eTable 1. Unweighted Number of Participants by Individual Sexual Orientation, Adult Respondents, Stratified by Sex and Survey Sample (2001-2005 CHIS, 2009 CHIS & 2013 NHIS)
eTable 2. Multivariate Adjusted Odds Ratio (aOR) for Prevalence of Skin Cancer by Sexual Orientation, Adult Respondents, Stratified by Sex and Survey Sample (2001-2005 CHIS & 2013 NHIS)
eTable 3. Multivariate Adjusted Odds Ratio (aOR) for Prevalence of 12-Month History of Indoor Tanning by Sexual Orientation, Adult Respondents, Stratified by Sex and Survey Sample (2009 CHIS & 2013 NHIS)
Mansh M, Katz KA, Linos E, Chren M, Arron S. Association of Skin Cancer and Indoor Tanning in Sexual Minority Men and Women. JAMA Dermatol. 2015;151(12):1308-1316. doi:10.1001/jamadermatol.2015.3126
Skin cancer, the most common cancer in the United States, is highly associated with outdoor and indoor tanning behaviors. Although indoor tanning has been suggested to be more common among sexual minority (self-reported as homosexual, gay, or bisexual) men compared with heterosexual men, whether rates of skin cancer vary by sexual orientation is unknown.
To investigate whether skin cancer prevalence and indoor tanning behaviors vary by sexual orientation in the general population.
Design, Setting, and Participants
We performed a cross-sectional study using data from the 2001, 2003, 2005, and 2009 California Health Interview Surveys (CHISs) and the 2013 National Health Interview Survey (NHIS) of population-based samples of the California and US noninstitutionalized civilian population. Participants included 192 575 men and women 18 years or older who identified as heterosexual or a sexual minority.
Main Outcomes and Measures
Self-reported lifetime history of skin cancer and 12-month history of indoor tanning.
The study included 78 487 heterosexual men, 3083 sexual minority men, 107 976 heterosexual women, and 3029 sexual minority women. Sexual minority men were more likely than heterosexual men to report having skin cancer (2001-2005 CHISs: adjusted odds ratio [aOR], 1.56; 95% CI, 1.18-2.06, P < .001; 2013 NHIS: aOR, 2.13; 95% CI, 1.14-3.96, P = .02) and having tanned indoors (2009 CHIS: aOR, 5.80; 95% CI, 2.90-11.60, P < .001; 2013 NHIS: aOR, 3.16; 95% CI, 1.77-5.64, P < .001). Sexual minority women were less likely than heterosexual women to report having had nonmelanoma skin cancer (2001-2005 CHIS: aOR, 0.56; 95% CI, 0.37-0.86, P = .008) and having tanned indoors (2009 CHIS: aOR, 0.43; 95% CI, 0.20-0.92, P = .03; 2013 NHIS: aOR, 0.46; 95% CI, 0.26-0.81, P = .007).
Conclusions and Relevance
Sexual minority men indoor tan more frequently and report higher rates of skin cancer than heterosexual men. Primary and secondary prevention efforts targeted at sexual minority men might reduce risk factors for, and consequences of, skin cancer.
