Indoor tanning is a risk factor for skin cancer, but the population-based prevalence of this behavior among adolescents is not clearly known.
To describe the prevalence of tanning booth use among white US adolescents and to assess its association with sociodemographic factors, health behaviors, and appearance.
Design, Setting, and Participants
A nationally representative sample of 6903 non-Hispanic white adolescents, aged 13 to 19 years, who participated in Wave II (between April 1, 1996, and August 31, 1996) of the National Longitudinal Study of Adolescent Health.
Main Outcome Measure
Frequency of lifetime use of indoor artificial tanning facilities.
With the data weighted to national levels, 36.8% of the white female adolescents and 11.2% of the white male adolescents have used a tanning booth at least once in their life, while 28.1% and 6.9% of the female and male adolescents, respectively, reported tanning booth use 3 or more times. The percentage of female adolescents using tanning booths 3 or more times increased with age, from 11.2% of the 13- to 14-year-olds to 47.0% of 18- to 19-year-olds and also increased with greater tanning ability, from 12.6% of the poor tanners to 38.1% of those with a strong tan response. After multivariate adjustment, those residing in the Midwest (adjusted odds ratio [aOR], 2.38; 95% confidence interval [CI], 1.53-3.68) or South (aOR, 2.91; 95% CI, 1.89-4.53), attending a rural high school (aOR, 1.80; 95% CI, 1.09-2.98), and reporting the use of 2 or 3 substances (aOR, 3.06; 95% CI, 2.44-3.82) were more likely to use indoor tanning facilities, as were dieters (aOR, 1.26; 95% CI, 1.01-1.57) regardless of their body mass index. Decreased odds of indoor tanning were observed among those with a college-educated mother (aOR, 0.68; 95% CI, 0.51-0.90) and greater cognitive ability (per 10-point score increase; aOR, 0.89; 95% CI, 0.82-0.96) while routine participation in physical activity significantly lowered the odds of indoor tanning only among female adolescents.
Indoor tanning is prevalent, particularly among female adolescents, and aligns with other risk behaviors, appearance-related factors, and intentional sunbathing. The risks of artificial tanning need increased emphasis among adolescents, especially in the Midwest and South where extremes in the availability of natural light appear to send intentional tanners indoors.
THE UV RADIATION component of natural light is a risk factor for skin cancer of all types1 and artificial light from sunbeds, sunlamps, and commercial tanning facilities, referred to here as "tanning booth use," also contain UV radiation. Studies using detailed exposure measurements have shown an association between artificial tanning devices and cutaneous malignant melanoma while other studies have reported mixed results.2- 4 Use of these devices by adolescents, particularly older female teenagers, has been documented in US5- 8 and international studies9,10 and short-term adverse health effects have been reported by adults and adolescents.5,10,11 Despite the potential for both short- and long-term adverse effects, the continued use of these devices demonstrates some adolescents' strong motivation for a tanned appearance.
Tanning booth use, as other behaviors, is likely to be influenced by cognitive, psychological, social, behavioral, and physical factors and several theories of health and social behaviors have been invoked to frame the development of sun-related behavior models.12- 15 Previous studies among adolescents have demonstrated an association between tanning booth behaviors and age, sex, favorable attitudes about tanning, a resident adult who also tans indoors, and smoking.5- 10 Understanding the position of tanning booth use in relationship to other teenage behaviors may identify a potential intervention point to influence the desire for a tan and decrease the tan-seeking behavior. To our knowledge, this is the first national study to investigate the relationship between tanning booth use and other health-promoting and health-risk behaviors. It is also the largest nationally representative sample to permit multivariate analysis of these correlates with tanning booth use.
Using the National Longitudinal Study of Adolescent Health (Add Health) Study (AHS) Wave II data set, this study aims to estimate the frequency of tanning booth use among white US adolescents and to examine the association of indoor tanning with demographic characteristics; tobacco, alcohol, and marijuana use; good nutritional and physical activity behaviors; cognitive ability; skin solar response; and body mass index (BMI) (calculated as weight in kilograms divided by the square of height in meters).
The AHS was designed to collect longitudinal data on health-related behaviors of adolescents in grades 7 through 12 and to explore the effects of individual characteristics and social environments on those behaviors.16 One hundred thirty-two schools in 80 communities were enrolled based on a sampling design that ensured representation of US schools for region of country, urbanicity, school type, ethnicity, and school size. The design and administration of the study surveys have been described elsewhere.16,17 The current analysis is based on Wave II data collected between April 1, 1996, and August 31, 1996, with the exception of maternal education that was obtained previously from the parent survey. The Wave II total response rate was 88.2%16 and the Wave II grand sample weights provided by the AHS simultaneously adjusted for nonresponse and the different selection probabilities of sample members.
