Jonathan I. Silverberg, MD, PhD, MPH

Diversity, equity, and inclusion (DEI) programs seek to improve dermatologic education and clinical care for an increasingly diverse patient population as well as to recruit and sustain a physician workforce that reflects the diversity of the patients they serve.^1,2 In dermatology, only 4.2% and 3.0% of practicing dermatologists self-identify as being of Hispanic and African American ethnicity, respectively, compared with 18.5% and 13.4% of the general population, respectively.³ Creating an educational system that works to meet the goals of DEI is essential to improve health outcomes and address disparities. The lack of robust DEI-related curricula during residency training may limit the ability of practicing dermatologists to provide comprehensive and culturally sensitive care. It has been shown that racial concordance between patients and physicians has a positive impact on patient satisfaction by fostering a trusting patient-physician relationship.⁴

It is the responsibility of all dermatologists to create an environment where patients from any background can feel comfortable, which can be cultivated by establishing patient-centered communication and cultural humility.⁵ These skills can be strengthened via the implementation of DEI-related curricula during residency training. Augmenting exposure of these topics during training can optimize the delivery of dermatologic care by providing residents with the tools and confidence needed to care for patients of culturally diverse backgrounds. Enhancing DEI education is crucial to not only improve the recognition and treatment of dermatologic conditions in all skin and hair types but also to minimize misconceptions, stigma, health disparities, and discrimination faced by historically marginalized communities. Creating a culture of inclusion is of paramount importance to build successful relationships with patients and colleagues of culturally diverse backgrounds.⁶

There are multiple efforts underway to increase DEI education across the field of dermatology, including the development of DEI task forces in professional organizations and societies that serve to expand DEI-related research, mentorship, and education. The American Academy of Dermatology has been leading efforts to create a curriculum focused on skin of color, particularly addressing inadequate educational training on how dermatologic conditions manifest in this population.⁷ The Skin of Color Society has similar efforts underway and is developing a speakers bureau to give leading experts a platform to lecture dermatology trainees as well as patient and community audiences on various topics in skin of color.⁸ These are just 2 of many professional dermatology organizations that are advocating for expanded education on DEI; however, consistently integrating DEI-related topics into dermatology residency training curricula remains a gap in pedagogy. To identify the DEI-related topics of greatest relevance to the dermatology resident curricula, we implemented a modified electronic Delphi (e-Delphi) consensus process to provide standardized recommendations.

Methods

A 2-round modified e-Delphi method was utilized (Figure). An initial list of potential curricular topics was formulated by an expert panel consisting of 5 dermatologists from the Association of Professors of Dermatology DEI subcommittee and the American Academy of Dermatology Diversity Task Force (A.M.A., S.B., R.V., S.D.W., J.I.S.). Initial topics were selected via several meetings among the panel members to discuss existing DEI concerns and issues that were deemed relevant due to education gaps in residency training. The list of topics was further expanded with recommendations obtained via an email sent to dermatology program directors on the Association of Professors of Dermatology listserve, which solicited voluntary participation of academic dermatologists, including program directors and dermatology residents.

There were 2 voting rounds, with each round consisting of questions scored on a Likert scale ranging from 1 to 5 (1=not essential, 2=probably not essential, 3=neutral, 4=probably essential, 5=definitely essential). The inclusion criteria to classify a topic as necessary for integration into the dermatology residency curriculum included 95% (18/19) or more of respondents rating the topic as probably essential or definitely essential; if more than 90% (17/19) of respondents rated the topic as probably essential or definitely essential and less than 10% (2/19) rated it as not essential or probably not essential, the topic was still included as part of the suggested curriculum. Topics that received ratings of probably essential or definitely essential by less than 80% (15/19) of respondents were removed from consideration. The topics that did not meet inclusion or exclusion criteria during the first round of voting were refined by the e-Delphi steering committee (V.S.E-C. and F-A.R.) based on open-ended feedback from the voting group provided at the end of the survey and subsequently passed to the second round of voting.

Results

Participants—A total of 19 respondents participated in both voting rounds, the majority (80% [15/19]) of whom were program directors or dermatologists affiliated with academia or development of DEI education; the remaining 20% [4/19]) were dermatology residents.

Open-Ended Feedback—Voting group members were able to provide open-ended feedback for each of the sets of topics after the survey, which the steering committee utilized to modify the topics as needed for the final voting round. For example, “structural racism/discrimination” was originally mentioned as a topic, but several participants suggested including specific types of racism; therefore, the wording was changed to “racism: types, definitions” to encompass broader definitions and types of racism.

Survey Results—Two genres of topics were surveyed in each voting round: clinical and nonclinical. Participants voted on a total of 61 topics, with 23 ultimately selected in the final list of consensus curricular topics. Of those, 9 were clinical and 14 nonclinical. All topics deemed necessary for inclusion in residency curricula are presented in eTables 1 and 2.

During the first round of voting, the e-Delphi panel reached a consensus to include the following 17 topics as essential to dermatology residency training (along with the percentage of voters who classified them as probably essential or definitely essential): how to mitigate bias in clinical and workplace settings (100% [40/40]); social determinants of health-related disparities in dermatology (100% [40/40]); hairstyling practices across different hair textures (100% [40/40]); definitions and examples of microaggressions (97.50% [39/40]); definition, background, and types of bias (97.50% [39/40]); manifestations of bias in the clinical setting (97.44% [38/39]); racial and ethnic disparities in dermatology (97.44% [38/39]); keloids (97.37% [37/38]); differences in dermoscopic presentations in skin of color (97.30% [36/37]); skin cancer in patients with skin of color (97.30% [36/37]); disparities due to bias (95.00% [38/40]); how to apply cultural humility and safety to patients of different cultural backgrounds (94.87% [37/40]); best practices in providing care to patients with limited English proficiency (94.87% [37/40]); hair loss in patients with textured hair (94.74% [36/38]); pseudofolliculitis barbae and acne keloidalis nuchae (94.60% [35/37]); disparities regarding people experiencing homelessness (92.31% [36/39]); and definitions and types of racism and other forms of discrimination (92.31% [36/39]). eTable 1 provides a list of suggested resources to incorporate these topics into the educational components of residency curricula. The resources provided were not part of the voting process, and they were not considered in the consensus analysis; they are included here as suggested educational catalysts.

During the second round of voting, 25 topics were evaluated. Of those, the following 6 topics were proposed to be included as essential in residency training: differences in prevalence and presentation of common inflammatory disorders (100% [29/29]); manifestations of bias in the learning environment (96.55%); antiracist action and how to decrease the effects of structural racism in clinical and educational settings (96.55% [28/29]); diversity of images in dermatology education (96.55% [28/29]); pigmentary disorders and their psychological effects (96.55% [28/29]); and LGBTQ (lesbian, gay, bisexual, transgender, and queer) dermatologic health care (96.55% [28/29]). eTable 2 includes these topics as well as suggested resources to help incorporate them into training.

Comment

This study utilized a modified e-Delphi technique to identify relevant clinical and nonclinical DEI topics that should be incorporated into dermatology residency curricula. The panel members reached a consensus for 9 clinical DEI-related topics. The respondents agreed that the topics related to skin and hair conditions in patients with skin of color as well as textured hair were crucial to residency education. Skin cancer, hair loss, pseudofolliculitis barbae, acne keloidalis nuchae, keloids, pigmentary disorders, and their varying presentations in patients with skin of color were among the recommended topics. The panel also recommended educating residents on the variable visual presentations of inflammatory conditions in skin of color. Addressing the needs of diverse patients—for example, those belonging to the LGBTQ community—also was deemed important for inclusion.

The remaining 14 chosen topics were nonclinical items addressing concepts such as bias and health care disparities as well as cultural humility and safety.⁹ Cultural humility and safety focus on developing cultural awareness by creating a safe setting for patients rather than encouraging power relationships between them and their physicians. Various topics related to racism also were recommended to be included in residency curricula, including education on implementation of antiracist action in the workplace.

Many of the nonclinical topics are intertwined; for instance, learning about health care disparities in patients with limited English proficiency allows for improved best practices in delivering care to patients from this population. The first step in overcoming bias and subsequent disparities is acknowledging how the perpetuation of bias leads to disparities after being taught tools to recognize it.

Our group’s guidance on DEI topics should help dermatology residency program leaders as they design and refine program curricula. There are multiple avenues for incorporating education on these topics, including lectures, interactive workshops, role-playing sessions, book or journal clubs, and discussion circles. Many of these topics/programs may already be included in programs’ didactic curricula, which would minimize the burden of finding space to educate on these topics. Institutional cultural change is key to ensuring truly diverse, equitable, and inclusive workplaces. Educating tomorrow’s dermatologists on these topics is a first step toward achieving that cultural change.

Limitations—A limitation of this e-Delphi survey is that only a selection of experts in this field was included. Additionally, we were concerned that the Likert scale format and the bar we set for inclusion and exclusion may have failed to adequately capture participants’ nuanced opinions. As such, participants were able to provide open-ended feedback, and suggestions for alternate wording or other changes were considered by the steering committee. Finally, inclusion recommendations identified in this survey were developed specifically for US dermatology residents.

Conclusion

In this e-Delphi consensus assessment of DEI-related topics, we recommend the inclusion of 23 topics into dermatology residency program curricula to improve medical training and the patient-physician relationship as well as to create better health outcomes. We also provide specific sample resource recommendations in eTables 1 and 2 to facilitate inclusion of these topics into residency curricula across the country.

Diversity, equity, and inclusion (DEI) programs seek to improve dermatologic education and clinical care for an increasingly diverse patient population as well as to recruit and sustain a physician workforce that reflects the diversity of the patients they serve.^1,2 In dermatology, only 4.2% and 3.0% of practicing dermatologists self-identify as being of Hispanic and African American ethnicity, respectively, compared with 18.5% and 13.4% of the general population, respectively.³ Creating an educational system that works to meet the goals of DEI is essential to improve health outcomes and address disparities. The lack of robust DEI-related curricula during residency training may limit the ability of practicing dermatologists to provide comprehensive and culturally sensitive care. It has been shown that racial concordance between patients and physicians has a positive impact on patient satisfaction by fostering a trusting patient-physician relationship.⁴

It is the responsibility of all dermatologists to create an environment where patients from any background can feel comfortable, which can be cultivated by establishing patient-centered communication and cultural humility.⁵ These skills can be strengthened via the implementation of DEI-related curricula during residency training. Augmenting exposure of these topics during training can optimize the delivery of dermatologic care by providing residents with the tools and confidence needed to care for patients of culturally diverse backgrounds. Enhancing DEI education is crucial to not only improve the recognition and treatment of dermatologic conditions in all skin and hair types but also to minimize misconceptions, stigma, health disparities, and discrimination faced by historically marginalized communities. Creating a culture of inclusion is of paramount importance to build successful relationships with patients and colleagues of culturally diverse backgrounds.⁶

There are multiple efforts underway to increase DEI education across the field of dermatology, including the development of DEI task forces in professional organizations and societies that serve to expand DEI-related research, mentorship, and education. The American Academy of Dermatology has been leading efforts to create a curriculum focused on skin of color, particularly addressing inadequate educational training on how dermatologic conditions manifest in this population.⁷ The Skin of Color Society has similar efforts underway and is developing a speakers bureau to give leading experts a platform to lecture dermatology trainees as well as patient and community audiences on various topics in skin of color.⁸ These are just 2 of many professional dermatology organizations that are advocating for expanded education on DEI; however, consistently integrating DEI-related topics into dermatology residency training curricula remains a gap in pedagogy. To identify the DEI-related topics of greatest relevance to the dermatology resident curricula, we implemented a modified electronic Delphi (e-Delphi) consensus process to provide standardized recommendations.

Methods

A 2-round modified e-Delphi method was utilized (Figure). An initial list of potential curricular topics was formulated by an expert panel consisting of 5 dermatologists from the Association of Professors of Dermatology DEI subcommittee and the American Academy of Dermatology Diversity Task Force (A.M.A., S.B., R.V., S.D.W., J.I.S.). Initial topics were selected via several meetings among the panel members to discuss existing DEI concerns and issues that were deemed relevant due to education gaps in residency training. The list of topics was further expanded with recommendations obtained via an email sent to dermatology program directors on the Association of Professors of Dermatology listserve, which solicited voluntary participation of academic dermatologists, including program directors and dermatology residents.

There were 2 voting rounds, with each round consisting of questions scored on a Likert scale ranging from 1 to 5 (1=not essential, 2=probably not essential, 3=neutral, 4=probably essential, 5=definitely essential). The inclusion criteria to classify a topic as necessary for integration into the dermatology residency curriculum included 95% (18/19) or more of respondents rating the topic as probably essential or definitely essential; if more than 90% (17/19) of respondents rated the topic as probably essential or definitely essential and less than 10% (2/19) rated it as not essential or probably not essential, the topic was still included as part of the suggested curriculum. Topics that received ratings of probably essential or definitely essential by less than 80% (15/19) of respondents were removed from consideration. The topics that did not meet inclusion or exclusion criteria during the first round of voting were refined by the e-Delphi steering committee (V.S.E-C. and F-A.R.) based on open-ended feedback from the voting group provided at the end of the survey and subsequently passed to the second round of voting.

Results

Participants—A total of 19 respondents participated in both voting rounds, the majority (80% [15/19]) of whom were program directors or dermatologists affiliated with academia or development of DEI education; the remaining 20% [4/19]) were dermatology residents.

Open-Ended Feedback—Voting group members were able to provide open-ended feedback for each of the sets of topics after the survey, which the steering committee utilized to modify the topics as needed for the final voting round. For example, “structural racism/discrimination” was originally mentioned as a topic, but several participants suggested including specific types of racism; therefore, the wording was changed to “racism: types, definitions” to encompass broader definitions and types of racism.

Survey Results—Two genres of topics were surveyed in each voting round: clinical and nonclinical. Participants voted on a total of 61 topics, with 23 ultimately selected in the final list of consensus curricular topics. Of those, 9 were clinical and 14 nonclinical. All topics deemed necessary for inclusion in residency curricula are presented in eTables 1 and 2.

During the first round of voting, the e-Delphi panel reached a consensus to include the following 17 topics as essential to dermatology residency training (along with the percentage of voters who classified them as probably essential or definitely essential): how to mitigate bias in clinical and workplace settings (100% [40/40]); social determinants of health-related disparities in dermatology (100% [40/40]); hairstyling practices across different hair textures (100% [40/40]); definitions and examples of microaggressions (97.50% [39/40]); definition, background, and types of bias (97.50% [39/40]); manifestations of bias in the clinical setting (97.44% [38/39]); racial and ethnic disparities in dermatology (97.44% [38/39]); keloids (97.37% [37/38]); differences in dermoscopic presentations in skin of color (97.30% [36/37]); skin cancer in patients with skin of color (97.30% [36/37]); disparities due to bias (95.00% [38/40]); how to apply cultural humility and safety to patients of different cultural backgrounds (94.87% [37/40]); best practices in providing care to patients with limited English proficiency (94.87% [37/40]); hair loss in patients with textured hair (94.74% [36/38]); pseudofolliculitis barbae and acne keloidalis nuchae (94.60% [35/37]); disparities regarding people experiencing homelessness (92.31% [36/39]); and definitions and types of racism and other forms of discrimination (92.31% [36/39]). eTable 1 provides a list of suggested resources to incorporate these topics into the educational components of residency curricula. The resources provided were not part of the voting process, and they were not considered in the consensus analysis; they are included here as suggested educational catalysts.