Skin cancer is the most common cancer in the United States, with approximately 5 million Americans treated annually at a cost of $8.1 billion.1 Incidence of nonmelanoma skin cancer (NMSC) and melanoma in the United States has increased rapidly at 4.2%2 and 1.9%3 annually, respectively. Skin cancers are more common in persons with fair skin or a family history of skin cancer but are also strongly associated with preventable risk behaviors related to UV radiation exposure,4,5 including outdoor sun exposure6 and indoor tanning.7,8 The World Health Organization considers indoor tanning devices to be “carcinogenic to humans.”9(p751) The US Food and Drug Administration mandates that all indoor tanning device manufacturers include a warning that people younger than 18 years not use these devices.10 In 2013, the US Surgeon General identified skin cancer as a major public health problem and recommended increasing prevention efforts aimed at identifying at-risk populations and reducing risk behaviors.11
Conceptual and empirical data suggest that sexual minority men might be more likely to engage in indoor tanning and to develop skin cancer. Sexual minorities include, but are not limited to, those who identify as lesbian, gay, or bisexual. Indoor tanning is strongly associated with body image12 and appearance concerns.13 Indoor tanning may also be addictive14,15 and is more common among individuals who engage in other addictive behaviors, such as tobacco or alcohol abuse.16 Compared with heterosexual men, gay and bisexual men are more likely to report body dissatisfaction17 and substance abuse issues.18 Those factors might predispose gay and bisexual men to engage in indoor tanning that, in turn, raises their likelihood of developing skin cancer. Analysis of a nationally representative, longitudinal study of US adolescents found that sexual minority men were significantly more likely than heterosexual men to have ever engaged in indoor tanning by 16 years of age (27.0% vs 8.6%; odds ratio [OR], 3.9; 95% CI, 1.6-9.8).19 To our knowledge, skin cancer prevalence among sexual minority men compared with heterosexual men has not been studied, and no study has assessed skin cancer prevalence or indoor tanning behaviors by sexual orientation in women. In this study, we investigated the prevalence of a lifetime history of skin cancer and 12-month history of indoor tanning among sexual minority men and women compared with heterosexuals in the general adult population.
In this cross-sectional study, we used data from the adult questionnaires of the 2001, 2003, 2005, and 2009 California Health Interview Surveys (CHISs)20 and the adult questionnaire of the 2013 National Health Interview Survey (NHIS).21 Survey instruments and years were selected based on the availability of data on sexual orientation, skin cancer, and indoor tanning behaviors. The adult questionnaire of the CHIS is a cross-sectional, population-based survey of California residents conducted every 2 years by the University of California, Los Angeles, Center for Health Policy Research that uses a 2-stage, geographically stratified, random-digit-dial sample of households, surveying 1 randomly selected adult from each sampled household. The adult questionnaire of the NHIS is a cross-sectional, population-based survey of the national population conducted annually by the National Center for Health Statistics that uses a multistage area probability design permitting the representative sampling of household and noninstitutional group quarters (ie, college dormitories), surveying 1 randomly selected adult from each sampled household.22 This study was deemed exempt from institutional review board review at Stanford University and the University of California, San Francisco, because it uses publicly available, deidentified data; therefore, no informed consent was required.
We restricted all analyses to persons 18 years or older who were asked their sexual orientation (n = 195 051) and identified as heterosexual (straight) or as a sexual minority (homosexual, gay, lesbian, or bisexual). We excluded individuals who identified as asexual or celibate (n = 820) or other (n = 92), who did not know or refused to identify their sexual orientation (n = 779), or for whom sexual orientation information was missing (n = 785). This resulted in a final unweighted study population of 192 575 across 3 analysis sample groups: the 2001-2005 CHISs, 2009 CHIS, and 2013 NHIS.
Covariates evaluated were factors associated with skin cancer23 and/or indoor tanning,24 including age,25 race/ethnicity,26 body mass index,27 household income,28 health care use29 (≥1 physician visit within 12 months), smoking history30 (never, current, former), and current alcohol consumption31 (nondrinker, light drinker, or moderate to heavy drinker). Information on US Census region32 and self-reported history of immunosuppression33 was collected only in the 2013 NHIS. History of immunosuppression was classified as due to a medical illness, such as human immunodeficiency virus (HIV), leukemia, or lymphoma, or due to all other causes (prescription medication, medical treatment, chronic disease, poor nutrition, general frailty, and/or history of cancer).
Skin cancer history was assessed in the 2001, 2003, and 2005 CHISs and the 2013 NHIS and was defined as self-reported, physician-diagnosed history of NMSC, melanoma, or an unknown skin cancer type. For the CHIS analyses, we also separately assessed NMSC and melanoma. For the NHIS analyses, skin cancer types were not individually evaluated because of insufficient sample size.
Indoor tanning history was assessed in the 2009 CHIS and the 2013 NHIS and defined as any self-reported use of an indoor tanning device (sunlamp, sunbed, or tanning booth, excluding spray-on tans) within the past 12 months.
