory baseline were older (median age 5.7 years, IQR 3.1-11.0, vs 4.9 years, IQR 2.6-9.7; P = .04) and more likely to be privately insured (54.2% vs 43.4%; P = .04), compared with children discharged at respiratory baseline (Table 1). Children discharged before return to respiratory baseline were also more likely to have a respiratory CCC (59.9% vs 30.9%; P < .001), have a respiratory technology dependence diagnosis code (44.8% vs 24.1%; P < .001), and have baseline respiratory support needs on EMR review (67.7% vs 37.7%; P < .001), compared with children discharged at baseline (Table 1).

Children discharged before return to respiratory baseline required significantly greater escalation in respiratory support during hospitalization, compared with children discharged at respiratory baseline, including higher rates of initiation of CPAP or BiPAP, increased pressure settings from baseline (for home CPAP, BiPAP, or ventilator users), and initiation of full mechanical ventilation (Table 1). Hospitalizations in which children were discharged before return to respiratory baseline were also more likely to include ICU care than were those for children discharged at baseline (52.1% vs 35.2%; P < .001; Table 1).

Clinical Outcomes and Utilization

Reutilization within 30 days occurred after 32.1% of hospitalizations, with 26.1% requiring hospital readmission and 6.0% requiring ED revisit (Table 2). There was no statistical association in either unadjusted (Table 2) or adjusted (Table 3) analysis between children discharged before return to respiratory baseline and 30-day all-cause hospital reutilizations, hospital readmissions, or ED revisits.

In analysis of secondary outcomes, 30-day reutilization because of ARI occurred after 21.5% of hospitalizations, with 19.0% requiring hospital readmission and 2.5% requiring ED revisit. Median hospital LOS for the cohort was 4 days (IQR 2-8; Table 2). Hospitalizations in which children were discharged before return to respiratory baseline were longer than in those discharged at baseline (median 6 days, IQR 3-11, vs 4 days, IQR 2-7; P < .001; Table 2).

For hospitalizations of children discharged at respiratory baseline, the median time to return to respiratory baseline was 3 days (IQR 1-5, complete range 0-80). In these encounters, discharge occurred soon after return to respiratory baseline (median 1 day, IQR 0-1.5, complete range 0-54).

In subgroup analysis excluding the 18 hospitalizations in which children were lost to follow-up (2.8% of the total cohort), discharge before return to respiratory baseline was not associated with 30-day all-cause hospital reutilization (Table 4).

In this retrospective cohort study, children with NI hospitalized with ARI were frequently discharged using increased respiratory support from baseline. However, those discharged before return to respiratory baseline, despite their greater clinical complexity and acute illness severity, did not have increased hospital reutilization, compared with children discharged at respiratory baseline. Our findings suggest that discharge before return to baseline respiratory support after ARI may be clinically appropriate in some children with NI.

With the growing emphasis on decreasing hospital costs, concern exists that patients are being discharged from hospitals “quicker and sicker,”^18,19 with shortening lengths of stay and higher patient instability at discharge. Clinical instability at discharge has been associated with adverse postdischarge outcomes in adults with pneumonia^20-23; however, studies evaluating discharge readiness have not examined the population of children with NI. Our findings of no difference in reutilization for children with NI discharged before return to respiratory baseline, which would be expected to approximate one or more clinical instabilities, contrast these concerns.

Clinicians caring for children with NI hospitalized with ARI may find it difficult to determine a child’s discharge readiness, in part because many children with NI have longer times to return to respiratory baseline and some never return to their pre-­illness baseline.²⁴ In otherwise healthy children hospitalized with respiratory infections such as pneumonia, discharge criteria typically include complete wean from respiratory support prior to discharge.^4,25 In our study’s more complex children, whose parents already manage respiratory support at home, we hypothesize that discharging providers may be comfortable with discharge when the child has certain types of increased respiratory support compatible with home equipment, a parent skilled with monitoring the child’s respiratory status, and the support of an experienced outpatient provider and home nursing providers. At our institution, outpatient respiratory support weans are primarily performed by pediatric pulmonologists and, for isolated weaning of supplemental oxygen or time using support, by parents and outpatient pediatricians.

Another important factor in determining a child’s discharge readiness is the perspective of the child’s parent. Berry et al found that children whose parents believe they are not healthy enough for discharge are more likely to experience unplanned hospital readmissions,²⁴ signaling the role of child- and family-­specific factors in safe discharge decisions. Therefore, parents of children with NI should be proactively involved throughout the multidisciplinary discharge process,^26,27 including the decision to discharge before return to respiratory baseline. Parents have identified ongoing provider support, opportunities to practice home care skills, and written instructions with contingency plans as important components of discharge readiness.²⁸ Further work to create partnerships with these highly skilled caregivers in discharge decision making and transition planning are needed to promote safe discharge practices in this complex population.

In our study, children discharged before return to respiratory baseline were more likely to be older and privately insured compared with children discharged at respiratory baseline. Prior studies have found that social factors including low socioeconomic status influence ED provider admissions decisions for children with pneumonia.^29,30 However, the role of socioeconomic factors in provider discharge decisions for children with NI has not been assessed. These traits may also be proxy markers of other sociodemographic factors, such as parent education level, financial hardship influencing ability to participate in a child’s care at the bedside, access to comprehensive outpatient primary care, and availability of private home nursing. We hypothesize that these related characteristics directly and indirectly influence provider discharge decisions.

Discharging providers are likely more comfortable with discharge prior to return to respiratory baseline when the family has private duty nursing in the home. Home nurses can assist families in providing increased respiratory support and recognizing respiratory problems that may arise following discharge. However, home nursing shortages are common nationwide.^31,32 Low-income children, children with respiratory technology use, and children without Medicaid have been found to have larger gaps in home nursing availability.^31,32 Further studies are needed to understand the role of home nursing availability in provider discharge decisions in this population.

This study has several limitations. The retrospective design of this study creates the potential for residual confounding; there may be other clinical or demographic factors influencing hospital discharge decisions that we are unable to capture using EMR review, including parental knowledge and comfort managing illness, quality of discharge instructions, frequency of follow-up visits, and presence of skilled home nursing services. Categorization of children based on respiratory support status at discharge lends potential for misclassification of exposure; however, our substantial interrater reliability suggests that misclassification bias is small. This study’s primary finding indicated no difference between exposure groups; although we may be unable to detect small differences, we had sufficient power with our sample size to detect meaningful differences in reutilization outcomes.

This study was not designed to capture outpatient time to return to respiratory baseline; prospective studies are needed to identify rates of return to respiratory baseline following ARI in children with NI. We did not measure the level of respiratory support used by children at the time of discharge and, therefore, are unable to estimate the amount of respiratory support weaning needed following discharge or the compatibility of support with home equipment using our data. In addition, this study focused on respiratory support modalities and, thus, did not measure inpatient utilization of mucociliary clearance technologies that might be hypothesized to decrease the time to return to baseline respiratory support. Next steps in evaluating treatment of ARI include investigating the effect of mucociliary clearance on both exposure and outcome in this population.

There may be other clinically meaningful outcomes for this population apart from reutilization that we have not assessed in this study, including increased respiratory support required following discharge, primary care reutilization, healthcare costs, or parent satisfaction with timing of and outcomes after discharge. Finally, although our hospital has reutilization rates for children with NI that are similar to other institutions in the United States,³³ our results may not be generalizable to children with NI hospitalized at other institutions because local discharge processes and systems of care may be different. Prospective, multicenter investigation is needed to evaluate the clinical consequences of discharge before return to respiratory baseline more broadly.

At our institution, approximately one-quarter of children with NI hospitalized with ARI were discharged before return to respiratory baseline, but these children were not at increased risk of reutilization, compared with children discharged at respiratory baseline. Our findings suggest that return to baseline respiratory support might not be a necessary component of hospital discharge criteria. In otherwise clinically stable children with NI, discharge before return to respiratory baseline may be reasonable if their parents are comfortable managing respiratory support at home.

The authors thank Jonathan Rodean of the Children’s Hospital Association for his assistance with abstraction of PHIS data.

Children with neurologic impairment (NI; eg, hypoxic-­ischemic encephalopathy, muscular dystrophy) are characterized by functional and/or intellectual impairments resulting from a variety of neurologic diseases.¹ These children commonly have respiratory comorbidities, including central hypoventilation, impaired cough, and oromotor dysfunction, that may lead to chronic respiratory insufficiency and a need for chronic respiratory support at baseline.^2,3 Baseline respiratory support modalities can include supplemental oxygen, noninvasive positive pressure ventilation, or invasive mechanical ventilation.

Acute respiratory infections (ARI; eg, pneumonia, bronchiolitis) are the most common cause of hospitalization, intensive care unit (ICU) admission, and death for children with NI.^1,3 Discharge criteria for otherwise healthy children admitted to the hospital with ARI often include return to respiratory baseline.⁴ Children with complex chronic conditions have longer hospitalizations when hospitalized with respiratory infections,^5-7 because, in part, comorbidities and complications prolong the time to return to baseline. This prolonged return to respiratory baseline in combination with family knowledge, comfort, and skill in managing respiratory support and other complexities at home may alter discharge practices in the population of children with NI. In our clinical experience, discharge before return to baseline respiratory support occurs more frequently in children with NI than in otherwise healthy children when hospitalized with ARI. However, the consequences of discharging children with NI prior to return to respiratory baseline are unknown.

In this study, we sought to determine if discharge prior to return to baseline respiratory support is associated with reutilization among children with NI hospitalized with ARI. We hypothesized that patients discharged prior to return to respiratory baseline would have higher rates of 30-day hospital reutilization.

Study Design and Data Source

This single-center, retrospective cohort study of children hospitalized at Cincinnati Children’s Hospital Medical Center (CCHMC) used data from the Pediatric Health Information System (PHIS) and the electronic medical record (EMR). PHIS, an administrative database of 45 not-for-profit, tertiary care, US pediatric hospitals managed by Children’s Hospital Association (Lenexa, Kansas), was used to identify eligible children, examine demographic and clinical variables, and define outcomes. PHIS contains data regarding patient demographics, inpatient resource utilization, and diagnoses. Encrypted medical record numbers in PHIS allowed for local identification of patients’ medical records to complete EMR review to confirm eligibility and obtain detailed patient-level clinical information (eg, respiratory support needs) not available in PHIS.

Pilot medical record reviews allowed for standardized study definitions and procedures. All study staff underwent training with the primary investigator, including detailed review of 10 initial abstractions. Two investigators (K.M. and S.C.) performed repeat abstractions from 40 randomly selected records to enable assessment of interrater reliability. Average reliability, indicated by the κ statistic, indicated substantial to near-perfect reliability⁸ (κ = 0.97, 95% CI 0.90-1.00) for the primary exposure. EMR data were managed using Research Electronic Data Capture (REDCap, Nashville, Tennessee)⁹ and subsequently merged with PHIS data.

Study Population

Hospitalizations of children with NI aged 1 to 18 years at CCHMC between January 2010 and September 2015 were eligible for inclusion if they had a principal discharge diagnosis indicative of ARI and required increased respiratory support from baseline during hospitalization. NI was defined as a high-intensity, chronic neurological diagnosis with substantial functional impairments according to previously defined diagnosis codes.^1,10 ARI was identified using codes in the Clinical Classification Software (Agency for Healthcare Research and Quality, Rockville, MD) respiratory group indicative of ARI (eg, pneumonia, bronchiolitis, influenza; Appendix Table).

Children transferred to CCHMC were excluded because records from their initial illness presentation and management were not available. Because of expected differences in management and outcomes, children with a known diagnosis of tuberculosis or human immunodeficiency virus were excluded. Because exposure criteria were dependent on hospital discharge status, hospitalizations for children who died during admission (4 of 632 hospitalizations, 0.63%) were excluded from the final cohort (Appendix Figure).

Study Definitions

Baseline respiratory support (ie, “respiratory baseline”) was defined as the child’s highest level of respiratory support needed prior to admission when well (ie, no support, supplemental oxygen, continuous positive airway pressure [CPAP] or bilevel positive airway pressure [BiPAP], or ventilator support), and further characterized by night or day/night requirement. Respiratory baseline was identified using EMR documentation of home respiratory support use at the time of index admission. Return to respiratory baseline was defined as the date on which the child achieved documented home respiratory support settings, regardless of clinical symptoms.

Children may have required increased respiratory support from baseline at any time during hospitalization. Maximum respiratory support required was categorized as one of the following: (1) initiation of supplemental oxygen or increase in oxygen flow or duration; (2) initiation of CPAP or BiPAP; (3) increase in pressure settings or duration of pressure support for those with baseline CPAP, BiPAP, or ventilator use; and (4) initiation of full mechanical ventilation. Respiratory support categories were mutually exclusive: children requiring multiple types of increased respiratory support were classified for analysis by the most invasive form of respiratory support used (eg, a child requiring increase in both oxygen flow and pressure settings was categorized as an increase in pressure settings). Children who received heated high-flow nasal cannula therapy as maximum support were categorized as initiation or increase in oxygen support.