During the second round of voting, 25 topics were evaluated. Of those, the following 6 topics were proposed to be included as essential in residency training: differences in prevalence and presentation of common inflammatory disorders (100% [29/29]); manifestations of bias in the learning environment (96.55%); antiracist action and how to decrease the effects of structural racism in clinical and educational settings (96.55% [28/29]); diversity of images in dermatology education (96.55% [28/29]); pigmentary disorders and their psychological effects (96.55% [28/29]); and LGBTQ (lesbian, gay, bisexual, transgender, and queer) dermatologic health care (96.55% [28/29]). eTable 2 includes these topics as well as suggested resources to help incorporate them into training.

Comment

This study utilized a modified e-Delphi technique to identify relevant clinical and nonclinical DEI topics that should be incorporated into dermatology residency curricula. The panel members reached a consensus for 9 clinical DEI-related topics. The respondents agreed that the topics related to skin and hair conditions in patients with skin of color as well as textured hair were crucial to residency education. Skin cancer, hair loss, pseudofolliculitis barbae, acne keloidalis nuchae, keloids, pigmentary disorders, and their varying presentations in patients with skin of color were among the recommended topics. The panel also recommended educating residents on the variable visual presentations of inflammatory conditions in skin of color. Addressing the needs of diverse patients—for example, those belonging to the LGBTQ community—also was deemed important for inclusion.

The remaining 14 chosen topics were nonclinical items addressing concepts such as bias and health care disparities as well as cultural humility and safety.⁹ Cultural humility and safety focus on developing cultural awareness by creating a safe setting for patients rather than encouraging power relationships between them and their physicians. Various topics related to racism also were recommended to be included in residency curricula, including education on implementation of antiracist action in the workplace.

Many of the nonclinical topics are intertwined; for instance, learning about health care disparities in patients with limited English proficiency allows for improved best practices in delivering care to patients from this population. The first step in overcoming bias and subsequent disparities is acknowledging how the perpetuation of bias leads to disparities after being taught tools to recognize it.

Our group’s guidance on DEI topics should help dermatology residency program leaders as they design and refine program curricula. There are multiple avenues for incorporating education on these topics, including lectures, interactive workshops, role-playing sessions, book or journal clubs, and discussion circles. Many of these topics/programs may already be included in programs’ didactic curricula, which would minimize the burden of finding space to educate on these topics. Institutional cultural change is key to ensuring truly diverse, equitable, and inclusive workplaces. Educating tomorrow’s dermatologists on these topics is a first step toward achieving that cultural change.

Limitations—A limitation of this e-Delphi survey is that only a selection of experts in this field was included. Additionally, we were concerned that the Likert scale format and the bar we set for inclusion and exclusion may have failed to adequately capture participants’ nuanced opinions. As such, participants were able to provide open-ended feedback, and suggestions for alternate wording or other changes were considered by the steering committee. Finally, inclusion recommendations identified in this survey were developed specifically for US dermatology residents.

Conclusion

In this e-Delphi consensus assessment of DEI-related topics, we recommend the inclusion of 23 topics into dermatology residency program curricula to improve medical training and the patient-physician relationship as well as to create better health outcomes. We also provide specific sample resource recommendations in eTables 1 and 2 to facilitate inclusion of these topics into residency curricula across the country.

Diversity, equity, and inclusion (DEI) programs seek to improve dermatologic education and clinical care for an increasingly diverse patient population as well as to recruit and sustain a physician workforce that reflects the diversity of the patients they serve.^1,2 In dermatology, only 4.2% and 3.0% of practicing dermatologists self-identify as being of Hispanic and African American ethnicity, respectively, compared with 18.5% and 13.4% of the general population, respectively.³ Creating an educational system that works to meet the goals of DEI is essential to improve health outcomes and address disparities. The lack of robust DEI-related curricula during residency training may limit the ability of practicing dermatologists to provide comprehensive and culturally sensitive care. It has been shown that racial concordance between patients and physicians has a positive impact on patient satisfaction by fostering a trusting patient-physician relationship.⁴

It is the responsibility of all dermatologists to create an environment where patients from any background can feel comfortable, which can be cultivated by establishing patient-centered communication and cultural humility.⁵ These skills can be strengthened via the implementation of DEI-related curricula during residency training. Augmenting exposure of these topics during training can optimize the delivery of dermatologic care by providing residents with the tools and confidence needed to care for patients of culturally diverse backgrounds. Enhancing DEI education is crucial to not only improve the recognition and treatment of dermatologic conditions in all skin and hair types but also to minimize misconceptions, stigma, health disparities, and discrimination faced by historically marginalized communities. Creating a culture of inclusion is of paramount importance to build successful relationships with patients and colleagues of culturally diverse backgrounds.⁶

There are multiple efforts underway to increase DEI education across the field of dermatology, including the development of DEI task forces in professional organizations and societies that serve to expand DEI-related research, mentorship, and education. The American Academy of Dermatology has been leading efforts to create a curriculum focused on skin of color, particularly addressing inadequate educational training on how dermatologic conditions manifest in this population.⁷ The Skin of Color Society has similar efforts underway and is developing a speakers bureau to give leading experts a platform to lecture dermatology trainees as well as patient and community audiences on various topics in skin of color.⁸ These are just 2 of many professional dermatology organizations that are advocating for expanded education on DEI; however, consistently integrating DEI-related topics into dermatology residency training curricula remains a gap in pedagogy. To identify the DEI-related topics of greatest relevance to the dermatology resident curricula, we implemented a modified electronic Delphi (e-Delphi) consensus process to provide standardized recommendations.

Methods

A 2-round modified e-Delphi method was utilized (Figure). An initial list of potential curricular topics was formulated by an expert panel consisting of 5 dermatologists from the Association of Professors of Dermatology DEI subcommittee and the American Academy of Dermatology Diversity Task Force (A.M.A., S.B., R.V., S.D.W., J.I.S.). Initial topics were selected via several meetings among the panel members to discuss existing DEI concerns and issues that were deemed relevant due to education gaps in residency training. The list of topics was further expanded with recommendations obtained via an email sent to dermatology program directors on the Association of Professors of Dermatology listserve, which solicited voluntary participation of academic dermatologists, including program directors and dermatology residents.

There were 2 voting rounds, with each round consisting of questions scored on a Likert scale ranging from 1 to 5 (1=not essential, 2=probably not essential, 3=neutral, 4=probably essential, 5=definitely essential). The inclusion criteria to classify a topic as necessary for integration into the dermatology residency curriculum included 95% (18/19) or more of respondents rating the topic as probably essential or definitely essential; if more than 90% (17/19) of respondents rated the topic as probably essential or definitely essential and less than 10% (2/19) rated it as not essential or probably not essential, the topic was still included as part of the suggested curriculum. Topics that received ratings of probably essential or definitely essential by less than 80% (15/19) of respondents were removed from consideration. The topics that did not meet inclusion or exclusion criteria during the first round of voting were refined by the e-Delphi steering committee (V.S.E-C. and F-A.R.) based on open-ended feedback from the voting group provided at the end of the survey and subsequently passed to the second round of voting.

Results

Participants—A total of 19 respondents participated in both voting rounds, the majority (80% [15/19]) of whom were program directors or dermatologists affiliated with academia or development of DEI education; the remaining 20% [4/19]) were dermatology residents.

Open-Ended Feedback—Voting group members were able to provide open-ended feedback for each of the sets of topics after the survey, which the steering committee utilized to modify the topics as needed for the final voting round. For example, “structural racism/discrimination” was originally mentioned as a topic, but several participants suggested including specific types of racism; therefore, the wording was changed to “racism: types, definitions” to encompass broader definitions and types of racism.

Survey Results—Two genres of topics were surveyed in each voting round: clinical and nonclinical. Participants voted on a total of 61 topics, with 23 ultimately selected in the final list of consensus curricular topics. Of those, 9 were clinical and 14 nonclinical. All topics deemed necessary for inclusion in residency curricula are presented in eTables 1 and 2.

During the first round of voting, the e-Delphi panel reached a consensus to include the following 17 topics as essential to dermatology residency training (along with the percentage of voters who classified them as probably essential or definitely essential): how to mitigate bias in clinical and workplace settings (100% [40/40]); social determinants of health-related disparities in dermatology (100% [40/40]); hairstyling practices across different hair textures (100% [40/40]); definitions and examples of microaggressions (97.50% [39/40]); definition, background, and types of bias (97.50% [39/40]); manifestations of bias in the clinical setting (97.44% [38/39]); racial and ethnic disparities in dermatology (97.44% [38/39]); keloids (97.37% [37/38]); differences in dermoscopic presentations in skin of color (97.30% [36/37]); skin cancer in patients with skin of color (97.30% [36/37]); disparities due to bias (95.00% [38/40]); how to apply cultural humility and safety to patients of different cultural backgrounds (94.87% [37/40]); best practices in providing care to patients with limited English proficiency (94.87% [37/40]); hair loss in patients with textured hair (94.74% [36/38]); pseudofolliculitis barbae and acne keloidalis nuchae (94.60% [35/37]); disparities regarding people experiencing homelessness (92.31% [36/39]); and definitions and types of racism and other forms of discrimination (92.31% [36/39]). eTable 1 provides a list of suggested resources to incorporate these topics into the educational components of residency curricula. The resources provided were not part of the voting process, and they were not considered in the consensus analysis; they are included here as suggested educational catalysts.

During the second round of voting, 25 topics were evaluated. Of those, the following 6 topics were proposed to be included as essential in residency training: differences in prevalence and presentation of common inflammatory disorders (100% [29/29]); manifestations of bias in the learning environment (96.55%); antiracist action and how to decrease the effects of structural racism in clinical and educational settings (96.55% [28/29]); diversity of images in dermatology education (96.55% [28/29]); pigmentary disorders and their psychological effects (96.55% [28/29]); and LGBTQ (lesbian, gay, bisexual, transgender, and queer) dermatologic health care (96.55% [28/29]). eTable 2 includes these topics as well as suggested resources to help incorporate them into training.

Comment

This study utilized a modified e-Delphi technique to identify relevant clinical and nonclinical DEI topics that should be incorporated into dermatology residency curricula. The panel members reached a consensus for 9 clinical DEI-related topics. The respondents agreed that the topics related to skin and hair conditions in patients with skin of color as well as textured hair were crucial to residency education. Skin cancer, hair loss, pseudofolliculitis barbae, acne keloidalis nuchae, keloids, pigmentary disorders, and their varying presentations in patients with skin of color were among the recommended topics. The panel also recommended educating residents on the variable visual presentations of inflammatory conditions in skin of color. Addressing the needs of diverse patients—for example, those belonging to the LGBTQ community—also was deemed important for inclusion.

The remaining 14 chosen topics were nonclinical items addressing concepts such as bias and health care disparities as well as cultural humility and safety.⁹ Cultural humility and safety focus on developing cultural awareness by creating a safe setting for patients rather than encouraging power relationships between them and their physicians. Various topics related to racism also were recommended to be included in residency curricula, including education on implementation of antiracist action in the workplace.

Many of the nonclinical topics are intertwined; for instance, learning about health care disparities in patients with limited English proficiency allows for improved best practices in delivering care to patients from this population. The first step in overcoming bias and subsequent disparities is acknowledging how the perpetuation of bias leads to disparities after being taught tools to recognize it.

Our group’s guidance on DEI topics should help dermatology residency program leaders as they design and refine program curricula. There are multiple avenues for incorporating education on these topics, including lectures, interactive workshops, role-playing sessions, book or journal clubs, and discussion circles. Many of these topics/programs may already be included in programs’ didactic curricula, which would minimize the burden of finding space to educate on these topics. Institutional cultural change is key to ensuring truly diverse, equitable, and inclusive workplaces. Educating tomorrow’s dermatologists on these topics is a first step toward achieving that cultural change.

Limitations—A limitation of this e-Delphi survey is that only a selection of experts in this field was included. Additionally, we were concerned that the Likert scale format and the bar we set for inclusion and exclusion may have failed to adequately capture participants’ nuanced opinions. As such, participants were able to provide open-ended feedback, and suggestions for alternate wording or other changes were considered by the steering committee. Finally, inclusion recommendations identified in this survey were developed specifically for US dermatology residents.

Conclusion

In this e-Delphi consensus assessment of DEI-related topics, we recommend the inclusion of 23 topics into dermatology residency program curricula to improve medical training and the patient-physician relationship as well as to create better health outcomes. We also provide specific sample resource recommendations in eTables 1 and 2 to facilitate inclusion of these topics into residency curricula across the country.

Impressive progress has been made in recent years in the management and treatment of atopic dermatitis (AD) and its comorbidities; however, there is a major need for state-of-the-art, evidence-based, multidisciplinary education for AD management. To address this need, the first Revolutionizing Atopic Dermatitis (RAD) Conference was held in April 2019 in Chicago, Illinois, featuring cutting-edge research presented by globally recognized experts in dermatology, allergy and immunology, sleep medicine, ophthalmology, and nursing care. The following is a recap of the latest topics in AD research presented at the conference.

Diagnosis and Assessment of AD: Jonathan I. Silverberg, MD, PhD, MPH

Although diagnosis of AD typically is straightforward in children, it can be challenging in adults, even for expert clinicians. These challenges stem from the different lesional distribution and morphology of AD in adults vs children.^1,2 Additionally, the conditions included in the differential diagnosis of AD (eg, allergic contact dermatitis, cutaneous T-cell lymphoma, psoriasis) are far more common in adults than in children. Formal diagnostic criteria can be useful to improve the diagnosis of AD in clinical practice.³ It is important to note that flexural lesions and early disease onset are diagnostic criteria in AD; nevertheless, neither are essential nor sufficient on their own to make the diagnosis.

Patch Testing: Jacob P. Thyssen, MD, PhD, DmSci, and Noreen Heer Nicol, PhD, RN, FNP, NEA-BC

Patch testing can be used in AD patients to rule out contact dermatitis as an alternative or comorbid diagnosis.^4-6 Because contact dermatitis can mimic AD, patch testing is recommended for all patients with adolescent and adult-onset AD.⁵ Additionally, refractory cases of AD across all ages, especially prior to initiation of systemic therapy, warrant patch testing. The unique challenges of patch testing in AD patients were reviewed.

Patient Panel

Atopic dermatitis can be a considerable disease burden on both patients and society in general. At the 2019 RAD Conference, a panel of patients bravely shared their AD journeys. Their eye-opening stories highlighted opportunities for improving real-world assessment and management of AD. Some key takeaways included the importance of adequately assessing the symptom burden of AD and not merely relying on visual inspection of the skin. The need for long-term treatment approaches beyond quick fixes with steroids also was discussed.

