Comparison of Dermatologist Density Between Urban and Rural Counties in the United States

Two-step hierarchical binary classification of cancerous skin lesions using transfer learning and the random forest algorithm

Skin lesion classification plays a crucial role in the early detection and diagnosis of various skin conditions. Recent advances in computer-aided diagnostic techniques have been instrumental in timely intervention, thereby improving patient outcomes, particularly in rural communities lacking specialized expertise. Despite the widespread adoption of convolutional neural networks (CNNs) in skin disease detection, their effectiveness has been hindered by the limited size and data imbalance of publicly accessible skin lesion datasets. In this context, a two-step hierarchical binary classification approach is proposed utilizing hybrid machine and deep learning (DL) techniques. Experiments conducted on the International Skin Imaging Collaboration (ISIC 2017) dataset demonstrate the effectiveness of the hierarchical approach in handling large class imbalances. Specifically, employing DenseNet121 (DNET) as a feature extractor and random forest (RF) as a classifier yielded the most promising results, achieving a balanced multiclass accuracy (BMA) of 91.07% compared to the pure deep-learning model (end-to-end DNET) with a BMA of 88.66%. The RF ensemble exhibited significantly greater efficiency than other machine-learning classifiers in aiding DL to address the challenge of learning with limited data. Furthermore, the implemented predictive hybrid hierarchical model demonstrated enhanced performance while significantly reducing computational time, indicating its potential efficiency in real-world applications for the classification of skin lesions.

Integration of a deep learning basal cell carcinoma detection and tumor mapping algorithm into the Mohs micrographic surgery workflow and effects on clinical staffing: A simulated, retrospective study

Background

Artificial intelligence (AI) enabled tools have been proposed as 1 solution to improve health care delivery. However, research on downstream effects of AI integration into the clinical workflow is lacking.

Objective

We aim to analyze how integration of an automated basal cell carcinoma detection and tumor mapping algorithm in a Mohs micrographic surgery unit impacts the work efficiency of clinical and laboratory staff.

Methods

Slide, staff, and histotechnician waiting times were analyzed over a 20-day period in a Mohs micrographic surgery unit. A simulated AI workflow was created and the time differences between the real and simulated workflows were compared.

Results

Simulated nonautonomous algorithm integration led to savings of 35.6% of slide waiting time, 18.4% of staff waiting time, and 18.6% of histotechnician waiting time per day. Algorithm integration on days with increased reconstruction complexity resulted in the greatest time savings.

Limitations

One Mohs micrographic surgery unit was analyzed and simulated AI integration was performed retrospectively.

Conclusions

AI integration results in reduced staff waiting times, enabling increased productivity and a streamlined clinical workflow. Schedules containing surgical cases with either increased repair complexity or numerous tumor removal stages stand to benefit most. However, significant logistical challenges must be addressed before broad adoption into clinical practice is realistic.

Persistent poverty and incidence-based melanoma mortality in Texas

PURPOSE

Previous studies have shown that individuals living in areas with persistent poverty (PP) experience worse cancer outcomes compared to those living in areas with transient or no persistent poverty (nPP). The association between PP and melanoma outcomes remains unexplored. We hypothesized that melanoma patients living in PP counties (defined as counties with ≥ 20% of residents living at or below the federal poverty level for the past two decennial censuses) would exhibit higher rates of incidence-based melanoma mortality (IMM).

METHODS

We used Texas Cancer Registry data to identify the patients diagnosed with invasive melanoma or melanoma in situ (stages 0 through 4) between 2000 and 2018 (n = 82,458). Each patient's PP status was determined by their county of residence at the time of diagnosis.

RESULTS

After adjusting for demographic variables, logistic regression analyses revealed that melanoma patients in PP counties had statistically significant higher IMM compared to those in nPP counties (17.4% versus 11.3%) with an adjusted odds ratio of 1.35 (95% CI 1.25-1.47).

CONCLUSION

These findings highlight the relationship between persistent poverty and incidence-based melanoma mortality rates, revealing that melanoma patients residing in counties with persistent poverty have higher melanoma-specific mortality compared to those residing in counties with transient or no poverty. This study further emphasizes the importance of considering area-specific socioeconomic characteristics when implementing place-based interventions to facilitate early melanoma diagnosis and improve melanoma treatment outcomes.

The Distribution of Dermatologists in the Philippines

BACKGROUND

Equitable health manpower distribution is essential for the successful implementation of the Universal Health Care (UHC) program by the Philippine Department of Health. Mapping the distribution and profile of dermatologists in the Philippines can improve Filipinos' access to skin disease treatment.

METHODS

A review of the database of dermatologists from the Philippine Dermatological Society (PDS) members' directory (as of November 2023), as well as the Philippine Health Insurance Corporation (PhilHealth) database (as of July 2023), was conducted. The distribution of PDS-accredited dermatologists was analyzed by geographic location, demographic profile (age and sex), density (per 100,000 people), and the dermatologist-to-general practitioner (GP) ratio. Heatmaps illustrating the distribution of dermatologists in the Philippines and the ratio of PhilHealth-accredited PDS board-certified dermatologists to GPs were created.

RESULTS

Out of 1389 PDS board-certified dermatologists, 1345 resided in the Philippines. The majority were women (1221/1345, 90.78%), with a median age of 47 years (range: 23 to 85). More than half were practicing in the National Capital Region (NCR) (684/1345, 50.86%). The overall dermatologist density was approximately 1 per 100,000 people (1.19); it was highest for the Luzon Island group (1.54) (NCR, 4.80) and lowest for the Mindanao Island group (0.27; the Bangsamoro Autonomous Region of Muslim Mindanao or BARMM, 0.04). Less than one-third (396/1345, 29.44%) of dermatologists were PhilHealth-accredited, with a density of 0.35 dermatologists per 100,000 people. Out of 45218 PhilHealth-accredited physicians, 396 (0.88%) were dermatologists while 11748 (25.98%) were GPs. The overall dermatologist-to-GP ratio among PhilHealth-accredited physicians was 1:30; it was highest in the Luzon Island group (1:20) and lowest in the Mindanao Island group (1:118).