Time to return to respiratory baseline was defined as the difference in days between date of return to respiratory baseline and date of admission. Time to return to respiratory baseline was determined only for children who were discharged at respiratory baseline.

Primary Exposure and Outcome Measures

The primary exposure was hospital discharge before return to respiratory baseline (ie, discharge on higher respiratory support than at baseline settings). At our institution, standardized discharge criteria for children with NI do not exist. The primary outcome was all-cause, 30-day hospital reutilization, including hospital readmissions and emergency department (ED) revisits. Secondary outcomes included 30-day reutilization for ARI and hospital length of stay (LOS) in days.

Patient Demographics and Clinical Characteristics

Demographic and patient characteristics that might influence hospital discharge before return to respiratory baseline or readmission were obtained from PHIS (eg, demographic information, age, insurance type, measures of clinical complexity, illness severity) and by EMR review (eg, baseline respiratory support needs, maximum respiratory support during hospitalization). Measures of clinical complexity included comorbid complex chronic conditions (CCCs)^11-14 and technology dependence^14-16 using previously defined diagnostic codes. Measures of illness severity included sepsis¹⁷ and ICU-level care. At our institution, children with baseline ventilator use do not require admission to the ICU unless they are clinically unstable.

Statistical Analysis

Continuous variables were described using medians and interquartile ranges (IQR). Categorical variables were described using counts and percentages. Patient characteristics and outcomes were stratified by primary exposure and compared using chi-square test or Fisher exact test for categorical variables and Wilcoxon rank sum test for continuous variables.

To examine the independent association between discharge before return to respiratory baseline and hospital reutilization, a generalized estimating equation was used that included potential confounders while accounting for within-patient clustering. Patient demographics included age, race, ethnicity, and insurance type; measures of clinical complexity included number of CCCs, technology dependence, and baseline respiratory support; and measures of acute illness severity included ARI diagnosis, degree of increase in respiratory support during hospitalization, and ICU-level care. LOS was also included in the model as a covariate because of its expected association with both exposure and outcome.

Secondary analyses were conducted using the outcome of 30-day reutilization for ARI. Subgroup analysis excluding hospitalizations of children lost to follow-up (ie, no encounters in the 6 months after hospital discharge) was also conducted. All analyses were performed with SAS v9.3 (SAS Institute, Cary, North Carolina). P values < .05 were considered statistically significant. This study was approved by the Institutional Review Board.

Study Cohort

A total of 632 hospitalizations experienced by 366 children with NI who were admitted with ARI were included (Appendix Figure). Most children (66.4%) in the cohort experienced only one hospitalization, 17.5% had two hospitalizations, 7.9% had three hospitalizations, and 8.2% had four or more hospitalizations. The median age at hospitalization was 5.0 years (IQR 2.8-10.5) and most hospitalizations were for children who were male (56.6%), white (78.3%), non-Hispanic (96.0%), and publicly insured (51.7%; Table 1). More than one-quarter (28.6%) of hospitalizations were for children with four or more CCCs, and in 73.4% of hospitalizations, children were technology dependent (Table 1). Baseline respiratory support was common (46.8%), including home mechanical ventilation in 11.1% of hospitalizations (Table 1). Bacterial pneumonia, including aspiration pneumonia, was the most common discharge diagnosis (50.5%, Table 1).

Demographic and Clinical Characteristics

Children were discharged before return to respiratory baseline in 30.4% of hospitalizations (Appendix Figure). Children discharged before return to respiratory baseline were older (median age 5.7 years, IQR 3.1-11.0, vs 4.9 years, IQR 2.6-9.7; P = .04) and more likely to be privately insured (54.2% vs 43.4%; P = .04), compared with children discharged at respiratory baseline (Table 1). Children discharged before return to respiratory baseline were also more likely to have a respiratory CCC (59.9% vs 30.9%; P < .001), have a respiratory technology dependence diagnosis code (44.8% vs 24.1%; P < .001), and have baseline respiratory support needs on EMR review (67.7% vs 37.7%; P < .001), compared with children discharged at baseline (Table 1).

Children discharged before return to respiratory baseline required significantly greater escalation in respiratory support during hospitalization, compared with children discharged at respiratory baseline, including higher rates of initiation of CPAP or BiPAP, increased pressure settings from baseline (for home CPAP, BiPAP, or ventilator users), and initiation of full mechanical ventilation (Table 1). Hospitalizations in which children were discharged before return to respiratory baseline were also more likely to include ICU care than were those for children discharged at baseline (52.1% vs 35.2%; P < .001; Table 1).

Clinical Outcomes and Utilization

Reutilization within 30 days occurred after 32.1% of hospitalizations, with 26.1% requiring hospital readmission and 6.0% requiring ED revisit (Table 2). There was no statistical association in either unadjusted (Table 2) or adjusted (Table 3) analysis between children discharged before return to respiratory baseline and 30-day all-cause hospital reutilizations, hospital readmissions, or ED revisits.

In analysis of secondary outcomes, 30-day reutilization because of ARI occurred after 21.5% of hospitalizations, with 19.0% requiring hospital readmission and 2.5% requiring ED revisit. Median hospital LOS for the cohort was 4 days (IQR 2-8; Table 2). Hospitalizations in which children were discharged before return to respiratory baseline were longer than in those discharged at baseline (median 6 days, IQR 3-11, vs 4 days, IQR 2-7; P < .001; Table 2).

For hospitalizations of children discharged at respiratory baseline, the median time to return to respiratory baseline was 3 days (IQR 1-5, complete range 0-80). In these encounters, discharge occurred soon after return to respiratory baseline (median 1 day, IQR 0-1.5, complete range 0-54).

In subgroup analysis excluding the 18 hospitalizations in which children were lost to follow-up (2.8% of the total cohort), discharge before return to respiratory baseline was not associated with 30-day all-cause hospital reutilization (Table 4).

In this retrospective cohort study, children with NI hospitalized with ARI were frequently discharged using increased respiratory support from baseline. However, those discharged before return to respiratory baseline, despite their greater clinical complexity and acute illness severity, did not have increased hospital reutilization, compared with children discharged at respiratory baseline. Our findings suggest that discharge before return to baseline respiratory support after ARI may be clinically appropriate in some children with NI.

With the growing emphasis on decreasing hospital costs, concern exists that patients are being discharged from hospitals “quicker and sicker,”^18,19 with shortening lengths of stay and higher patient instability at discharge. Clinical instability at discharge has been associated with adverse postdischarge outcomes in adults with pneumonia^20-23; however, studies evaluating discharge readiness have not examined the population of children with NI. Our findings of no difference in reutilization for children with NI discharged before return to respiratory baseline, which would be expected to approximate one or more clinical instabilities, contrast these concerns.

Clinicians caring for children with NI hospitalized with ARI may find it difficult to determine a child’s discharge readiness, in part because many children with NI have longer times to return to respiratory baseline and some never return to their pre-­illness baseline.²⁴ In otherwise healthy children hospitalized with respiratory infections such as pneumonia, discharge criteria typically include complete wean from respiratory support prior to discharge.^4,25 In our study’s more complex children, whose parents already manage respiratory support at home, we hypothesize that discharging providers may be comfortable with discharge when the child has certain types of increased respiratory support compatible with home equipment, a parent skilled with monitoring the child’s respiratory status, and the support of an experienced outpatient provider and home nursing providers. At our institution, outpatient respiratory support weans are primarily performed by pediatric pulmonologists and, for isolated weaning of supplemental oxygen or time using support, by parents and outpatient pediatricians.

Another important factor in determining a child’s discharge readiness is the perspective of the child’s parent. Berry et al found that children whose parents believe they are not healthy enough for discharge are more likely to experience unplanned hospital readmissions,²⁴ signaling the role of child- and family-­specific factors in safe discharge decisions. Therefore, parents of children with NI should be proactively involved throughout the multidisciplinary discharge process,^26,27 including the decision to discharge before return to respiratory baseline. Parents have identified ongoing provider support, opportunities to practice home care skills, and written instructions with contingency plans as important components of discharge readiness.²⁸ Further work to create partnerships with these highly skilled caregivers in discharge decision making and transition planning are needed to promote safe discharge practices in this complex population.

In our study, children discharged before return to respiratory baseline were more likely to be older and privately insured compared with children discharged at respiratory baseline. Prior studies have found that social factors including low socioeconomic status influence ED provider admissions decisions for children with pneumonia.^29,30 However, the role of socioeconomic factors in provider discharge decisions for children with NI has not been assessed. These traits may also be proxy markers of other sociodemographic factors, such as parent education level, financial hardship influencing ability to participate in a child’s care at the bedside, access to comprehensive outpatient primary care, and availability of private home nursing. We hypothesize that these related characteristics directly and indirectly influence provider discharge decisions.

Discharging providers are likely more comfortable with discharge prior to return to respiratory baseline when the family has private duty nursing in the home. Home nurses can assist families in providing increased respiratory support and recognizing respiratory problems that may arise following discharge. However, home nursing shortages are common nationwide.^31,32 Low-income children, children with respiratory technology use, and children without Medicaid have been found to have larger gaps in home nursing availability.^31,32 Further studies are needed to understand the role of home nursing availability in provider discharge decisions in this population.

This study has several limitations. The retrospective design of this study creates the potential for residual confounding; there may be other clinical or demographic factors influencing hospital discharge decisions that we are unable to capture using EMR review, including parental knowledge and comfort managing illness, quality of discharge instructions, frequency of follow-up visits, and presence of skilled home nursing services. Categorization of children based on respiratory support status at discharge lends potential for misclassification of exposure; however, our substantial interrater reliability suggests that misclassification bias is small. This study’s primary finding indicated no difference between exposure groups; although we may be unable to detect small differences, we had sufficient power with our sample size to detect meaningful differences in reutilization outcomes.

This study was not designed to capture outpatient time to return to respiratory baseline; prospective studies are needed to identify rates of return to respiratory baseline following ARI in children with NI. We did not measure the level of respiratory support used by children at the time of discharge and, therefore, are unable to estimate the amount of respiratory support weaning needed following discharge or the compatibility of support with home equipment using our data. In addition, this study focused on respiratory support modalities and, thus, did not measure inpatient utilization of mucociliary clearance technologies that might be hypothesized to decrease the time to return to baseline respiratory support. Next steps in evaluating treatment of ARI include investigating the effect of mucociliary clearance on both exposure and outcome in this population.

There may be other clinically meaningful outcomes for this population apart from reutilization that we have not assessed in this study, including increased respiratory support required following discharge, primary care reutilization, healthcare costs, or parent satisfaction with timing of and outcomes after discharge. Finally, although our hospital has reutilization rates for children with NI that are similar to other institutions in the United States,³³ our results may not be generalizable to children with NI hospitalized at other institutions because local discharge processes and systems of care may be different. Prospective, multicenter investigation is needed to evaluate the clinical consequences of discharge before return to respiratory baseline more broadly.

At our institution, approximately one-quarter of children with NI hospitalized with ARI were discharged before return to respiratory baseline, but these children were not at increased risk of reutilization, compared with children discharged at respiratory baseline. Our findings suggest that return to baseline respiratory support might not be a necessary component of hospital discharge criteria. In otherwise clinically stable children with NI, discharge before return to respiratory baseline may be reasonable if their parents are comfortable managing respiratory support at home.

The authors thank Jonathan Rodean of the Children’s Hospital Association for his assistance with abstraction of PHIS data.

Increasing numbers of children and adolescents are presenting to children’s hospitals with acute mental health crises requiring emergent or inpatient treatment.^1-5 As a result, children’s hospitals are experiencing additional financial challenges because specialty mental health services are often reimbursed at lower rates than other medical services.^6-9 Poor reimbursement has also been cited as a deterrent to the provision of mental health specialty care, including emergency mental health crisis services.¹⁰ The cumulative financial impact of recent trends in the provision of mental health crisis services at children’s hospitals, however, is unknown. We conducted this study to assess children’s hospitals’ costs, reimbursement, and net profits or losses when delivering inpatient mental health care.

We conducted a retrospective cohort study of the Children’s Hospital Association’s Pediatric Health Information System (PHIS) and Revenue Management Program (RMP) databases. PHIS is an administrative and billing database that collects International Classification of Disease, 10th Revision (ICD-10) diagnoses, procedure codes, and hospital charges from encounters at 52 US children’s hospitals. Costs are estimated from charges using hospital-, year-, and department-specific cost-to-charge ratios. The RMP database is an add-on module to the PHIS database that captures reimbursement data submitted quarterly from 17 participating hospitals based on actual reimbursement amounts collected for each encounter.

