Ecological Analysis of Broadband Access and Dermatologist Density in the United States

Technical Summary: Ecological Analysis of Broadband Access and Dermatologist Density in the United States

Problem Statement
The United States faces a growing shortage of dermatologists, a disparity that is particularly acute in rural regions. While teledermatology is increasingly utilized to bridge this gap by facilitating real-time video visits and store-and-forward transmission of high-resolution images, its efficacy is contingent upon high-speed broadband infrastructure. Although broadband access serves as a critical proxy for teledermatology availability, there is a lack of published data characterizing the geographic distribution of broadband access relative to the local availability of dermatologic care. This study addresses the need to investigate the association between household broadband access and dermatologist density across US counties, while simultaneously examining how broadband access varies by county-level demographics.

Methodology
This study employed an ecological, cross-sectional design utilizing 2019 data from four federal sources: the American Community Survey (ACS), the Area Health Resources File (AHRF), the Small Area Income and Poverty Estimates Program, and the National Center for Health Statistics (NCHS). The analysis included data from 3,220 of 3,230 US counties (99.7%).

• Variables: The primary exposure was household broadband access (cable, fiber optic, or DSL), treated as a continuous variable representing the proportion of households with access. The primary outcome was dermatologist density, defined as the number of practicing dermatologists per 100,000 persons.
• Covariates: The analysis adjusted for numerous sociodemographic factors, including urbanicity (categorized via the NCHS 6-level classification scheme), median household income, poverty levels, unemployment rates, and population proportions regarding race (Black, Hispanic), age (≥65), education (<high school), gender, and insurance status.
• Statistical Analysis: Researchers utilized simple linear regression to assess the bivariate relationship between broadband access and dermatologist density. Multiple linear regression models were constructed to adjust for potential confounding variables. Additionally, a second multiple linear regression model investigated broadband access as the outcome predicted by county demographics. Data visualization and heatmaps were generated using RStudio® and 2019 US Census Bureau TIGER/LINE Shapefiles.

Key Results
The study found a significant positive association between broadband access and dermatologist density.

• Descriptive Statistics: The median broadband access across all counties was 55.0%. A stark disparity was observed between urban and rural settings: in the most urban areas (NCHS code 1), 72.2% of households had broadband access and the dermatologist density was 6.5 per 100,000 people. Conversely, in the most rural areas (NCHS code 6), broadband access dropped to 49.1%, with a dermatologist density of only 0.21 per 100,000 people.
• Regression Analysis: In simple linear regression, broadband access and dermatologist density were significantly associated ($\beta = 0.75$; 95% CI: 0.66 to 0.84). This association remained significant after adjusting for sociodemographic covariates in the multiple linear regression model ($\beta = 0.48$; 95% CI: 0.35 to 0.62).
• Demographic Correlations: Lower broadband access was significantly associated with higher rurality, lower educational attainment, higher unemployment, lower median household income, a higher proportion of residents aged ≥65, a higher proportion of Black residents, and a higher proportion of uninsured individuals. Conversely, greater broadband access was associated with higher median income, higher percentages of female and Hispanic residents.

Significance and Claims
The authors conclude that the potential of teledermatology to improve access to dermatologic care is inherently limited by the requirement for broadband internet. The study identifies a significant inequity: the populations with the greatest need for dermatologic care—specifically those in rural communities and those characterized by Black, uninsured, unemployed, older, lower-income, and less-educated demographics—are the same populations with the poorest broadband access.

The paper posits that these findings highlight structural barriers to teledermatology adoption. The authors suggest that expanding rural broadband coverage could increase access to care in underserved communities. Furthermore, for areas lacking traditional broadband infrastructure, they propose that satellite-based internet connectivity may serve as a viable alternative. The study explicitly acknowledges its limitations, noting that as a cross-sectional ecological analysis, it cannot determine temporal sequence or causation, and its results are specific to the 2019 data period and dermatology.