CONCLUSION

The Philippines lacks dermatologists in regions outside the NCR. The majority are women, and few are PhilHealth-accredited. The dermatologist-to-GP ratio among PhilHealth-accredited physicians is low. Dermatology training programs should encourage more applicants, especially men, and prioritize applicants from underserved regions.

Comparing preferences for skin cancer screening: AI-enabled app vs dermatologist

BACKGROUND AND AIM

Skin cancer is a major public health issue. While self-examinations and professional screenings are recommended, they are rarely performed. Mobile health (mHealth) apps utilising artificial intelligence (AI) for skin cancer screening offer a potential solution to aid self-examinations; however, their uptake is low. Therefore, the aim of this research was to examine provider and user characteristics influencing people's decisions to seek skin cancer screening performed by a mHealth app or a dermatologist.

METHODS

Two forced-choice conjoint experiments with Nmain = 1591 and Nreplication = 308 participants from the United States were conducted online to investigate preferences for screening providers. In addition to the provider type (mHealth app vs dermatologist), the following provider attributes were manipulated: costs, expertise, privacy policy, and result details. Subsequently, a questionnaire assessed various user characteristics, including demographics, attitudes toward AI technology and medical mistrust.

RESULTS

Outcomes were consistent across the two studies. The provider type was the most influential factor, with the dermatologist being selected more often than the mHealth app. Cost, expertise, and privacy policy also significantly impacted decisions. Demographic subgroup analyses showed rather consistent preference trends across various age, gender, and ethnicity groups. Individuals with greater medical mistrust were more inclined to choose the mHealth app. Trust, accuracy, and quality ratings were higher for the dermatologist, whether selected or not.

CONCLUSION

Our results offer valuable insights for technology developers, healthcare providers, and policymakers, contributing to unlocking the potential of skin cancer screening apps in bridging healthcare gaps in underserved communities.

Rural Melanoma Patients Have Less Surgery and Higher Melanoma-Specific Mortality

BACKGROUND

Melanoma causes most skin cancer-related deaths, and disparities in mortality persist. Rural communities, compared to urban, face higher levels of poverty and more barriers to care, leading to higher stage at presentation and shorter survival in melanoma. To further evaluate these disparities, we sought to assess the association between rurality and melanoma cause-specific mortality and receipt of recommended surgery in a national cohort.

METHODS

Patients with primary non-ocular, cutaneous melanoma from the SEER database, 2000-2017, were included. Outcomes included melanoma-specific survival and receipt of recommended surgery. Rurality was based on Rural-Urban Continuum Codes. Variables included age, sex, race, ethnicity, income, and stage. Multivariate regression models assessed the effect of rurality on survival and receipt of recommended surgery.

RESULTS

103,606 patients diagnosed with non-ocular cutaneous primary melanoma met criteria during this period. 93.3% (n = 96620) were in urban areas and 6.7% (n = 6986) were in rural areas. On multivariate regression controlling for age, sex, race, ethnicity, and stage patients living in a rural area were less likely to receive recommended surgery (aOR .52, 95% CI: .29-.90, P = .02) and had increased hazard of melanoma-specific mortality (aHR 1.19, 95% CI: 1.02-1.40, P = .03) even after additionally controlling for surgery receipt.

CONCLUSION

Using a large national cohort, our study found that rural patients were less likely to receive recommended surgery and had shorter melanoma cause-specific survival. Our findings highlight the importance of access to cancer care in rural areas and how this ultimately effects survival for these patients.

Artificial intelligence and skin cancer

Artificial intelligence is poised to rapidly reshape many fields, including that of skin cancer screening and diagnosis, both as a disruptive and assistive technology. Together with the collection and availability of large medical data sets, artificial intelligence will become a powerful tool that can be leveraged by physicians in their diagnoses and treatment plans for patients. This comprehensive review focuses on current progress toward AI applications for patients, primary care providers, dermatologists, and dermatopathologists, explores the diverse applications of image and molecular processing for skin cancer, and highlights AI's potential for patient self-screening and improving diagnostic accuracy for non-dermatologists. We additionally delve into the challenges and barriers to clinical implementation, paths forward for implementation and areas of active research.

Implementation of a postgraduate dermatology fellowship program for nurse practitioners

ABSTRACT

Transition to practice programs (also referred to as fellowship, residency, or postgraduate training programs) for nurse practitioners (NPs) are becoming more popular, especially in specialties such as dermatology. A nationwide shortage of dermatology clinicians, which had led to long appointment wait times and inadequate patient access to care, has led to more NPs practicing dermatology to help meet the demand for care. New graduate NPs may struggle in their transition to practice, and fellowship programs have been shown to support NPs as they transition from novice to expert. In this article, the University of Rochester Medical Center shares its experience in developing, implementing, and managing a postgraduate dermatology fellowship program for NPs.

Melanoma Incidence Rates Among Non-Hispanic American Indian/Alaska Native Individuals, 1999-2019

Importance

Non-Hispanic American Indian/Alaska Native people have the second highest incidence rate of invasive cutaneous melanoma in the US after non-Hispanic White people.

Objective

To examine invasive cutaneous melanoma incidence rates and trends over time among non-Hispanic American Indian/Alaska Native people.

Design, Setting, and Participants

This descriptive, observational cross-sectional study used population-based cancer registry data (US Cancer Statistics AI/AN Incidence Analytic Database) linked to the Indian Health Service administrative database to examine incidence rates by age, sex, region, histology, tumor site, stage, and other demographic and clinical characteristics. The study examined trends from 1999 to 2019 time period by age, sex, stage at diagnosis, and region. Non-Hispanic American Indian/Alaska Native people 15 years and older who received a diagnosis of invasive cutaneous melanoma from 1999 to 2019 who were members of federally recognized tribes and resided in Indian Health Service purchased/referred care delivery areas were included in this study to reduce racial misclassification and provide more accurate rates. The data were analyzed in 2022.

Exposures

Demographic and clinical characteristics, such as age, sex, geographic region, histology, stage, and tumor site.

Main Outcomes and Measures

Invasive cutaneous melanoma incidence rates by age group, sex, region, resident county characteristics (poverty level, rurality, education level, and socioeconomic status), stage at diagnosis, tumor site, and histology. Trends over time by age, sex, region, and stage.

