Changes in skin cancer screening rates in the United States from 2005-2015

To the Editor: Despite the rising incidence of melanoma in the United States,1Adamson A.S. Welch H. Welch H.G. Association of UV radiation exposure, diagnostic scrutiny, and melanoma incidence in US counties.JAMA Intern Med. 2022; 182: 1181-1189https://doi.org/10.1001/jamainternmed.2022.4342Crossref PubMed Scopus (3) Google Scholar the current US Preventive Services Task Force guidelines state there is insufficient evidence to recommend population-based skin cancer screening by total body skin examination (TBSE). This results in screening being driven by patient request or physician suggestion and potentially a discordance between who is screened and who may benefit most from screening. Increased diagnostic scrutiny, including skin cancer screening, has been implicated as a potential cause of the increase in melanoma incidence.1Adamson A.S. Welch H. Welch H.G. Association of UV radiation exposure, diagnostic scrutiny, and melanoma incidence in US counties.JAMA Intern Med. 2022; 182: 1181-1189https://doi.org/10.1001/jamainternmed.2022.4342Crossref PubMed Scopus (3) Google Scholar Measuring screening rates using claims data is challenging as skin cancer screening is not a billable service. To better estimate trends in population-based skin cancer screening (vs surveillance in high-risk patients) we evaluated the rates of self-reported TBSEs within the previous 5 years among participants 18 years or older without a personal history of skin cancer (melanoma or non-melanoma skin cancer) or a first-degree relative with a history of melanoma in 2005 (28,233 participants), 2010 (24,113 participants), and 2015 (29,902 participants) (Table I) utilizing the National Health Interview Survey data. Subpopulation logistic regression models were performed utilizing STATA/MP 17.0 to explore changes in TBSE rates in 2005 versus 2015.Table ICharacteristics of individuals (aged ≥18 years) in National Health Interview Survey 2005, 2010, and 2015 that have received a total body skin examination in the last 5 yearsCharacteristic200520102015TBSE†TBSE status was assessed by the question “Have you ever had all of your skin from head to toe checked for cancer either by a dermatologist or some other kind of doctor within the last 5 years?”–yes weighted %∗Weighted percentage was calculated using NHIS survey design parameters. (95% CI)TBSE†TBSE status was assessed by the question “Have you ever had all of your skin from head to toe checked for cancer either by a dermatologist or some other kind of doctor within the last 5 years?”–yes weighted %∗Weighted percentage was calculated using NHIS survey design parameters. (95% CI)TBSE†TBSE status was assessed by the question “Have you ever had all of your skin from head to toe checked for cancer either by a dermatologist or some other kind of doctor within the last 5 years?”–yes weighted %∗Weighted percentage was calculated using NHIS survey design parameters. (95% CI)Sample size (n/N, weighted %)1,304/28,233 (4.8)1,171/24,113 (5.3)1,678/29,902 (6.1)Sex Male45.2 (42.2-48.2)46.0 (42.6-49.5)42.6 (39.6-45.7) Female54.8 (51.8-57.8)54.0 (50.5-57.5)57.4 (54.3-60.4)Age 18-3011.6 (9.6-13.9)13.3 (11.0-15.9)12.2 (9.9-14.9) 31-4015.0 (12.8-17.5)12.9 (10.9-15.2)14.8 (12.7-17.1) 41-5022.5 (19.9-25.3)19.7 (16.9-22.9)14.4 (12.3-16.8) 51-6427.2 (24.6-30.0)30.2 (27.4-33.2)32.3 (29.3-35.4) ≥6523.8 (21.3-26.4)23.9 (21.5-26.6)26.4 (23.8-29.0)Annual household income‡Low income was defined as <$75,000 in 2005 and <$90,000 in 2010 and 2015. Low59.3 (56.0-62.5)62.8 (59.1-66.4)54.1 (50.6-57.5) High40.7 (37.5-44.0)37.2 (33.6-40.9)45.9 (42.5-49.4)Race/ethnicity Non-Hispanic White85.7 (83.4-87.7)83.6 (81.0-85.8)85.9 (83.9-87.7) Non-Hispanic Black6.6 (5.1-8.4)6.4 (5.0-8.1)4.5 (3.6-5.7) Hispanic5.8 (4.5-7.4)6.3 (5.1-7.9)6.1 (5.0-7.4) Non-Hispanic all other races1.9 (1.2-3.1)3.8 (2.7-5.2)3.5 (2.6-4.8)Region Northeast24.1 (21.3-27.0)24.9 (21.8-28.3)24.2 (21.4-27.3) Midwest21.3 (18.5-24.5)20.5 (17.6-23.8)19.4 (17.0-22.1) South35.8 (32.8-38.8)33.1 (29.8-36.6)33.7 (30.6-37.0) West18.9 (16.2-21.8)21.5 (18.9-24.3)22.7 (20.0-25.5)Education§Education status was assessed by the question “Education of adult with highest education in family?”,Eight subjects missing in 2005, 4 subjects missing in 2010, and 1 subject missing in 2015. No/some HS education5.7 (4.6-6.9)4.3 (3.3-5.5)3.0 (2.2-4.0) HS graduate/GED recipient18.2 (15.3-21.4)15.7 (13.6-18.1)12.3 (10.6-14.3) Some college/AA degree27.5 (24.9-30.3)30.0 (26.8-33.3)30.0 (27.0-33.2) College graduate/postgraduate48.7 (45.2-52.2)50.1 (46.7-53.5)54.7 (51.3-58.1)Sunscreen use in sun‖Sunscreen usage was assessed by the question “When you go outside on a warm sunny day for more than 1 hour, how often do you use sunscreen?”,Eighteen subjects missing in 2005 and 5 subjects missing in 2015. Always/most of the time41.3 (38.2-44.4)42.5 (39.1-46.0)48.5 (44.8-52.1) Sometimes/rarely26.6 (23.8-29.5)28.4 (25.5-31.4)26.3 (23.5-29.4) Never25.6 (23.0-28.4)23.5 (20.7-26.6)20.8 (18.4-23.4) No sun exposure6.6 (5.3-8.1)5.6 (4.4-7.1)4.4 (3.3-5.9)Effect of sun on skin¶Effect of sun on skin was assessed by the question “After several months of not being in the sun very much, if you went out in the sun for an hour without sunscreen, a hat, or protective clothing, which one of these best describes what would happen to your skin?”,Thirty-nine subjects missing in 2005, 9 subjects missing in 2010, and 15 subjects missing in 2015. Severe/moderate#Severe/moderate was defined as a response of “get a severe sunburn with blisters” or “have a moderate sunburn with peeling.”34.1 (31.0-37.3)32.3 (29.2-35.7)37.7 (34.3-41.2) MildMild was defined as a response of “burn mildly with some or no darkening/tanning.”28.2 (25.2-31.4)29.5 (26.5-32.7)26.9 (24.2-29.7) Turn darker without sunburn19.2 (16.8-21.8)22.1 (19.6-24.8)20.1 (17.5-22.9) No reactionNo reaction was defined as a response of “nothing would happen to my skin.”9.8 (8.1-11.9)9.5 (7.7-11.6)9.4 (7.7-11.5) No sun exposure8.7 (7.2-10.4)6.6 (5.4-8.1)6.0 (4.5-7.9)AA, Associate degree; GED, General Educational Development; HS, high school; TBSE, total body skin exam.∗ Weighted percentage was calculated using NHIS survey design parameters.† TBSE status was assessed by the question “Have you ever had all of your skin from head to toe checked for cancer either by a dermatologist or some other kind of doctor within the last 5 years?”‡ Low income was defined as <$75,000 in 2005 and <$90,000 in 2010 and 2015.§ Education status was assessed by the question “Education of adult with highest education in family?”‖ Sunscreen usage was assessed by the question “When you go outside on a warm sunny day for more than 1 hour, how often do you use sunscreen?”¶ Effect of sun on skin was assessed by the question “After several months of not being in the sun very much, if you went out in the sun for an hour without sunscreen, a hat, or protective clothing, which one of these best describes what would happen to your skin?”# Severe/moderate was defined as a response of “get a severe sunburn with blisters” or “have a moderate sunburn with peeling.”∗∗ Mild was defined as a response of “burn mildly with some or no darkening/tanning.”†† No reaction was defined as a response of “nothing would happen to my skin.”