Among the 17 participating hospitals, we included all medical (ie, not surgical or intensive care) encounters during calendar year 2017 for children older than 6 years. We stratified encounters into three diagnosis types: primary mental health diagnosis,⁵ suicide attempt,¹¹ or other medical hospitalizations. We separated suicide attempts since these encounters often require care for both mental health concerns and medical complications. Eating disorders were excluded because these programs at children’s hospitals primarily focus on medical complications, require complex multispecialty support, have significantly longer hospitalizations and made up a small volume of overall mental health hospitalizations.

We stratified all analyses by inpatient or observation encounter and determined the proportion of encounters and hospital days attributed to primary mental health, suicide attempt, and other medical conditions at each hospital. One of the 17 children’s hospitals does not use observation status billing, so the observation encounters dataset includes 16 hospitals.

We summarized patients’ demographic and clinical characteristics using frequencies and percentages, comparing across diagnosis groups using chi-square tests. We calculated mean cost per day as (total cost) ÷ (total length of stay [LOS]), reimbursement per day as (total reimbursement) ÷ (total LOS) for each hospital and patient group, and margin per day as (reimbursement per day) – (cost per day). We then determined the total margin difference of caring for mental health vs caring for other medical encounters as ([margin per day for mental health] – [margin per day other medical]) × (number of mental health days). Similarly, we calculated the total margin loss for suicide attempts vs other medical encounters. After calculating profits and losses at individual hospitals, we summed total annual profits and losses to calculate cumulative annual differences. We summarized these profits and losses across all hospitals with medians and interquartile ranges (IQR).

This study of deidentified administrative data was approved by the Internal Review Board at Vanderbilt University as non-human subjects research. All statistical analyses were performed using SAS v.9.4 (SAS Institute, Cary, North Carolina), and P values < .05 were considered statistically significant.

Study Population

Across the 17 included children’s hospitals, there were 8,521 (7.6%) mental health encounters, 3,247 (2.9%) suicide attempt encounters, and 99,937 (89.5%) other medical encounters. LOS was significantly longer for mental health hospitalizations than for suicide attempts and for other medical hospitalizations.

Hospital Characteristics

All 17 free-standing children’s hospitals in the study had an inpatient behavioral health/psychiatric consultation service, and 7 of the 17 had an inpatient behavioral health/psychiatric unit. The total number of discharges for mental health, suicide attempt, and other medical conditions per year varied (range, 2,868-13,214) across the hospitals.

Hospital Daily Profits and Losses for Mental Health, Suicide Attempt, and Other Medical Admissions

For inpatient status mental health hospitalizations, the median margin was $376/day (IQR, $23-$618). For inpatient status suicide attempt hospitalizations, the median margin was $685/day (IQR, $3-$1,117), and for other medical hospitalizations the median margin was $603/day (IQR, $240-$991). With regard to observation status admissions, mental health hospitalizations had a median margin of –$453/day (IQR, –$806 to $362), suicide attempts of –$103/day (IQR, –$639 to $264), and other medical conditions of $353/day (IQR, –$616 to $658; Figure).

Hospital Annual Profits and Losses for Mental Health and Suicide Attempt Admissions, Compared With Other Medical Admissions

The Table shows daily and annual profits and losses for inpatient and observation status. The total annual loss across all hospitals for mental health and suicide attempt hospitalizations, compared with other medical hospitalizations, including both inpatient and observation status, was –$26,658,255 when taking both profits and losses into account. For the seven hospitals with net profits for mental health and suicide attempt hospitalizations, compared with other medical hospitalizations, the median net profit for combined inpatient and observation status encounters was $119,361 (IQR, $82,818-$195,543), and the total net profit was $5,872,665. For the 10 hospitals with net losses for mental health and suicide attempt hospitalizations, compared with other medical hospitalizations, the median net loss for combined inpatient and observation status was –$2,169,357 (IQR, –$4,034,085 to –$511,755), and the total net loss was –$27,419,379.

Hospitalizations for mental health disorders and suicide attempts accounted for 10.5% of hospitalizations at 17 US children’s hospitals in 2017. Overall, mental health and suicide attempt hospitalizations had lower financial margins than did other medical hospitalizations, and they accounted for a total margin loss of more than $26 million across 17 hospitals. Seven hospitals generated a profit for mental health and suicide attempt admissions; 10 hospitals reported losses. Only three hospitals generated a higher net profit for mental health admissions than for other medical admissions. More hospitals had net profits for inpatient status mental health and suicide attempt admissions than for observation status mental health and suicide attempt admissions.

For a minority of children’s hospitals, mental health hospitalizations had higher profit margins than for other medical hospitalizations. This raises questions about patient outcomes and the type of care models employed. One potential explanation is that these hospitals have negotiated favorable agreements with payers. Another possibility could be variations in case-mix and payer mix. Certain mental health services, such as crisis response teams, social workers, and child life specialists, may also be funded from nonpayer sources, so estimates may not fully reflect the cost of providing mental health services. A worst-case view is that hospitals with higher profit margins are providing less or poorer care because of lower reimbursement.

Mental health and suicide attempt hospitalizations were associated with smaller margins but counterintuitively generally wider IQRs for cost. This might be related to variation in care models, but our study was not positioned to examine reasons for this variation. The relationship between reimbursement or margins and patient outcomes, as well as specific mechanisms which may drive costs and outcomes, are areas for future research.

Health insurance plays a crucial role in mental health care. In our study, hospitals were more likely to report positive margins from inpatient status mental health hospitalizations rather than from observation status ones. This is unsurprising because payments for observation status are generally lower than for inpatient status.¹² Less is known about what influences billing and payment for inpatient versus observation at individual hospitals, particularly for mental health hospitalizations. In many cases, billing status is not strictly under the hospital’s control and may be determined by payers during or after the hospitalization. Significant variability in the percentage of patients billed as observation status and the impact of lower, often negative, margins for observation mental health encounters, will have a disproportionate effect on some hospitals. Future work could investigate how these differences may influence overall costs and delivery of care.

This study has several limitations that deserve attention. Costs reported are based on cost to charge ratios, which may generate imperfect estimates. Data was limited to 17 freestanding children’s hospitals, and our findings may not generalize to other hospitals. We also compared mental health and suicide attempt hospitalizations with “other medical” hospitalizations. This broad group contains certain medical conditions that may have higher or lower profit margins than average, and estimates of the margins could be over- or underestimated. We assumed that mental health and suicide attempt admissions were displacing admissions with non–mental health medical conditions (ie, not an empty bed). If those beds would otherwise be unoccupied, raw margins are better estimates of the financial impact than margin differences between mental health/suicide attempt and other medical hospitalizations.

Children’s hospitals are more likely to have significantly lower financial margins for mental health and suicide attempt hospitalizations than for other medical hospitalizations. Future work to investigate how quality of care is associated with reimbursement can help ensure that funding for children’s acute mental health care services is commensurate with resources required to provide high quality services.

The authors had no financial relationships relevant to this article to disclose.

Research reported in this publication was supported by the National Institute of Mental Health of the National Institutes of Health under Award Number K23MH115162 (Doupnik).

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Increasing numbers of children and adolescents are presenting to children’s hospitals with acute mental health crises requiring emergent or inpatient treatment.^1-5 As a result, children’s hospitals are experiencing additional financial challenges because specialty mental health services are often reimbursed at lower rates than other medical services.^6-9 Poor reimbursement has also been cited as a deterrent to the provision of mental health specialty care, including emergency mental health crisis services.¹⁰ The cumulative financial impact of recent trends in the provision of mental health crisis services at children’s hospitals, however, is unknown. We conducted this study to assess children’s hospitals’ costs, reimbursement, and net profits or losses when delivering inpatient mental health care.

We conducted a retrospective cohort study of the Children’s Hospital Association’s Pediatric Health Information System (PHIS) and Revenue Management Program (RMP) databases. PHIS is an administrative and billing database that collects International Classification of Disease, 10th Revision (ICD-10) diagnoses, procedure codes, and hospital charges from encounters at 52 US children’s hospitals. Costs are estimated from charges using hospital-, year-, and department-specific cost-to-charge ratios. The RMP database is an add-on module to the PHIS database that captures reimbursement data submitted quarterly from 17 participating hospitals based on actual reimbursement amounts collected for each encounter.

Among the 17 participating hospitals, we included all medical (ie, not surgical or intensive care) encounters during calendar year 2017 for children older than 6 years. We stratified encounters into three diagnosis types: primary mental health diagnosis,⁵ suicide attempt,¹¹ or other medical hospitalizations. We separated suicide attempts since these encounters often require care for both mental health concerns and medical complications. Eating disorders were excluded because these programs at children’s hospitals primarily focus on medical complications, require complex multispecialty support, have significantly longer hospitalizations and made up a small volume of overall mental health hospitalizations.

We stratified all analyses by inpatient or observation encounter and determined the proportion of encounters and hospital days attributed to primary mental health, suicide attempt, and other medical conditions at each hospital. One of the 17 children’s hospitals does not use observation status billing, so the observation encounters dataset includes 16 hospitals.

We summarized patients’ demographic and clinical characteristics using frequencies and percentages, comparing across diagnosis groups using chi-square tests. We calculated mean cost per day as (total cost) ÷ (total length of stay [LOS]), reimbursement per day as (total reimbursement) ÷ (total LOS) for each hospital and patient group, and margin per day as (reimbursement per day) – (cost per day). We then determined the total margin difference of caring for mental health vs caring for other medical encounters as ([margin per day for mental health] – [margin per day other medical]) × (number of mental health days). Similarly, we calculated the total margin loss for suicide attempts vs other medical encounters. After calculating profits and losses at individual hospitals, we summed total annual profits and losses to calculate cumulative annual differences. We summarized these profits and losses across all hospitals with medians and interquartile ranges (IQR).

This study of deidentified administrative data was approved by the Internal Review Board at Vanderbilt University as non-human subjects research. All statistical analyses were performed using SAS v.9.4 (SAS Institute, Cary, North Carolina), and P values < .05 were considered statistically significant.

Study Population

Across the 17 included children’s hospitals, there were 8,521 (7.6%) mental health encounters, 3,247 (2.9%) suicide attempt encounters, and 99,937 (89.5%) other medical encounters. LOS was significantly longer for mental health hospitalizations than for suicide attempts and for other medical hospitalizations.

Hospital Characteristics

All 17 free-standing children’s hospitals in the study had an inpatient behavioral health/psychiatric consultation service, and 7 of the 17 had an inpatient behavioral health/psychiatric unit. The total number of discharges for mental health, suicide attempt, and other medical conditions per year varied (range, 2,868-13,214) across the hospitals.

Hospital Daily Profits and Losses for Mental Health, Suicide Attempt, and Other Medical Admissions

For inpatient status mental health hospitalizations, the median margin was $376/day (IQR, $23-$618). For inpatient status suicide attempt hospitalizations, the median margin was $685/day (IQR, $3-$1,117), and for other medical hospitalizations the median margin was $603/day (IQR, $240-$991). With regard to observation status admissions, mental health hospitalizations had a median margin of –$453/day (IQR, –$806 to $362), suicide attempts of –$103/day (IQR, –$639 to $264), and other medical conditions of $353/day (IQR, –$616 to $658; Figure).

Hospital Annual Profits and Losses for Mental Health and Suicide Attempt Admissions, Compared With Other Medical Admissions

The Table shows daily and annual profits and losses for inpatient and observation status. The total annual loss across all hospitals for mental health and suicide attempt hospitalizations, compared with other medical hospitalizations, including both inpatient and observation status, was –$26,658,255 when taking both profits and losses into account. For the seven hospitals with net profits for mental health and suicide attempt hospitalizations, compared with other medical hospitalizations, the median net profit for combined inpatient and observation status encounters was $119,361 (IQR, $82,818-$195,543), and the total net profit was $5,872,665. For the 10 hospitals with net losses for mental health and suicide attempt hospitalizations, compared with other medical hospitalizations, the median net loss for combined inpatient and observation status was –$2,169,357 (IQR, –$4,034,085 to –$511,755), and the total net loss was –$27,419,379.

Hospitalizations for mental health disorders and suicide attempts accounted for 10.5% of hospitalizations at 17 US children’s hospitals in 2017. Overall, mental health and suicide attempt hospitalizations had lower financial margins than did other medical hospitalizations, and they accounted for a total margin loss of more than $26 million across 17 hospitals. Seven hospitals generated a profit for mental health and suicide attempt admissions; 10 hospitals reported losses. Only three hospitals generated a higher net profit for mental health admissions than for other medical admissions. More hospitals had net profits for inpatient status mental health and suicide attempt admissions than for observation status mental health and suicide attempt admissions.