All analyses were stratified by sex, with heterosexuals as the reference group. Data for the CHIS analyses were weighted using replicate weights with jackknife variance estimates. Data for the NHIS analyses were weighted using the final person weights with linearized variance estimates based on the predefined stratum and primary sampling units.
Using the previously defined covariates associated with skin cancer and/or indoor tanning, we first compared characteristics of heterosexual and sexual minority respondents for each survey sample and sex-specific analysis group. We assessed statistical significance using likelihood ratio tests for categorical variables and adjusted Wald tests for continuous variables.
We calculated age-adjusted prevalence rates for self-reported lifetime history of skin cancer and 12-month history of indoor tanning using the direct method for age standardization.34 Age-adjusted NMSC and melanoma prevalence rates were calculated only for the 2001-2005 CHIS analyses. Among all individuals with a lifetime history of any skin cancer in the unweighted 2001-2005 CHIS sample (n = 4570), those who did not know their skin cancer type (n = 328) were excluded from these analyses. We found no evidence of differential missing data of skin cancer type by sexual orientation status among men or women.
We used multivariate logistic regression to calculate the adjusted OR for (aOR) for prevalence, by sexual orientation, of self-reported lifetime history of skin cancer and 12-month history of indoor tanning. The aORs for prevalence, by sexual orientation, of self-reported NMSC and melanoma were calculated only for the 2001-2005 CHIS analyses with exclusion criteria as above. Covariates for multivariate analysis were selected a priori based on their known association with skin cancer or indoor tanning, respectively. All skin cancer analyses were adjusted for age, race/ethnicity, body mass index, annual household income, health care use, smoking history, and current alcohol consumption. Analyses of the 2013 NHIS data were also adjusted for US Census region and history of immunosuppression. All indoor tanning analyses controlled for age, race/ethnicity, and annual household income. Analyses of the 2013 NHIS data were also adjusted for US Census region. For indoor tanning, we also conducted subpopulation analyses that restricted each survey sample– and sex-specific analysis group to only respondents aged 18 to 34 years, among whom indoor tanning is more strongly associated with increased skin cancer risk.35
All statistical analyses were performed using STATA software, version 13.1 (StataCorp), with a 2-sided α < .05.
The study included 78 487 heterosexual men, 3083 sexual minority men, 107 976 heterosexual women, and 3029 sexual minority women. Participant characteristics by sexual orientation in the weighted sample are presented for men (Table 1) and women (Table 2). The unweighted number of participants according to sexual orientation are listed in eTable 1 in the Supplement. Compared with heterosexuals, sexual minority men were younger; had higher household incomes; were more likely to use health care, smoke, drink alcohol, and identify as non-Hispanic white; and were less likely to be overweight or obese. In the 2013 NHIS, sexual minority men were also more likely to report a history of immunosuppression due to a medical illness and to report living in the Western US Census region. Compared with heterosexuals, sexual minority women were younger and more likely to be obese, smoke, drink alcohol, and identify as non-Hispanic white. In the 2013 NHIS, sexual minority women were also more likely to report living in the Northeast or Western US Census region.
Age-adjusted prevalence rates and aORs of lifetime history of skin cancer by sexual orientation are given in Table 3 (detailed regression models are available in eTable 2 in the Supplement). In the 2001-2005 CHISs, skin cancer prevalence was higher among sexual minority men (4.3%; 95% CI, 3.4%-5.4%) compared with heterosexual men (2.7%; 95% CI, 2.5%-2.8%), with a significantly increased odds of reporting a history of skin cancer (P = .002), including both NMSC (P = .04) and melanoma (P = .02). In the 2013 NHIS, we found comparably increased rates of skin cancer among sexual minority men (6.7%; 95% CI, 4.2%-10.7%) compared with heterosexual men (3.2%; 95% CI, 2.8%-3.5%), with a significantly increased odds of reporting a history of skin cancer (P = .02).