Results

From 1999 to 2019, 2151 non-Hispanic American Indian/Alaska Native people (1021 female individuals [47.5%]) received a diagnosis of incident cutaneous melanoma (rate, 10.7 per 100 000; 95% CI, 10.3-11.2). Rates were higher among male than female individuals (13.0 [95% CI, 12.2-13.8] vs 9.2 [95% CI, 8.6-9.8]) and for people 55 years and older (24.2; 95% CI, 22.8-25.7) compared with those aged 15 to 39 years (3.5; 95% CI, 3.2-3.9). Rates were highest for male individuals 55 years and older (34.5; 95% CI, 31.8-37.3) and people living in the Southern Plains (male individuals: 23.8; 95% CI, 21.5-26.2; female individuals: 15.5; 95% CI, 14.0-17.2) and Pacific Coast region (male individuals: 16.5; 95% CI, 14.5-18.7; female individuals: 12.3; 95% CI, 10.9-13.9). Rates increased among female individuals from 1999 to 2019 (average annual percent change [AAPC], 2.5; P < .001); among regional/distant stage tumors (AAPC, 2.5; P = .01) and people 55 years and older (AAPC, 2.8; P = .001).

Conclusions and Relevance

The results of this study suggest that additional studies could potentially identify risk factors among non-Hispanic American Indian/Alaska Native people.

The Social and Home Environment: Impacts of Determinants of Health on Atopic Dermatitis, Pathways Toward Solutions, and Unique Considerations for Rural and Remote North American Indigenous Populations

Disparities in environmental and social determinants of health (DOH) are associated with morbidity in atopic dermatitis (AD). The socioecological model (SEM) is a framework that can be applied to better understand how such DOH impacts patients with AD. We include a case scenario of a remote Indigenous patient reflective of real-world situations of living with AD and examine relevant impact, gaps in knowledge, and further research needs. This review highlights a variety of social and environmental exposures as important DOH which must be addressed to achieve optimal management in AD. The "rainbow model" is a modified framework to help illustrate how complex environmental and social forces impact both AD presentation and therapeutic success. However, practical applications and outcome metrics for health promotion are limited. An inter- and transdisciplinary approach is paramount to address the complex challenges associated with AD care, as well as multistakeholder approach integrating culturally-competent equitable health frameworks. This review underscores the importance of expanding the focus of AD management beyond basic science and clinical trials to recognize and address health disparities and to promote optimal health and well-being in patients with AD, and contributes a working approach to mapping the complex interventions and patient-oriented research needed using a focus on remote North American Indigenous patients affected by AD.

CR-Conformer: a fusion network for clinical skin lesion classification

Deep convolutional neural network (DCNN) models have been widely used to diagnose skin lesions, and some of them have achieved diagnostic results comparable to or even better than dermatologists. Most publicly available skin lesion datasets used to train DCNN were dermoscopic images. Expensive dermoscopic equipment is rarely available in rural clinics or small hospitals in remote areas. Therefore, it is of great significance to rely on clinical images for computer-aided diagnosis of skin lesions. This paper proposes an improved dual-branch fusion network called CR-Conformer. It integrates a DCNN branch that can effectively extract local features and a Transformer branch that can extract global features to capture more valuable features in clinical skin lesion images. In addition, we improved the DCNN branch to extract enhanced features in four directions through the convolutional rotation operation, further improving the classification performance of clinical skin lesion images. To verify the effectiveness of our proposed method, we conducted comprehensive tests on a private dataset named XJUSL, which contains ten types of clinical skin lesions. The test results indicate that our proposed method reduced the number of parameters by 11.17 M and improved the accuracy of clinical skin lesion image classification by 1.08%. It has the potential to realize automatic diagnosis of skin lesions in mobile devices.

Community Dermatology: Current Status and the Way Forward

Community health interventions in dermatology have the potential to deliver healthcare to large populations. Such interventions can bridge the gap between increasing demands from vast populations and limited availability and asymmetric geographical distribution of specialist doctors. This narrative review aims to delineate various areas in which community dermatology initiatives are useful, the different approaches used, and the factors which could determine their effectiveness. We did a PubMed search using the search terms, "Community Health Services"[Mesh]) AND "Dermatology"[Mesh]. A further literature search was performed by cross-referencing these articles. In addition to its primary purpose of increasing access, community involvement can augment medical education, foster research, and help to develop more patient-centered clinical guidelines, understand disease trends and influences of the environment on various diseases, and formulate population-centered health policy. Task shifting, development of culture-sensitive and gender-sensitive community healthcare teams, disease-specific patient support groups, and use of advances in technology such as telemedicine are some of the important methods used in community dermatology. Task shifting may be performed at different levels - general practitioners, pharmacists, nurses, and community health workers. Community interventions harness volunteerism among health professionals as well as various stakeholders in the community. Partnering with non-governmental organizations, involving members of the community, and being a part of a long-term public health program help to sustain community interventions. The interventions should adapt to the ethical and cultural norms of the community. Continuity of care, fidelity, and evidence support are crucial for the success of community interventions in dermatology.

Using the RE-AIM framework to assess national teledermatology expansion

Background

Teledermatology has been utilized in the United States Department of Veterans Affairs (VA) for decades but continues to have incomplete penetration. VA has funded an initiative to enhance access to dermatology services since 2017 to support asynchronous teledermatology for Veterans living in rural areas. As part of an ongoing evaluation of this program, we assessed the teledermatology activity between the fiscal years 2020 and 2022. We focused on the second cohort of the initiative, comprising six VA facilities and their 54 referral clinics.

Methods

We studied teledermatology programs at cohort facilities using the reach, effectiveness, adoption, implementation, and maintenance framework. We used a mixed-methods design including annual online reports completed by participating facilities and VA administrative data. When possible, we compared the data from the 3 years of teledermatology funding with the baseline year prior to the start of funding.

Findings

Reach: Compared with the baseline year, there was a 100% increase in encounters and a 62% increase in patients seen at the funded facilities. Over 500 clinicians and support staff members were trained. Effectiveness: In FY 2022, primary or specialty care clinics affiliated with the funded facilities had more dermatology programs than primary or specialty care clinics across the VA (83% vs. 71% of sites). Adoption: By the end of the funding period, teledermatology constituted 16% of dermatology encounters at the funded facilities compared with 12% nationally. This reflected an increase from 9.2% at the funded facilities and 10.3% nationally prior to the funding period. Implementation: The continued funding for staff and equipment facilitated the expansion to rural areas. Maintenance: By the end of the funding period, all facilities indicated that they had fully implemented their program for patients of targeted primary care providers. The Program Sustainability Index scores generally increased during the funding period.