For a minority of children’s hospitals, mental health hospitalizations had higher profit margins than for other medical hospitalizations. This raises questions about patient outcomes and the type of care models employed. One potential explanation is that these hospitals have negotiated favorable agreements with payers. Another possibility could be variations in case-mix and payer mix. Certain mental health services, such as crisis response teams, social workers, and child life specialists, may also be funded from nonpayer sources, so estimates may not fully reflect the cost of providing mental health services. A worst-case view is that hospitals with higher profit margins are providing less or poorer care because of lower reimbursement.

Mental health and suicide attempt hospitalizations were associated with smaller margins but counterintuitively generally wider IQRs for cost. This might be related to variation in care models, but our study was not positioned to examine reasons for this variation. The relationship between reimbursement or margins and patient outcomes, as well as specific mechanisms which may drive costs and outcomes, are areas for future research.

Health insurance plays a crucial role in mental health care. In our study, hospitals were more likely to report positive margins from inpatient status mental health hospitalizations rather than from observation status ones. This is unsurprising because payments for observation status are generally lower than for inpatient status.¹² Less is known about what influences billing and payment for inpatient versus observation at individual hospitals, particularly for mental health hospitalizations. In many cases, billing status is not strictly under the hospital’s control and may be determined by payers during or after the hospitalization. Significant variability in the percentage of patients billed as observation status and the impact of lower, often negative, margins for observation mental health encounters, will have a disproportionate effect on some hospitals. Future work could investigate how these differences may influence overall costs and delivery of care.

This study has several limitations that deserve attention. Costs reported are based on cost to charge ratios, which may generate imperfect estimates. Data was limited to 17 freestanding children’s hospitals, and our findings may not generalize to other hospitals. We also compared mental health and suicide attempt hospitalizations with “other medical” hospitalizations. This broad group contains certain medical conditions that may have higher or lower profit margins than average, and estimates of the margins could be over- or underestimated. We assumed that mental health and suicide attempt admissions were displacing admissions with non–mental health medical conditions (ie, not an empty bed). If those beds would otherwise be unoccupied, raw margins are better estimates of the financial impact than margin differences between mental health/suicide attempt and other medical hospitalizations.

Children’s hospitals are more likely to have significantly lower financial margins for mental health and suicide attempt hospitalizations than for other medical hospitalizations. Future work to investigate how quality of care is associated with reimbursement can help ensure that funding for children’s acute mental health care services is commensurate with resources required to provide high quality services.

The authors had no financial relationships relevant to this article to disclose.

Research reported in this publication was supported by the National Institute of Mental Health of the National Institutes of Health under Award Number K23MH115162 (Doupnik).

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Increasing numbers of children and adolescents are presenting to children’s hospitals with acute mental health crises requiring emergent or inpatient treatment.^1-5 As a result, children’s hospitals are experiencing additional financial challenges because specialty mental health services are often reimbursed at lower rates than other medical services.^6-9 Poor reimbursement has also been cited as a deterrent to the provision of mental health specialty care, including emergency mental health crisis services.¹⁰ The cumulative financial impact of recent trends in the provision of mental health crisis services at children’s hospitals, however, is unknown. We conducted this study to assess children’s hospitals’ costs, reimbursement, and net profits or losses when delivering inpatient mental health care.

We conducted a retrospective cohort study of the Children’s Hospital Association’s Pediatric Health Information System (PHIS) and Revenue Management Program (RMP) databases. PHIS is an administrative and billing database that collects International Classification of Disease, 10th Revision (ICD-10) diagnoses, procedure codes, and hospital charges from encounters at 52 US children’s hospitals. Costs are estimated from charges using hospital-, year-, and department-specific cost-to-charge ratios. The RMP database is an add-on module to the PHIS database that captures reimbursement data submitted quarterly from 17 participating hospitals based on actual reimbursement amounts collected for each encounter.

Among the 17 participating hospitals, we included all medical (ie, not surgical or intensive care) encounters during calendar year 2017 for children older than 6 years. We stratified encounters into three diagnosis types: primary mental health diagnosis,⁵ suicide attempt,¹¹ or other medical hospitalizations. We separated suicide attempts since these encounters often require care for both mental health concerns and medical complications. Eating disorders were excluded because these programs at children’s hospitals primarily focus on medical complications, require complex multispecialty support, have significantly longer hospitalizations and made up a small volume of overall mental health hospitalizations.

We stratified all analyses by inpatient or observation encounter and determined the proportion of encounters and hospital days attributed to primary mental health, suicide attempt, and other medical conditions at each hospital. One of the 17 children’s hospitals does not use observation status billing, so the observation encounters dataset includes 16 hospitals.

We summarized patients’ demographic and clinical characteristics using frequencies and percentages, comparing across diagnosis groups using chi-square tests. We calculated mean cost per day as (total cost) ÷ (total length of stay [LOS]), reimbursement per day as (total reimbursement) ÷ (total LOS) for each hospital and patient group, and margin per day as (reimbursement per day) – (cost per day). We then determined the total margin difference of caring for mental health vs caring for other medical encounters as ([margin per day for mental health] – [margin per day other medical]) × (number of mental health days). Similarly, we calculated the total margin loss for suicide attempts vs other medical encounters. After calculating profits and losses at individual hospitals, we summed total annual profits and losses to calculate cumulative annual differences. We summarized these profits and losses across all hospitals with medians and interquartile ranges (IQR).

This study of deidentified administrative data was approved by the Internal Review Board at Vanderbilt University as non-human subjects research. All statistical analyses were performed using SAS v.9.4 (SAS Institute, Cary, North Carolina), and P values < .05 were considered statistically significant.

Study Population

Across the 17 included children’s hospitals, there were 8,521 (7.6%) mental health encounters, 3,247 (2.9%) suicide attempt encounters, and 99,937 (89.5%) other medical encounters. LOS was significantly longer for mental health hospitalizations than for suicide attempts and for other medical hospitalizations.

Hospital Characteristics

All 17 free-standing children’s hospitals in the study had an inpatient behavioral health/psychiatric consultation service, and 7 of the 17 had an inpatient behavioral health/psychiatric unit. The total number of discharges for mental health, suicide attempt, and other medical conditions per year varied (range, 2,868-13,214) across the hospitals.

Hospital Daily Profits and Losses for Mental Health, Suicide Attempt, and Other Medical Admissions

For inpatient status mental health hospitalizations, the median margin was $376/day (IQR, $23-$618). For inpatient status suicide attempt hospitalizations, the median margin was $685/day (IQR, $3-$1,117), and for other medical hospitalizations the median margin was $603/day (IQR, $240-$991). With regard to observation status admissions, mental health hospitalizations had a median margin of –$453/day (IQR, –$806 to $362), suicide attempts of –$103/day (IQR, –$639 to $264), and other medical conditions of $353/day (IQR, –$616 to $658; Figure).

Hospital Annual Profits and Losses for Mental Health and Suicide Attempt Admissions, Compared With Other Medical Admissions

The Table shows daily and annual profits and losses for inpatient and observation status. The total annual loss across all hospitals for mental health and suicide attempt hospitalizations, compared with other medical hospitalizations, including both inpatient and observation status, was –$26,658,255 when taking both profits and losses into account. For the seven hospitals with net profits for mental health and suicide attempt hospitalizations, compared with other medical hospitalizations, the median net profit for combined inpatient and observation status encounters was $119,361 (IQR, $82,818-$195,543), and the total net profit was $5,872,665. For the 10 hospitals with net losses for mental health and suicide attempt hospitalizations, compared with other medical hospitalizations, the median net loss for combined inpatient and observation status was –$2,169,357 (IQR, –$4,034,085 to –$511,755), and the total net loss was –$27,419,379.

Hospitalizations for mental health disorders and suicide attempts accounted for 10.5% of hospitalizations at 17 US children’s hospitals in 2017. Overall, mental health and suicide attempt hospitalizations had lower financial margins than did other medical hospitalizations, and they accounted for a total margin loss of more than $26 million across 17 hospitals. Seven hospitals generated a profit for mental health and suicide attempt admissions; 10 hospitals reported losses. Only three hospitals generated a higher net profit for mental health admissions than for other medical admissions. More hospitals had net profits for inpatient status mental health and suicide attempt admissions than for observation status mental health and suicide attempt admissions.

For a minority of children’s hospitals, mental health hospitalizations had higher profit margins than for other medical hospitalizations. This raises questions about patient outcomes and the type of care models employed. One potential explanation is that these hospitals have negotiated favorable agreements with payers. Another possibility could be variations in case-mix and payer mix. Certain mental health services, such as crisis response teams, social workers, and child life specialists, may also be funded from nonpayer sources, so estimates may not fully reflect the cost of providing mental health services. A worst-case view is that hospitals with higher profit margins are providing less or poorer care because of lower reimbursement.

Mental health and suicide attempt hospitalizations were associated with smaller margins but counterintuitively generally wider IQRs for cost. This might be related to variation in care models, but our study was not positioned to examine reasons for this variation. The relationship between reimbursement or margins and patient outcomes, as well as specific mechanisms which may drive costs and outcomes, are areas for future research.

Health insurance plays a crucial role in mental health care. In our study, hospitals were more likely to report positive margins from inpatient status mental health hospitalizations rather than from observation status ones. This is unsurprising because payments for observation status are generally lower than for inpatient status.¹² Less is known about what influences billing and payment for inpatient versus observation at individual hospitals, particularly for mental health hospitalizations. In many cases, billing status is not strictly under the hospital’s control and may be determined by payers during or after the hospitalization. Significant variability in the percentage of patients billed as observation status and the impact of lower, often negative, margins for observation mental health encounters, will have a disproportionate effect on some hospitals. Future work could investigate how these differences may influence overall costs and delivery of care.

This study has several limitations that deserve attention. Costs reported are based on cost to charge ratios, which may generate imperfect estimates. Data was limited to 17 freestanding children’s hospitals, and our findings may not generalize to other hospitals. We also compared mental health and suicide attempt hospitalizations with “other medical” hospitalizations. This broad group contains certain medical conditions that may have higher or lower profit margins than average, and estimates of the margins could be over- or underestimated. We assumed that mental health and suicide attempt admissions were displacing admissions with non–mental health medical conditions (ie, not an empty bed). If those beds would otherwise be unoccupied, raw margins are better estimates of the financial impact than margin differences between mental health/suicide attempt and other medical hospitalizations.

Children’s hospitals are more likely to have significantly lower financial margins for mental health and suicide attempt hospitalizations than for other medical hospitalizations. Future work to investigate how quality of care is associated with reimbursement can help ensure that funding for children’s acute mental health care services is commensurate with resources required to provide high quality services.

The authors had no financial relationships relevant to this article to disclose.

Research reported in this publication was supported by the National Institute of Mental Health of the National Institutes of Health under Award Number K23MH115162 (Doupnik).

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Adverse Childhood Experiences, or ACEs, include exposure to abuse, neglect, or household dysfunction (eg, having a parent who is mentally ill) as a child.¹ Exposure to ACEs affects health into adulthood, with a dose-response relationship between ACEs and a range of comorbidities.¹ Adults with 6 or more ACEs have a 20-year shorter life expectancy than do those with no ACEs.¹ Still, ACEs are static; once experienced, that experience cannot be undone. However, resilience, or positive adaptation in the context of adversity, can be protective, buffering the negative effects of ACEs.^2,3 Protective factors that promote resilience include social capital, such as positive relationships with caregivers and peers.³

With their clear link to health outcomes across the life-course, there is a movement for pediatricians to screen children for ACEs⁴ and to develop strategies that promote resilience in children, parents, and families. However, screening a child for adversity has challenges because younger children may not have experienced an adverse exposure, or they may be unable to voice their experiences. Studies have demonstrated that parental adversity, or ACEs, may be a marker for childhood adversity.^5,6 Biological models also support this potential intergenerational effect of ACEs. Chronic exposure to stress, including ACEs, results in elevated cortisol via a dysregulated hypothalamic-pituitary-adrenal axis, which results in chronic inflammation.⁷ This “toxic stress” is prolonged, severe in intensity, and can lead to epigenetic changes that may be passed on to the next generation.^8,9

Hospitalization of an ill child, and the transition to home after that hospitalization, is a stressful event for children and families.¹⁰ This stress may be relevant to parents that have a history of a high rate of ACEs or a current low degree of resilience. Our previous work demonstrated that, in the inpatient setting, parents with high ACEs (≥4) or low resilience have increased coping difficulty 14 days after their child’s hospital discharge.¹¹ Our objective here was to evaluate whether a parent’s ACEs and/or resilience would also be associated with that child’s likelihood of reutilization. We hypothesized that more parental ACEs and/or lower parental resilience would be associated with revisits the emergency room, urgent care, or hospital readmissions.

Participants and Study Design

We conducted a prospective cohort study of parents of hospitalized children recruited from the “Hospital-to-Home Outcomes” Studies (H2O I and H2O II).^12,13 H2O I and II were prospective, single-center, randomized controlled trials designed to determine the effectiveness of either a nurse-led transitional home visit (H2O I) or telephone call (H2O II) on 30-day unplanned healthcare reutilization. The trials and this study were approved by the Cincinnati Children’s Institutional Review Board. All parents provided written informed consent.