The AHS data were collected with active parental consent (Waves I and II) and under strict guidelines of confidentiality to guard against deductive disclosure. Human subject protection during both the primary data collection (conducted by the University of North Carolina, Chapel Hill) and the secondary use of these data by us was reviewed and approved by our respective institutional review boards.
Participants included in this analysis (1) had a Wave II grand sample weight, (2) were between 13 and 19 years old, (3) self-identified as non-Hispanic white, and (4) answered all sun-related outcome questions. The final sample size for analysis was 6903 of the 7573 white respondents in Wave II. The analysis was restricted to white adolescents because they are the primary users of the artificial tanning facilities and are at significantly greater risk for both short- and long-term adverse effects from overexposure to UV radiation.
The study outcome—lifetime frequency of artificial tanning light use—was measured by 1 question: "How many times in your life have you used a sunlamp or a tanning booth or a tanning parlor or salon?" The 6 response categories (Table 1) were dichotomized into nonusers (<3 times, 82.4%) and users (≥3 times, 17.6%).
Age, sex, geographic region of residence, urban location of the school, maternal educational level, and student income or allowance were recorded for each respondent. The following 2 questions assessed skin solar response type: "After several months of not being in the sun, when you go out in the sun without sunscreen or protective clothing for the first time for at least an hour, do you get . . . " a severe sunburn with blisters; a bad sunburn that peels later (combined for burn type 1/2); a mild sunburn that becomes a tan (burn type 3); or a suntan without burning (burn type 4). The tan response was measured by the question: "When you go out in the sun every day for 2 weeks, do you get . . . " very dark and deeply tanned (tan type 4); moderately tanned (tan type 3); mildly tanned (tan type 2); only freckled, no suntan at all or repeated sunburns (tan type 1). Both questions included an "Other" category without additional specification. Intentional sunbathing was measured by the question: "During the summer, how often do you sunbathe, or lie in the sun, to get a tan?" with response categories of frequently, occasionally, rarely, or never (referent).
Any use of tobacco, alcohol, or marijuana since the Wave I interview (mean [SD] time since Wave I was 10.9 [1.6] months) was combined and categorized as no substance use (referent), 1 substance used, or 2 or 3 substances used. A nutrition behavior index was constructed from 3 questions regarding eating behaviors in the last 7 days: (1) eating breakfast on 5 or more days (yes, 1; no, 0); (2) eating fast food on 2 or more days (no, 1; yes, 0); and (3) taking vitamin supplements (yes, 1; no, 0). Responses were summed and categorized as 0, poor behaviors (referent); 1, moderate behaviors; 2 or 3, good behaviors. Physical activity was included as a composite score from 3 questions regarding the frequency per week of participating in a variety of active sports or exercises. Responses were summed over the 3 questions with a range of 0 to 15 or more times per week and analyzed as a continuous variable.
Body mass index was calculated from physical measurements obtained at the time of the Wave II interview (n = 6774) or from self-reported height and weight (n = 98). Missing cases (n = 31) were imputed with the Stata18 imputation command using the sample BMI, age, and sex. Respondents were then categorized according to the gender-specific BMI-for-age Centers for Disease Control and Prevention growth curves.19 Body image was measured with "How do you think of yourself in terms of weight?" with 5 response categories ranging from very underweight to very overweight. Teenagers reported a self-assessment of physical maturity with responses ranging from "younger than most my age" through "older than most my age."
Dieting behavior was measured by self-report of trying to lose weight, gain weight, or neither (referent). Responses from 2 questions about a permanent tattoo (yes, 1; no, 0) and body piercing, other than ears (yes, 1; no, 0) were summed and dichotomized as neither (0, referent) and either or both (1).
The Add Health–Picture Vocabulary Test, used as a measure of cognitive ability, is an abridged version of the Peabody Picture Vocabulary Test-Revised20 and its validation and use in the AHS is reported elsewhere.21 Age-standardized scores for this sample of teenagers ranged from 77 to 129 (weighted mean [SE], 104.7 [0.48]). One item assessed thoughtful decision making on a 5-item Likert scale of strongly agree (5) to strongly disagree (1) regarding "making decisions without thinking too much about the consequences."