Pathogenesis of AD: Mark Boguniewicz, MD

There have been many advances in our understanding of the complex pathogenesis of AD,^7-11 which is characterized by an altered skin barrier and immune dysregulation. Filaggrin deficiency in the skin has structural and biophysical consequences. A subset of patients with AD has filaggrin loss-of-function genetic polymorphisms inherited in an autosomal-semidominant pattern; however, many other genetic polymorphisms have been identified that affect different components of the skin architecture and immune system. Many cytokine pathways have been found to be upregulated in AD lesions, including IL-13, IL-4, IL-31, and IL-5 in acute and chronic lesions, and IFN-γ and other helper T cell (T_H1) cytokines in chronic lesions. IL-4 and IL-13 (T_H2 cytokines) have been shown to decrease epidermal expression of filaggrin and lead to lipid abnormalities in the skin of patients with AD. Even normal-appearing, nonlesional skin has substantial immune activation and barrier abnormalities in patients with moderate to severe AD. Activation of different immune pathways may contribute to the heterogeneous clinical presentation of AD. There also is an increasingly recognized role of superantigen-producing Staphylococcus aureus and decreased microbial diversity in AD.

Therapies for AD

The advances in our understanding of AD pathophysiology have led to the development of 2 recently approved therapeutic agents.^7-10 Crisaborole ointment 2% is a topical phosphodiesterase 4 inhibitor that was approved by the US Food and Drug Administration in 2016 for treatment of mild to moderate AD. Treatment with crisaborole ointment 2% demonstrated improvement in lesion severity, itch, and quality of life in children and adults with AD. Dupilumab, an injectable biologic therapy that inhibits IL-4 and IL-13 signaling, was approved by the US Food and Drug Administration in 2017 for adults and in 2019 for adolescents aged 12 to 17 years with moderate to severe AD. The expert panel of speakers at the 2019 RAD Conference discussed many practical clinical pearls regarding patient education, optimization of both short- and long-term efficacy, and prevention and management of treatment-related adverse events. The discussion included evidence-based guidelines for bathing practices and topical therapy in AD, as well as practical pearls for patient and provider education in AD, reviewed by Dr. Nicol. Evidence-based guidelines for use of phototherapy and systemic and biologic therapy for AD also were highlighted by Dr. Silverberg.

After decades of limited therapeutic options, there is a large therapeutic pipeline of topical, oral, and biologic agents in development for the treatment AD.^7-9 Dr. Boguniewicz reviewed the state-of-the-art treatments that are the furthest advanced in development. Many of these agents may be approved within the next couple of years and look promising in terms of their potential to improve the care of patients with AD.

Comorbidities of AD

The impact of AD is not just skin deep. Atopic dermatitis is associated with myriad comorbid health conditions.^12-16 Dr. Boguniewicz reviewed the relationship between AD and atopic comorbidities, including asthma, hay fever, and food allergies, which are common across all AD patients. In addition, a subset of children with AD demonstrated the atopic march, in which AD first appears early in life followed by the development of other atopic comorbidities in later childhood or adulthood. In particular, children with filaggrin null mutations were found to be at increased risk of early-onset, severe, persistent AD with asthma and allergic sensitization.¹⁷ More recently, eosinophilic esophagitis was demonstrated to be a late-onset comorbidity of the atopic march.¹⁸ The allergy guidelines for which patients are appropriate candidates for food and/or aeroallergen testing were discussed,¹⁹ and it was emphasized that patients with AD should not routinely receive this testing.

Atopic dermatitis is associated with many other comorbidities, including sleep disturbances. Phyllis C. Zee, MD, PhD, provided a brilliant review of circadian regulation of physiology and the immune system. Sleep is one of the most important determinants of patients’ health and well-being. Atopic dermatitis is associated with disturbances of sleep and circadian rhythms. Sleep disturbances are gaining recognition as an important end point to assess for improvement in clinical practice and trials.

Patients with AD have long been recognized to have increased ophthalmic comorbidities, including allergic conjunctivitis, atopic keratoconjunctivitis, and cataracts. More recently, conjunctivitis has emerged as an important adverse event with dupilumab treatment.²⁰ Jeanine Baqai, MD, reviewed the various ophthalmic comorbidities and shared numerous clinical signs of ophthalmic comorbidities that dermatologists can assess with the naked eye (no slit-lamp examination needed). Pearls to manage dupilumab-related conjunctivitis shared by Dr. Baqai and the speaker panel included elimination of eye rubbing, cold compresses, avoidance of exacerbating factors, artificial tears, and timely referral to an ophthalmologist. Medications discussed were mast cell stabilizers, antihistamines, and corticosteroids and calcineurin inhibitors.

Final Thoughts

There has been an explosion of new research that has increased our understanding of all aspects of AD, and the standard of care is truly being revolutionized. Clinicians should stay tuned to a wealth of new evidence-based recommendations coming down the pike.

Impressive progress has been made in recent years in the management and treatment of atopic dermatitis (AD) and its comorbidities; however, there is a major need for state-of-the-art, evidence-based, multidisciplinary education for AD management. To address this need, the first Revolutionizing Atopic Dermatitis (RAD) Conference was held in April 2019 in Chicago, Illinois, featuring cutting-edge research presented by globally recognized experts in dermatology, allergy and immunology, sleep medicine, ophthalmology, and nursing care. The following is a recap of the latest topics in AD research presented at the conference.

Diagnosis and Assessment of AD: Jonathan I. Silverberg, MD, PhD, MPH

Although diagnosis of AD typically is straightforward in children, it can be challenging in adults, even for expert clinicians. These challenges stem from the different lesional distribution and morphology of AD in adults vs children.^1,2 Additionally, the conditions included in the differential diagnosis of AD (eg, allergic contact dermatitis, cutaneous T-cell lymphoma, psoriasis) are far more common in adults than in children. Formal diagnostic criteria can be useful to improve the diagnosis of AD in clinical practice.³ It is important to note that flexural lesions and early disease onset are diagnostic criteria in AD; nevertheless, neither are essential nor sufficient on their own to make the diagnosis.

Patch Testing: Jacob P. Thyssen, MD, PhD, DmSci, and Noreen Heer Nicol, PhD, RN, FNP, NEA-BC

Patch testing can be used in AD patients to rule out contact dermatitis as an alternative or comorbid diagnosis.^4-6 Because contact dermatitis can mimic AD, patch testing is recommended for all patients with adolescent and adult-onset AD.⁵ Additionally, refractory cases of AD across all ages, especially prior to initiation of systemic therapy, warrant patch testing. The unique challenges of patch testing in AD patients were reviewed.

Patient Panel

Atopic dermatitis can be a considerable disease burden on both patients and society in general. At the 2019 RAD Conference, a panel of patients bravely shared their AD journeys. Their eye-opening stories highlighted opportunities for improving real-world assessment and management of AD. Some key takeaways included the importance of adequately assessing the symptom burden of AD and not merely relying on visual inspection of the skin. The need for long-term treatment approaches beyond quick fixes with steroids also was discussed.

Pathogenesis of AD: Mark Boguniewicz, MD

There have been many advances in our understanding of the complex pathogenesis of AD,^7-11 which is characterized by an altered skin barrier and immune dysregulation. Filaggrin deficiency in the skin has structural and biophysical consequences. A subset of patients with AD has filaggrin loss-of-function genetic polymorphisms inherited in an autosomal-semidominant pattern; however, many other genetic polymorphisms have been identified that affect different components of the skin architecture and immune system. Many cytokine pathways have been found to be upregulated in AD lesions, including IL-13, IL-4, IL-31, and IL-5 in acute and chronic lesions, and IFN-γ and other helper T cell (T_H1) cytokines in chronic lesions. IL-4 and IL-13 (T_H2 cytokines) have been shown to decrease epidermal expression of filaggrin and lead to lipid abnormalities in the skin of patients with AD. Even normal-appearing, nonlesional skin has substantial immune activation and barrier abnormalities in patients with moderate to severe AD. Activation of different immune pathways may contribute to the heterogeneous clinical presentation of AD. There also is an increasingly recognized role of superantigen-producing Staphylococcus aureus and decreased microbial diversity in AD.

Therapies for AD

The advances in our understanding of AD pathophysiology have led to the development of 2 recently approved therapeutic agents.^7-10 Crisaborole ointment 2% is a topical phosphodiesterase 4 inhibitor that was approved by the US Food and Drug Administration in 2016 for treatment of mild to moderate AD. Treatment with crisaborole ointment 2% demonstrated improvement in lesion severity, itch, and quality of life in children and adults with AD. Dupilumab, an injectable biologic therapy that inhibits IL-4 and IL-13 signaling, was approved by the US Food and Drug Administration in 2017 for adults and in 2019 for adolescents aged 12 to 17 years with moderate to severe AD. The expert panel of speakers at the 2019 RAD Conference discussed many practical clinical pearls regarding patient education, optimization of both short- and long-term efficacy, and prevention and management of treatment-related adverse events. The discussion included evidence-based guidelines for bathing practices and topical therapy in AD, as well as practical pearls for patient and provider education in AD, reviewed by Dr. Nicol. Evidence-based guidelines for use of phototherapy and systemic and biologic therapy for AD also were highlighted by Dr. Silverberg.

After decades of limited therapeutic options, there is a large therapeutic pipeline of topical, oral, and biologic agents in development for the treatment AD.^7-9 Dr. Boguniewicz reviewed the state-of-the-art treatments that are the furthest advanced in development. Many of these agents may be approved within the next couple of years and look promising in terms of their potential to improve the care of patients with AD.

Comorbidities of AD

The impact of AD is not just skin deep. Atopic dermatitis is associated with myriad comorbid health conditions.^12-16 Dr. Boguniewicz reviewed the relationship between AD and atopic comorbidities, including asthma, hay fever, and food allergies, which are common across all AD patients. In addition, a subset of children with AD demonstrated the atopic march, in which AD first appears early in life followed by the development of other atopic comorbidities in later childhood or adulthood. In particular, children with filaggrin null mutations were found to be at increased risk of early-onset, severe, persistent AD with asthma and allergic sensitization.¹⁷ More recently, eosinophilic esophagitis was demonstrated to be a late-onset comorbidity of the atopic march.¹⁸ The allergy guidelines for which patients are appropriate candidates for food and/or aeroallergen testing were discussed,¹⁹ and it was emphasized that patients with AD should not routinely receive this testing.

Atopic dermatitis is associated with many other comorbidities, including sleep disturbances. Phyllis C. Zee, MD, PhD, provided a brilliant review of circadian regulation of physiology and the immune system. Sleep is one of the most important determinants of patients’ health and well-being. Atopic dermatitis is associated with disturbances of sleep and circadian rhythms. Sleep disturbances are gaining recognition as an important end point to assess for improvement in clinical practice and trials.

Patients with AD have long been recognized to have increased ophthalmic comorbidities, including allergic conjunctivitis, atopic keratoconjunctivitis, and cataracts. More recently, conjunctivitis has emerged as an important adverse event with dupilumab treatment.²⁰ Jeanine Baqai, MD, reviewed the various ophthalmic comorbidities and shared numerous clinical signs of ophthalmic comorbidities that dermatologists can assess with the naked eye (no slit-lamp examination needed). Pearls to manage dupilumab-related conjunctivitis shared by Dr. Baqai and the speaker panel included elimination of eye rubbing, cold compresses, avoidance of exacerbating factors, artificial tears, and timely referral to an ophthalmologist. Medications discussed were mast cell stabilizers, antihistamines, and corticosteroids and calcineurin inhibitors.

Final Thoughts

There has been an explosion of new research that has increased our understanding of all aspects of AD, and the standard of care is truly being revolutionized. Clinicians should stay tuned to a wealth of new evidence-based recommendations coming down the pike.

Impressive progress has been made in recent years in the management and treatment of atopic dermatitis (AD) and its comorbidities; however, there is a major need for state-of-the-art, evidence-based, multidisciplinary education for AD management. To address this need, the first Revolutionizing Atopic Dermatitis (RAD) Conference was held in April 2019 in Chicago, Illinois, featuring cutting-edge research presented by globally recognized experts in dermatology, allergy and immunology, sleep medicine, ophthalmology, and nursing care. The following is a recap of the latest topics in AD research presented at the conference.

Diagnosis and Assessment of AD: Jonathan I. Silverberg, MD, PhD, MPH

Although diagnosis of AD typically is straightforward in children, it can be challenging in adults, even for expert clinicians. These challenges stem from the different lesional distribution and morphology of AD in adults vs children.^1,2 Additionally, the conditions included in the differential diagnosis of AD (eg, allergic contact dermatitis, cutaneous T-cell lymphoma, psoriasis) are far more common in adults than in children. Formal diagnostic criteria can be useful to improve the diagnosis of AD in clinical practice.³ It is important to note that flexural lesions and early disease onset are diagnostic criteria in AD; nevertheless, neither are essential nor sufficient on their own to make the diagnosis.

Patch Testing: Jacob P. Thyssen, MD, PhD, DmSci, and Noreen Heer Nicol, PhD, RN, FNP, NEA-BC

Patch testing can be used in AD patients to rule out contact dermatitis as an alternative or comorbid diagnosis.^4-6 Because contact dermatitis can mimic AD, patch testing is recommended for all patients with adolescent and adult-onset AD.⁵ Additionally, refractory cases of AD across all ages, especially prior to initiation of systemic therapy, warrant patch testing. The unique challenges of patch testing in AD patients were reviewed.

Patient Panel

Atopic dermatitis can be a considerable disease burden on both patients and society in general. At the 2019 RAD Conference, a panel of patients bravely shared their AD journeys. Their eye-opening stories highlighted opportunities for improving real-world assessment and management of AD. Some key takeaways included the importance of adequately assessing the symptom burden of AD and not merely relying on visual inspection of the skin. The need for long-term treatment approaches beyond quick fixes with steroids also was discussed.

Pathogenesis of AD: Mark Boguniewicz, MD

There have been many advances in our understanding of the complex pathogenesis of AD,^7-11 which is characterized by an altered skin barrier and immune dysregulation. Filaggrin deficiency in the skin has structural and biophysical consequences. A subset of patients with AD has filaggrin loss-of-function genetic polymorphisms inherited in an autosomal-semidominant pattern; however, many other genetic polymorphisms have been identified that affect different components of the skin architecture and immune system. Many cytokine pathways have been found to be upregulated in AD lesions, including IL-13, IL-4, IL-31, and IL-5 in acute and chronic lesions, and IFN-γ and other helper T cell (T_H1) cytokines in chronic lesions. IL-4 and IL-13 (T_H2 cytokines) have been shown to decrease epidermal expression of filaggrin and lead to lipid abnormalities in the skin of patients with AD. Even normal-appearing, nonlesional skin has substantial immune activation and barrier abnormalities in patients with moderate to severe AD. Activation of different immune pathways may contribute to the heterogeneous clinical presentation of AD. There also is an increasingly recognized role of superantigen-producing Staphylococcus aureus and decreased microbial diversity in AD.