Details of H2O I and II recruitment and design have been described previously.^12,13 Briefly, children were eligible for inclusion in either study if they were admitted to our institution’s general Hospital Medicine or the Hospital Medicine Complex Care Services; for H2O I, children hospitalized on the Neurology and Neurosurgery services were also eligible.^12,13 Patients were excluded if they were discharged to a residential facility, if they lived outside the home healthcare nurse service area, if they were eligible for skilled home healthcare services (eg, intravenous antibiotics), or if the participating caregiver was non-English speaking.^12,13 In H2O I, families were randomized either to receive a single nurse home visit within 96 hours of discharge or standard of care. In H2O II, families enrolled were randomized to receive a telephone call by a nurse within 96 hours of discharge or standard of care. As we have previously published, randomization in both trials successfully balanced the intervention and control arms with respect to key demographic characteristics.^12,13 For the analyses presented here, we focused on a subset of caregivers 18 years and older whose children were enrolled in either H2O I or II between August 2015 and October 2016. In both H2O trials, face-to-face and paper-based questionnaires were completed by parents during the index hospitalization.

Outcome and Predictors

Our primary outcome was unanticipated healthcare reutilization defined as return to the emergency room, urgent care, or unplanned readmission within 30 days of hospital discharge, consistent with the H2O trials. This was measured using the primary institution’s administrative data supplemented by a utilization database shared across regional hospitals.¹⁴ Readmissions were identified as “unplanned” using a previously validated algorithm,¹⁵ and treated as a dichotomous yes/no variable.

Our primary predictors were parental ACEs and resilience (see Appendix Tables). The ACE questionnaire addresses abuse, neglect, and household dysfunction in the first 18 years of life.¹ It is composed of 10 questions, each with a yes/no response.¹ We defined parents as low (ACE 0), moderate (ACE 1-3), or high (ACE ≥4) risk a priori because previous literature has described poor outcomes in adults with 4 or more ACEs.¹⁶

Given the sensitive nature of the questions, respondents independently completed the ACE questionnaire on paper instead of via the face-to-face survey. Respondents returned the completed questionnaire to the research assistant in a sealed envelope. All families received educational information on relevant hospital and community-based resources (eg, social work).

Parental resilience was measured using the Brief Resilience Scale (BRS). The BRS is 6 items, each on a 5-point Likert scale. Responses were averaged, providing a total score of 1-5; higher scores are representative of higher resilience.¹⁷ We treated the BRS score as a continuous variable. BRS has been used in clinical settings; it has demonstrated positive correlation with social support and negative correlation with fatigue.¹⁷ Parents answered BRS questions during the index pediatric hospitalization in a face-to-face interview.

Parent and Child Characteristics

Parent and child sociodemographic variables were also obtained during the face-to-face interview. Parental variables included age, gender, educational attainment, household income, employment status, and financial and social strain.¹¹ Educational attainment was analyzed in 2 categories—high school or less vs more than high school—because most discharge instructions are written at a high school reading level.¹⁸ Parents reported their annual household income in the following categories: <$15,000; $15,000-$29,999; $30,000-$44,999; $45,000-$59,999; $60,000-$89,999; $90,000-$119,999; ≥$120,000. Employment was dichotomized as not employed/student vs any employment. Financial and social strain were assessed using a series of 9 previously described questions.¹⁹ These questions assessed, via self-report, a family’s ability to make ends meet, ability to pay rent/mortgage or utilities, need to move in with others because of financial reasons, and ability to borrow money if needed, as well as home ownership and parental marital status.^15,19 Strain questions were all dichotomous (yes/no, single/not single). A composite variable was then constructed that categorized those reporting no strain items, 1 to 2 items, 3 to 4 items, and 5 or more items.²⁰

Child variables included race, ethnicity, age, primary care access,²¹ payer, and H2O treatment arm. Race categories were white/Caucasian, black/African American, American Indian or Alaskan Native, Asian or Pacific Islander, and other; ethnicity categories were Hispanic/Latino, non-Hispanic/Latino, and unknown. Given relatively low numbers of children reported to be Hispanic/Latino, we combined race and ethnicity into a single variable, categorized as non-Hispanic/white, non-Hispanic/black, and multiracial/Hispanic/other. Primary care access was assessed using the access subscale to the Parent’s Perception of Primary Care questionnaire. This includes assessment of a family’s ability to travel to their doctor, to see their doctor for routine or sick care, and to get help or advice on evenings or weekends. Scores were categorized as always adequate, almost always adequate, or sometimes/never adequate.²¹ Payer was dichotomized to private or public/self-pay.

Statistical Analyses

We examined the distribution of outcomes, predictors, and covariates. We compared sociodemographic characteristics of those respondents and nonrespondents to the ACE screen using the chi-square test for categorical variables or the t test for continuous variables. We used logistic regression to assess for associations between the independent variables of interest and reutilization, adjusting for potential confounders. To build our adjusted, multivariable model, we decided a priori to include child race/ethnicity, primary care access, financial and social strain, and trial treatment arm. We treated the H2O I control group as the referent group. Other covariates considered for inclusion were caregiver education, household income, employment, and payer. These were included in multivariable models if bivariate associations were significant at the P < .1 level. We assessed an ACE-by-resilience interaction term because we hypothesized that those with more ACEs and lower resilience may have more reutilization outcomes than parents with fewer ACEs and higher resilience. We also evaluated interaction terms between trial arm assignment and predictors to assess effects that may be introduced by the randomization. Predictors in the final logistic regression model were significant at the P < .05 level. Logistic regression assumption of little or no multicollinearity among the independent variables was verified in the final models. All analyses were performed with Stata v16 (Stata Corp, College Station, Texas).

There were a total of 1,787 parent-child dyads enrolled in the H2O I and II during the study period; 1,320 parents (74%) completed the ACE questionnaire and were included the analysis. Included parents were primarily female and employed, as well as educated beyond high school (Table 1). Overall, 64% reported one or more ACEs (range 0 to 9); 45% reported 1to 3, and 19% reported 4 or more ACEs. The most commonly reported ACEs were divorce (n = 573, 43%), exposure to alcoholism (n = 306, 23%), and exposure to mental illness (n = 281, 21%; Figure 1). Parents had a mean BRS score of 3.97 (range 1.17-5.00), with the distribution shown in Figure 2.

Of the 1,320 included patients, the average length of stay was 2.5 days, and 82% of hospitalizations were caused by acute medical issues (eg, bronchiolitis). A total of 211 children experienced a reutilization event within 30 days of discharge. In bivariate analysis, children with parents with 4 or more ACEs had a 2.02-times (95% CI 1.35-3.02) higher odds of experiencing a reutilization event than did those with parents reporting no ACEs. Parents with higher resilience scores had children with a lower odds of reutilization (odds ratio [OR] 0.77 95% CI 0.63-0.95).

In addition to our a priori variables, parental education, employment, and insurance met our significance threshold for inclusion in the multivariable model. The ACE-by-resilience interaction term was not significant and not included in the model. Similarly, there was no significant interaction between ACE and resilience and H2O treatment arm; the interaction terms were not included in the final adjusted model, but treatment arm assignment was kept as a covariate. A total of 1,292 children, out of the 1,320 respondents, remained in the final multivariable model; the excluded 28 had incomplete covariate data but were not otherwise different. In this final adjusted model, children with parents reporting 4 or more ACEs had a 1.69-times (95% CI 1.11-2.60) greater odds of reutilization than did those with parents reporting no ACEs (Table 2). Resilience failed to reach statistical significance in the adjusted model (OR 0.86, 95% CI 0.70-1.07).

We found that high-risk parents (4 or more ACEs) had children with an increased odds of healthcare reutilization, suggesting intergenerational effects of ACEs. We did not find a similar effect relating to parental resilience. We also did not find an interaction between parental ACEs and resilience, suggesting that a parent’s reported degree of resilience does not modify the effect of ACEs on reutilization risk.

Parental adversity may be a risk factor for a child’s unanticipated reutilization. We previously demonstrated that parents with 4 or more ACEs have more coping difficulty than a parent with no ACEs after a child’s hospitalization.¹¹ It is possible that parents with high adversity may have poorer coping mechanisms when dealing with a stressful situation, such as a child’s hospitalization. This may have resulted in inequitable outcomes (eg, increased reutilization) for their children. Other studies have confirmed such an intergenerational effect of adversity, linking a parent’s ACEs with poor developmental, behavioral, and health outcomes in their children.^6,22,23 O’Malley et al showed an association of parental ACEs to current adversities,²⁴ such as insurance or housing concerns, that affect the entirety of the household, including children. In short, it appears that parental ACEs may be a compelling predictor of current childhood adversity.

Resilience buffers the negative effects of ACEs; however, we did not find significant associations between resilience and reutilization or an interaction between ACEs and resilience. The factors that may contribute to reutilization are complex. In our previous work, parental resilience was associated with coping difficulty after discharge; but again, did not interact with parental ACEs.¹¹ Here, we suggest that while resilience may buffer the negative effects of ACEs, that buffering may not affect the likelihood of reutilization. It is also possible that the BRS tool is of less relevance on how one handles the stress of a child’s hospitalization. While the BRS is one measure of resilience, there are many other relevant constructs to resilience, such as connection to social supports, that also may also contribute to risk of reutilization.²⁵

Reducing the stress of a hospitalization itself and promoting a safe transition from hospital to home is critical to improving child health outcomes. Our data here, and in our previous work, demonstrate that a history of adversity and one’s current coping ability may drive a parent’s response to a child’s hospitalization and affect their capacity to care for that child after hospital discharge.¹¹ Additional in-hospital supports like child life, behavioral health, or pastoral care could reduce the stress of the hospitalization while also building positive coping mechanisms.^26-29 A meta-analysis demonstrated that such coping interventions can help alleviate the stress of a hospitalization.³⁰ Hill et al demonstrated successful stress reduction in parents of hospitalized children using a “Coping Kit for Parents.”³¹ Further studies are warranted to understand which interventions are most effective for children and families and whether they could be more effectively deployed if the inpatient team knew more about parental ACEs.

Screening for parental ACEs could help to identify patients at highest risk for a poor transition to home. Therefore, screening for parental adversity in clinical settings, including inpatient settings, may be relevant and valuable.³² Additionally, by recognizing the high prevalence of ACEs in an inpatient setting, hospitals and healthcare organizations could be motivated to develop and enact trauma-informed approaches. A trauma-informed care approach recognizes the intersection of trauma with health and social problems. With this recognition, care teams can more sensitively address the trauma as they provide relevant services.³³ Trauma-informed care is a secondary public health prevention approach that would help team members identify the prevalence and effects of trauma via screening, recognize the signs of a maladaptive response to stress, and respond by integrating awareness of trauma into practice management.^28,34 Both the National Academy of Medicine and the Agency for Healthcare Research and Quality have called for such a trauma-informed approach in primary care.³⁵ In response, many healthcare organizations have developed trauma-informed practices to better address the needs of the populations they serve. For example, provider training on this approach has led to improved rapport in patient-provider relationships.³⁶

Although ACE awareness is a component of trauma-informed care, there are still limitations of the original ACE questionnaire developed by Felitti et al. The existing tool is not inclusive of all adversities a parent or child may face. Moreover, its focus is on past exposures and experiences and not current health-related social needs (eg, food insecurity) which have known linkages with a range of health outcomes and health disparities.³⁷ Additionally, the original ACE questionnaire was created as a population level tool and not as a screening tool. If used as a screening tool, providers may view the questions as too sensitive to ask, and parents may have difficulty responding to and understanding the relevance to their child’s care. Therefore, we suggest that more evidence is required to understand how to best adapt ACE questions into a screening processes that may be implemented in a medical setting.

More evidence is also needed to determine when and where such screening may be most useful. A primary care provider would be best equipped to screen caregivers for ACEs given their established relationship with parents and patients. Given the potential relevance of such information for inpatient care provision, information could then flow from primary care to the inpatient team. However, because not all patients have established primary care providers and only 4% of pediatricians screen for ACEs,³⁸ it is important for inpatient medical teams to understand their role in identifying and addressing ACEs during hospital stays. Development of a screening tool, with input from all stakeholders—including parents—that is valid and feasible for use in a pediatric inpatient setting would be an important step forward. This tool should be paired with training in how to discuss these topics in a trauma-informed, nonjudgmental, empathic manner. We see this as a way in which providers can more effectively elicit an accurate response while simultaneously educating parents on the relevance of such sensitive topics during an acute hospital stay. We also recommend that screening should always be paired with response capabilities that connect those who screen positive with resources that could help them to navigate the stress experienced during and after a child’s hospitalization. Furthermore, communication with primary care providers about parents that screen positive should be integrated into the transition process.