Conclusions

Targeted funding to support asynchronous teledermatology implementation for rural Veterans increased its reach, adoption, and implementation, ultimately improving access. Providing program guidance with staffing and training resources can increase the impact of these programs. Ongoing efforts to maintain and increase communication between primary care and dermatology will be needed to sustain success.

Principles, applications, and future of artificial intelligence in dermatology

This paper provides an overview of artificial-intelligence (AI), as applied to dermatology. We focus our discussion on methodology, AI applications for various skin diseases, limitations, and future opportunities. We review how the current image-based models are being implemented in dermatology across disease subsets, and highlight the challenges facing widespread adoption. Additionally, we discuss how the future of AI in dermatology might evolve and the emerging paradigm of large language, and multi-modal models to emphasize the importance of developing responsible, fair, and equitable models in dermatology.

Differential patient travel distance and time to psoriasis clinical trial sites

Considering the known disparities in racial representation in psoriasis clinical trials, this study sought to characterize travel distance and time to reach a psoriasis clinical trial site as a potential barrier to trial participation for multiple demographic and geographic variables. We determined travel distance and time from every census tract population center in the United States to the nearest psoriasis clinical trial site using ArcGIS and linked travel estimates to demographic characteristics in each census tract based on 2020 American Community Survey. The average distance and time traveled to reach a psoriasis clinical trial site nationally were 45.6 miles and 51.8 min, respectively. Urban residence and Northeast location had significantly lower travel distance and time relative to their geographic counterparts. Travel burden was significantly greater among Native American and Black races, individuals without college education and Veterans Affairs beneficiaries relative to their counterparts. These findings reveal disparate access regarding rurality, race, education and insurance type, which may encourage investigators to increase travel funding for underrepresented groups and diversity recruitment efforts to promote access to psoriasis clinical trials.

Trends Over Time in Medicare for Advanced Practice Clinicians in Dermatology, 2013-2020

Importance

The number of advanced practice clinicians (APCs, including nurse practitioners and physician assistants) in the US is increasing. The effect this has on dermatology is unclear.

Objective

To develop a method to identify APCs practicing dermatology in claims data and to evaluate the contribution of dermatology APCs to the dermatology workforce and how this has changed over time.

Design, Setting, and Participants

This retrospective cohort study used the Medicare Provider Utilization and Payment Data Public Use files (2013 to 2020). As APCs are not listed by specialty, a method to identify APCs practicing dermatology was developed and validated using common dermatology procedural codes. The data were analyzed from November 2022 to April 2023.

Main Outcomes and Measures

The proportion of clinicians and office visits by dermatology APCs and physician dermatologists were evaluated using Mann-Kendall tests. Joinpoint analysis was also used to compare the average annual percentage change of dermatology procedures and clinicians in rural-urban areas between dermatology APCs and physician dermatologists.

Results

The method to identify APCs practicing dermatology had 96% positive predictive value, 100% negative predictive value, 100% sensitivity, and 100% specificity. Between 2013 and 2020, 8444 dermatology APCs and 14 402 physician dermatologists were identified. They provided 109 366 704 office visits in Medicare. The percentage of dermatology clinicians who were APCs increased over time, from 27.7% in 2013 to 37.0% in 2020 (P = .002). The proportion of dermatologic office visits provided by APCs also increased over time, from 15.5% in 2013 to 27.4% in 2020 (P = .002). For all procedure categories, the average annual percentage change was positive for dermatology APCs (range, 10.05%-12.65%) and was higher than that of physician dermatologists. For all rural-urban designations, the average annual percentage change was positive for dermatology APCs (range, 2.03%-8.69%) and was higher than metropolitan, micropolitan, and small-town areas from that of physician dermatologists.

Conclusions and Relevance

In this retrospective cohort study, there was a temporal increase in the amount of dermatologic care provided by APCs in Medicare. These findings demonstrate changes in the dermatology workforce and may have implications for dermatology as a specialty.

Factors facilitating shared decision making in eczema: Met and unmet needs from the patient perspective

Background

Research has shown that eczema patients prefer some degree of shared control over treatment decisions, but little is known about factors perceived to be important to facilitate shared decision making (SDM).

Objective

To determine factors eczema patients and caregivers consider to be important for SDM, and how often they experience them with their eczema healthcare provider (HCP).

Methods

A cross-sectional survey study (64 questions) was conducted, which included factors related to SDM rated by respondents on a Likert scale for importance, and how often these factors were true with their current eczema HCP.

Results

Respondents (840, response rate 62.4%) most frequently rated their health literacy and communication skills as important for SDM. Factors which indicated a strong provider-patient relationship, and HCPs who initiate treatment conversations were also deemed beneficial. Low importance was placed on concordant HCP race/ethnicity, however, of those who did rate it as important, 53/91 identified as Black (half of all Black respondents).

Limitations

A high proportion of respondents were aware of the term SDM prior to the survey.

Conclusions

SDM is more likely to be facilitated when patient education and empowerment are coupled with HCPs who initiate treatment discussions, maintain compassion resilience, and listen to patient perspectives.

Cultural Competence and Humility

The increasing diversification of the United States has led to more racially and ethnically discordant visits between health care providers and patients; this is especially true in dermatology due to the lack of diversity in the field. Diversifying the health care workforce has been shown to reduce health care disparities and is an ongoing goal of dermatology. Improving cultural competence and humility among physicians is an important part of addressing health care inequities. This article reviews cultural competence, cultural humility, and practices dermatologists can incorporate to address this challenge.

The Social Determinants of Health and Their Impact on Dermatologic Health, Part 1: The Social Determinants of Health and Their Dermatologic Implications

The social determinants of health (SDoH) have significant influences on health and lead to health disparities in a variety of complex and intersecting ways. They are the nonmedical factors that must be addressed to improve health outcomes and achieve greater health equity. They are shaped by the structural determinants of health and impact individual socioeconomic status as well as the health of entire communities. Part 1 of this 2-part review aims to shed light on how the SDoH impact health and their specific implications on dermatologic health disparities.