Therapies for AD

The advances in our understanding of AD pathophysiology have led to the development of 2 recently approved therapeutic agents.^7-10 Crisaborole ointment 2% is a topical phosphodiesterase 4 inhibitor that was approved by the US Food and Drug Administration in 2016 for treatment of mild to moderate AD. Treatment with crisaborole ointment 2% demonstrated improvement in lesion severity, itch, and quality of life in children and adults with AD. Dupilumab, an injectable biologic therapy that inhibits IL-4 and IL-13 signaling, was approved by the US Food and Drug Administration in 2017 for adults and in 2019 for adolescents aged 12 to 17 years with moderate to severe AD. The expert panel of speakers at the 2019 RAD Conference discussed many practical clinical pearls regarding patient education, optimization of both short- and long-term efficacy, and prevention and management of treatment-related adverse events. The discussion included evidence-based guidelines for bathing practices and topical therapy in AD, as well as practical pearls for patient and provider education in AD, reviewed by Dr. Nicol. Evidence-based guidelines for use of phototherapy and systemic and biologic therapy for AD also were highlighted by Dr. Silverberg.

After decades of limited therapeutic options, there is a large therapeutic pipeline of topical, oral, and biologic agents in development for the treatment AD.^7-9 Dr. Boguniewicz reviewed the state-of-the-art treatments that are the furthest advanced in development. Many of these agents may be approved within the next couple of years and look promising in terms of their potential to improve the care of patients with AD.

Comorbidities of AD

The impact of AD is not just skin deep. Atopic dermatitis is associated with myriad comorbid health conditions.^12-16 Dr. Boguniewicz reviewed the relationship between AD and atopic comorbidities, including asthma, hay fever, and food allergies, which are common across all AD patients. In addition, a subset of children with AD demonstrated the atopic march, in which AD first appears early in life followed by the development of other atopic comorbidities in later childhood or adulthood. In particular, children with filaggrin null mutations were found to be at increased risk of early-onset, severe, persistent AD with asthma and allergic sensitization.¹⁷ More recently, eosinophilic esophagitis was demonstrated to be a late-onset comorbidity of the atopic march.¹⁸ The allergy guidelines for which patients are appropriate candidates for food and/or aeroallergen testing were discussed,¹⁹ and it was emphasized that patients with AD should not routinely receive this testing.

Atopic dermatitis is associated with many other comorbidities, including sleep disturbances. Phyllis C. Zee, MD, PhD, provided a brilliant review of circadian regulation of physiology and the immune system. Sleep is one of the most important determinants of patients’ health and well-being. Atopic dermatitis is associated with disturbances of sleep and circadian rhythms. Sleep disturbances are gaining recognition as an important end point to assess for improvement in clinical practice and trials.

Patients with AD have long been recognized to have increased ophthalmic comorbidities, including allergic conjunctivitis, atopic keratoconjunctivitis, and cataracts. More recently, conjunctivitis has emerged as an important adverse event with dupilumab treatment.²⁰ Jeanine Baqai, MD, reviewed the various ophthalmic comorbidities and shared numerous clinical signs of ophthalmic comorbidities that dermatologists can assess with the naked eye (no slit-lamp examination needed). Pearls to manage dupilumab-related conjunctivitis shared by Dr. Baqai and the speaker panel included elimination of eye rubbing, cold compresses, avoidance of exacerbating factors, artificial tears, and timely referral to an ophthalmologist. Medications discussed were mast cell stabilizers, antihistamines, and corticosteroids and calcineurin inhibitors.

Final Thoughts

There has been an explosion of new research that has increased our understanding of all aspects of AD, and the standard of care is truly being revolutionized. Clinicians should stay tuned to a wealth of new evidence-based recommendations coming down the pike.

Atopic dermatitis (AD) is a vexing multisystem disorder characterized by frequently recurrent, intrusive, and sometimes disabling itch and dermatitis. The itch may be present throughout the day but crescendos at bedtime or 1 to 2 hours after sleep initiation, resulting in disrupted sleep cycles, lack of rest, more hours scratching, daytime somnolence, poor work attendance and performance, and poor school attendance and performance.¹

Atopic dermatitis is a lifelong disease that only remits in approximately half of patients.² There is a need for a disease-specific systemic drug in AD. Phototherapy, cyclosporine, methotrexate, and azathioprine are nonspecific immunosuppressive agents that can be used off label for AD but may or may not be effective.³ Oral or intramuscular corticosteroids are associated with problematic side effects such as weight gain, osteoporosis, fractures, psychological problems, striae, buffalo hump, and steroid withdrawal symptoms and disease aggravation upon withdrawal (ie, flaring to a state worse than prior to steroid initiation).^3,4

A biologic medication for AD has been long overdue. Psoriatic biologic medications have been tried in AD with occasional benefit in case reports but no major response in larger trials. Belloni et al⁵ reviewed early data on off-label usage of biologics approved by the US Food and Drug Administration for psoriasis or other indications applied to AD patients. In their review of cases, they make the point that results are variable and anti-B-cell activity may hold the greatest promise.⁵ On the other hand, a recent series of 3 patients showed limited response to rituximab in chronic AD,⁶ while a combination of omalizumab, an anti-IgE medication, and rituximab was helpful in some patients.⁷ Ultimately, the issue is that nonspecific biologics may or may not address the underlying disease factors in AD. Therefore, there has been a true need for biologic intervention targeted directly at the pathogenic mechanism of AD. Furthermore, the desire for a biologic targeted at AD is paired with the true need to have a medication so targeted that the drug would have little effect on the rest of the immune system, resulting in targeted immunomodulation without secondary risk of infections.

Wait no longer, that era arrived a few months ago with the rapid US Food and Drug Administration approval of dupilumab, an injectable medication used every 2 weeks for the therapy of moderate to severe AD. This fully human monoclonal antibody against the IL-4Rα subunit blocks IL-4 and IL-13, key inflammatory agents in the triggering of production of IgE and eosinophil activation. Even better than the fact that it is targeted are the excellent outcomes in the therapy of moderate to severe AD in adults and the minimal side-effect profile resulting in no requirements for laboratory screening or ongoing monitoring.⁸

Dupilumab seems to perform well, both clinically and in improving the lives of AD patients. Meta-analysis of trials involving dupilumab has shown improved health-related quality of life outcomes.^9,10 Usage of dupilumab alone in clinical trials for 16 weeks (SOLO 1 and SOLO 2) has resulted in stunning reduction in disease severity with a limited side-effect profile, with patients most commonly reporting conjunctivitis.¹¹ In real-world models where dupilumab is added into a regimen of topical corticosteroid usage (LIBERTY AD CHRONOS trial), patients fared even better with the combination, highlighting that this medication may best be used adjunctively to our skin care guidance as dermatologists.¹²

A new era for AD patients has arrived and we as practitioners are now fortunate to be able to therapeutically reach the worst cases of AD. The new era has only begun with dozens of new agents addressing a variety of interleukin pathways including IL-17 and IL-22 still under development. Ultimately, we hope that ongoing pediatric trials will allow us to glean the role of early disease intervention at the root cause of AD and address our abilities to prevent comorbidities and disease persistence. Will we be able to avert years of disabling disease? The future holds immense hope.

Atopic dermatitis (AD) is a vexing multisystem disorder characterized by frequently recurrent, intrusive, and sometimes disabling itch and dermatitis. The itch may be present throughout the day but crescendos at bedtime or 1 to 2 hours after sleep initiation, resulting in disrupted sleep cycles, lack of rest, more hours scratching, daytime somnolence, poor work attendance and performance, and poor school attendance and performance.¹

Atopic dermatitis is a lifelong disease that only remits in approximately half of patients.² There is a need for a disease-specific systemic drug in AD. Phototherapy, cyclosporine, methotrexate, and azathioprine are nonspecific immunosuppressive agents that can be used off label for AD but may or may not be effective.³ Oral or intramuscular corticosteroids are associated with problematic side effects such as weight gain, osteoporosis, fractures, psychological problems, striae, buffalo hump, and steroid withdrawal symptoms and disease aggravation upon withdrawal (ie, flaring to a state worse than prior to steroid initiation).^3,4

A biologic medication for AD has been long overdue. Psoriatic biologic medications have been tried in AD with occasional benefit in case reports but no major response in larger trials. Belloni et al⁵ reviewed early data on off-label usage of biologics approved by the US Food and Drug Administration for psoriasis or other indications applied to AD patients. In their review of cases, they make the point that results are variable and anti-B-cell activity may hold the greatest promise.⁵ On the other hand, a recent series of 3 patients showed limited response to rituximab in chronic AD,⁶ while a combination of omalizumab, an anti-IgE medication, and rituximab was helpful in some patients.⁷ Ultimately, the issue is that nonspecific biologics may or may not address the underlying disease factors in AD. Therefore, there has been a true need for biologic intervention targeted directly at the pathogenic mechanism of AD. Furthermore, the desire for a biologic targeted at AD is paired with the true need to have a medication so targeted that the drug would have little effect on the rest of the immune system, resulting in targeted immunomodulation without secondary risk of infections.

Wait no longer, that era arrived a few months ago with the rapid US Food and Drug Administration approval of dupilumab, an injectable medication used every 2 weeks for the therapy of moderate to severe AD. This fully human monoclonal antibody against the IL-4Rα subunit blocks IL-4 and IL-13, key inflammatory agents in the triggering of production of IgE and eosinophil activation. Even better than the fact that it is targeted are the excellent outcomes in the therapy of moderate to severe AD in adults and the minimal side-effect profile resulting in no requirements for laboratory screening or ongoing monitoring.⁸

Dupilumab seems to perform well, both clinically and in improving the lives of AD patients. Meta-analysis of trials involving dupilumab has shown improved health-related quality of life outcomes.^9,10 Usage of dupilumab alone in clinical trials for 16 weeks (SOLO 1 and SOLO 2) has resulted in stunning reduction in disease severity with a limited side-effect profile, with patients most commonly reporting conjunctivitis.¹¹ In real-world models where dupilumab is added into a regimen of topical corticosteroid usage (LIBERTY AD CHRONOS trial), patients fared even better with the combination, highlighting that this medication may best be used adjunctively to our skin care guidance as dermatologists.¹²

A new era for AD patients has arrived and we as practitioners are now fortunate to be able to therapeutically reach the worst cases of AD. The new era has only begun with dozens of new agents addressing a variety of interleukin pathways including IL-17 and IL-22 still under development. Ultimately, we hope that ongoing pediatric trials will allow us to glean the role of early disease intervention at the root cause of AD and address our abilities to prevent comorbidities and disease persistence. Will we be able to avert years of disabling disease? The future holds immense hope.

Atopic dermatitis (AD) is a vexing multisystem disorder characterized by frequently recurrent, intrusive, and sometimes disabling itch and dermatitis. The itch may be present throughout the day but crescendos at bedtime or 1 to 2 hours after sleep initiation, resulting in disrupted sleep cycles, lack of rest, more hours scratching, daytime somnolence, poor work attendance and performance, and poor school attendance and performance.¹

Atopic dermatitis is a lifelong disease that only remits in approximately half of patients.² There is a need for a disease-specific systemic drug in AD. Phototherapy, cyclosporine, methotrexate, and azathioprine are nonspecific immunosuppressive agents that can be used off label for AD but may or may not be effective.³ Oral or intramuscular corticosteroids are associated with problematic side effects such as weight gain, osteoporosis, fractures, psychological problems, striae, buffalo hump, and steroid withdrawal symptoms and disease aggravation upon withdrawal (ie, flaring to a state worse than prior to steroid initiation).^3,4

A biologic medication for AD has been long overdue. Psoriatic biologic medications have been tried in AD with occasional benefit in case reports but no major response in larger trials. Belloni et al⁵ reviewed early data on off-label usage of biologics approved by the US Food and Drug Administration for psoriasis or other indications applied to AD patients. In their review of cases, they make the point that results are variable and anti-B-cell activity may hold the greatest promise.⁵ On the other hand, a recent series of 3 patients showed limited response to rituximab in chronic AD,⁶ while a combination of omalizumab, an anti-IgE medication, and rituximab was helpful in some patients.⁷ Ultimately, the issue is that nonspecific biologics may or may not address the underlying disease factors in AD. Therefore, there has been a true need for biologic intervention targeted directly at the pathogenic mechanism of AD. Furthermore, the desire for a biologic targeted at AD is paired with the true need to have a medication so targeted that the drug would have little effect on the rest of the immune system, resulting in targeted immunomodulation without secondary risk of infections.

Wait no longer, that era arrived a few months ago with the rapid US Food and Drug Administration approval of dupilumab, an injectable medication used every 2 weeks for the therapy of moderate to severe AD. This fully human monoclonal antibody against the IL-4Rα subunit blocks IL-4 and IL-13, key inflammatory agents in the triggering of production of IgE and eosinophil activation. Even better than the fact that it is targeted are the excellent outcomes in the therapy of moderate to severe AD in adults and the minimal side-effect profile resulting in no requirements for laboratory screening or ongoing monitoring.⁸

Dupilumab seems to perform well, both clinically and in improving the lives of AD patients. Meta-analysis of trials involving dupilumab has shown improved health-related quality of life outcomes.^9,10 Usage of dupilumab alone in clinical trials for 16 weeks (SOLO 1 and SOLO 2) has resulted in stunning reduction in disease severity with a limited side-effect profile, with patients most commonly reporting conjunctivitis.¹¹ In real-world models where dupilumab is added into a regimen of topical corticosteroid usage (LIBERTY AD CHRONOS trial), patients fared even better with the combination, highlighting that this medication may best be used adjunctively to our skin care guidance as dermatologists.¹²

A new era for AD patients has arrived and we as practitioners are now fortunate to be able to therapeutically reach the worst cases of AD. The new era has only begun with dozens of new agents addressing a variety of interleukin pathways including IL-17 and IL-22 still under development. Ultimately, we hope that ongoing pediatric trials will allow us to glean the role of early disease intervention at the root cause of AD and address our abilities to prevent comorbidities and disease persistence. Will we be able to avert years of disabling disease? The future holds immense hope.

Atopic dermatitis (AD) is a multifactorial inflammatory disorder with an estimated prevalence of 279,889,120 cases worldwide.1 Most cases of AD begin in early childhood (with almost 85% developing by 5 years of age),² but recent studies have found that 40% to over 80% of cases persist into adulthood.^1,3,4 Although a previous study focused largely on T helper type 1/T helper type 2 (Th2) immune dysregulation as the pathogenesis of the disease,⁵ disruption of the skin barrier and systemic inflammation are at the center of current AD research. In AD, breakdown of the skin barrier results in increased transepidermal water loss, reduced skin hydration, and increased antigen presentation by Langerhans cells initiating inflammation.6-8 The cascade largely activated is the Th2 and T helper type 22 cascade with resultant cytokine release (ie, IL-4, IL-13, IL-2, IL-8, IL-10, IL-17, IL-22, tumor necrosis factor α, interferon γ).^9,10 In active AD, Th2 inflammation and barrier breakdown result in reduced filaggrin and claudin 1 expression, resulting in further exacerbation of the barrier defect and enhancing the risk of development of asthma and hay fever as well as transcutaneous sensitization to a variety of food allergens (eg, peanuts).^9,11,12 Although all of these immunologic features are well established in AD, controversy remains as to whether AD is caused by systemic inflammation triggering barrier dysfunction (the “inside-out” hypothesis) or from the epidermal skin barrier disruption triggering immunologic imbalance (the “outside-in” hypothesis).