This work has several limitations. First, our study was a part of randomized controlled trials conducted in one academic setting, which thereby limits generalizability. For example, we limited our cohort to those who were English-speaking patients only. This may bias our results because respondents with limited English proficiency may have different risk profiles than their English-speaking peers. In addition, the administration of the both the ACE and resilience questionnaires occurred during an acutely stressful period, which may influence how a parent responds to these questions. Also, both of the surveys are self-reported by parents, which may be susceptible to memory and response biases. Relatedly, we had a high number of nonrespondents, particularly to the ACE questionnaire. Our results are therefore only relevant to those who chose to respond and cannot be applied to nonrespondents. Further work assessing why one does or does not respond to such sensitive questions is an important area for future inquiry. Lastly, our cohort had limited medical complexity; future studies may consider links between parental ACEs (and resilience) and morbidity experienced by children with medical complexity.

Parents history of adversity is linked to their children’s unanticipated healthcare reutilization after a hospital discharge. Screening for parental stressors during a hospitalization may be an important first step to connecting parents and children to evidence-based interventions capable of mitigating the stress of hospitalization and promoting better, more seamless transitions from hospital to home.

Group Members: The following H2O members are nonauthor contributors: JoAnne Bachus, BSN, RN; Monica Borell, BSN, RN; Lenisa V Chang, MA, PhD; Patricia Crawford, RN; Sarah Ferris, BA; Jennifer Gold, BSN, RN; Judy A Heilman, BSN, RN; Jane C Khoury, PhD; Pierce Kuhnell, MS; Karen Lawley, BSN, RN; Margo Moore, MS, BSN, RN; Lynne O’Donnell, BSN, RN; Sarah Riddle, MD; Susan N Sherman, DPA; Angela M Statile, MD, MEd; Karen P Sullivan, BSN, RN; Heather Tubbs-Cooley, PhD, RN; Susan Wade-Murphy, MSN, RN; and Christine M White, MD, MAT.

The authors also thank David Keller, MD, for his guidance on the study.

The authors have no financial relationships or conflicts of interest relevant to this article to disclose.

Supported by funds from the Academic Pediatric Young Investigator Award (Dr A Shah) and the Patient-Centered Outcomes Research Institute Award (IHS-1306-0081, to Dr K Auger, Dr S Shah, Dr H Sucharew, Dr J Simmons), the National Institutes of Health (1K23AI112916, to Dr AF Beck), and the Agency for Healthcare Research and Quality (1K12HS026393-01, to Dr A Shah, K08-HS024735- 01A1, to Dr K Auger). Dr J Haney received Summer Undergraduate Research Fellowship funding through the Summer Undergraduate Research Fellowship at Cincinnati Children’s Hospital Medical Center.

All statements in this report, including findings and conclusions, are solely those of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute, its Board of Governors, or the Methodology Committee.

Adverse Childhood Experiences, or ACEs, include exposure to abuse, neglect, or household dysfunction (eg, having a parent who is mentally ill) as a child.¹ Exposure to ACEs affects health into adulthood, with a dose-response relationship between ACEs and a range of comorbidities.¹ Adults with 6 or more ACEs have a 20-year shorter life expectancy than do those with no ACEs.¹ Still, ACEs are static; once experienced, that experience cannot be undone. However, resilience, or positive adaptation in the context of adversity, can be protective, buffering the negative effects of ACEs.^2,3 Protective factors that promote resilience include social capital, such as positive relationships with caregivers and peers.³

With their clear link to health outcomes across the life-course, there is a movement for pediatricians to screen children for ACEs⁴ and to develop strategies that promote resilience in children, parents, and families. However, screening a child for adversity has challenges because younger children may not have experienced an adverse exposure, or they may be unable to voice their experiences. Studies have demonstrated that parental adversity, or ACEs, may be a marker for childhood adversity.^5,6 Biological models also support this potential intergenerational effect of ACEs. Chronic exposure to stress, including ACEs, results in elevated cortisol via a dysregulated hypothalamic-pituitary-adrenal axis, which results in chronic inflammation.⁷ This “toxic stress” is prolonged, severe in intensity, and can lead to epigenetic changes that may be passed on to the next generation.^8,9

Hospitalization of an ill child, and the transition to home after that hospitalization, is a stressful event for children and families.¹⁰ This stress may be relevant to parents that have a history of a high rate of ACEs or a current low degree of resilience. Our previous work demonstrated that, in the inpatient setting, parents with high ACEs (≥4) or low resilience have increased coping difficulty 14 days after their child’s hospital discharge.¹¹ Our objective here was to evaluate whether a parent’s ACEs and/or resilience would also be associated with that child’s likelihood of reutilization. We hypothesized that more parental ACEs and/or lower parental resilience would be associated with revisits the emergency room, urgent care, or hospital readmissions.

Participants and Study Design

We conducted a prospective cohort study of parents of hospitalized children recruited from the “Hospital-to-Home Outcomes” Studies (H2O I and H2O II).^12,13 H2O I and II were prospective, single-center, randomized controlled trials designed to determine the effectiveness of either a nurse-led transitional home visit (H2O I) or telephone call (H2O II) on 30-day unplanned healthcare reutilization. The trials and this study were approved by the Cincinnati Children’s Institutional Review Board. All parents provided written informed consent.

Details of H2O I and II recruitment and design have been described previously.^12,13 Briefly, children were eligible for inclusion in either study if they were admitted to our institution’s general Hospital Medicine or the Hospital Medicine Complex Care Services; for H2O I, children hospitalized on the Neurology and Neurosurgery services were also eligible.^12,13 Patients were excluded if they were discharged to a residential facility, if they lived outside the home healthcare nurse service area, if they were eligible for skilled home healthcare services (eg, intravenous antibiotics), or if the participating caregiver was non-English speaking.^12,13 In H2O I, families were randomized either to receive a single nurse home visit within 96 hours of discharge or standard of care. In H2O II, families enrolled were randomized to receive a telephone call by a nurse within 96 hours of discharge or standard of care. As we have previously published, randomization in both trials successfully balanced the intervention and control arms with respect to key demographic characteristics.^12,13 For the analyses presented here, we focused on a subset of caregivers 18 years and older whose children were enrolled in either H2O I or II between August 2015 and October 2016. In both H2O trials, face-to-face and paper-based questionnaires were completed by parents during the index hospitalization.

Outcome and Predictors

Our primary outcome was unanticipated healthcare reutilization defined as return to the emergency room, urgent care, or unplanned readmission within 30 days of hospital discharge, consistent with the H2O trials. This was measured using the primary institution’s administrative data supplemented by a utilization database shared across regional hospitals.¹⁴ Readmissions were identified as “unplanned” using a previously validated algorithm,¹⁵ and treated as a dichotomous yes/no variable.

Our primary predictors were parental ACEs and resilience (see Appendix Tables). The ACE questionnaire addresses abuse, neglect, and household dysfunction in the first 18 years of life.¹ It is composed of 10 questions, each with a yes/no response.¹ We defined parents as low (ACE 0), moderate (ACE 1-3), or high (ACE ≥4) risk a priori because previous literature has described poor outcomes in adults with 4 or more ACEs.¹⁶

Given the sensitive nature of the questions, respondents independently completed the ACE questionnaire on paper instead of via the face-to-face survey. Respondents returned the completed questionnaire to the research assistant in a sealed envelope. All families received educational information on relevant hospital and community-based resources (eg, social work).

Parental resilience was measured using the Brief Resilience Scale (BRS). The BRS is 6 items, each on a 5-point Likert scale. Responses were averaged, providing a total score of 1-5; higher scores are representative of higher resilience.¹⁷ We treated the BRS score as a continuous variable. BRS has been used in clinical settings; it has demonstrated positive correlation with social support and negative correlation with fatigue.¹⁷ Parents answered BRS questions during the index pediatric hospitalization in a face-to-face interview.

Parent and Child Characteristics

Parent and child sociodemographic variables were also obtained during the face-to-face interview. Parental variables included age, gender, educational attainment, household income, employment status, and financial and social strain.¹¹ Educational attainment was analyzed in 2 categories—high school or less vs more than high school—because most discharge instructions are written at a high school reading level.¹⁸ Parents reported their annual household income in the following categories: <$15,000; $15,000-$29,999; $30,000-$44,999; $45,000-$59,999; $60,000-$89,999; $90,000-$119,999; ≥$120,000. Employment was dichotomized as not employed/student vs any employment. Financial and social strain were assessed using a series of 9 previously described questions.¹⁹ These questions assessed, via self-report, a family’s ability to make ends meet, ability to pay rent/mortgage or utilities, need to move in with others because of financial reasons, and ability to borrow money if needed, as well as home ownership and parental marital status.^15,19 Strain questions were all dichotomous (yes/no, single/not single). A composite variable was then constructed that categorized those reporting no strain items, 1 to 2 items, 3 to 4 items, and 5 or more items.²⁰

Child variables included race, ethnicity, age, primary care access,²¹ payer, and H2O treatment arm. Race categories were white/Caucasian, black/African American, American Indian or Alaskan Native, Asian or Pacific Islander, and other; ethnicity categories were Hispanic/Latino, non-Hispanic/Latino, and unknown. Given relatively low numbers of children reported to be Hispanic/Latino, we combined race and ethnicity into a single variable, categorized as non-Hispanic/white, non-Hispanic/black, and multiracial/Hispanic/other. Primary care access was assessed using the access subscale to the Parent’s Perception of Primary Care questionnaire. This includes assessment of a family’s ability to travel to their doctor, to see their doctor for routine or sick care, and to get help or advice on evenings or weekends. Scores were categorized as always adequate, almost always adequate, or sometimes/never adequate.²¹ Payer was dichotomized to private or public/self-pay.

Statistical Analyses

We examined the distribution of outcomes, predictors, and covariates. We compared sociodemographic characteristics of those respondents and nonrespondents to the ACE screen using the chi-square test for categorical variables or the t test for continuous variables. We used logistic regression to assess for associations between the independent variables of interest and reutilization, adjusting for potential confounders. To build our adjusted, multivariable model, we decided a priori to include child race/ethnicity, primary care access, financial and social strain, and trial treatment arm. We treated the H2O I control group as the referent group. Other covariates considered for inclusion were caregiver education, household income, employment, and payer. These were included in multivariable models if bivariate associations were significant at the P < .1 level. We assessed an ACE-by-resilience interaction term because we hypothesized that those with more ACEs and lower resilience may have more reutilization outcomes than parents with fewer ACEs and higher resilience. We also evaluated interaction terms between trial arm assignment and predictors to assess effects that may be introduced by the randomization. Predictors in the final logistic regression model were significant at the P < .05 level. Logistic regression assumption of little or no multicollinearity among the independent variables was verified in the final models. All analyses were performed with Stata v16 (Stata Corp, College Station, Texas).

There were a total of 1,787 parent-child dyads enrolled in the H2O I and II during the study period; 1,320 parents (74%) completed the ACE questionnaire and were included the analysis. Included parents were primarily female and employed, as well as educated beyond high school (Table 1). Overall, 64% reported one or more ACEs (range 0 to 9); 45% reported 1to 3, and 19% reported 4 or more ACEs. The most commonly reported ACEs were divorce (n = 573, 43%), exposure to alcoholism (n = 306, 23%), and exposure to mental illness (n = 281, 21%; Figure 1). Parents had a mean BRS score of 3.97 (range 1.17-5.00), with the distribution shown in Figure 2.

Of the 1,320 included patients, the average length of stay was 2.5 days, and 82% of hospitalizations were caused by acute medical issues (eg, bronchiolitis). A total of 211 children experienced a reutilization event within 30 days of discharge. In bivariate analysis, children with parents with 4 or more ACEs had a 2.02-times (95% CI 1.35-3.02) higher odds of experiencing a reutilization event than did those with parents reporting no ACEs. Parents with higher resilience scores had children with a lower odds of reutilization (odds ratio [OR] 0.77 95% CI 0.63-0.95).

In addition to our a priori variables, parental education, employment, and insurance met our significance threshold for inclusion in the multivariable model. The ACE-by-resilience interaction term was not significant and not included in the model. Similarly, there was no significant interaction between ACE and resilience and H2O treatment arm; the interaction terms were not included in the final adjusted model, but treatment arm assignment was kept as a covariate. A total of 1,292 children, out of the 1,320 respondents, remained in the final multivariable model; the excluded 28 had incomplete covariate data but were not otherwise different. In this final adjusted model, children with parents reporting 4 or more ACEs had a 1.69-times (95% CI 1.11-2.60) greater odds of reutilization than did those with parents reporting no ACEs (Table 2). Resilience failed to reach statistical significance in the adjusted model (OR 0.86, 95% CI 0.70-1.07).