Data were analyzed using the design-based survey commands in Stata Statistical Software, Release 6.018 to adjust for the effects of the complex survey design used in the AHS and to produce unbiased estimates of the SEs. Crude associations between tanning booth use and categorical study variables were examined by sex using χ2 tests of independence implemented with survey design-based corrections. Age, physical activity level, thoughtful decision making, and Add Health–Picture Vocabulary Test score were examined as continuous variables and compared between tanning booth users and nonusers within gender with weighted t tests.
Hypothesized risk factors for tanning booth use were organized within a multivariate framework that examined the relationship between tanning booth use and (1) sociodemographic factors, (2) skin type and sunbathing practices, (3) health behaviors, (4) appearance-related behaviors, and (5) psychosocial factors. Study variables significant in the χ2 analyses at P<.05 were included in preliminary logistic regression models to assess the independent contribution of candidate variables within these domains to tanning booth use. A final logistic model was determined by a stepwise backward elimination approach using significant predictors from all 5 domains, including 2-way interaction terms between BMI and dieting behavior, sex and nutritional behavior, sex and physical activity, and sex and physical maturity. The contribution of each independent variable to the model was examined using an adjusted Wald statistic. Nonsignificant variables remained in the model if they proved to be confounders (ie, if their exclusion altered other parameter coefficients by >10%) or improved the fit of the model based on the Hosmer-Lemeshow χ2 Goodness-of-Fit statistic,22 resulting in the most parsimonious model in terms of individual variables and overall model fit. Adjusted odds ratios and 95% CIs were reported from the multivariate analysis.
The final sample of 6903 respondents represented a weighted total of about 12.03 million non-Hispanic white adolescents in the United States between the ages of 13 and 19 years. Male and female adolescents were equally represented with weighted mean age of 15.9 years (95% CI, 15.6-16.1). Almost two thirds of the teenagers attended suburban schools, and 22.0% had a college-degreed mother. Most teenagers classified their unprotected skin response as burn type 1/2 or burn type 3 (19.4% and 47.3%, respectively). A larger majority of white teenagers reported at least a mild tan or better after a 2-week exposure (tan types 2-4, 87.6%), suggesting that most white adolescents perceive themselves as capable of obtaining a tan.
Overall, 24.1% of the respondents used a tanning facility at least once in their life, representing 2.9 million adolescents (Table 1). Six and a half percent of the teenagers reported indoor tanning only once or twice, while 17.6% used tanning booths 3 or more times. Among female adolescents who have ever used a tanning booth, 76.2% did so 3 or more times, while 50.8% used them 10 or more times. Thus, multiple tanning sessions appear to be the norm. For this analysis, tanning booth users were defined as those who reported indoor tanning 3 or more times.
As given in Table 2, more female than male adolescents have used tanning facilities (28.1% vs 6.9%) and the frequency of tanning booth use increased with age. The proportion of tanning booth users increased with the increasing ability to tan and with a decreasing tendency to burn. The proportion of indoor tanners was different by region, urban locale, student income or allowance, and maternal educational level but not by the month of the survey.
Weighted percentages were obtained for the distribution of all other study variables between tanning booth users and nonusers by sex (Table 3). Tanning booth users were more frequent sunbathers and substance users. Female tanning booth users had a higher mean (SE) participation in physical activities (5.1 [0.14] times per week) and more indoor tanners reported 2 to 3 good nutritional behaviors (40.7%), both compared with female nonusers. Overall, nonusers had a higher average score on the Add Health–Picture Vocabulary Test and more female nonusers agreed with a thoughtful decision-making process compared with indoor tanners. Because several variables demonstrated associations with tanning booth use that differed by sex, interaction terms with these variables and sex were included in the logistic regression analyses.
From the initial domain-specific models, burn type, month of interview, and body image were not significantly associated with tanning booth use and were not confounders. The interaction of BMI and dieting behavior was also not significant. Frequent sunbathers were 8 times more likely to use a tanning booth than those who never sunbathed (adjusted odds ratio, 8.29; 95% CI, 6.0-11.44). However, the frequency of sunbathing also redundantly controlled for other independent variables (for example being female and skin type) and its presence significantly reduced the fit of the model. Ultimately, sunbathing, interview month, body image, and burn type were not retained in the final model.