History of Race in America

Racial and ethnic disparities exist across a wide range of disease areas and clinical services. Becoming familiar with the history of race in America, and how it has been used to structure laws or policies that drive inequities in the social determinants of health, even today, is necessary to mitigate these disparities across medicine.

A comparison of family physician and dermatologist topical corticosteroid prescriptions: A population-based cross-sectional study

BACKGROUND

Topical corticosteroids (TCS) are commonly prescribed to treat inflammatory skin diseases, and appropriate prescription is necessary for treatment success.

OBJECTIVE

To quantify differences between TCS prescribed by dermatologists at consultation and family physicians for patients treated for any skin condition.

METHODS

Using administrative health data in Ontario, we included all Ontario Drug Benefit recipients who filled at least one TCS prescription from a dermatologist at consultation and a family physician in the year prior between January 2014 and December 2019. We estimated mean differences and 95% confidence intervals in amount (in grams) and potency between the index dermatologist prescription and the highest and most recent family physician prescription amounts and potencies in the preceding year using linear mixed-effect models.

RESULTS

A total of 69,335 persons were included. The mean dermatologist amount was 34% larger than the highest amount and 54% larger than the most recent amount prescribed by family physicians. There were small but statistically significant differences in potency using established 7-category and 4-category potency classification systems.

CONCLUSIONS

Compared to family physicians, dermatologists prescribed substantially larger amounts and similarly potent TCS at consultation. Further research is needed to determine the effect of these differences on clinical outcomes.

Travel distance and time to dermatology clinical trial sites: a cross-sectional geospatial analysis

Despite increasing calls for diversity in clinical trial recruitment, data are lacking regarding disparities in access to dermatologic clinical trials. The objective of this study was to characterize travel distance and time to reach a dermatology clinical trial site considering patient demographic and location characteristics. We determined travel distance and time from every census tract population center in the United States to the nearest dermatologic clinical trial site using ArcGIS and linked travel estimates to demographic characteristics in each census tract based on 2020 American Community Survey. Nationally, patients travel an average of 14.3 miles and 19.7 min to reach a dermatologic clinical trial site. Significantly shorter travel distance and time were observed for urban and Northeast residence, White and Asian race and private insurance relative to rural and Southern residence, Native American and Black race and public insurance (p < 0.001). These findings reveal disparate access regarding geographic region, rurality, race and insurance type, which may encourage investigators to allocate funding for travel assistance for underrepresented and disadvantaged groups to promote access and diversity in dermatologic clinical trials.

Interpretable Skin Cancer Classification based on Incremental Domain Knowledge Learning

The recent advances in artificial intelligence have led to the rapid development of computer-aided skin cancer diagnosis applications that perform on par with dermatologists. However, the black-box nature of such applications makes it difficult for physicians to trust the predicted decisions, subsequently preventing the proliferation of such applications in the clinical workflow. In this work, we aim to address this challenge by developing an interpretable skin cancer diagnosis approach using clinical images. Accordingly, a skin cancer diagnosis model consolidated with two interpretability methods is developed. The first interpretability method integrates skin cancer diagnosis domain knowledge, characterized by a skin lesion taxonomy, into model development, whereas the other method focuses on visualizing the decision-making process by highlighting the dominant of interest regions of skin lesion images. The proposed model is trained and validated on clinical images since the latter are easily obtainable by non-specialist healthcare providers. The results demonstrate the effectiveness of incorporating lesion taxonomy in improving model classification accuracy, where our model can predict the skin lesion origin as melanocytic or non-melanocytic with an accuracy of 87%, predict lesion malignancy with 77% accuracy, and provide disease diagnosis with an accuracy of 71%. In addition, the implemented interpretability methods assist understand the model's decision-making process and detecting misdiagnoses. This work is a step toward achieving interpretability in skin cancer diagnosis using clinical images. The developed approach can assist general practitioners to make an early diagnosis, thus reducing the redundant referrals that expert dermatologists receive for further investigations.

Association of Patient Proximity to Dermatologic Care With Melanoma Stage at Diagnosis and Outcome

This cohort study evaluates the association of proximity to dermatologic clinicians with stage at diagnosis and cancer-specific survival among adults with cutaneous melanoma in Iowa.

DermX: An end-to-end framework for explainable automated dermatological diagnosis

Dermatological diagnosis automation is essential in addressing the high prevalence of skin diseases and critical shortage of dermatologists. Despite approaching expert-level diagnosis performance, convolutional neural network (ConvNet) adoption in clinical practice is impeded by their limited explainability, and by subjective, expensive explainability validations. We introduce DermX, an end-to-end framework for explainable automated dermatological diagnosis. DermX is a clinically-inspired explainable dermatological diagnosis ConvNet, trained using DermXDB, a 554 image dataset annotated by eight dermatologists with diagnoses, supporting explanations, and explanation attention maps. DermX+ extends DermX with guided attention training for explanation attention maps. Both methods achieve near-expert diagnosis performance, with DermX, DermX+, and dermatologist F1 scores of 0.79, 0.79, and 0.87, respectively. We assess the explanation performance in terms of identification and localization by comparing model-selected with dermatologist-selected explanations, and gradient-weighted class-activation maps with dermatologist explanation maps, respectively. DermX obtained an identification F1 score of 0.77, while DermX+ obtained 0.79. The localization F1 score is 0.39 for DermX and 0.35 for DermX+. These results show that explainability does not necessarily come at the expense of predictive power, as our high-performance models provide expert-inspired explanations for their diagnoses without lowering their diagnosis performance.

Urban versus rural utilization of teledermoscopy in self-skin examinations: preliminary results of a cohort study in the states of Oregon and Washington

In this letter, we examine the application of smartphone dermoscopy in teledermatology skin cancer screenings for individuals in rural communities. There is currently a paucity of literature that examines teledermoscopy in rural areas of the United States. This is significant in consideration of the barriers posed to individuals in rural areas seeking access to dermatology care. This manuscript addresses the intersection of technology and expanding access to underserved communities − issues paramount to the future of dermatology.