We found that high-risk parents (4 or more ACEs) had children with an increased odds of healthcare reutilization, suggesting intergenerational effects of ACEs. We did not find a similar effect relating to parental resilience. We also did not find an interaction between parental ACEs and resilience, suggesting that a parent’s reported degree of resilience does not modify the effect of ACEs on reutilization risk.

Parental adversity may be a risk factor for a child’s unanticipated reutilization. We previously demonstrated that parents with 4 or more ACEs have more coping difficulty than a parent with no ACEs after a child’s hospitalization.¹¹ It is possible that parents with high adversity may have poorer coping mechanisms when dealing with a stressful situation, such as a child’s hospitalization. This may have resulted in inequitable outcomes (eg, increased reutilization) for their children. Other studies have confirmed such an intergenerational effect of adversity, linking a parent’s ACEs with poor developmental, behavioral, and health outcomes in their children.^6,22,23 O’Malley et al showed an association of parental ACEs to current adversities,²⁴ such as insurance or housing concerns, that affect the entirety of the household, including children. In short, it appears that parental ACEs may be a compelling predictor of current childhood adversity.

Resilience buffers the negative effects of ACEs; however, we did not find significant associations between resilience and reutilization or an interaction between ACEs and resilience. The factors that may contribute to reutilization are complex. In our previous work, parental resilience was associated with coping difficulty after discharge; but again, did not interact with parental ACEs.¹¹ Here, we suggest that while resilience may buffer the negative effects of ACEs, that buffering may not affect the likelihood of reutilization. It is also possible that the BRS tool is of less relevance on how one handles the stress of a child’s hospitalization. While the BRS is one measure of resilience, there are many other relevant constructs to resilience, such as connection to social supports, that also may also contribute to risk of reutilization.²⁵

Reducing the stress of a hospitalization itself and promoting a safe transition from hospital to home is critical to improving child health outcomes. Our data here, and in our previous work, demonstrate that a history of adversity and one’s current coping ability may drive a parent’s response to a child’s hospitalization and affect their capacity to care for that child after hospital discharge.¹¹ Additional in-hospital supports like child life, behavioral health, or pastoral care could reduce the stress of the hospitalization while also building positive coping mechanisms.^26-29 A meta-analysis demonstrated that such coping interventions can help alleviate the stress of a hospitalization.³⁰ Hill et al demonstrated successful stress reduction in parents of hospitalized children using a “Coping Kit for Parents.”³¹ Further studies are warranted to understand which interventions are most effective for children and families and whether they could be more effectively deployed if the inpatient team knew more about parental ACEs.

Screening for parental ACEs could help to identify patients at highest risk for a poor transition to home. Therefore, screening for parental adversity in clinical settings, including inpatient settings, may be relevant and valuable.³² Additionally, by recognizing the high prevalence of ACEs in an inpatient setting, hospitals and healthcare organizations could be motivated to develop and enact trauma-informed approaches. A trauma-informed care approach recognizes the intersection of trauma with health and social problems. With this recognition, care teams can more sensitively address the trauma as they provide relevant services.³³ Trauma-informed care is a secondary public health prevention approach that would help team members identify the prevalence and effects of trauma via screening, recognize the signs of a maladaptive response to stress, and respond by integrating awareness of trauma into practice management.^28,34 Both the National Academy of Medicine and the Agency for Healthcare Research and Quality have called for such a trauma-informed approach in primary care.³⁵ In response, many healthcare organizations have developed trauma-informed practices to better address the needs of the populations they serve. For example, provider training on this approach has led to improved rapport in patient-provider relationships.³⁶

Although ACE awareness is a component of trauma-informed care, there are still limitations of the original ACE questionnaire developed by Felitti et al. The existing tool is not inclusive of all adversities a parent or child may face. Moreover, its focus is on past exposures and experiences and not current health-related social needs (eg, food insecurity) which have known linkages with a range of health outcomes and health disparities.³⁷ Additionally, the original ACE questionnaire was created as a population level tool and not as a screening tool. If used as a screening tool, providers may view the questions as too sensitive to ask, and parents may have difficulty responding to and understanding the relevance to their child’s care. Therefore, we suggest that more evidence is required to understand how to best adapt ACE questions into a screening processes that may be implemented in a medical setting.

More evidence is also needed to determine when and where such screening may be most useful. A primary care provider would be best equipped to screen caregivers for ACEs given their established relationship with parents and patients. Given the potential relevance of such information for inpatient care provision, information could then flow from primary care to the inpatient team. However, because not all patients have established primary care providers and only 4% of pediatricians screen for ACEs,³⁸ it is important for inpatient medical teams to understand their role in identifying and addressing ACEs during hospital stays. Development of a screening tool, with input from all stakeholders—including parents—that is valid and feasible for use in a pediatric inpatient setting would be an important step forward. This tool should be paired with training in how to discuss these topics in a trauma-informed, nonjudgmental, empathic manner. We see this as a way in which providers can more effectively elicit an accurate response while simultaneously educating parents on the relevance of such sensitive topics during an acute hospital stay. We also recommend that screening should always be paired with response capabilities that connect those who screen positive with resources that could help them to navigate the stress experienced during and after a child’s hospitalization. Furthermore, communication with primary care providers about parents that screen positive should be integrated into the transition process.

This work has several limitations. First, our study was a part of randomized controlled trials conducted in one academic setting, which thereby limits generalizability. For example, we limited our cohort to those who were English-speaking patients only. This may bias our results because respondents with limited English proficiency may have different risk profiles than their English-speaking peers. In addition, the administration of the both the ACE and resilience questionnaires occurred during an acutely stressful period, which may influence how a parent responds to these questions. Also, both of the surveys are self-reported by parents, which may be susceptible to memory and response biases. Relatedly, we had a high number of nonrespondents, particularly to the ACE questionnaire. Our results are therefore only relevant to those who chose to respond and cannot be applied to nonrespondents. Further work assessing why one does or does not respond to such sensitive questions is an important area for future inquiry. Lastly, our cohort had limited medical complexity; future studies may consider links between parental ACEs (and resilience) and morbidity experienced by children with medical complexity.

Parents history of adversity is linked to their children’s unanticipated healthcare reutilization after a hospital discharge. Screening for parental stressors during a hospitalization may be an important first step to connecting parents and children to evidence-based interventions capable of mitigating the stress of hospitalization and promoting better, more seamless transitions from hospital to home.

Group Members: The following H2O members are nonauthor contributors: JoAnne Bachus, BSN, RN; Monica Borell, BSN, RN; Lenisa V Chang, MA, PhD; Patricia Crawford, RN; Sarah Ferris, BA; Jennifer Gold, BSN, RN; Judy A Heilman, BSN, RN; Jane C Khoury, PhD; Pierce Kuhnell, MS; Karen Lawley, BSN, RN; Margo Moore, MS, BSN, RN; Lynne O’Donnell, BSN, RN; Sarah Riddle, MD; Susan N Sherman, DPA; Angela M Statile, MD, MEd; Karen P Sullivan, BSN, RN; Heather Tubbs-Cooley, PhD, RN; Susan Wade-Murphy, MSN, RN; and Christine M White, MD, MAT.

The authors also thank David Keller, MD, for his guidance on the study.

The authors have no financial relationships or conflicts of interest relevant to this article to disclose.

Supported by funds from the Academic Pediatric Young Investigator Award (Dr A Shah) and the Patient-Centered Outcomes Research Institute Award (IHS-1306-0081, to Dr K Auger, Dr S Shah, Dr H Sucharew, Dr J Simmons), the National Institutes of Health (1K23AI112916, to Dr AF Beck), and the Agency for Healthcare Research and Quality (1K12HS026393-01, to Dr A Shah, K08-HS024735- 01A1, to Dr K Auger). Dr J Haney received Summer Undergraduate Research Fellowship funding through the Summer Undergraduate Research Fellowship at Cincinnati Children’s Hospital Medical Center.

All statements in this report, including findings and conclusions, are solely those of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute, its Board of Governors, or the Methodology Committee.

Adverse Childhood Experiences, or ACEs, include exposure to abuse, neglect, or household dysfunction (eg, having a parent who is mentally ill) as a child.¹ Exposure to ACEs affects health into adulthood, with a dose-response relationship between ACEs and a range of comorbidities.¹ Adults with 6 or more ACEs have a 20-year shorter life expectancy than do those with no ACEs.¹ Still, ACEs are static; once experienced, that experience cannot be undone. However, resilience, or positive adaptation in the context of adversity, can be protective, buffering the negative effects of ACEs.^2,3 Protective factors that promote resilience include social capital, such as positive relationships with caregivers and peers.³

With their clear link to health outcomes across the life-course, there is a movement for pediatricians to screen children for ACEs⁴ and to develop strategies that promote resilience in children, parents, and families. However, screening a child for adversity has challenges because younger children may not have experienced an adverse exposure, or they may be unable to voice their experiences. Studies have demonstrated that parental adversity, or ACEs, may be a marker for childhood adversity.^5,6 Biological models also support this potential intergenerational effect of ACEs. Chronic exposure to stress, including ACEs, results in elevated cortisol via a dysregulated hypothalamic-pituitary-adrenal axis, which results in chronic inflammation.⁷ This “toxic stress” is prolonged, severe in intensity, and can lead to epigenetic changes that may be passed on to the next generation.^8,9

Hospitalization of an ill child, and the transition to home after that hospitalization, is a stressful event for children and families.¹⁰ This stress may be relevant to parents that have a history of a high rate of ACEs or a current low degree of resilience. Our previous work demonstrated that, in the inpatient setting, parents with high ACEs (≥4) or low resilience have increased coping difficulty 14 days after their child’s hospital discharge.¹¹ Our objective here was to evaluate whether a parent’s ACEs and/or resilience would also be associated with that child’s likelihood of reutilization. We hypothesized that more parental ACEs and/or lower parental resilience would be associated with revisits the emergency room, urgent care, or hospital readmissions.

Participants and Study Design

We conducted a prospective cohort study of parents of hospitalized children recruited from the “Hospital-to-Home Outcomes” Studies (H2O I and H2O II).^12,13 H2O I and II were prospective, single-center, randomized controlled trials designed to determine the effectiveness of either a nurse-led transitional home visit (H2O I) or telephone call (H2O II) on 30-day unplanned healthcare reutilization. The trials and this study were approved by the Cincinnati Children’s Institutional Review Board. All parents provided written informed consent.

Details of H2O I and II recruitment and design have been described previously.^12,13 Briefly, children were eligible for inclusion in either study if they were admitted to our institution’s general Hospital Medicine or the Hospital Medicine Complex Care Services; for H2O I, children hospitalized on the Neurology and Neurosurgery services were also eligible.^12,13 Patients were excluded if they were discharged to a residential facility, if they lived outside the home healthcare nurse service area, if they were eligible for skilled home healthcare services (eg, intravenous antibiotics), or if the participating caregiver was non-English speaking.^12,13 In H2O I, families were randomized either to receive a single nurse home visit within 96 hours of discharge or standard of care. In H2O II, families enrolled were randomized to receive a telephone call by a nurse within 96 hours of discharge or standard of care. As we have previously published, randomization in both trials successfully balanced the intervention and control arms with respect to key demographic characteristics.^12,13 For the analyses presented here, we focused on a subset of caregivers 18 years and older whose children were enrolled in either H2O I or II between August 2015 and October 2016. In both H2O trials, face-to-face and paper-based questionnaires were completed by parents during the index hospitalization.

Outcome and Predictors

Our primary outcome was unanticipated healthcare reutilization defined as return to the emergency room, urgent care, or unplanned readmission within 30 days of hospital discharge, consistent with the H2O trials. This was measured using the primary institution’s administrative data supplemented by a utilization database shared across regional hospitals.¹⁴ Readmissions were identified as “unplanned” using a previously validated algorithm,¹⁵ and treated as a dichotomous yes/no variable.

Our primary predictors were parental ACEs and resilience (see Appendix Tables). The ACE questionnaire addresses abuse, neglect, and household dysfunction in the first 18 years of life.¹ It is composed of 10 questions, each with a yes/no response.¹ We defined parents as low (ACE 0), moderate (ACE 1-3), or high (ACE ≥4) risk a priori because previous literature has described poor outcomes in adults with 4 or more ACEs.¹⁶

Given the sensitive nature of the questions, respondents independently completed the ACE questionnaire on paper instead of via the face-to-face survey. Respondents returned the completed questionnaire to the research assistant in a sealed envelope. All families received educational information on relevant hospital and community-based resources (eg, social work).

Parental resilience was measured using the Brief Resilience Scale (BRS). The BRS is 6 items, each on a 5-point Likert scale. Responses were averaged, providing a total score of 1-5; higher scores are representative of higher resilience.¹⁷ We treated the BRS score as a continuous variable. BRS has been used in clinical settings; it has demonstrated positive correlation with social support and negative correlation with fatigue.¹⁷ Parents answered BRS questions during the index pediatric hospitalization in a face-to-face interview.

