Exposure to ultraviolet radiation is an important risk factor in the development of most skin cancers (U.S. Department of Health and Human Services, 2014). The DNA damage caused by ultraviolet exposure is cumulative, and most skin cancers are diagnosed in older adulthood (U.S. Department of Health and Human Services, 2014). For example, among non-Hispanic white adults, more than 70% of melanomas are diagnosed among adults aged 55 years and older (Holman, Freeman, & Shoemaker, 2018). Epidemiological research findings indicate that the number of sunburns during adulthood and the lifetime number of sunburns are both important risk factors for melanoma (Dennis et al., 2008). Given that adults in the United States who reach the age of 65 years can expect to live, on average, another two decades (Murphy et al., 2017; Social Security Administration, 2019), a person's skin cancer risk over a lifetime can be influenced by events that occur during older adulthood (White et al., 2019). Yet, primary skin cancer prevention efforts have largely ignored this demographic group (U.S. Department of Health and Human Services, 2014). Non-Hispanic white older adults are more likely to have been sunburned (Holman et al., 2019) and more likely to be diagnosed with skin cancer compared with older adults of other races and ethnicities (Rogers et al., 2015; U.S. Department of Health and Human Services, 2014). When used as directed, sunscreen can prevent sunburn and reduce skin cancer risk and is recommended for children (over the age of 6 months) and adults of all ages, races, and ethnicities (U.S. Food and Drug Administration, 2021). In this study, we describe sunscreen use among non-Hispanic white adults aged 50 years or older and examine beliefs about sunscreen as possible predictors of sunscreen use within this demographic group, after adjusting for age, gender, education, geographic location, and skin cancer risk score.
METHODS
Survey questions to assess beliefs about sunscreen use were grounded in the Integrative Model of Behavioral Change and Prediction (Ajzen, 2020) and based on findings from semistructured interviews with non-Hispanic white adults ages 18-49 years. Specific constructs related to sunscreen use included beliefs about outcomes (i.e., behavioral beliefs), beliefs about what others think (i.e., injunctive normative beliefs), and beliefs about what others do (i.e., descriptive normative beliefs). The corresponding survey questions are provided in Table 1. Additional details about the formative research used to elicit relevant beliefs associated with sunscreen use and develop the survey questions were previously published (Bleakley et al., 2018). To assess sunscreen use (the primary outcome), participants were asked to use a 5-point scale to indicate how often they used sunscreen when outside during the summer on a warm sunny day for more than 1 hour (1 = never, 2 = rarely, 3 = sometimes, 4 = often, and 5 = always). This item has been used in previous research (Glanz et al., 2008). We used the Brief Skin Cancer Risk Assessment Tool and the corresponding risk tertiles that were previously established to categorize participants' skin cancer risk as low, medium, or high (Glanz et al., 2003). This tool quantifies skin cancer risk based on skin cancer history, number of moles, skin color, hair color, ability to tan, ease of burning, latitude of childhood residence, sunburn history, and the presence of freckles (Glanz et al., 2003). Demographic characteristics included in the analyses were age, gender, education, and geographic location.
Survey respondents were recruited from a probability-based online sample called KnowledgePanel (IPSOS Knowledge Panel, 2021). The sampling frame for this online panel is the United States Postal Service Delivery Sequence File, which covers nearly 100% of the U.S. population. The survey was conducted during October-December 2015 and March 2016 among 1,742 U.S. adults aged 18 years or older. The overall response rate was 44%. Findings for non-Hispanic white participants under the age of 50 years were previously reported (Bleakley et al., 2018). The present analysis focuses on non-Hispanic white respondents aged 50 years or older (N = 238) and excludes one participant with missing data resulting in a final analytic sample of 237. The study was approved by the University of Pennsylvania Institutional Review Board.
Statistical Analyses
We computed descriptive statistics for all study variables. Differences between demographic groups were tested using chi-square tests. To check for multicollinearity across the study variables, we calculated correlation coefficients and variance inflation factor for each variable. All correlation coefficients were less than 7, and all variables had a variance inflation factor of less than 3. Multiple logistic regression analyses were conducted to examine which beliefs were associated with often or always using sunscreen, after adjusting for all variables in the model. Scales were treated as ordinal variables in the regression analyses, so the odds ratios for the beliefs about outcomes and beliefs about what others think indicated the increased (or decreased) likelihood of often or always using sunscreen for each point increase on the belief scale. Differences in adjusted odds ratios for each predictor variable in the model were tested by exploring Wald tests. All the statistical analyses were performed using SAS v. 9.4 (Cary, NC).
RESULTS
Participant ages ranged from 50 to 87 years, with 34.6% aged 50-59 years, 39.2% aged 60-69 years, and 25.2% aged 70 years and older. Just over half (54.0%) were women, and most (60.3%) had at least some college education. Slightly more than half (54.4%) were in the northern part of the United States. Participants were evenly distributed across the three skin cancer risk categories, with 35.4% categorized as being at a low risk for skin cancer, 33.3% at a medium risk, and 31.2% at a high risk.