Factors Associated With Dermatologic Follow-up vs Emergency Department Return in Patients With Hidradenitis Suppurativa After an Initial Emergency Department Visit

Importance

Emergency department (ED) visitation is common for the treatment of hidradenitis suppurativa (HS), whereas dermatology outpatient care is low. The reasons underlying this differential follow-up have not been elucidated.

Objective

To assess the interventions and patient factors associated with ED return following an initial ED visit for HS.

Design, Setting, and Participants

This retrospective cohort study used data from the IBM® MarketScan® Commercial and Multi-State Medicaid databases (trademark symbols retained per database owner requirement). An HS cohort was formed from patients who had 2 or more claims for HS during the study period of 2010 to 2019 and with at least 1 ED visit for their HS or a defined proxy. Data were analyzed from November 2021 to May 2022.

Exposures

Factors analyzed included those associated with the ED visit and patient characteristics.

Main Outcomes and Measures

Primary outcomes were return to the ED or dermatology outpatient follow-up for HS or related proxy within 30 or 180 days of index ED visit.

Results

This retrospective cohort study included 20 269 patients with HS (median [IQR] age, 32 [25-41] years; 16 804 [82.9%] female patients), of which 7455 (36.8%) had commercial insurance and 12 814 (63.2%) had Medicaid. A total of 9737 (48.0%) patients had incision and drainage performed at the index ED visit, 14 725 (72.6%) received an oral antibiotic prescription, and 9913 (48.9%) received an opioid medication prescription. A total of 3484 (17.2%) patients had at least 1 return ED visit for HS or proxy within 30 days, in contrast with 483 (2.4%) who had a dermatology visit (P < .001). Likewise, 6893 (34.0%) patients had a return ED visit for HS or proxy within 180 days, as opposed to 1374 (6.8%) with a dermatology visit (P < .001). Patients with Medicaid and patients who had an opioid prescribed were more likely to return to the ED for treatment of their disease (odds ratio [OR], 1.48; 95% CI, 1.38-1.58; and OR, 1.48; 95% CI, 1.39-1.58, respectively, within 180 days) and, conversely, less likely to have dermatology follow-up (OR, 0.16; 95% CI, 0.14-0.18; and OR, 0.81; 95% CI, 0.71-0.91, respectively, within 180 days).

Conclusions and Relevance

This cohort study suggests that many patients with HS frequent the ED for their disease but are not subsequently seen in the dermatology clinic for ongoing care. The findings in this study raise the opportunity for cross-specialty interventions that could be implemented to better connect patients with HS to longitudinal care.

Socioeconomic predictors of cost and length of stay for erythroderma: a cross-sectional analysis of the national inpatient sample

Erythroderma is an uncommon but serious dermatologic disorder that often requires hospitalization for diagnosis and treatment. However, little is known about predictors influencing cost and patient outcomes. The present study sought to characterize the sociodemographic factors that predict patient outcomes and hospital cost. Data were obtained from the 2016-2018 National Inpatient Sample (NIS) provided by the Healthcare Cost and Utilization Project from the Agency for Healthcare Research and Quality for patients of any age with a primary or secondary diagnosis of exfoliative dermatitis. Regression analyses were performed to find predictors for hospital costs and patient outcomes, represented by the length of stay (LOS). Univariate analysis of LOS revealed urban teaching hospitals were associated with prolonged LOS (p = 0.023). Univariate analysis of hospital cost yielded the following factors associated with increased hospital cost: Black and Asian patients (p = .045), urban teaching hospitals (p = .035), and northeast or south geographic location (p = .004). Multivariable regression analysis revealed prolonged LOS was associated with female sex (p = .043) and large bed capacity (p = .044) while shorter LOS was associated with increased age (p = .025); lower hospital costs were associated with private-owned hospitals -  (p = .025). In patients diagnosed with erythroderma, there appear to be racial, economic, and geographic disparities for patients that lead to greater hospital costs and longer LOS.

Use of convolutional neural networks in skin lesion analysis using real world image and non-image data

Background

Understanding performance of convolutional neural networks (CNNs) for binary (benign vs. malignant) lesion classification based on real world images is important for developing a meaningful clinical decision support (CDS) tool.

Methods

We developed a CNN based on real world smartphone images with histopathological ground truth and tested the utility of structured electronic health record (EHR) data on model performance. Model accuracy was compared against three board-certified dermatologists for clinical validity.

Results

At a classification threshold of 0.5, the sensitivity was 79 vs. 77 vs. 72%, and specificity was 64 vs. 65 vs. 57% for image-alone vs. combined image and clinical data vs. clinical data-alone models, respectively. The PPV was 68 vs. 69 vs. 62%, AUC was 0.79 vs. 0.79 vs. 0.69, and AP was 0.78 vs. 0.79 vs. 0.64 for image-alone vs. combined data vs. clinical data-alone models. Older age, male sex, and number of prior dermatology visits were important positive predictors for malignancy in the clinical data-alone model.

Conclusion

Additional clinical data did not significantly improve CNN image model performance. Model accuracy for predicting malignant lesions was comparable to dermatologists (model: 71.31% vs. 3 dermatologists: 77.87, 69.88, and 71.93%), validating clinical utility. Prospective validation of the model in primary care setting will enhance understanding of the model’s clinical utility.

Association of UV Radiation Exposure, Diagnostic Scrutiny, and Melanoma Incidence in US Counties

Importance

Although UV radiation exposure is the conventionally reported risk factor for cutaneous melanoma, an alternative exposure is diagnostic scrutiny: the more physicians look for and biopsy moles, the more melanoma they find.

Objective

To assess the association of proxies for UV radiation exposure and diagnostic scrutiny with geographical patterns of melanoma incidence.

Design, Setting, and Participants

This was a cross-sectional ecological analysis of the 727 continental US counties reporting to the Surveillance, Epidemiology, and End Results (SEER) Program (among a total of 3108 counties). Environmental data relevant to UV radiation exposure (from a variety of sources), Health Resources and Services Administration data relevant to diagnostic scrutiny, and SEER data on melanoma incidence among the non-Hispanic White population diagnosed with melanoma from 2012 through 2016 were combined. Data analysis was performed between January 2020 and July 2022.