‡‡ Eight subjects missing in 2005, 4 subjects missing in 2010, and 1 subject missing in 2015.§§ Eighteen subjects missing in 2005 and 5 subjects missing in 2015.‖‖ Thirty-nine subjects missing in 2005, 9 subjects missing in 2010, and 15 subjects missing in 2015. Open table in a new tab AA, Associate degree; GED, General Educational Development; HS, high school; TBSE, total body skin exam. The weighted prevalence of participants reporting having a TBSE within the last 5 years increased from 4.8% in 2005 to 6.1% in 2015 (adjusted odds ratio, 1.09; 95% CI, 1.04-1.14; P < .001). Non-Hispanic Whites were the only race/ethnicity with a significant increase in TBSE (adjusted odds ratio, 1.15; 95% CI, 1.09-1.21; P < .001). Additionally, significant increases in TBSE were associated with higher educational attainment, female sex, and higher income. TBSE rates increased significantly among those aged 31 to 40 and 51 to 64 years (Fig 1). While we found skin cancer screening rates increased from 2005 to 2015, we did not see a significant increase among those at highest risk of melanoma mortality, including males and individuals ≥65 years.2Lakhani N.A. Saraiya M. Thompson T.D. King S.C. Guy Jr., G.P. Total body skin examination for skin cancer screening among U.S. adults from 2000 to 2010.Prev Med. 2014; 61: 75-80https://doi.org/10.1016/j.ypmed.2014.01.003Crossref PubMed Scopus (52) Google Scholar Although lower income and educational attainment are associated with a higher risk of mortality among individuals with melanoma,3Sitenga J.L. Aird G. Ahmed A. Walters R. Silberstein P.T. Socioeconomic status and survival for patients with melanoma in the United States: an NCDB analysis.Int J Dermatol. 2018; 57: 1149-1156https://doi.org/10.1111/ijd.14026Crossref PubMed Scopus (31) Google Scholar screening increased significantly only among people with higher income and at least some college education. Lakhani et al2Lakhani N.A. Saraiya M. Thompson T.D. King S.C. Guy Jr., G.P. Total body skin examination for skin cancer screening among U.S. adults from 2000 to 2010.Prev Med. 2014; 61: 75-80https://doi.org/10.1016/j.ypmed.2014.01.003Crossref PubMed Scopus (52) Google Scholar previously demonstrated a rise in ever having a TBSE among US adults from 2000 to 2010. Our findings add to theirs by providing updated data and better measuring population-based screening by limiting our analysis to lower-risk patients and to TBSE within the past 5 years. Limitations to our study include self-reporting of TBSE, lack of the 2020 National Health Interview Survey data due to the COVID-19 pandemic, and the inability to link survey results with the medical record to correlate screening with clinical outcomes. Our findings suggest that despite an absence of guidelines, skin cancer screening is increasing, but not among those most likely to benefit from it. This will dilute potential benefits of screening and impact the outcomes of data that drive US Preventive Services Task Force recommendations. Guidelines could help by more effectively targeting screening to those most likely to die from melanoma, lead to better utilization of the limited resources of the dermatology work force,4Hartman R.I. Xue Y. Singer S. Markossian T.W. Joyce C. Mostaghimi A. Modelling the value of risk-stratified skin cancer screening of asymptomatic patients by dermatologists.Br J Dermatol. 2020; 183: 509-515Crossref PubMed Scopus (9) Google Scholar and assist with reducing melanoma disparities.. Dr Ferris is a consultant for DermTech and an investigator for Castle Biosciences, SkinAnalytics, and DermTech. Author Smith, Author Smith, and Dr Demanelis have no conflicts of interest to declare.