Racial differences in sun-protective behaviors: a retrospective, cross-sectional analysis of the National Health and Nutrition Examination Survey (NHANES) 2003-2018.

To the Editor: Sun-protective behaviors (SPBs), such as wearing long sleeve shirts, seeking shade, and using sunscreen, are protective against UV-dependent skin conditions.1Centers for Disease Control and PreventionSun Safety.https://www.cdc.gov/cancer/skin/basic_info/sun-safety.htmDate accessed: November 29, 2020Google Scholar People with skin of color face increased morbidity and mortality from skin conditions due in part to different SPBs, secondary to reduced risk perception, inconvenience, and cost.2Diehl K. Schneider S. Seuffert S. Greinert R. Görig T. Who are the nonusers of sunscreen, and what are their reasons? Development of a new item set.J Cancer Educ. 2020; https://doi.org/10.1007/s13187-020-01732-2Crossref Scopus (3) Google Scholar,3Buster K.J. Stevens E.I. Elmets C.A. Dermatologic health disparities.Dermatol Clin. 2012; 30 (viii): 53-59Abstract Full Text Full Text PDF PubMed Scopus (112) Google Scholar Few studies characterize these differences and fewer track their evolution over time. In this retrospective cross-sectional analysis, we use National Health and Nutrition Examination Survey (NHANES) data from 2003-2006, 2009-2012, and 2015-2018 to track changes in SPBs in 3 CDC-defined racial groups over a 15-year period.Three SPB questions were assessed in the NHANES dermatology questionnaire: "When you go outside on a very sunny day, for more than one hour, how often you: (1) Stay in the shade? (2) Wear a long-sleeved shirt? (3) Use sunscreen?" All NHANES data are publicly available at https://www.cdc.gov/nchs/nhanes/. The relationship between race (Hispanic, non-Hispanic White, and non-Hispanic Black) and frequent engagement in SPBs (always, most of the time, or sometimes) was adjusted by age, sex, annual household income, education level completed, marriage status, sun sensitivity, and sunburns in the past year. We used a multinomial logistic regression adjusted with a Bonferroni correction and weighted according to NHANES analytical guidelines; adjusted odds ratios were reported. The complete, restricted data set included 16,859 individuals.SPBs increased across all racial groups, although differences in behavior persist. Congruent with past findings, we observed that across all years, Hispanics and non-Hispanic Blacks were more likely than non-Hispanic Whites to stay in the shade (OR: 1.792, 95% CI: 1.790, 1.794; OR: 2.432, 95% CI: 2.428, 2.435; respectively) and wear a long sleeve shirt (OR: 2.199, 95% CI: 2.196, 2.201; OR: 1.666, 95% CI: 1.663, 1.668; respectively) but less likely to use sunscreen (OR: 0.946, 95% CI: 0.945, 0.947; OR: 0.228, 95% CI: 0.228, 0.229; respectively) (Table I).4Summers P. Bena J. Arrigain S. Alexis A.F. Cooper K. Bordeaux J.S. Sunscreen use: non-Hispanic Blacks compared with other racial and/or ethnic groups.Arch Dermatol. 2011; 147: 863-864Crossref PubMed Scopus (25) Google Scholar Interestingly, Hispanics reported using sunscreen less frequently than non-Hispanic Whites in 2003-2006 (31.1% vs 54.7%, respectively) but equally as frequently in 2015-2018 (44.6% vs 54.5%). After conducting a subgroup analysis, we found that this effect was not driven by Hispanics whose skin burned or peeled after sun exposure (OR: 0.893, 95% CI: 0.890, 0.896) but was rather driven by Hispanics whose skin darkened but did not sunburn (OR: 1.340 95% CI: 1.337, 1.343) when compared with non-Hispanic Whites with the corresponding skin reaction (Table II). These data suggested that sunscreen use among Hispanics may be motivated by factors other than sunburn and skin cancer risk and is supported by another focus group research.5Buchanan Lunsford N. Berktold J. Holman D.M. Stein K. Prempeh A. Yerkes A. Skin cancer knowledge, awareness, beliefs and preventive behaviors among black and hispanic men and women.Prev Med Rep. 2018; 12: 203-209Crossref PubMed Scopus (31) Google Scholar Non-Hispanic Blacks remained