Exposures

Three UV radiation proxies (UV daily dose, cloud variability, and temperature variability) and 3 diagnostic scrutiny proxies (median household income, dermatologists, and primary care physician supply).

Main Outcomes and Measures

Melanoma incidence (in situ and invasive cancers).

Results

In total, 235 333 melanomas were diagnosed. Proxies for UV radiation exposure changed gradually across geography, while melanoma incidence and proxies for diagnostic scrutiny changed abruptly across contiguous counties. The UV daily dose, a variable the National Cancer Institute specifically developed for melanoma analyses, was uncorrelated with incidence (r = 0.03; P = .42). For context, smoking prevalence was highly correlated with lung cancer incidence in the same counties (r = 0.81; P < .001). Melanoma incidence was correlated with median household income (r = 0.43; P < .001). Counties with no dermatologists and shortages of primary care physicians had the lowest incidence, while counties amply supplied with both had the highest, despite having lower mean UV daily dose. There was little association between melanoma incidence and melanoma mortality (r = 0.09; P = .05), while the analogous association in lung cancer was strong (r = 0.96; P < .001).

Conclusions and Relevance

In this cross-sectional ecological study, the current geographical pattern of melanoma incidence across US counties was less associated with proxies for UV radiation exposure and more so with proxies for diagnostic scrutiny. Incidence-the fundamental epidemiologic measure of disease frequency-now had little association with the feared outcome of melanoma: death.

Racial/ethnic health disparities in dermatology in the United States, part 1: Overview of contributing factors and management strategies

Racial or ethnic disparities are prevalent in the field of dermatology. Part 1 of this continuing medical education series aims to elucidate contributors to racial and ethnic disparities within dermatology and highlight potential actionable steps to combat these disparities. We review access to care, workforce diversity, cultural competency, implicit bias, dermatologic education material, patient education, and clinical research. Part 2 of the continuing medical education series will address disease-specific inequities that influence the clinical practice of dermatology.

Orthopaedic Surgeon Distribution in the United States

BACKGROUND

There is limited research on the supply and distribution of orthopaedic surgeons in the United States. The goal of this study was to analyze the association of orthopaedic surgeon distribution in the United States with geographic and sociodemographic factors.

METHODS

County-level data from the US Department of Health and Human Services Area Health Resources Files were used to determine the density of orthopaedic surgeons across the United States on a county level. Data were examined from 2000 to 2019 to analyze trends over time. Bivariate and multivariable negative binomial regression models were constructed to identify county-level sociodemographic factors associated with orthopaedic surgeon density.

RESULTS

In 2019, 51% of the counties in the United States did not have an orthopaedic surgeon. Metropolitan counties had a mean of 22 orthopaedic surgeons per 100,000 persons while nonmetropolitan and rural counties had a mean of 2 and 0.1 orthopaedic surgeons per 100,000 persons, respectively. Over the past 2 decades, there was a significant increase in the percentage of orthopaedic surgeons in metropolitan counties (77% in 2000 vs 93% in 2019, P < 0.001) and in the proportion of orthopaedic surgeons 55 years and older (32% in 2000 vs 39% in 2019, P < 0.001). Orthopaedic surgeon density increased with increasing median home value (P < 0.001) and median household income (P < 0.001). Counties with a higher percentage of persons in poverty (P < 0.001) and higher unemployment rate (P < 0.001) and nonmetropolitan (P < 0.001) and rural (P < 0.001) counties had a lower density of orthopaedic surgeons. On multivariable analysis, a model consisting of median home value (P < 0.001), rural counties (P < 0.001), percentage of noninsured persons (P < 0.001), and percentage of foreign-born persons (P < 0.001) predicted orthopaedic surgeon density.

CONCLUSION

Access to orthopaedic surgeons in the United States in rural areas is decreasing over time. County-level socioeconomic factors such as wealth and urbanization were found to be closely related with surgeon density.

Dermatology Workforce in the United States - Part 1: Overview, Transformations, and Implications

The dermatology workforce continues to evolve to meet the growing and diversified demands of the United States population. Part 1 of this continuing medical education (CME) series is designed to provide an overview of the dermatology workforce as well as delineate the motivators and socio-economic implications of significant workforce transformations which are impacting dermatologic health care. Part 2 of the series will consider the impact of workforce challenges on patient outcomes and discuss potential actions that may help to optimize workforce organization and care delivery.

Skin Cancer Education Interventions for Primary Care Providers: A Scoping Review

Primary care physicians (PCPs) are often the first line of defense against skin cancers. Despite this, many PCPs do not receive a comprehensive training in skin conditions. Educational interventions aimed at skin cancer screening instruction for PCPs offer an opportunity to detect skin cancer at earlier stages and subsequent improved morbidity and mortality. A scoping review was conducted to collect data about previously reported skin cancer screening interventions for PCPs. A structured literature search found 51 studies describing 37 unique educational interventions. Curriculum elements utilized by the interventions were divided into categories that would facilitate comparison including curriculum components, delivery format, delivery timing, and outcome measures. The interventions varied widely in design, including literature-based interventions, live teaching sessions, and online courses with durations ranging from 5 min to 24 months. While several interventions demonstrated improvements in skin cancer knowledge and competency by written exams, only a few revealed positive clinical practice changes by biopsy review or referral analysis. Examining successful interventions could aid in developing a skin cancer detection curriculum for PCPs that can produce positive clinical practice and population-based changes in the management of skin cancer.

Racial and Ethnic Healthcare Disparities in Skin Cancer in the United States: A Review of Existing Inequities, Contributing Factors, and Potential Solutions

Objective

Racial and ethnic health disparities affect the diagnosis and management of melanoma and nonmelanoma skin cancers, leading to deleterious outcomes. Non-Hispanic White patients make up the majority of skin cancers cases, yet racial and ethnic minorities have poorer prognoses and outcomes. The skin cancer literature is fragmented with regards to potential contributors to these healthcare disparities. In this article, we provide a comprehensive review of the skin cancer literature to briefly quantify racial and ethnic inequities, highlight contributing factors, and propose practical changes that can be made.

Methods

A PubMed search was completed to identify articles related to racial and ethnic health care disparities in the context of melanoma, basal cell carcinoma, squamous cell carcinoma, Merkel cell carcinoma, and dermatofibrosarcoma protuberans.