Table 2 shows the results from bivariate analyses to examine sunscreen use by demographic characteristics. Overall, 30.8% of participants reported often or always using sunscreen "when outside during the summer on a warm sunny day for more than an hour." There were no significant differences in sunscreen use by age group or geographic location. However, there were differences in sunscreen use by gender, education, and skin cancer risk category. Women were more likely to report often or always using sunscreen (39.8%) compared with men (20.2%). Those with some college education were more likely to report often or always using sunscreen (39.7%) compared with those with high school or lower education (21.3%). Those with a high risk for skin cancer were more likely to report often or always using sunscreen (46.0%) compared with those at a medium (26.6%) or low (21.4%) risk.
Although there were demographic differences in sunscreen use in the bivariate analyses, the demographic differences did not remain statistically significant in the multivariate regression models (Table 3). Two of the beliefs about outcomes were significantly associated with sunscreen use. Specifically, the more likely a respondent thought that sunscreen use would prevent them from getting sunburned, the more likely they were to report often or always using sunscreen (odds ratio [OR] = 1.84, 95% CI [1.16, 2.93]). The more likely a respondent thought that sunscreen use would "take too much time," the less likely they were to report often or always using sunscreen (OR = 0.65, 95% CI [0.43, 0.98]). The only belief about what others think was statistically significant related to respondents' romantic partner. Individuals who believed that their romantic partners thought they should use sunscreen were more likely to report often or always using sunscreen (OR = 1.72, 95% CI [1.07, 2.77]). Respondents who believed that most or all men or women their age use sunscreen were more likely to report often or always using sunscreen (OR = 2.75, 95% CI [0.88, 8.38]), but this finding was not statistically significant.
DISCUSSION
This study provides new information about sunscreen use among non-Hispanic white adults aged 50 years and older, a demographic group at an elevated risk for skin cancer. Although some national surveillance data on sunscreen use are available for this demographic group, this is the first study to examine their sunscreen use in the context of sunscreen-related beliefs. The prevalence of sunscreen use was similar to findings from other national data (Holman, Ding, et al., 2018; Holman et al., 2019), with only under one third of participants reporting the behavior.
Only two behavioral beliefs were significantly associated with sunscreen use. Although most skin cancers are diagnosed among older, non-Hispanic white adults (Centers for Disease Control and Prevention, 2019; Rogers et al., 2015), the belief that sunscreen use could prevent skin cancer was not associated with a greater likelihood of sunscreen use. In contrast, the belief that sunscreen use could prevent sunburn was associated with a greater likelihood of sunscreen use. This finding suggests that messages focused on the immediate benefits of sunscreen use, and in particular, the prevention of sunburn, may be more salient to older adults than messages about reducing their skin cancer risk.
The belief that sunscreen use takes too much time was significantly associated with reduced frequency in sunscreen use. Given the wide range of sunscreen products featuring different characteristics on the market, there may be a benefit to tools or information that helps consumers find a sunscreen product that fits with their personal application preferences. For example, those concerned about sunscreen use taking too much time might prefer a sunscreen in a solid stick formulation that could be quickly rubbed over skin (similar to popular deodorant applications) or a spray-on formulation to quickly spray large sections of skin at one time. Such communications could also advise sunscreen users about potential concerns with certain application methods. For example, aerosol spray on sunscreens can be challenging to apply in outdoor settings, particularly in windy weather, and should not be applied directly to the face because of inhalation concerns (U.S. Food and Drug Administration, 2021).
Of the beliefs examined about what others think, only the belief that one's romantic partner thinks they should use sunscreen was significantly associated with sunscreen use. These results suggest that beliefs about the opinions of friends, people who have had skin cancer, and medical professions may be less influential on this demographic group's sunscreen use. Future research could evaluate messages that invite older adults to encourage their partner to engage in sun-safe behaviors with them.
Study Limitations and Strengths
This study has important limitations and strengths to consider when interpreting the results. The online survey had a low response rate, and the final sample of non-Hispanic white adults aged 50 years and older was small. This small sample size limits the statistical power of the study and the ability to measure effects as statistically significant. The small sample size also limits the generalizability of the study findings. The data were self-reported and thus subject to social desirability bias. The beliefs we examined were based on data from structured interviews with adults ages 18-49 years and may not fully capture the most salient beliefs regarding sunscreen use among older age groups. However, very little work has been done to examine sun protection behaviors among older U.S. adults, and this is the first study of its kind to examine beliefs associated with sunscreen use among older U.S. non-Hispanic white adults. These study findings begin to fill a gap in the current literature and expand the sun-safety dialogue to include opportunities to promote sun safety among older age groups.
CONCLUSION
Given the elevated incidence rates of skin cancer at older ages in the United States, effective sun-safety intervention strategies are needed for older adults to lower their risk of developing skin cancer. The current findings can help to shape and inform the focus of such efforts, including the evaluation of messaging strategies that highlight the benefits of sunscreen use for sunburn prevention and address concerns about sunscreen use taking too much time. In addition, the findings point to the potential for older adults to have an important influence on the sunscreen use of their romantic partners-a concept that could also be further examined through the evaluation of novel messaging and intervention approaches that tap into this influence.
REFERENCES