Odds of reporting a history of skin cancer overall did not differ significantly between sexual minority women and heterosexual women in the 2001-2005 CHISs (P = .38) or the 2013 NHIS (P = .36). In the 2001-2005 CHISs, we found that NMSC prevalence was lower among sexual minority women (1.1%; 95% CI, 0.7%-1.7%) compared with heterosexual women (1.8%; 95% CI, 1.7%-1.9%), with a significantly decreased odds of reporting a history of NSMC (P = .008). No significant differences in history of melanoma were found.
Age-adjusted prevalence rates and aORs of 12-month history of indoor tanning by sexual orientation are given in Table 4 (detailed regression models are available in eTable 3 in the Supplement). Compared with heterosexuals, sexual minority men had a significantly increased odds of reporting a history of indoor tanning in the 2009 CHIS (P < .001) and the 2013 NHIS (P < .001) for men of all ages and in the 2009 CHIS (P = .001) and the 2013 NHIS (P = .004) for only men aged 18 to 34 years.
Compared with heterosexuals, sexual minority women had a significantly decreased odds of reporting a history of indoor tanning in the 2009 CHIS (P = .03) and the 2013 NHIS (P = .007) for women of all ages and in the 2013 NHIS (P = .003) for only women aged 18 to 34 years.
This study indicates that skin cancer prevalence and indoor tanning behaviors vary by sexual orientation among men and women. In this population-based, representative study using both state-based and national, cross-sectional data, we found that sexual minority men had approximately twice the odds of reporting a history of skin cancer, including both NMSC and melanoma, compared with heterosexual men. Those differences persisted even after controlling for history of immunosuppression, which included HIV infection status. Furthermore, sexual minority women had approximately half the odds of reporting a history of NMSC compared with heterosexual women.
It is likely that differences in skin cancer history reflect, in large part, differences in UV light exposure behaviors that vary by sexual orientation that could not be controlled in our analyses. Sexual minority men had up to an approximately 6.0-fold increased odds of reporting indoor tanning in the last year, including men aged 18 to 34 years, when use in adults is most associated with increased skin cancer risk.35 The magnitude of these disparities is similar to a previous study.19 Conversely, sexual minority women had approximately half the odds of indoor tanning in the last year compared with heterosexual women in our California- and US-based representative samples. Data on outdoor tanning practices were unavailable for our analyses. However, a previous study36 found that indoor tanners are also more likely to engage in outdoor tanning behaviors. We predict that disparities in outdoor tanning might also exist by sexual orientation among men and women.
Our findings have several important implications. In 2011, the Institute of Medicine called for improved research to better identify unmet health needs of sexual minorities.37 Our study highlights disparities by sexual orientation in skin cancer history and indoor tanning, a known behavioral risk factor for skin cancer. This finding may inform practice by increasing awareness among practitioners and the larger medical community and by raising the need for better public health efforts,38,39 specifically, targeting risk behaviors among sexual minority men. The US Preventive Services Task Force does not currently recommend routine skin cancer screening in the general population by primary care physicians because of insufficient evidence concerning its benefits and harms.40 However, other studies41- 43 suggest skin cancer screening can significantly increase early detection of skin cancer and decrease disease-related mortality. Identifying sexual minority men as a high-risk population for skin cancer may help physicians risk stratify those patients in whom more diligent screening might be more beneficial. These proposed efforts fall in line with public health recommendations to address increasing rates of skin cancer in the United States.11
These findings also have important implications for skin cancer research, particularly for studies examining the link between skin cancer and HIV infection, which disproportionately affects sexual minority men in the United States.44 Two studies found that an increased risk of melanoma45 and basal cell carcinoma46 in HIV-positive compared with HIV-negative men was limited to or significantly elevated among men who have sex with men. The authors hypothesized that increased rates of skin cancer in HIV-positive men might be due to higher rates of recreational sun exposure or indoor tanning behaviors among men who have sex with men because these variables were not controlled for. Our study indicates that sexual minority men are more likely to develop skin cancer, even after controlling for history of immunosuppression, and are more likely to indoor tan. Thus, data on sexual orientation might be important to consider in epidemiologic studies to better control for UV exposure when data on related behaviors are unavailable or incomplete.