Parent and Child Characteristics

Parent and child sociodemographic variables were also obtained during the face-to-face interview. Parental variables included age, gender, educational attainment, household income, employment status, and financial and social strain.¹¹ Educational attainment was analyzed in 2 categories—high school or less vs more than high school—because most discharge instructions are written at a high school reading level.¹⁸ Parents reported their annual household income in the following categories: <$15,000; $15,000-$29,999; $30,000-$44,999; $45,000-$59,999; $60,000-$89,999; $90,000-$119,999; ≥$120,000. Employment was dichotomized as not employed/student vs any employment. Financial and social strain were assessed using a series of 9 previously described questions.¹⁹ These questions assessed, via self-report, a family’s ability to make ends meet, ability to pay rent/mortgage or utilities, need to move in with others because of financial reasons, and ability to borrow money if needed, as well as home ownership and parental marital status.^15,19 Strain questions were all dichotomous (yes/no, single/not single). A composite variable was then constructed that categorized those reporting no strain items, 1 to 2 items, 3 to 4 items, and 5 or more items.²⁰

Child variables included race, ethnicity, age, primary care access,²¹ payer, and H2O treatment arm. Race categories were white/Caucasian, black/African American, American Indian or Alaskan Native, Asian or Pacific Islander, and other; ethnicity categories were Hispanic/Latino, non-Hispanic/Latino, and unknown. Given relatively low numbers of children reported to be Hispanic/Latino, we combined race and ethnicity into a single variable, categorized as non-Hispanic/white, non-Hispanic/black, and multiracial/Hispanic/other. Primary care access was assessed using the access subscale to the Parent’s Perception of Primary Care questionnaire. This includes assessment of a family’s ability to travel to their doctor, to see their doctor for routine or sick care, and to get help or advice on evenings or weekends. Scores were categorized as always adequate, almost always adequate, or sometimes/never adequate.²¹ Payer was dichotomized to private or public/self-pay.

Statistical Analyses

We examined the distribution of outcomes, predictors, and covariates. We compared sociodemographic characteristics of those respondents and nonrespondents to the ACE screen using the chi-square test for categorical variables or the t test for continuous variables. We used logistic regression to assess for associations between the independent variables of interest and reutilization, adjusting for potential confounders. To build our adjusted, multivariable model, we decided a priori to include child race/ethnicity, primary care access, financial and social strain, and trial treatment arm. We treated the H2O I control group as the referent group. Other covariates considered for inclusion were caregiver education, household income, employment, and payer. These were included in multivariable models if bivariate associations were significant at the P < .1 level. We assessed an ACE-by-resilience interaction term because we hypothesized that those with more ACEs and lower resilience may have more reutilization outcomes than parents with fewer ACEs and higher resilience. We also evaluated interaction terms between trial arm assignment and predictors to assess effects that may be introduced by the randomization. Predictors in the final logistic regression model were significant at the P < .05 level. Logistic regression assumption of little or no multicollinearity among the independent variables was verified in the final models. All analyses were performed with Stata v16 (Stata Corp, College Station, Texas).

There were a total of 1,787 parent-child dyads enrolled in the H2O I and II during the study period; 1,320 parents (74%) completed the ACE questionnaire and were included the analysis. Included parents were primarily female and employed, as well as educated beyond high school (Table 1). Overall, 64% reported one or more ACEs (range 0 to 9); 45% reported 1to 3, and 19% reported 4 or more ACEs. The most commonly reported ACEs were divorce (n = 573, 43%), exposure to alcoholism (n = 306, 23%), and exposure to mental illness (n = 281, 21%; Figure 1). Parents had a mean BRS score of 3.97 (range 1.17-5.00), with the distribution shown in Figure 2.

Of the 1,320 included patients, the average length of stay was 2.5 days, and 82% of hospitalizations were caused by acute medical issues (eg, bronchiolitis). A total of 211 children experienced a reutilization event within 30 days of discharge. In bivariate analysis, children with parents with 4 or more ACEs had a 2.02-times (95% CI 1.35-3.02) higher odds of experiencing a reutilization event than did those with parents reporting no ACEs. Parents with higher resilience scores had children with a lower odds of reutilization (odds ratio [OR] 0.77 95% CI 0.63-0.95).

In addition to our a priori variables, parental education, employment, and insurance met our significance threshold for inclusion in the multivariable model. The ACE-by-resilience interaction term was not significant and not included in the model. Similarly, there was no significant interaction between ACE and resilience and H2O treatment arm; the interaction terms were not included in the final adjusted model, but treatment arm assignment was kept as a covariate. A total of 1,292 children, out of the 1,320 respondents, remained in the final multivariable model; the excluded 28 had incomplete covariate data but were not otherwise different. In this final adjusted model, children with parents reporting 4 or more ACEs had a 1.69-times (95% CI 1.11-2.60) greater odds of reutilization than did those with parents reporting no ACEs (Table 2). Resilience failed to reach statistical significance in the adjusted model (OR 0.86, 95% CI 0.70-1.07).

We found that high-risk parents (4 or more ACEs) had children with an increased odds of healthcare reutilization, suggesting intergenerational effects of ACEs. We did not find a similar effect relating to parental resilience. We also did not find an interaction between parental ACEs and resilience, suggesting that a parent’s reported degree of resilience does not modify the effect of ACEs on reutilization risk.

Parental adversity may be a risk factor for a child’s unanticipated reutilization. We previously demonstrated that parents with 4 or more ACEs have more coping difficulty than a parent with no ACEs after a child’s hospitalization.¹¹ It is possible that parents with high adversity may have poorer coping mechanisms when dealing with a stressful situation, such as a child’s hospitalization. This may have resulted in inequitable outcomes (eg, increased reutilization) for their children. Other studies have confirmed such an intergenerational effect of adversity, linking a parent’s ACEs with poor developmental, behavioral, and health outcomes in their children.^6,22,23 O’Malley et al showed an association of parental ACEs to current adversities,²⁴ such as insurance or housing concerns, that affect the entirety of the household, including children. In short, it appears that parental ACEs may be a compelling predictor of current childhood adversity.

Resilience buffers the negative effects of ACEs; however, we did not find significant associations between resilience and reutilization or an interaction between ACEs and resilience. The factors that may contribute to reutilization are complex. In our previous work, parental resilience was associated with coping difficulty after discharge; but again, did not interact with parental ACEs.¹¹ Here, we suggest that while resilience may buffer the negative effects of ACEs, that buffering may not affect the likelihood of reutilization. It is also possible that the BRS tool is of less relevance on how one handles the stress of a child’s hospitalization. While the BRS is one measure of resilience, there are many other relevant constructs to resilience, such as connection to social supports, that also may also contribute to risk of reutilization.²⁵

Reducing the stress of a hospitalization itself and promoting a safe transition from hospital to home is critical to improving child health outcomes. Our data here, and in our previous work, demonstrate that a history of adversity and one’s current coping ability may drive a parent’s response to a child’s hospitalization and affect their capacity to care for that child after hospital discharge.¹¹ Additional in-hospital supports like child life, behavioral health, or pastoral care could reduce the stress of the hospitalization while also building positive coping mechanisms.^26-29 A meta-analysis demonstrated that such coping interventions can help alleviate the stress of a hospitalization.³⁰ Hill et al demonstrated successful stress reduction in parents of hospitalized children using a “Coping Kit for Parents.”³¹ Further studies are warranted to understand which interventions are most effective for children and families and whether they could be more effectively deployed if the inpatient team knew more about parental ACEs.

Screening for parental ACEs could help to identify patients at highest risk for a poor transition to home. Therefore, screening for parental adversity in clinical settings, including inpatient settings, may be relevant and valuable.³² Additionally, by recognizing the high prevalence of ACEs in an inpatient setting, hospitals and healthcare organizations could be motivated to develop and enact trauma-informed approaches. A trauma-informed care approach recognizes the intersection of trauma with health and social problems. With this recognition, care teams can more sensitively address the trauma as they provide relevant services.³³ Trauma-informed care is a secondary public health prevention approach that would help team members identify the prevalence and effects of trauma via screening, recognize the signs of a maladaptive response to stress, and respond by integrating awareness of trauma into practice management.^28,34 Both the National Academy of Medicine and the Agency for Healthcare Research and Quality have called for such a trauma-informed approach in primary care.³⁵ In response, many healthcare organizations have developed trauma-informed practices to better address the needs of the populations they serve. For example, provider training on this approach has led to improved rapport in patient-provider relationships.³⁶

Although ACE awareness is a component of trauma-informed care, there are still limitations of the original ACE questionnaire developed by Felitti et al. The existing tool is not inclusive of all adversities a parent or child may face. Moreover, its focus is on past exposures and experiences and not current health-related social needs (eg, food insecurity) which have known linkages with a range of health outcomes and health disparities.³⁷ Additionally, the original ACE questionnaire was created as a population level tool and not as a screening tool. If used as a screening tool, providers may view the questions as too sensitive to ask, and parents may have difficulty responding to and understanding the relevance to their child’s care. Therefore, we suggest that more evidence is required to understand how to best adapt ACE questions into a screening processes that may be implemented in a medical setting.

More evidence is also needed to determine when and where such screening may be most useful. A primary care provider would be best equipped to screen caregivers for ACEs given their established relationship with parents and patients. Given the potential relevance of such information for inpatient care provision, information could then flow from primary care to the inpatient team. However, because not all patients have established primary care providers and only 4% of pediatricians screen for ACEs,³⁸ it is important for inpatient medical teams to understand their role in identifying and addressing ACEs during hospital stays. Development of a screening tool, with input from all stakeholders—including parents—that is valid and feasible for use in a pediatric inpatient setting would be an important step forward. This tool should be paired with training in how to discuss these topics in a trauma-informed, nonjudgmental, empathic manner. We see this as a way in which providers can more effectively elicit an accurate response while simultaneously educating parents on the relevance of such sensitive topics during an acute hospital stay. We also recommend that screening should always be paired with response capabilities that connect those who screen positive with resources that could help them to navigate the stress experienced during and after a child’s hospitalization. Furthermore, communication with primary care providers about parents that screen positive should be integrated into the transition process.

This work has several limitations. First, our study was a part of randomized controlled trials conducted in one academic setting, which thereby limits generalizability. For example, we limited our cohort to those who were English-speaking patients only. This may bias our results because respondents with limited English proficiency may have different risk profiles than their English-speaking peers. In addition, the administration of the both the ACE and resilience questionnaires occurred during an acutely stressful period, which may influence how a parent responds to these questions. Also, both of the surveys are self-reported by parents, which may be susceptible to memory and response biases. Relatedly, we had a high number of nonrespondents, particularly to the ACE questionnaire. Our results are therefore only relevant to those who chose to respond and cannot be applied to nonrespondents. Further work assessing why one does or does not respond to such sensitive questions is an important area for future inquiry. Lastly, our cohort had limited medical complexity; future studies may consider links between parental ACEs (and resilience) and morbidity experienced by children with medical complexity.

Parents history of adversity is linked to their children’s unanticipated healthcare reutilization after a hospital discharge. Screening for parental stressors during a hospitalization may be an important first step to connecting parents and children to evidence-based interventions capable of mitigating the stress of hospitalization and promoting better, more seamless transitions from hospital to home.

Group Members: The following H2O members are nonauthor contributors: JoAnne Bachus, BSN, RN; Monica Borell, BSN, RN; Lenisa V Chang, MA, PhD; Patricia Crawford, RN; Sarah Ferris, BA; Jennifer Gold, BSN, RN; Judy A Heilman, BSN, RN; Jane C Khoury, PhD; Pierce Kuhnell, MS; Karen Lawley, BSN, RN; Margo Moore, MS, BSN, RN; Lynne O’Donnell, BSN, RN; Sarah Riddle, MD; Susan N Sherman, DPA; Angela M Statile, MD, MEd; Karen P Sullivan, BSN, RN; Heather Tubbs-Cooley, PhD, RN; Susan Wade-Murphy, MSN, RN; and Christine M White, MD, MAT.

The authors also thank David Keller, MD, for his guidance on the study.

The authors have no financial relationships or conflicts of interest relevant to this article to disclose.

Supported by funds from the Academic Pediatric Young Investigator Award (Dr A Shah) and the Patient-Centered Outcomes Research Institute Award (IHS-1306-0081, to Dr K Auger, Dr S Shah, Dr H Sucharew, Dr J Simmons), the National Institutes of Health (1K23AI112916, to Dr AF Beck), and the Agency for Healthcare Research and Quality (1K12HS026393-01, to Dr A Shah, K08-HS024735- 01A1, to Dr K Auger). Dr J Haney received Summer Undergraduate Research Fellowship funding through the Summer Undergraduate Research Fellowship at Cincinnati Children’s Hospital Medical Center.

All statements in this report, including findings and conclusions, are solely those of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute, its Board of Governors, or the Methodology Committee.