Inside-Out Hypothesis

While barrier impairment appears to occur in all patients with AD, it still is unclear how AD begins. The inside-out hypothesis suggests that cutaneous inflammation precedes barrier impairment and in fact may result in an impaired skin barrier. It has previously been reported that inflammatory states weaken the barrier by downregulating filaggrin production in the skin.¹³ Barrier disruption may be accompanied by transcutaneous penetration of allergens and increased Staphylococcus aureus counts. Recently, mutations and polymorphisms of inflammatory genes have been linked to AD (eg, single nucleotide polymorphisms of the IL4RA [interleukin 4 receptor, alpha] and CD14 [cluster of differentiation 14] genes, the serine protease inhibitor SPINK5 [serine peptidase inhibitor, Kazal type 5], RANTES [chemokine (C-C motif) ligand 5], IL-4, IL-13).¹⁴ These alterations highlight the role of systemic inflammation in triggering AD.

Outside-In Hypothesis

The outside-in hypothesis suggests that the impaired skin barrier precedes AD and is required for immune dysregulation to occur. This hypothesis was largely advanced by a study demonstrating that deactivating mutations of the filaggrin gene were linked to nearly 20% of AD cases in Northern European populations.15 Filaggrin (chromosome 1q21.3) performs an essential function in the skin barrier through its differential cleavage and the breakdown and release of natural moisturizing factor.¹⁶ Filaggrin gene mutations are associated with persistent AD, and it has been posited that environmental factors such as temperature and humidity also can affect filaggrin production as it relates to barrier function.17-19 Skin barrier disruption results in increased cutaneous and systemic Th2 responses (ie, IL-4/13), with thymic stromal lymphopoietin as the potential mechanism of Th2 cell recruitment.^10,20 Inflammatory Th2 cells triggered by an impaired skin barrier also may predispose patients to the development of allergic diseases such as asthma, in line with Atopic March, or the progression of AD to other forms of atopy (eg, food allergy, asthma).^5,7,21-23

The outside-in hypothesis may only explain the root pathogenesis of AD in a subset of patients, however, as only 1 in 5 cases of AD in Northern European and Asian populations are associated with underlying filaggrin mutations (which are only present in about 10% of those who are unaffected by AD).¹⁵ Filaggrin does not appear to account for the basis of AD in all cases. In a study of 762 newborns in Cincinnati, Ohio, 39% of children with at least one parent with atopy developed AD by 3 years of age, about quadruple of what would be projected based on filaggrin defects in general population studies, which are noted in only about 10% of white individuals.²⁴ Furthermore, less than 5% of patients of African descent have mutations of the filaggrin 1 gene.²⁵

Implications for the Prevention and Treatment of Atopic Dermatitis

Preventative strategies for AD currently are in development. Atopic dermatitis may be unpreventable because the in utero environment triggers some of the barrier alterations, which can be noted as early as 2 days following birth and will predict early-onset AD. The putative mechanism is via Th2 cytokines (IL-4, IL-13).²⁶

Certainly, application of over-the-counter and prescription emollients are mainstays of treatment for AD and may suffice as monotherapy in cases of mild disease. In a recent randomized trial in the United States and the United Kingdom, emollients were used in newborns considered at high risk for AD (family history of atopy) until 6 months of age.²⁷ The risk of AD development was reduced by half, irrespective of the emollient used. Unfortunately, 21.8% of children without a family history of atopy will develop AD; therefore, not all cases can be prevented if use of emollients is limited to newborns with a family history of atopy.²⁸ Long-term follow-up is needed to track whether emollient use in newborns will prevent AD indefinitely.

Prevention of AD onset using systemic interventions has also been investigated. Probiotics have been suggested as a means to modify the gut microbiota and reduce systemic and mucosal inflammation. Lactobacillus reuteri taken prenatally by pregnant women and by newborns has shown mild benefit in preventing some forms of AD.29 Although they are not approved by the US Food and Drug Administration for this indication, systemic interventions for moderate-to-severe AD such as methotrexate and cyclosporine certainly have shown benefit in managing ongoing illness and breaking the cycle of disease.³⁰ The efficacy of these agents points to the role of systemic inflammation in ongoing AD activity. Moreover, the inside-out hypothesis recently has led to the proliferation of promising new therapeutic agents in the pipeline to treat the systemic Th2 inflammation that occurs in severe AD (eg, anti–IL-4/13 receptor antibody, anti–IL-13 antibodies, and biologics targeting IL-12/23, IL-22, and IL-31 receptors).³¹

Final Thoughts

Atopic dermatitis is a multifactorial disease associated with barrier disruption and intense systemic inflammation. It is likely that both the inside-out and outside-in hypotheses hold true in different subsets of AD patients. It is clear that some individuals are born with filaggrin defects that sufficiently trigger systemic inflammation, resulting in AD. On the other hand, there are clearly some individuals with inflammatory dysregulation that results in systemic inflammation and secondary barrier disruption. Until we can determine the genomic triggering or promoting event in each individual patient, large-scale introduction of active prevention and severity reduction strategies may not be realistic. In the meantime, we can approach AD in childhood from the inside out, through appropriate treatment of systemic inflammation of AD, and from the outside in, with treatment and prevention via emollient use in newborns.

Atopic dermatitis (AD) is a multifactorial inflammatory disorder with an estimated prevalence of 279,889,120 cases worldwide.1 Most cases of AD begin in early childhood (with almost 85% developing by 5 years of age),² but recent studies have found that 40% to over 80% of cases persist into adulthood.^1,3,4 Although a previous study focused largely on T helper type 1/T helper type 2 (Th2) immune dysregulation as the pathogenesis of the disease,⁵ disruption of the skin barrier and systemic inflammation are at the center of current AD research. In AD, breakdown of the skin barrier results in increased transepidermal water loss, reduced skin hydration, and increased antigen presentation by Langerhans cells initiating inflammation.6-8 The cascade largely activated is the Th2 and T helper type 22 cascade with resultant cytokine release (ie, IL-4, IL-13, IL-2, IL-8, IL-10, IL-17, IL-22, tumor necrosis factor α, interferon γ).^9,10 In active AD, Th2 inflammation and barrier breakdown result in reduced filaggrin and claudin 1 expression, resulting in further exacerbation of the barrier defect and enhancing the risk of development of asthma and hay fever as well as transcutaneous sensitization to a variety of food allergens (eg, peanuts).^9,11,12 Although all of these immunologic features are well established in AD, controversy remains as to whether AD is caused by systemic inflammation triggering barrier dysfunction (the “inside-out” hypothesis) or from the epidermal skin barrier disruption triggering immunologic imbalance (the “outside-in” hypothesis).

Inside-Out Hypothesis

While barrier impairment appears to occur in all patients with AD, it still is unclear how AD begins. The inside-out hypothesis suggests that cutaneous inflammation precedes barrier impairment and in fact may result in an impaired skin barrier. It has previously been reported that inflammatory states weaken the barrier by downregulating filaggrin production in the skin.¹³ Barrier disruption may be accompanied by transcutaneous penetration of allergens and increased Staphylococcus aureus counts. Recently, mutations and polymorphisms of inflammatory genes have been linked to AD (eg, single nucleotide polymorphisms of the IL4RA [interleukin 4 receptor, alpha] and CD14 [cluster of differentiation 14] genes, the serine protease inhibitor SPINK5 [serine peptidase inhibitor, Kazal type 5], RANTES [chemokine (C-C motif) ligand 5], IL-4, IL-13).¹⁴ These alterations highlight the role of systemic inflammation in triggering AD.

Outside-In Hypothesis

The outside-in hypothesis suggests that the impaired skin barrier precedes AD and is required for immune dysregulation to occur. This hypothesis was largely advanced by a study demonstrating that deactivating mutations of the filaggrin gene were linked to nearly 20% of AD cases in Northern European populations.15 Filaggrin (chromosome 1q21.3) performs an essential function in the skin barrier through its differential cleavage and the breakdown and release of natural moisturizing factor.¹⁶ Filaggrin gene mutations are associated with persistent AD, and it has been posited that environmental factors such as temperature and humidity also can affect filaggrin production as it relates to barrier function.17-19 Skin barrier disruption results in increased cutaneous and systemic Th2 responses (ie, IL-4/13), with thymic stromal lymphopoietin as the potential mechanism of Th2 cell recruitment.^10,20 Inflammatory Th2 cells triggered by an impaired skin barrier also may predispose patients to the development of allergic diseases such as asthma, in line with Atopic March, or the progression of AD to other forms of atopy (eg, food allergy, asthma).^5,7,21-23

The outside-in hypothesis may only explain the root pathogenesis of AD in a subset of patients, however, as only 1 in 5 cases of AD in Northern European and Asian populations are associated with underlying filaggrin mutations (which are only present in about 10% of those who are unaffected by AD).¹⁵ Filaggrin does not appear to account for the basis of AD in all cases. In a study of 762 newborns in Cincinnati, Ohio, 39% of children with at least one parent with atopy developed AD by 3 years of age, about quadruple of what would be projected based on filaggrin defects in general population studies, which are noted in only about 10% of white individuals.²⁴ Furthermore, less than 5% of patients of African descent have mutations of the filaggrin 1 gene.²⁵

Implications for the Prevention and Treatment of Atopic Dermatitis

Preventative strategies for AD currently are in development. Atopic dermatitis may be unpreventable because the in utero environment triggers some of the barrier alterations, which can be noted as early as 2 days following birth and will predict early-onset AD. The putative mechanism is via Th2 cytokines (IL-4, IL-13).²⁶

Certainly, application of over-the-counter and prescription emollients are mainstays of treatment for AD and may suffice as monotherapy in cases of mild disease. In a recent randomized trial in the United States and the United Kingdom, emollients were used in newborns considered at high risk for AD (family history of atopy) until 6 months of age.²⁷ The risk of AD development was reduced by half, irrespective of the emollient used. Unfortunately, 21.8% of children without a family history of atopy will develop AD; therefore, not all cases can be prevented if use of emollients is limited to newborns with a family history of atopy.²⁸ Long-term follow-up is needed to track whether emollient use in newborns will prevent AD indefinitely.

Prevention of AD onset using systemic interventions has also been investigated. Probiotics have been suggested as a means to modify the gut microbiota and reduce systemic and mucosal inflammation. Lactobacillus reuteri taken prenatally by pregnant women and by newborns has shown mild benefit in preventing some forms of AD.29 Although they are not approved by the US Food and Drug Administration for this indication, systemic interventions for moderate-to-severe AD such as methotrexate and cyclosporine certainly have shown benefit in managing ongoing illness and breaking the cycle of disease.³⁰ The efficacy of these agents points to the role of systemic inflammation in ongoing AD activity. Moreover, the inside-out hypothesis recently has led to the proliferation of promising new therapeutic agents in the pipeline to treat the systemic Th2 inflammation that occurs in severe AD (eg, anti–IL-4/13 receptor antibody, anti–IL-13 antibodies, and biologics targeting IL-12/23, IL-22, and IL-31 receptors).³¹

Final Thoughts

Atopic dermatitis is a multifactorial disease associated with barrier disruption and intense systemic inflammation. It is likely that both the inside-out and outside-in hypotheses hold true in different subsets of AD patients. It is clear that some individuals are born with filaggrin defects that sufficiently trigger systemic inflammation, resulting in AD. On the other hand, there are clearly some individuals with inflammatory dysregulation that results in systemic inflammation and secondary barrier disruption. Until we can determine the genomic triggering or promoting event in each individual patient, large-scale introduction of active prevention and severity reduction strategies may not be realistic. In the meantime, we can approach AD in childhood from the inside out, through appropriate treatment of systemic inflammation of AD, and from the outside in, with treatment and prevention via emollient use in newborns.

Atopic dermatitis (AD) is a multifactorial inflammatory disorder with an estimated prevalence of 279,889,120 cases worldwide.1 Most cases of AD begin in early childhood (with almost 85% developing by 5 years of age),² but recent studies have found that 40% to over 80% of cases persist into adulthood.^1,3,4 Although a previous study focused largely on T helper type 1/T helper type 2 (Th2) immune dysregulation as the pathogenesis of the disease,⁵ disruption of the skin barrier and systemic inflammation are at the center of current AD research. In AD, breakdown of the skin barrier results in increased transepidermal water loss, reduced skin hydration, and increased antigen presentation by Langerhans cells initiating inflammation.6-8 The cascade largely activated is the Th2 and T helper type 22 cascade with resultant cytokine release (ie, IL-4, IL-13, IL-2, IL-8, IL-10, IL-17, IL-22, tumor necrosis factor α, interferon γ).^9,10 In active AD, Th2 inflammation and barrier breakdown result in reduced filaggrin and claudin 1 expression, resulting in further exacerbation of the barrier defect and enhancing the risk of development of asthma and hay fever as well as transcutaneous sensitization to a variety of food allergens (eg, peanuts).^9,11,12 Although all of these immunologic features are well established in AD, controversy remains as to whether AD is caused by systemic inflammation triggering barrier dysfunction (the “inside-out” hypothesis) or from the epidermal skin barrier disruption triggering immunologic imbalance (the “outside-in” hypothesis).

Inside-Out Hypothesis

While barrier impairment appears to occur in all patients with AD, it still is unclear how AD begins. The inside-out hypothesis suggests that cutaneous inflammation precedes barrier impairment and in fact may result in an impaired skin barrier. It has previously been reported that inflammatory states weaken the barrier by downregulating filaggrin production in the skin.¹³ Barrier disruption may be accompanied by transcutaneous penetration of allergens and increased Staphylococcus aureus counts. Recently, mutations and polymorphisms of inflammatory genes have been linked to AD (eg, single nucleotide polymorphisms of the IL4RA [interleukin 4 receptor, alpha] and CD14 [cluster of differentiation 14] genes, the serine protease inhibitor SPINK5 [serine peptidase inhibitor, Kazal type 5], RANTES [chemokine (C-C motif) ligand 5], IL-4, IL-13).¹⁴ These alterations highlight the role of systemic inflammation in triggering AD.