All variables that demonstrated significance in the preliminary models were included in a final model (Table 4). Being female, increasing age, residing in the Midwest or South, having a personal income or allowance and nonurban locale all increased the odds of tanning booth use among white teenagers. Compared with all other educational attainment, having a college-degreed mother decreased the likelihood of tanning booth use (adjusted odds ratio, 0.68; 95% CI, 0.51-0.91).
Those reporting the use of 2 or 3 substances were more than 3 times as likely to be tanning booth users compared with those who reported no substance use. Those attempting weight loss and female adolescents who perceived themselves as more physically mature than their peers were also more likely to be indoor tanners. Poor tanning ability, greater cognitive ability, and higher BMI all remained protective factors. Increasing frequency of physical activity also decreased the odds of indoor tanning among female adolescents. In the final model, body piercing/tattoos and thoughtful decision making, and the modification of nutritional behaviors by sex were no longer independent predictors of tanning booth use.
The current analysis presents national estimates of the prevalence of lifetime tanning booth use in the white adolescent US population as well as the relationship of this activity to other risk behaviors. Tanning booth use increased with age among adolescents regardless of sex, but female adolescents consistently reported more frequent indoor tanning than male adolescents. In conjunction with 2 recent national reports,7,8 these results consistently estimate that indoor tanning among white teenagers is significant, with 30% to 40% of 16- through 18-year-old white female adolescents using tanning booths, many of them repeatedly.
Adolescents differed in tanning booth use by skin type, sunbathing frequency, region of the country, and urban locale. Adolescents with the best tan response were more likely to be tanning booths patrons, but use among the more sensitive skin type was not trivial. Self-rated tanning ability more strongly influenced the use of indoor tanning facilities than did the initial burn response. Thus, mild to moderate tanners with an initial burn response will attempt tanning indoors. However, the 2-week tan response question does not specify the use of sun protection or sun-enhancing products, which may lead to the underestimation of sun sensitivity. Furthermore, one study23 suggests that people often perceive themselves as less sun susceptible than would be scored by skin reflectance measurements. This underestimation of sun sensitivity would result in sun-sensitive individuals populating the less-sensitive categories and using tanning booths at higher frequencies than were detected.
Adolescents in the Midwest and South were 2 to 3 times more likely to use tanning booths compared with the rest of the country. Increased use of indoor tanning beds in areas of higher ambient UV radiation (South) might be explained by discomfort during sunbathing in outdoor heat or the desire to "base tan" indoors prior to intense natural light exposure. The significant association of outdoor sunbathing and indoor tanning is of concern because of intense exposure to UV-A rays from indoor tanning and UV-B rays from natural sunlight. Moreover, indoor tanning lamps still produce UV-B emissions contributing to DNA damage24,25 in addition to the predominant UV-A fraction, which increasingly is considered to be important in the carcinogenic process.8,24,25
To our knowledge, this analysis reveals for the first time the strong positive association of other health-risk behaviors with tanning booth use among US adolescents and is congruent with the reported negative associations of smoking, alcohol, and marijuana use with sunscreen use.26- 28 If intentional tanning aligns with other health-risk behaviors, incorporating antitanning messages into comprehensive prevention messages could supplement topic-specific curricula on sun-related behaviors. In addition, physician health behavior counseling to adolescents should include the potential risk of tanning booth use.29
Appearance-related motives (eg, looking attractive, perceived attractiveness, and self-presentation motives) have been identified as strong psychosocial indicators of intentional sun exposure among adolescents.30,31 Dieting activity and body piercing/tattoos are behaviors generally intended to enhance attractiveness and were hypothesized to be positively associated with tanning bed use to improve appearance. Trying to lose weight moderately increased the odds of tanning booth use while an actual greater BMI was associated with decreased indoor tanning. Thus, these dieters include many average-weight individuals (generally female adolescents) who may also be tanning to increase attractiveness. Truly overweight adolescents seem less likely to engage in this behavior, possibly because they may not be comfortable entering a tanning facility.
