Photodynamic therapy (PDT) is used in dermatology as “field therapy” for diffuse premalignant actinic keratoses and as treatment for superficial nonmelanoma skin cancers (NMSCs).1 Photodynamic therapy may be a useful strategy in response to the increasing incidence of NMSCs, especially for the Medicare population, in which treatment of NMSCs is a major expenditure.2,3 There are, however, challenges associated with the use of PDT, including the need for incubation time with the photosensitizer. This challenge may be most relevant in rural areas where patients may live far from the nearest dermatologist. Work studying other specialized dermatologic treatments suggests that patients in rural areas have limited access to certain treatments.4,5 Nevertheless, the temporal and geographical trends in the use and cost of PDT have not been characterized, to our knowledge. This cross-sectional study evaluates the geographical variation in the availability of PDT for the Medicare population in the United States and characterizes the temporal trends of PDT cost and use from 2012 to 2017.
The data originated from detailed data files from the Medicare Provider Utilization and Payment Data: Physician and Other Supplier Public Use Files for calendar years 2012 to 2017. These files include claims submitted to the Medicare Part B program. Claims were included if they were submitted by a dermatologist for PDT services. This study was deemed not human participants research by the Yale University School of Medicine Institutional Review Board.
Counties were assigned a Rural-Urban Continuum Code as defined by the Urban-Rural Classification Scheme for Counties.6 Claims and counties from US territories were excluded from this analysis. Data analysis was performed from November 11, 2019, to April 28, 2020, using STATA, version 13 (StataCorp LP).
The year-over-year trends of the number of dermatologists providing PDT and the total number of dermatologists for both metropolitan and nonmetropolitan counties are outlined in Table 1. The increased number of dermatologists offering PDT has led to a rise in the density of such physicians from 0.540 per 100 000 individuals in 2012 to 0.725 per 100 000 individuals in 2017 within metropolitan regions and from 0.270 per 100 000 individuals in 2012 to 0.344 per 100 000 individuals in 2017 within nonmetropolitan regions. Although PDT services were offered in 41.6% of metropolitan counties (483 of 1162) in 2017, they were available in only 5.4% of nonmetropolitan counties (105 of 1957) in 2017.
Between 2012 and 2017, overall Medicare reimbursements for PDT services in metropolitan regions increased by $1 601 015 (average annual growth rate [AAGR], 4.0%), with 19 219 (AAGR, 4.7%) additional claims; overall Medicare reimbursements for PDT services in nonmetropolitan regions increased by $122 076 (AAGR, 3.2%), with 1625 (AAGR, 3.9%) additional claims (Table 2). As of 2017, metropolitan counties accounted for 90.9% of Medicare reimbursements for PDT claims ($9 260 322 of $10 184 822).
In this study, we present data characterizing the trends in use, accessibility, and cost of PDT among the US Medicare population. Despite the consistent growth in the number of dermatologists offering PDT, the density of dermatologists providing PDT in nonmetropolitan counties is less than half of that in metropolitan counties in the most recent study year; this difference has widened from 2012 to 2017.
With the incidence of NMSCs rising, PDT could be important for prevention and treatment of some NMSCs while maintaining low costs.2 Strategies leading to more dermatologists offering PDT, such as recent increases to reimbursements, may, therefore, ultimately lower Medicare expenditures on skin cancer treatment, especially in nonmetropolitan areas, where our study suggests that PDT access is particularly lacking.
This study has several limitations. Our use of Medicare claims data precludes evaluation of PDT use for patients with commercial insurance. Likewise, while PDT is generally performed in outpatient settings, exclusion of institutional providers (such as hospitals) in the data set may limit our ability to capture all PDT services.
We found that the use of PDT increased in both metropolitan and nonmetropolitan regions from 2012 to 2017 but that there remains notably less access to PDT in nonmetropolitan regions in comparison with metropolitan regions in the United States.
Accepted for Publication: May 14, 2020.
Corresponding Author: Jeffrey M. Cohen, MD, Department of Dermatology, Yale University School of Medicine, 15 York St, New Haven, CT 06510 (jeffrey.m.cohen@yale.edu).
Published Online: July 22, 2020. doi:10.1001/jamadermatol.2020.2427
Author Contributions: Mr Cheraghlou and Dr Cohen had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: All authors.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Cheraghlou, Cohen.
Critical revision of the manuscript for important intellectual content: Feng, Cohen.
Statistical analysis: Cheraghlou.
Administrative, technical, or material support: Feng, Cohen.
Supervision: Feng, Cohen.
Conflict of Interest Disclosures: None reported.
6.Ingram
DD, Franco
SJ. 2013 NCHS Urban-Rural Classification Scheme for Counties.
Vital Health Stat 2. 2014;(166):1-73.
PubMedGoogle Scholar