Outside-In Hypothesis

The outside-in hypothesis suggests that the impaired skin barrier precedes AD and is required for immune dysregulation to occur. This hypothesis was largely advanced by a study demonstrating that deactivating mutations of the filaggrin gene were linked to nearly 20% of AD cases in Northern European populations.15 Filaggrin (chromosome 1q21.3) performs an essential function in the skin barrier through its differential cleavage and the breakdown and release of natural moisturizing factor.¹⁶ Filaggrin gene mutations are associated with persistent AD, and it has been posited that environmental factors such as temperature and humidity also can affect filaggrin production as it relates to barrier function.17-19 Skin barrier disruption results in increased cutaneous and systemic Th2 responses (ie, IL-4/13), with thymic stromal lymphopoietin as the potential mechanism of Th2 cell recruitment.^10,20 Inflammatory Th2 cells triggered by an impaired skin barrier also may predispose patients to the development of allergic diseases such as asthma, in line with Atopic March, or the progression of AD to other forms of atopy (eg, food allergy, asthma).^5,7,21-23

The outside-in hypothesis may only explain the root pathogenesis of AD in a subset of patients, however, as only 1 in 5 cases of AD in Northern European and Asian populations are associated with underlying filaggrin mutations (which are only present in about 10% of those who are unaffected by AD).¹⁵ Filaggrin does not appear to account for the basis of AD in all cases. In a study of 762 newborns in Cincinnati, Ohio, 39% of children with at least one parent with atopy developed AD by 3 years of age, about quadruple of what would be projected based on filaggrin defects in general population studies, which are noted in only about 10% of white individuals.²⁴ Furthermore, less than 5% of patients of African descent have mutations of the filaggrin 1 gene.²⁵

Implications for the Prevention and Treatment of Atopic Dermatitis

Preventative strategies for AD currently are in development. Atopic dermatitis may be unpreventable because the in utero environment triggers some of the barrier alterations, which can be noted as early as 2 days following birth and will predict early-onset AD. The putative mechanism is via Th2 cytokines (IL-4, IL-13).²⁶

Certainly, application of over-the-counter and prescription emollients are mainstays of treatment for AD and may suffice as monotherapy in cases of mild disease. In a recent randomized trial in the United States and the United Kingdom, emollients were used in newborns considered at high risk for AD (family history of atopy) until 6 months of age.²⁷ The risk of AD development was reduced by half, irrespective of the emollient used. Unfortunately, 21.8% of children without a family history of atopy will develop AD; therefore, not all cases can be prevented if use of emollients is limited to newborns with a family history of atopy.²⁸ Long-term follow-up is needed to track whether emollient use in newborns will prevent AD indefinitely.

Prevention of AD onset using systemic interventions has also been investigated. Probiotics have been suggested as a means to modify the gut microbiota and reduce systemic and mucosal inflammation. Lactobacillus reuteri taken prenatally by pregnant women and by newborns has shown mild benefit in preventing some forms of AD.29 Although they are not approved by the US Food and Drug Administration for this indication, systemic interventions for moderate-to-severe AD such as methotrexate and cyclosporine certainly have shown benefit in managing ongoing illness and breaking the cycle of disease.³⁰ The efficacy of these agents points to the role of systemic inflammation in ongoing AD activity. Moreover, the inside-out hypothesis recently has led to the proliferation of promising new therapeutic agents in the pipeline to treat the systemic Th2 inflammation that occurs in severe AD (eg, anti–IL-4/13 receptor antibody, anti–IL-13 antibodies, and biologics targeting IL-12/23, IL-22, and IL-31 receptors).³¹

Final Thoughts

Atopic dermatitis is a multifactorial disease associated with barrier disruption and intense systemic inflammation. It is likely that both the inside-out and outside-in hypotheses hold true in different subsets of AD patients. It is clear that some individuals are born with filaggrin defects that sufficiently trigger systemic inflammation, resulting in AD. On the other hand, there are clearly some individuals with inflammatory dysregulation that results in systemic inflammation and secondary barrier disruption. Until we can determine the genomic triggering or promoting event in each individual patient, large-scale introduction of active prevention and severity reduction strategies may not be realistic. In the meantime, we can approach AD in childhood from the inside out, through appropriate treatment of systemic inflammation of AD, and from the outside in, with treatment and prevention via emollient use in newborns.

Vitiligo is the loss of skin pigmentation caused by autoimmune destruction of melanocytes. Multiple pathogenic factors for vitiligo have been described, including CD8⁺ T lymphocyte/T helper 1 infiltrates in lesional skin^1,2 with increased expression of IFN-γ³ and tumor necrosis factor α,^3-6 decreased transforming growth factor β,⁷ and circulating autoantibodies against tyrosine hydroxylase.⁸ Additionally, several studies have found a high prevalence of antecedent psychological stressors in vitiligo patients, suggesting that specific stressors may trigger and/or exacerbate vitiligo.^9-12

The relationship between antecedent psychological stressors and vitiligo extent has not been well studied. Potential mechanisms for stress-triggered vitiligo include increased catecholamines¹³ and neuropeptides,¹⁴ which have been found in vitiligo patients. However, the complex relationship between stressors and subsequent vitiligo is not well defined. We hypothesized that persistent stressors are associated with increased vitiligo extent.

Vitiligo is classically considered to be a silent pigmentary disorder with few or no symptoms. Prior studies have demonstrated that one-third of vitiligo patients report skin symptoms (eg, pruritus, burning), which may be specifically associated with early-onset disease.^15-17 Further, we observed that some vitiligo patients report abdominal cramping associated with their disease. Few studies have described the burden of skin symptoms and other associated symptoms in vitiligo or their determinants.

We conducted a prospective questionnaire-based study of 1541 adult vitiligo patients to identify psychological factors that may precede vitiligo onset. We hypothesized that some types of stressors that occur within 2 years prior to disease onset would have specific associations with vitiligo and/or somatic symptoms.

Methods

Study Population and Questionnaire Distribution

This prospective questionnaire-based study was approved by the institutional review board at St. Luke’s-Roosevelt Hospital Center (now Mount Sinai St. Luke’s-Roosevelt) (New York, New York) for adults (>18 years; male or female) with vitiligo. The survey was validated in paper format at St. Luke’s-Roosevelt Hospital Center and distributed online to members of nonprofit support groups for vitiligo vulgaris, as previously described.¹⁵

Questionnaire

The a priori aim of this questionnaire was to identify psychological factors that may precede vitiligo onset. The questionnaire consisted of 77 items (55 closed questions and 22 open questions) pertaining to participant demographics/vitiligo phenotype and psychological stressors preceding vitiligo onset. The questions related to this study and response rates are listed in eTable 1. Responses were verified by screening for noninteger or implausible values (eg, <0 or >100 years of age).

Sample Size

The primary outcome used for sample size calculation was the potential association between vitiligo and the presence of antecedent psychological stressors. Using a 2-tailed test, we determined that a sample size of 1264 participants would have 90% power at α=.05 and a baseline proportion of 0.01 (1% presumed prevalence of vitiligo) to detect an odds ratio (OR) of 2.5 or higher.¹⁸

Data and Statistical Analysis

Closed question responses were analyzed using descriptive statistics. Open-ended question responses were analyzed using content analysis. Related comments were coded and grouped, with similarities and differences noted. All data processing and statistics were done with SAS version 9.2. Age at diagnosis (years) and number of anatomic sites affected were divided into tertiles for statistical analysis due to wide skewing.    

Logistic regression models were constructed with numbers of reported deaths or stressors per participant within the 2 years prior to vitiligo onset as independent variables (0, 1, or ≥2), and symptoms associated with vitiligo as dependent variables. Adjusted ORs were calculated from multivariate models that included sex, current age (continuous), and comorbid autoimmune disease (binary) as covariates. Linear interaction terms were tested and were included in final models if statistically significant (P<.05).

Ordinal logistic regression was used to analyze the relationship between stressors (and other independent variables) and number of anatomic sites affected with vitiligo (tertiles). Ordinal logistic regression models were constructed to examine the impact of psychological stressors on pruritus secondary to vitiligo (not relevant combined with not at all, a little, a lot, very much) as the dependent variable. The proportional odds assumption was met in both models, as judged by score testing (P>.05). Binary logistic regression was used to analyze laterality, body surface area (BSA) greater than 25%, and involvement of the face and/or body with vitiligo lesions (binary).

Binary logistic regression models were constructed with impact of psychological stressors preceding vitiligo onset on comorbid abdominal cramping and specific etiologies as the dependent variables. There were 20 candidate stressors occurring within the 2 years prior to vitiligo onset. Selection methods for predictors were used to identify significant covariates within the context of the other covariates included in the final models. The results of forward, backward, and stepwise approaches were similar, and the stepwise selection output was presented.

Missing values were encountered because some participants did not respond to all the questionnaire items. A complete case analysis was performed (ie, missing values were ignored throughout the study). Data imputation was considered by multiple imputations; however, there were few or no differences between the estimates from the 2 approaches. Therefore, final models did not involve data imputation.

The statistical significance for all estimates was considered to be P<.05. However, a P value near .05 should be interpreted with caution given the multiple dependent tests performed in this study with increased risk for falsely rejecting the null hypothesis.

Results

Survey Population Characteristics

One thousand seven hundred participants started the survey; 1632 completed the survey (96.0% completion rate) and 1553 had been diagnosed with vitiligo by a physician. Twelve participants were excluded because they were younger than 18 years, leaving 1541 evaluable participants. Five hundred thirty-eight participants (34.9%) had comorbid autoimmune disorders. Demographics and disease phenotypes of the study participants are listed in Table 1.

Stressors Preceding Vitiligo Onset

Eight hundred twenty-one participants (56.6%) experienced at least one death or stressor within 2 years prior to vitiligo onset (Table 2), including death of a loved one (16.6%) and stressful life events (51.0%) within the 2 years prior to the onset of vitiligo, especially work/financial problems (10.8%), end of a long-term relationship (10.2%), and family problems (not otherwise specified)(7.8%). Two hundred (13.5%) participants reported experiencing 1 death and 46 (3.1%) reported multiple deaths. Five hundred participants (33.6%) reported experiencing 1 stressor and 259 (17.4%) reported multiple stressors.

Stressors Not Associated With Vitiligo Extent

The number of deaths or stressors reported per participant within the 2 years prior to vitiligo onset were not associated with BSA, laterality, or distribution of lesions (Table 3 and eTable 2–eTable 4).

Symptoms Associated With Vitiligo

Five hundred twenty-two participants (34.5%) reported intermittent abdominal cramping, including premenstrual and/or menstrual cramping in women (9.7%), food-related abdominal cramping (4.4%), inflammatory bowel syndrome (IBS)(2.6%), anxiety-related abdominal cramping (1.5%), autoimmune gastrointestinal disorders (1.2%), and “other” etiologies (20.4%). Five hundred ten participants reported itching and/or burning associated with vitiligo lesions (35.1%).

Intermittent abdominal cramping overall was associated with a BSA greater than 75% (OR, 1.65; 95% confidence interval (CI), 1.17-2.32; P=.004). However, specific etiologies of abdominal cramping were not significantly associated with BSA (P≥.11). In contrast, itching and/or burning from vitiligo lesions was associated with a BSA greater than 25% (OR, 1.53; 95% CI, 1.23-1.90; P<.0001).

Association Between Number of Stressors and Symptoms in Vitiligo

A history of multiple stressors (≥2) within the 2 years prior to vitiligo onset was associated with intermittent abdominal cramping overall (OR, 1.84; 95% CI, 1.38-2.47; P<.0001), including premenstrual and/or menstrual cramping in women (OR, 1.84; 95% CI, 1.15-2.95; P=.01), IBS (OR, 3.29; 95% CI, 1.34-8.05; P=.01), and autoimmune gastrointestinal disorders (OR, 4.02; 95% CI, 1.27-12.80; P=.02)(eTable 5). These associations remained significant in multivariate models that included age, sex, and BSA as covariates. However, a history of 1 stressor or death or multiple deaths in the 2 years prior to vitiligo onset was not associated with any etiology of abdominal cramping.

Experiencing 1 (OR, 1.43; 95% CI, 1.12-1.82; P=.005) or multiple stressors (OR, 1.51; 95% CI,  1.12-2.04; P=.007) also was associated with itching and/or burning secondary to vitiligo. This association remained significant in a multivariate model that included age, sex, and BSA as covariates. However, a history of 1 or multiple deaths in the 2 years prior to vitiligo onset was not associated with itching and/or burning.

Association Between Specific Stressors and Vitiligo Symptoms

Perimenstrual (premenstrual and/or menstrual) cramping in women was associated with family problems (not otherwise specified) within the 2 years prior to vitiligo onset (Table 4). Food-related abdominal cramping was associated with school- and/or test-related stressors. Diagnosis of IBS was associated with health problems or surgery and being a victim of abuse within the 2 years prior to onset of vitiligo. Autoimmune gastrointestinal disorders were associated with moving to a new home/region, health problems or surgery, and witness to a violent crime or death. Finally, itching and/or burning of vitiligo lesions was associated with work and financial problems.

Comment

The present study found a high frequency of stressful life events and deaths of loved ones occurring within the 2 years preceding vitiligo onset. A history of multiple stressors but not deaths of loved ones was associated with more frequent symptoms in vitiligo patients, including itching and/or burning and intermittent abdominal pain. Specific stressors were associated with intermittent abdominal cramping, which occurred in approximately one-third of vitiligo patients. Abdominal cramping was related to menses in women, anxiety, foods, IBS, autoimmune gastrointestinal disorders, and other etiologies of abdominal cramping, which underscores the complex relationship between stressors, vitiligo, and inflammation. It is possible that stress-related immune abnormalities occur in vitiligo, which may influence the development of other autoimmune disorders. Alternatively, abdominal symptoms may precede and perhaps contribute to psychological stressors and impaired quality of life in vitiligo patients; however, the cross-sectional nature of the study did not allow us to elucidate this temporal relationship.

The present study found that 56.6% of participants experienced 1 or more deaths (17%) and/or stressful life events (51%) within the 2 years prior to vitiligo onset. These results are consistent with prior smaller studies that demonstrated a high frequency of stressful events preceding vitiligo onset. A case-controlled study found stressful events in 12 of 21 (57%) Romanian children with vitiligo, which was higher than controls.¹⁹ Another questionnaire-based, case-controlled study compared a heterogeneous group of 32 adolescent and adult Romanian patients with vitiligo and found higher odds of a stressful event in women preceding vitiligo diagnosis compared to controls.¹⁰ A retrospective analysis of 65 Croatian patients with vitiligo also reported that 56.9% (37/65) had some associated psychological factors.⁹ Another retrospective study of 31 adults with vitiligo found increased occurrence of 3 or more uncontrollable events, decreased perceived social support, and increased anxiety in vitiligo patients versus 116 other dermatologic disease controls.¹² A questionnaire-based study found increased bereavements, changes in sleeping and eating habits, and personal injuries/illnesses in 73 British adults with vitiligo compared to 73 other age- and sex-matched dermatologic disease controls.¹¹ All of these studies were limited by a small sample size, and the patient populations were localized to a regional dermatology referral center. The present study provided a larger analysis of stressful life events preceding vitiligo onset and included a diverse patient population.