This study has particular strengths. We used large, population-based study samples and statistically rigorous methods that controlled for many known risk factors for skin cancer and indoor tanning. Our results were consistent across 2 survey instruments in California- and US-based representative samples. Although a previous study19 has suggested increased rates of indoor tanning in sexual minority men, we have validated these findings in larger study samples controlling for known risk factors for indoor tanning. Furthermore, we have extended these findings by assessing indoor tanning behaviors by sexual orientation in women.
This study also has some limitations. First, other uncontrolled confounders that might vary by sexual orientation, including Fitzpatrick skin type, family history, chemical exposures (eg, arsenic), or differences in reporting bias or skin cancer screening (ie, ascertainment bias), could also explain these findings. Second, data on history of immunosuppression were unavailable in the 2001-2005 CHISs, and previous work47 has indicated that the prevalence of HIV infection among gay and bisexual men in California may be high. However, controlling for self-reported history of immunosuppression in the 2013 NHIS had only a minor effect on the increased odds of skin cancer in sexual minority men, and we would predict a similar relationship in the 2001-2005 CHISs. Third, we relied on self-reported skin cancer diagnoses that were not validated by medical record review. Some studies48,49 have found self-reported diagnoses may underestimate skin cancer prevalence rates, whereas one study50 found that they have validity and reliability. Fourth, although our overall sample size was large, some analyses were likely not adequately powered to detect differences. Fifth, our analysis approach combined gay and bisexual men and lesbian, gay, and bisexual women into a single sexual minority group for statistical power, which may limit the generalizability of our results when applied individually across these heterogeneous populations. Sixth, only a small proportion of sexual minority respondents identified as elderly (>65 years old), which may limit the generalizability of these findings to older adults. Seventh, this is an observational, cross-sectional study, which precludes proving casual inference between indoor tanning and skin cancer in sexual minority men.
Further research addressing sexual orientation, skin cancer, and related risk behaviors is warranted. Future prospective studies could be designed to better assess a causal relationship between UV light exposure behaviors and risk of skin cancer in sexual minorities. Furthermore, quantitative and qualitative studies could be designed to investigate correlates and motivations for indoor tanning and other UV light exposure behaviors among sexual minorities; such studies would provide valuable information to better inform public health interventions aimed at addressing these disparities.
We report an increased odds of skin cancer among sexual minority men and more frequent indoor tanning, a known, preventable cause of skin cancer. Targeted prevention efforts focused on the dangers of indoor tanning and early detection of skin cancer may help reduce the burden of skin cancer among sexual minority men.
Accepted for Publication: August 12, 2015.
Corresponding Author: Sarah Arron, MD, PhD, Department of Dermatology, University of California, San Francisco, 1701 Divisadero St, Third Floor, Box 316, San Francisco, CA 94115 (firstname.lastname@example.org).
Published Online: October 7, 2015. doi:10.1001/jamadermatol.2015.3126.
Author Contributions: Dr Mansh had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Mansh, Chren, Arron.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Mansh, Chren, Arron.
Critical revision of the manuscript for important intellectual content: All authors,
Statistical analysis: Mansh, Katz, Linos, Arron.
Obtained funding: Mansh, Chren.
Administrative, technical, or material support: Chren, Arron.
Study supervision: Linos, Chren, Arron.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported by a Stanford Medical Scholars Research Grant (Mr Mansh); grant KL2TR000143 from the National Center for Advancing Translational Sciences, National Institutes of Health to the University of California, San Francisco, Clinical and Translational Science Institute (Drs Linos and Arron); and grant K24 AR052667 from the National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health (Dr Chren).
Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and the decision to submit the manuscript for publication.
Disclaimer: The contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health.
Previous Presentations: This study was presented in abstract form at the 2015 American Academy of Dermatology Annual Meeting; March 20-24, 2015; San Francisco, California.
Additional Contributions: Amy J. Markowitz, JD, Clinical and Translational Science Institute, Career Development Program, University of California, San Francisco, provided editorial assistance in the development of the article. No compensation was provided.