Using a different paradigm, others have suggested that appearance-enhancing behaviors present alternatives from which persons will choose. Hillhouse et al14 determined that college students with strong positive attitudes toward artificial tanning held strong negative attitudes toward clothing as an alternate behavior to improve attractiveness, but found no relationship with exercise, weight control, financial status, or intelligence. Our framework included these (except clothing) and other variables in a sample of younger adolescents and found several associations with tanning bed use, suggesting that multiple choices for appearance enhancement are possible especially at younger ages when adolescents are experimenting with ways to increase peer acceptance. Appearance-related behaviors may include both positive and negative aspects and, thus, can be associated with both positive and negative behaviors. Physical activity and good nutrition may enhance appearance as well as overall good health, while weight control via smoking may improve appearance with poor health consequences. Thus, models of tanning behaviors should include diverse behaviors as well as psychosocial and cognitive variables to further elucidate the influential relationships and care must be taken when designing interventions surrounding appearance motivation not to present alternatives that may inadvertently encourage other risk behaviors.
We interpret using an artificial tanning facility once or twice to be experimental and not representative of the behavior. Among the youngest teenagers, this may be too conservative of an approach, that is, even 1 or 2 visits at this age may identify a highly motivated tanner who will eventually engage in additional tanning sessions given time. Thus, our observed associations may actually be biased toward the null since some tanning booth users may have been misclassified as nonusers. The period of tanning booth use was specified as "in your life." Thus, we do not know, for example, if the 10 to 20 sessions were over a 2-week period or spread out over the previous 2 years. While this represents a limitation in defining the behavior pattern more precisely, dichotomizing the frequencies as a response variable still permits cautious interpretation that certain independent variables associate with repeated indoor tanning.
Several limitations to our study are similar to previous reports. We relied on self-reported data without additional objective corroboration. However, at least one study has provided evidence that adolescents' self-report of sun protective practices was valid.32 The cross-sectional nature of the data does not permit causal inferences to be drawn nor does it allow examining an individual's behavior throughout adolescence. Future study design would benefit from a longitudinal component.
The present findings extend previous work that a substantial minority of adolescents engage in indoor tanning practices and further suggest that this behavior may align with other health-risk and health-promoting behaviors practiced by adolescents. Availability of tanning salons33 and economic marketing tactics2,34 give teenagers with strong intentions to tan another method to do so. Increased emphasis on interventions and strategies to prevent both short- and long-term adverse effects of unprotected UV exposure from artificial light among adolescents is warranted.
Corresponding author: Catherine A. Demko, PhD, Comprehensive Cancer Center at Case Western Reserve University and University Hospitals of Cleveland, 11100 Euclid Ave, WRN 152, Cleveland, OH 44106-5065 (e-mail: firstname.lastname@example.org).
Accepted for publication May 22, 2003.
This work was supported in part by training grant R25-CA90355-01 from the National Institutes of Health, Bethesda, Md (Dr Demko).
This research uses data from Add Health, a program project designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by grant P01-HD31921 from the National Institute of Child Health and Human Development, with cooperative funding from 17 other agencies. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Persons interested in obtaining data files from Add Health should contact Add Health, Carolina Population Center, 123 W Franklin St, Chapel Hill, NC 27516-2524 (also available at: http://www.cpc.unc.edu/addhealth/contract.html).
Dr Demko is a postdoctoral fellow and research associate with the Behavioral Cancer Control and Prevention Program at Case Western Reserve University, Cleveland, Ohio.
A tanned appearance continues to be considered healthy and attractive and a measurable segment of the US adolescent population engages in the use of artificial light sources (eg, tanning beds or sunlamps) to achieve a tan. Even as commercial facilities promote indoor tanning as a safe tan, both short- and long-term adverse effects have been associated with tanning bed use. Development of interventions targeting the use of indoor tanning facilities would benefit from a better understanding of the correlates of tanning booth use in a large representative survey of adolescents.
Prior surveys on artificial tanning in the United States focused on areas of lower ambient sunlight, but this analysis demonstrates that indoor tanning is practiced in areas of high sunlight as well, suggesting that advice to avoid indoor tanning should not be overlooked in sunnier regions. Indoor tanning generally involves multiple sessions and is strongly associated with sunbathing; thus, teenagers who maintain year-round tans or repeatedly base tan indoors prior to outdoor sunbathing risk exposure to multiple sources of UV radiation and the unknown consequences of that combination. Finally, tanning booth use among high school adolescents seems to align with appearance-related factors and other health-risk behaviors, providing potential intervention pathways and linkages to both novel and tested prevention strategies.
Demko CA, Borawski EA, Debanne SM, Cooper KD, Stange KC. Use of Indoor Tanning Facilities by White Adolescents in the United States. Arch Pediatr Adolesc Med. 2003;157(9):854-860. doi:10.1001/archpedi.157.9.854