significantly less likely to use sunscreen in 2015-2018, regardless of sun reaction.Table IWeighted odds of frequent or moderate engagement in sun-protective behaviorBehaviorRace∗White race as the reference category for corresponding years. All findings had P < .0001.Odds ratios (95% CI)2003-2006 (n = 5724)†All effects were adjusted for age, sex, annual household income, marital status, sun reactivity, and sunburns in the last year.2009-2012 (n = 6112)†All effects were adjusted for age, sex, annual household income, marital status, sun reactivity, and sunburns in the last year.2015-2018 (n = 5023)†All effects were adjusted for age, sex, annual household income, marital status, sun reactivity, and sunburns in the last year.All years (n = 16,859)†All effects were adjusted for age, sex, annual household income, marital status, sun reactivity, and sunburns in the last year.Stay in shadeHispanic1.774 (1.772, 1.776)1.480 (1.479, 1.482)1.937 (1.935, 1.939)1.792 (1.790, 1.794)Non-Hispanic Black2.528 (2.525, 2.531)2.439 (2.436, 2.442)2.163 (2.160, 2.166)2.432 (2.428, 2.435)Wear long-sleeve shirtHispanic2.314 (2.311, 2.317)1.680 (1.679, 1.682)2.334 (2.331, 2.336)2.199 (2.196, 2.201)Non-Hispanic Black2.075 (2.072, 2.078)1.185 (1.183, 1.186)1.744 (1.742, 1.746)1.666 (1.663, 1.668)Sunscreen useHispanic0.857 (0.856, 0.859)0.900 (0.899, 0.901)1.031 (1.029, 1.032)0.946 (0.945, 0.947)Non-Hispanic Black0.212 (0.212, 0.212)0.217 (0.217, 0.218)0.251 (0.250, 0.251)0.228 (0.228, 0.229)∗ White race as the reference category for corresponding years. All findings had P < .0001.† All effects were adjusted for age, sex, annual household income, marital status, sun reactivity, and sunburns in the last year. Open table in a new tab Table IIOdds of behavior engagement by skin reactivity and sunburn in the last yearBehaviorRace∗White race as the reference category for corresponding skin reactions. All findings had P < .0001.Skin reactivity, odds ratios (95% CI)Sunburn in last year†All effects were adjusted for age, sex, annual household income, and marital status.Severe (n = 1810)†All effects were adjusted for age, sex, annual household income, and marital status.Mild burn with some tanning (n = 4254)†All effects were adjusted for age, sex, annual household income, and marital status.Darkened, no sunburn (n = 3751)†All effects were adjusted for age, sex, annual household income, and marital status.Not reactive (n = 7044)†All effects were adjusted for age, sex, annual household income, and marital status.Stay in shadeHispanic2.627 (2.612, 2.642)1.688 (1.683, 1.692)1.898 (1.893, 1.902)1.777 (1.774, 1.780)2.247 (2.243, 2.251)Non-Hispanic Black3.396 (3.348, 3.445)2.783 (2.764, 2.802)2.436 (2.430, 2.443)2.481 (2.476, 2.485)2.189 (2.182, 2.195)Wear long-sleeve shirtHispanic1.824 (1.817, 1.830)2.209 (2.204, 2.214)2.575 (2.569, 2.581)2.073 (2.069, 2.077)2.205 (2.201, 2.210)Non-Hispanic Black1.812 (1.799, 1.825)2.191 (2.181, 2.201)2.046 (2.041, 2.051)1.539 (1.537, 1.542)3.025 (3.014, 3.037)Sunscreen useHispanic0.893 (0.890, 0.896)0.903 (0.901, 0.905)1.340 (1.337, 1.343)0.838 (0.836, 0.839)0.972 (0.971, 0.974)Non-Hispanic Black0.285 (0.283, 0.287)0.321 (0.319, 0.322)0.368 (0.367, 0.369)0.191 (0.191, 0.192)0.363 (0.362, 0.365)∗ White race as the reference category for corresponding skin reactions. All findings had P < .0001.† All effects were adjusted for age, sex, annual household income, and marital status. Open table in a new tab Our analysis demonstrates that SPBs have changed from 2003-2018. Understanding the factors that motivate SPBs may facilitate tailored efforts for skin self-examinations and may increase SPB compliance in all populations. Widespread engagement in SPBs can help narrow differences in dermatologic outcomes in patients with skin of color. To the Editor: Sun-protective behaviors (SPBs), such as wearing long sleeve shirts, seeking shade, and using sunscreen, are protective against UV-dependent skin conditions.1Centers for Disease Control and PreventionSun Safety.https://www.cdc.gov/cancer/skin/basic_info/sun-safety.htmDate