The present study found that stressful life events and deaths of a loved one are not associated with vitiligo extent and distribution. This finding suggests that stressful life events may act as vitiligo triggers in genetically predisposed individuals, but ultimately the disease course and prognosis are driven by other factors, such as increased systemic inflammation or other immunologic abnormalities. Indeed, Silverberg and Silverberg²⁰ and other investigators^21,22 reported relative deficiencies of 25-hydroxyvitamin D,²³ vitamins B₆ and B₁₂, and folic acid,²⁰ as well as elevated serum homocysteine levels in vitiligo patients. Increased serum homocysteine levels were associated with increased BSA of vitiligo lesions.²⁰ Elevated serum homocysteine levels also have been associated with increased inflammation in coronary artery disease,²⁴ psoriasis,^25,26 and in vitro.²⁷ These laboratory anomalies likely reflect an underlying predisposition toward vitiligo, which might be triggered by stress responses or secondarily altered immune responses.

The present study had several strengths, including being prospective with a large sample size. The patient population included a large sample of men and women with representation of various adult ages and vitiligo extent. However, this study also had potential limitations. Measures of vitiligo extent were self-reported and were not clinically assessed. To address this limitation, we validated the questionnaire before posting it online.¹⁵ Invitation to participate in the survey was distributed by vitiligo support groups, which may have resulted in a selection bias toward participants with greater disease severity or with a poorer quality of life associated with vitiligo. Invitation to participate in this study was sent to members of vitiligo support groups, which allowed for recruitment of a large number of vitiligo patients despite a relatively low prevalence of disease in the general population. However, there are several challenges using this approach for nonvitiligo controls. Using participants with another dermatological disease as a control group may yield spurious results. Ideally, a large randomized sample of healthy participants with minimization of bias should be used for controls, which is an ambitious undertaking that was beyond the scope of this pilot study and will be the subject of future studies. Finally, this analysis found associations between stressors that occurred in the 2 years prior to vitiligo onset with symptomatic disease. We chose a broad interval for stressors because early vitiligo lesions may go unnoticed, making recognition of stressors occurring within days or weeks of onset infeasible. Further, we considered that chronic and prolonged stressors are more likely to have harmful consequences than acute stressors. Thus, stressors occurring within a more narrow interval (eg, 2 months) may not have the same association with vitiligo. Future studies are warranted to precisely identify the type and timing of psychological stressors preceding vitiligo onset.

Conclusion

In conclusion, there is a high prevalence of stressful life events preceding vitiligo, which may play an important role as disease triggers as well as predict the presence of intermittent abdominal cramping and itching or burning of skin. These associations indicate that screening of vitiligo patients for psychological stressors, abdominal cramping, and itching and/or burning of skin should be included in the routine assessment of vitiligo patients.

Appendix

Please refer to the eTables in the PDF.

Vitiligo is the loss of skin pigmentation caused by autoimmune destruction of melanocytes. Multiple pathogenic factors for vitiligo have been described, including CD8⁺ T lymphocyte/T helper 1 infiltrates in lesional skin^1,2 with increased expression of IFN-γ³ and tumor necrosis factor α,^3-6 decreased transforming growth factor β,⁷ and circulating autoantibodies against tyrosine hydroxylase.⁸ Additionally, several studies have found a high prevalence of antecedent psychological stressors in vitiligo patients, suggesting that specific stressors may trigger and/or exacerbate vitiligo.^9-12

The relationship between antecedent psychological stressors and vitiligo extent has not been well studied. Potential mechanisms for stress-triggered vitiligo include increased catecholamines¹³ and neuropeptides,¹⁴ which have been found in vitiligo patients. However, the complex relationship between stressors and subsequent vitiligo is not well defined. We hypothesized that persistent stressors are associated with increased vitiligo extent.

Vitiligo is classically considered to be a silent pigmentary disorder with few or no symptoms. Prior studies have demonstrated that one-third of vitiligo patients report skin symptoms (eg, pruritus, burning), which may be specifically associated with early-onset disease.^15-17 Further, we observed that some vitiligo patients report abdominal cramping associated with their disease. Few studies have described the burden of skin symptoms and other associated symptoms in vitiligo or their determinants.

We conducted a prospective questionnaire-based study of 1541 adult vitiligo patients to identify psychological factors that may precede vitiligo onset. We hypothesized that some types of stressors that occur within 2 years prior to disease onset would have specific associations with vitiligo and/or somatic symptoms.

Methods

Study Population and Questionnaire Distribution

This prospective questionnaire-based study was approved by the institutional review board at St. Luke’s-Roosevelt Hospital Center (now Mount Sinai St. Luke’s-Roosevelt) (New York, New York) for adults (>18 years; male or female) with vitiligo. The survey was validated in paper format at St. Luke’s-Roosevelt Hospital Center and distributed online to members of nonprofit support groups for vitiligo vulgaris, as previously described.¹⁵

Questionnaire

The a priori aim of this questionnaire was to identify psychological factors that may precede vitiligo onset. The questionnaire consisted of 77 items (55 closed questions and 22 open questions) pertaining to participant demographics/vitiligo phenotype and psychological stressors preceding vitiligo onset. The questions related to this study and response rates are listed in eTable 1. Responses were verified by screening for noninteger or implausible values (eg, <0 or >100 years of age).

Sample Size

The primary outcome used for sample size calculation was the potential association between vitiligo and the presence of antecedent psychological stressors. Using a 2-tailed test, we determined that a sample size of 1264 participants would have 90% power at α=.05 and a baseline proportion of 0.01 (1% presumed prevalence of vitiligo) to detect an odds ratio (OR) of 2.5 or higher.¹⁸

Data and Statistical Analysis

Closed question responses were analyzed using descriptive statistics. Open-ended question responses were analyzed using content analysis. Related comments were coded and grouped, with similarities and differences noted. All data processing and statistics were done with SAS version 9.2. Age at diagnosis (years) and number of anatomic sites affected were divided into tertiles for statistical analysis due to wide skewing.    

Logistic regression models were constructed with numbers of reported deaths or stressors per participant within the 2 years prior to vitiligo onset as independent variables (0, 1, or ≥2), and symptoms associated with vitiligo as dependent variables. Adjusted ORs were calculated from multivariate models that included sex, current age (continuous), and comorbid autoimmune disease (binary) as covariates. Linear interaction terms were tested and were included in final models if statistically significant (P<.05).

Ordinal logistic regression was used to analyze the relationship between stressors (and other independent variables) and number of anatomic sites affected with vitiligo (tertiles). Ordinal logistic regression models were constructed to examine the impact of psychological stressors on pruritus secondary to vitiligo (not relevant combined with not at all, a little, a lot, very much) as the dependent variable. The proportional odds assumption was met in both models, as judged by score testing (P>.05). Binary logistic regression was used to analyze laterality, body surface area (BSA) greater than 25%, and involvement of the face and/or body with vitiligo lesions (binary).

Binary logistic regression models were constructed with impact of psychological stressors preceding vitiligo onset on comorbid abdominal cramping and specific etiologies as the dependent variables. There were 20 candidate stressors occurring within the 2 years prior to vitiligo onset. Selection methods for predictors were used to identify significant covariates within the context of the other covariates included in the final models. The results of forward, backward, and stepwise approaches were similar, and the stepwise selection output was presented.

Missing values were encountered because some participants did not respond to all the questionnaire items. A complete case analysis was performed (ie, missing values were ignored throughout the study). Data imputation was considered by multiple imputations; however, there were few or no differences between the estimates from the 2 approaches. Therefore, final models did not involve data imputation.

The statistical significance for all estimates was considered to be P<.05. However, a P value near .05 should be interpreted with caution given the multiple dependent tests performed in this study with increased risk for falsely rejecting the null hypothesis.

Results

Survey Population Characteristics

One thousand seven hundred participants started the survey; 1632 completed the survey (96.0% completion rate) and 1553 had been diagnosed with vitiligo by a physician. Twelve participants were excluded because they were younger than 18 years, leaving 1541 evaluable participants. Five hundred thirty-eight participants (34.9%) had comorbid autoimmune disorders. Demographics and disease phenotypes of the study participants are listed in Table 1.

Stressors Preceding Vitiligo Onset

Eight hundred twenty-one participants (56.6%) experienced at least one death or stressor within 2 years prior to vitiligo onset (Table 2), including death of a loved one (16.6%) and stressful life events (51.0%) within the 2 years prior to the onset of vitiligo, especially work/financial problems (10.8%), end of a long-term relationship (10.2%), and family problems (not otherwise specified)(7.8%). Two hundred (13.5%) participants reported experiencing 1 death and 46 (3.1%) reported multiple deaths. Five hundred participants (33.6%) reported experiencing 1 stressor and 259 (17.4%) reported multiple stressors.

Stressors Not Associated With Vitiligo Extent

The number of deaths or stressors reported per participant within the 2 years prior to vitiligo onset were not associated with BSA, laterality, or distribution of lesions (Table 3 and eTable 2–eTable 4).

Symptoms Associated With Vitiligo

Five hundred twenty-two participants (34.5%) reported intermittent abdominal cramping, including premenstrual and/or menstrual cramping in women (9.7%), food-related abdominal cramping (4.4%), inflammatory bowel syndrome (IBS)(2.6%), anxiety-related abdominal cramping (1.5%), autoimmune gastrointestinal disorders (1.2%), and “other” etiologies (20.4%). Five hundred ten participants reported itching and/or burning associated with vitiligo lesions (35.1%).

Intermittent abdominal cramping overall was associated with a BSA greater than 75% (OR, 1.65; 95% confidence interval (CI), 1.17-2.32; P=.004). However, specific etiologies of abdominal cramping were not significantly associated with BSA (P≥.11). In contrast, itching and/or burning from vitiligo lesions was associated with a BSA greater than 25% (OR, 1.53; 95% CI, 1.23-1.90; P<.0001).

Association Between Number of Stressors and Symptoms in Vitiligo

A history of multiple stressors (≥2) within the 2 years prior to vitiligo onset was associated with intermittent abdominal cramping overall (OR, 1.84; 95% CI, 1.38-2.47; P<.0001), including premenstrual and/or menstrual cramping in women (OR, 1.84; 95% CI, 1.15-2.95; P=.01), IBS (OR, 3.29; 95% CI, 1.34-8.05; P=.01), and autoimmune gastrointestinal disorders (OR, 4.02; 95% CI, 1.27-12.80; P=.02)(eTable 5). These associations remained significant in multivariate models that included age, sex, and BSA as covariates. However, a history of 1 stressor or death or multiple deaths in the 2 years prior to vitiligo onset was not associated with any etiology of abdominal cramping.

Experiencing 1 (OR, 1.43; 95% CI, 1.12-1.82; P=.005) or multiple stressors (OR, 1.51; 95% CI,  1.12-2.04; P=.007) also was associated with itching and/or burning secondary to vitiligo. This association remained significant in a multivariate model that included age, sex, and BSA as covariates. However, a history of 1 or multiple deaths in the 2 years prior to vitiligo onset was not associated with itching and/or burning.

Association Between Specific Stressors and Vitiligo Symptoms

Perimenstrual (premenstrual and/or menstrual) cramping in women was associated with family problems (not otherwise specified) within the 2 years prior to vitiligo onset (Table 4). Food-related abdominal cramping was associated with school- and/or test-related stressors. Diagnosis of IBS was associated with health problems or surgery and being a victim of abuse within the 2 years prior to onset of vitiligo. Autoimmune gastrointestinal disorders were associated with moving to a new home/region, health problems or surgery, and witness to a violent crime or death. Finally, itching and/or burning of vitiligo lesions was associated with work and financial problems.

Comment

The present study found a high frequency of stressful life events and deaths of loved ones occurring within the 2 years preceding vitiligo onset. A history of multiple stressors but not deaths of loved ones was associated with more frequent symptoms in vitiligo patients, including itching and/or burning and intermittent abdominal pain. Specific stressors were associated with intermittent abdominal cramping, which occurred in approximately one-third of vitiligo patients. Abdominal cramping was related to menses in women, anxiety, foods, IBS, autoimmune gastrointestinal disorders, and other etiologies of abdominal cramping, which underscores the complex relationship between stressors, vitiligo, and inflammation. It is possible that stress-related immune abnormalities occur in vitiligo, which may influence the development of other autoimmune disorders. Alternatively, abdominal symptoms may precede and perhaps contribute to psychological stressors and impaired quality of life in vitiligo patients; however, the cross-sectional nature of the study did not allow us to elucidate this temporal relationship.

The present study found that 56.6% of participants experienced 1 or more deaths (17%) and/or stressful life events (51%) within the 2 years prior to vitiligo onset. These results are consistent with prior smaller studies that demonstrated a high frequency of stressful events preceding vitiligo onset. A case-controlled study found stressful events in 12 of 21 (57%) Romanian children with vitiligo, which was higher than controls.¹⁹ Another questionnaire-based, case-controlled study compared a heterogeneous group of 32 adolescent and adult Romanian patients with vitiligo and found higher odds of a stressful event in women preceding vitiligo diagnosis compared to controls.¹⁰ A retrospective analysis of 65 Croatian patients with vitiligo also reported that 56.9% (37/65) had some associated psychological factors.⁹ Another retrospective study of 31 adults with vitiligo found increased occurrence of 3 or more uncontrollable events, decreased perceived social support, and increased anxiety in vitiligo patients versus 116 other dermatologic disease controls.¹² A questionnaire-based study found increased bereavements, changes in sleeping and eating habits, and personal injuries/illnesses in 73 British adults with vitiligo compared to 73 other age- and sex-matched dermatologic disease controls.¹¹ All of these studies were limited by a small sample size, and the patient populations were localized to a regional dermatology referral center. The present study provided a larger analysis of stressful life events preceding vitiligo onset and included a diverse patient population.

The present study found that stressful life events and deaths of a loved one are not associated with vitiligo extent and distribution. This finding suggests that stressful life events may act as vitiligo triggers in genetically predisposed individuals, but ultimately the disease course and prognosis are driven by other factors, such as increased systemic inflammation or other immunologic abnormalities. Indeed, Silverberg and Silverberg²⁰ and other investigators^21,22 reported relative deficiencies of 25-hydroxyvitamin D,²³ vitamins B₆ and B₁₂, and folic acid,²⁰ as well as elevated serum homocysteine levels in vitiligo patients. Increased serum homocysteine levels were associated with increased BSA of vitiligo lesions.²⁰ Elevated serum homocysteine levels also have been associated with increased inflammation in coronary artery disease,²⁴ psoriasis,^25,26 and in vitro.²⁷ These laboratory anomalies likely reflect an underlying predisposition toward vitiligo, which might be triggered by stress responses or secondarily altered immune responses.