Results

Relative to non-Hispanic White patients, patients of racial and ethnic minorities have differing clinical presentations of skin cancers and genetic risk factors. Insurance, access to specialty care, cultural beliefs, and available educational resources further contribute to racial and ethnic disparities.

Limitations

We are limited to the level of detail provided in the existing literature, and at some times are unable to distinguish race of Hispanic populations. We also acknowledge that there are different nationalities grouped under these broad labels as well as multi-racial populations that may not be accounted for.

Conclusion

Awareness of and familiarization with innate factors and potentially more modifiable contributors can help inform efforts to close the observed gap in racial and ethnic inequities.

Single Model Deep Learning on Imbalanced Small Datasets for Skin Lesion Classification

Deep convolutional neural network (DCNN) models have been widely explored for skin disease diagnosis and some of them have achieved the diagnostic outcomes comparable or even superior to those of dermatologists. However, broad implementation of DCNN in skin disease detection is hindered by small size and data imbalance of the publically accessible skin lesion datasets. This paper proposes a novel single-model based strategy for classification of skin lesions on small and imbalanced datasets. First, various DCNNs are trained on different small and imbalanced datasets to verify that the models with moderate complexity outperform the larger models. Second, regularization DropOut and DropBlock are added to reduce overfitting and a Modified RandAugment augmentation strategy is proposed to deal with the defects of sample underrepresentation in the small dataset. Finally, a novel Multi-Weighted New Loss (MWNL) function and an end-to-end cumulative learning strategy (CLS) are introduced to overcome the challenge of uneven sample size and classification difficulty and to reduce the impact of abnormal samples on training. By combining Modified RandAugment, MWNL and CLS, our single DCNN model method achieved the classification accuracy comparable or superior to those of multiple ensembling models on different dermoscopic image datasets. Our study shows that this method is able to achieve a high classification performance at a low cost of computational resources and inference time, potentially suitable to implement in mobile devices for automated screening of skin lesions and many other malignancies in low resource settings.

Disparities in access for melanoma screening by region, specialty, and insurance: A cross-sectional audit study

Background

Early detection of melanoma is critical for positive outcomes. However, access for the diagnosis of melanoma remains problematic for segments of the general population.

Objective

To compare the rates of dermatology and family medicine practitioner acceptances for a public insurance (Medicaid) versus private insurance (Anthem Blue Cross) and clinic wait times for an appointment for a changing pigmented skin lesion concerning melanoma in rural and urban regions in California.

Methods

Cross-sectional audit study between June 2017 and March 2019; scripted phone calls were made to dermatology and family medicine practices (FMPs).

Results

Family medicine and dermatology practices in both regions had significantly decreased acceptance of Medicaid. Dermatology practices had 11.3% to 13.0% Medicaid acceptance rates that were less than FMP rates of 28% to 36%. In both regions, FMP wait times were 2.4- to 3.2-fold longer for public versus private insurance; there were little differences in wait times for the 2 insurance types in dermatology practices, in both regions.

Limitations

Assessment of only 2 regions in the state of California.

Conclusion

Delays at FMPs and insurance types limit access to melanoma screening in California for underserved segments of the general population, which has implications for melanoma outcomes and health policy.

Dermatology workforce over a decade in Saudi Arabia: demographics, distributions, and future challenges

BACKGROUND

The dermatology workforce is an important topic, as many countries are facing an undersupply of dermatologists, while some are expecting a surplus. Therefore, we conducted this study to identify the current dermatology workforce demographics in Saudi Arabia (SA) and the changes in such demographics over the last 10 years to identify future workforce-related challenges.

METHODS

This study was conducted in SA, and it included all the practicing dermatologists in the country over the last decade (2010-2020). The number of practicing dermatologists, their gender, their nationality, and dermatology residency candidates and graduates were obtained from the Saudi Commission for Health Specialties (SCFHS). The geographic distribution of dermatologists was obtained from the Ministry of Health Statistical Yearbook 2018.

RESULTS

As of September 2020, there were 2678 practicing dermatologists in SA at a ratio of 7.82 dermatologists per 100 000 people. Of the 2678 dermatologists, only 24.8% were Saudis. The Saudi dermatologist ratio has been almost constant over 10 years, ranging from 1.3 to 1.9 per 100 000 people. Of all Saudi dermatologists, 42% were female. The number of residents who graduated from the residency program was not consistent for each year and ranged from 4 to 25. The number of dermatologists varied by region, with 9.2 in Riyadh and 3.4 in Najran per 100 000 people.

CONCLUSIONS

The results of our study revealed that a quarter of dermatologists in SA are Saudis. In addition, the number of non-Saudi dermatologists has increased in the last 10 years, while the number of Saudi dermatologists to the population has remained almost constant. There is also a geographic maldistribution of dermatologists, with urban areas having a higher number of dermatologists than rural areas. We encourage local studies that can elucidate the factors influencing the workforce, such as the dermatologist appointment waiting time, dermatologists' working hours, and the geographic maldistribution of dermatologists in the country.

Usefulness of Smartphones in Dermatology: A US-Based Review

(1) Background: As smartphones have become more widely used, they have become an appealing tool for health-related functions. For dermatology alone, hundreds of applications (apps) are available to download for both patients and providers. (2) Methods: The Google Play Store and Apple App Store were searched from the United States using dermatology-related terms. Apps were categorized based on description, and the number of reviews, download cost, target audience, and use of AI were recorded. The top apps from each category by number of reviews were reported. Additionally, literature on the benefits and limitations of using smartphones for dermatology were reviewed. (3) Results: A total of 632 apps were included in the study: 395 (62.5%) were marketed towards patients, 203 (32.1%) towards providers, and 34 (5.4%) towards both; 265 (41.9%) were available only on the Google Play Store, 146 (23.1%) only on the Apple App Store, and 221 (35.0%) were available on both; and 595 (94.1%) were free to download and 37 (5.9%) had a cost to download, ranging from USD 0.99 to USD 349.99 (median USD 37.49). A total of 99 apps (15.7%) reported the use of artificial intelligence. (4) Conclusions: Although there are many benefits of using smartphones for dermatology, lack of regulation and high-quality evidence supporting the efficacy and accuracy of apps hinders their potential.