accessed: November 29, 2020Google Scholar People with skin of color face increased morbidity and mortality from skin conditions due in part to different SPBs, secondary to reduced risk perception, inconvenience, and cost.2Diehl K. Schneider S. Seuffert S. Greinert R. Görig T. Who are the nonusers of sunscreen, and what are their reasons? Development of a new item set.J Cancer Educ. 2020; https://doi.org/10.1007/s13187-020-01732-2Crossref Scopus (3) Google Scholar,3Buster K.J. Stevens E.I. Elmets C.A. Dermatologic health disparities.Dermatol Clin. 2012; 30 (viii): 53-59Abstract Full Text Full Text PDF PubMed Scopus (112) Google Scholar Few studies characterize these differences and fewer track their evolution over time. In this retrospective cross-sectional analysis, we use National Health and Nutrition Examination Survey (NHANES) data from 2003-2006, 2009-2012, and 2015-2018 to track changes in SPBs in 3 CDC-defined racial groups over a 15-year period. Three SPB questions were assessed in the NHANES dermatology questionnaire: "When you go outside on a very sunny day, for more than one hour, how often you: (1) Stay in the shade? (2) Wear a long-sleeved shirt? (3) Use sunscreen?" All NHANES data are publicly available at https://www.cdc.gov/nchs/nhanes/. The relationship between race (Hispanic, non-Hispanic White, and non-Hispanic Black) and frequent engagement in SPBs (always, most of the time, or sometimes) was adjusted by age, sex, annual household income, education level completed, marriage status, sun sensitivity, and sunburns in the past year. We used a multinomial logistic regression adjusted with a Bonferroni correction and weighted according to NHANES analytical guidelines; adjusted odds ratios were reported. The complete, restricted data set included 16,859 individuals. SPBs increased across all racial groups, although differences in behavior persist. Congruent with past findings, we observed that across all years, Hispanics and non-Hispanic Blacks were more likely than non-Hispanic Whites to stay in the shade (OR: 1.792, 95% CI: 1.790, 1.794; OR: 2.432, 95% CI: 2.428, 2.435; respectively) and wear a long sleeve shirt (OR: 2.199, 95% CI: 2.196, 2.201; OR: 1.666, 95% CI: 1.663, 1.668; respectively) but less likely to use sunscreen (OR: 0.946, 95% CI: 0.945, 0.947; OR: 0.228, 95% CI: 0.228, 0.229; respectively) (Table I).4Summers P. Bena J. Arrigain S. Alexis A.F. Cooper K. Bordeaux J.S. Sunscreen use: non-Hispanic Blacks compared with other racial and/or ethnic groups.Arch Dermatol. 2011; 147: 863-864Crossref PubMed Scopus (25) Google Scholar Interestingly, Hispanics reported using sunscreen less frequently than non-Hispanic Whites in 2003-2006 (31.1% vs 54.7%, respectively) but equally as frequently in 2015-2018 (44.6% vs 54.5%). After conducting a subgroup analysis, we found that this effect was not driven by Hispanics whose skin burned or peeled after sun exposure (OR: 0.893, 95% CI: 0.890, 0.896) but was rather driven by Hispanics whose skin darkened but did not sunburn (OR: 1.340 95% CI: 1.337, 1.343) when compared with non-Hispanic Whites with the corresponding skin reaction (Table II). These data suggested that sunscreen use among Hispanics may be motivated by factors other than sunburn and skin cancer risk and is supported by another focus group research.5Buchanan Lunsford N. Berktold J. Holman D.M. Stein K. Prempeh A. Yerkes A. Skin cancer knowledge, awareness, beliefs and preventive behaviors among black and hispanic men and women.Prev Med Rep. 2018; 12: 203-209Crossref PubMed Scopus (31) Google Scholar Non-Hispanic Blacks remained significantly less likely to use sunscreen in 2015-2018, regardless of sun reaction. Our analysis demonstrates that SPBs have changed from 2003-2018. Understanding the factors that motivate SPBs may facilitate tailored efforts for skin self-examinations and may increase SPB compliance in all populations. Widespread engagement in SPBs can help narrow differences in dermatologic outcomes in patients with skin of color. None disclosed.