The present study had several strengths, including being prospective with a large sample size. The patient population included a large sample of men and women with representation of various adult ages and vitiligo extent. However, this study also had potential limitations. Measures of vitiligo extent were self-reported and were not clinically assessed. To address this limitation, we validated the questionnaire before posting it online.¹⁵ Invitation to participate in the survey was distributed by vitiligo support groups, which may have resulted in a selection bias toward participants with greater disease severity or with a poorer quality of life associated with vitiligo. Invitation to participate in this study was sent to members of vitiligo support groups, which allowed for recruitment of a large number of vitiligo patients despite a relatively low prevalence of disease in the general population. However, there are several challenges using this approach for nonvitiligo controls. Using participants with another dermatological disease as a control group may yield spurious results. Ideally, a large randomized sample of healthy participants with minimization of bias should be used for controls, which is an ambitious undertaking that was beyond the scope of this pilot study and will be the subject of future studies. Finally, this analysis found associations between stressors that occurred in the 2 years prior to vitiligo onset with symptomatic disease. We chose a broad interval for stressors because early vitiligo lesions may go unnoticed, making recognition of stressors occurring within days or weeks of onset infeasible. Further, we considered that chronic and prolonged stressors are more likely to have harmful consequences than acute stressors. Thus, stressors occurring within a more narrow interval (eg, 2 months) may not have the same association with vitiligo. Future studies are warranted to precisely identify the type and timing of psychological stressors preceding vitiligo onset.

Conclusion

In conclusion, there is a high prevalence of stressful life events preceding vitiligo, which may play an important role as disease triggers as well as predict the presence of intermittent abdominal cramping and itching or burning of skin. These associations indicate that screening of vitiligo patients for psychological stressors, abdominal cramping, and itching and/or burning of skin should be included in the routine assessment of vitiligo patients.

Appendix

Please refer to the eTables in the PDF.

Vitiligo is the loss of skin pigmentation caused by autoimmune destruction of melanocytes. Multiple pathogenic factors for vitiligo have been described, including CD8⁺ T lymphocyte/T helper 1 infiltrates in lesional skin^1,2 with increased expression of IFN-γ³ and tumor necrosis factor α,^3-6 decreased transforming growth factor β,⁷ and circulating autoantibodies against tyrosine hydroxylase.⁸ Additionally, several studies have found a high prevalence of antecedent psychological stressors in vitiligo patients, suggesting that specific stressors may trigger and/or exacerbate vitiligo.^9-12

The relationship between antecedent psychological stressors and vitiligo extent has not been well studied. Potential mechanisms for stress-triggered vitiligo include increased catecholamines¹³ and neuropeptides,¹⁴ which have been found in vitiligo patients. However, the complex relationship between stressors and subsequent vitiligo is not well defined. We hypothesized that persistent stressors are associated with increased vitiligo extent.

Vitiligo is classically considered to be a silent pigmentary disorder with few or no symptoms. Prior studies have demonstrated that one-third of vitiligo patients report skin symptoms (eg, pruritus, burning), which may be specifically associated with early-onset disease.^15-17 Further, we observed that some vitiligo patients report abdominal cramping associated with their disease. Few studies have described the burden of skin symptoms and other associated symptoms in vitiligo or their determinants.

We conducted a prospective questionnaire-based study of 1541 adult vitiligo patients to identify psychological factors that may precede vitiligo onset. We hypothesized that some types of stressors that occur within 2 years prior to disease onset would have specific associations with vitiligo and/or somatic symptoms.

Methods

Study Population and Questionnaire Distribution

This prospective questionnaire-based study was approved by the institutional review board at St. Luke’s-Roosevelt Hospital Center (now Mount Sinai St. Luke’s-Roosevelt) (New York, New York) for adults (>18 years; male or female) with vitiligo. The survey was validated in paper format at St. Luke’s-Roosevelt Hospital Center and distributed online to members of nonprofit support groups for vitiligo vulgaris, as previously described.¹⁵

Questionnaire

The a priori aim of this questionnaire was to identify psychological factors that may precede vitiligo onset. The questionnaire consisted of 77 items (55 closed questions and 22 open questions) pertaining to participant demographics/vitiligo phenotype and psychological stressors preceding vitiligo onset. The questions related to this study and response rates are listed in eTable 1. Responses were verified by screening for noninteger or implausible values (eg, <0 or >100 years of age).

Sample Size

The primary outcome used for sample size calculation was the potential association between vitiligo and the presence of antecedent psychological stressors. Using a 2-tailed test, we determined that a sample size of 1264 participants would have 90% power at α=.05 and a baseline proportion of 0.01 (1% presumed prevalence of vitiligo) to detect an odds ratio (OR) of 2.5 or higher.¹⁸

Data and Statistical Analysis

Closed question responses were analyzed using descriptive statistics. Open-ended question responses were analyzed using content analysis. Related comments were coded and grouped, with similarities and differences noted. All data processing and statistics were done with SAS version 9.2. Age at diagnosis (years) and number of anatomic sites affected were divided into tertiles for statistical analysis due to wide skewing.    

Logistic regression models were constructed with numbers of reported deaths or stressors per participant within the 2 years prior to vitiligo onset as independent variables (0, 1, or ≥2), and symptoms associated with vitiligo as dependent variables. Adjusted ORs were calculated from multivariate models that included sex, current age (continuous), and comorbid autoimmune disease (binary) as covariates. Linear interaction terms were tested and were included in final models if statistically significant (P<.05).

Ordinal logistic regression was used to analyze the relationship between stressors (and other independent variables) and number of anatomic sites affected with vitiligo (tertiles). Ordinal logistic regression models were constructed to examine the impact of psychological stressors on pruritus secondary to vitiligo (not relevant combined with not at all, a little, a lot, very much) as the dependent variable. The proportional odds assumption was met in both models, as judged by score testing (P>.05). Binary logistic regression was used to analyze laterality, body surface area (BSA) greater than 25%, and involvement of the face and/or body with vitiligo lesions (binary).

Binary logistic regression models were constructed with impact of psychological stressors preceding vitiligo onset on comorbid abdominal cramping and specific etiologies as the dependent variables. There were 20 candidate stressors occurring within the 2 years prior to vitiligo onset. Selection methods for predictors were used to identify significant covariates within the context of the other covariates included in the final models. The results of forward, backward, and stepwise approaches were similar, and the stepwise selection output was presented.

Missing values were encountered because some participants did not respond to all the questionnaire items. A complete case analysis was performed (ie, missing values were ignored throughout the study). Data imputation was considered by multiple imputations; however, there were few or no differences between the estimates from the 2 approaches. Therefore, final models did not involve data imputation.

The statistical significance for all estimates was considered to be P<.05. However, a P value near .05 should be interpreted with caution given the multiple dependent tests performed in this study with increased risk for falsely rejecting the null hypothesis.

Results

Survey Population Characteristics

One thousand seven hundred participants started the survey; 1632 completed the survey (96.0% completion rate) and 1553 had been diagnosed with vitiligo by a physician. Twelve participants were excluded because they were younger than 18 years, leaving 1541 evaluable participants. Five hundred thirty-eight participants (34.9%) had comorbid autoimmune disorders. Demographics and disease phenotypes of the study participants are listed in Table 1.

Stressors Preceding Vitiligo Onset

Eight hundred twenty-one participants (56.6%) experienced at least one death or stressor within 2 years prior to vitiligo onset (Table 2), including death of a loved one (16.6%) and stressful life events (51.0%) within the 2 years prior to the onset of vitiligo, especially work/financial problems (10.8%), end of a long-term relationship (10.2%), and family problems (not otherwise specified)(7.8%). Two hundred (13.5%) participants reported experiencing 1 death and 46 (3.1%) reported multiple deaths. Five hundred participants (33.6%) reported experiencing 1 stressor and 259 (17.4%) reported multiple stressors.

Stressors Not Associated With Vitiligo Extent

The number of deaths or stressors reported per participant within the 2 years prior to vitiligo onset were not associated with BSA, laterality, or distribution of lesions (Table 3 and eTable 2–eTable 4).

Symptoms Associated With Vitiligo

Five hundred twenty-two participants (34.5%) reported intermittent abdominal cramping, including premenstrual and/or menstrual cramping in women (9.7%), food-related abdominal cramping (4.4%), inflammatory bowel syndrome (IBS)(2.6%), anxiety-related abdominal cramping (1.5%), autoimmune gastrointestinal disorders (1.2%), and “other” etiologies (20.4%). Five hundred ten participants reported itching and/or burning associated with vitiligo lesions (35.1%).

Intermittent abdominal cramping overall was associated with a BSA greater than 75% (OR, 1.65; 95% confidence interval (CI), 1.17-2.32; P=.004). However, specific etiologies of abdominal cramping were not significantly associated with BSA (P≥.11). In contrast, itching and/or burning from vitiligo lesions was associated with a BSA greater than 25% (OR, 1.53; 95% CI, 1.23-1.90; P<.0001).

Association Between Number of Stressors and Symptoms in Vitiligo

A history of multiple stressors (≥2) within the 2 years prior to vitiligo onset was associated with intermittent abdominal cramping overall (OR, 1.84; 95% CI, 1.38-2.47; P<.0001), including premenstrual and/or menstrual cramping in women (OR, 1.84; 95% CI, 1.15-2.95; P=.01), IBS (OR, 3.29; 95% CI, 1.34-8.05; P=.01), and autoimmune gastrointestinal disorders (OR, 4.02; 95% CI, 1.27-12.80; P=.02)(eTable 5). These associations remained significant in multivariate models that included age, sex, and BSA as covariates. However, a history of 1 stressor or death or multiple deaths in the 2 years prior to vitiligo onset was not associated with any etiology of abdominal cramping.

Experiencing 1 (OR, 1.43; 95% CI, 1.12-1.82; P=.005) or multiple stressors (OR, 1.51; 95% CI,  1.12-2.04; P=.007) also was associated with itching and/or burning secondary to vitiligo. This association remained significant in a multivariate model that included age, sex, and BSA as covariates. However, a history of 1 or multiple deaths in the 2 years prior to vitiligo onset was not associated with itching and/or burning.

Association Between Specific Stressors and Vitiligo Symptoms

Perimenstrual (premenstrual and/or menstrual) cramping in women was associated with family problems (not otherwise specified) within the 2 years prior to vitiligo onset (Table 4). Food-related abdominal cramping was associated with school- and/or test-related stressors. Diagnosis of IBS was associated with health problems or surgery and being a victim of abuse within the 2 years prior to onset of vitiligo. Autoimmune gastrointestinal disorders were associated with moving to a new home/region, health problems or surgery, and witness to a violent crime or death. Finally, itching and/or burning of vitiligo lesions was associated with work and financial problems.

Comment

The present study found a high frequency of stressful life events and deaths of loved ones occurring within the 2 years preceding vitiligo onset. A history of multiple stressors but not deaths of loved ones was associated with more frequent symptoms in vitiligo patients, including itching and/or burning and intermittent abdominal pain. Specific stressors were associated with intermittent abdominal cramping, which occurred in approximately one-third of vitiligo patients. Abdominal cramping was related to menses in women, anxiety, foods, IBS, autoimmune gastrointestinal disorders, and other etiologies of abdominal cramping, which underscores the complex relationship between stressors, vitiligo, and inflammation. It is possible that stress-related immune abnormalities occur in vitiligo, which may influence the development of other autoimmune disorders. Alternatively, abdominal symptoms may precede and perhaps contribute to psychological stressors and impaired quality of life in vitiligo patients; however, the cross-sectional nature of the study did not allow us to elucidate this temporal relationship.

The present study found that 56.6% of participants experienced 1 or more deaths (17%) and/or stressful life events (51%) within the 2 years prior to vitiligo onset. These results are consistent with prior smaller studies that demonstrated a high frequency of stressful events preceding vitiligo onset. A case-controlled study found stressful events in 12 of 21 (57%) Romanian children with vitiligo, which was higher than controls.¹⁹ Another questionnaire-based, case-controlled study compared a heterogeneous group of 32 adolescent and adult Romanian patients with vitiligo and found higher odds of a stressful event in women preceding vitiligo diagnosis compared to controls.¹⁰ A retrospective analysis of 65 Croatian patients with vitiligo also reported that 56.9% (37/65) had some associated psychological factors.⁹ Another retrospective study of 31 adults with vitiligo found increased occurrence of 3 or more uncontrollable events, decreased perceived social support, and increased anxiety in vitiligo patients versus 116 other dermatologic disease controls.¹² A questionnaire-based study found increased bereavements, changes in sleeping and eating habits, and personal injuries/illnesses in 73 British adults with vitiligo compared to 73 other age- and sex-matched dermatologic disease controls.¹¹ All of these studies were limited by a small sample size, and the patient populations were localized to a regional dermatology referral center. The present study provided a larger analysis of stressful life events preceding vitiligo onset and included a diverse patient population.

The present study found that stressful life events and deaths of a loved one are not associated with vitiligo extent and distribution. This finding suggests that stressful life events may act as vitiligo triggers in genetically predisposed individuals, but ultimately the disease course and prognosis are driven by other factors, such as increased systemic inflammation or other immunologic abnormalities. Indeed, Silverberg and Silverberg²⁰ and other investigators^21,22 reported relative deficiencies of 25-hydroxyvitamin D,²³ vitamins B₆ and B₁₂, and folic acid,²⁰ as well as elevated serum homocysteine levels in vitiligo patients. Increased serum homocysteine levels were associated with increased BSA of vitiligo lesions.²⁰ Elevated serum homocysteine levels also have been associated with increased inflammation in coronary artery disease,²⁴ psoriasis,^25,26 and in vitro.²⁷ These laboratory anomalies likely reflect an underlying predisposition toward vitiligo, which might be triggered by stress responses or secondarily altered immune responses.

The present study had several strengths, including being prospective with a large sample size. The patient population included a large sample of men and women with representation of various adult ages and vitiligo extent. However, this study also had potential limitations. Measures of vitiligo extent were self-reported and were not clinically assessed. To address this limitation, we validated the questionnaire before posting it online.¹⁵ Invitation to participate in the survey was distributed by vitiligo support groups, which may have resulted in a selection bias toward participants with greater disease severity or with a poorer quality of life associated with vitiligo. Invitation to participate in this study was sent to members of vitiligo support groups, which allowed for recruitment of a large number of vitiligo patients despite a relatively low prevalence of disease in the general population. However, there are several challenges using this approach for nonvitiligo controls. Using participants with another dermatological disease as a control group may yield spurious results. Ideally, a large randomized sample of healthy participants with minimization of bias should be used for controls, which is an ambitious undertaking that was beyond the scope of this pilot study and will be the subject of future studies. Finally, this analysis found associations between stressors that occurred in the 2 years prior to vitiligo onset with symptomatic disease. We chose a broad interval for stressors because early vitiligo lesions may go unnoticed, making recognition of stressors occurring within days or weeks of onset infeasible. Further, we considered that chronic and prolonged stressors are more likely to have harmful consequences than acute stressors. Thus, stressors occurring within a more narrow interval (eg, 2 months) may not have the same association with vitiligo. Future studies are warranted to precisely identify the type and timing of psychological stressors preceding vitiligo onset.

Conclusion

In conclusion, there is a high prevalence of stressful life events preceding vitiligo, which may play an important role as disease triggers as well as predict the presence of intermittent abdominal cramping and itching or burning of skin. These associations indicate that screening of vitiligo patients for psychological stressors, abdominal cramping, and itching and/or burning of skin should be included in the routine assessment of vitiligo patients.

Appendix

Please refer to the eTables in the PDF.