https://orcid.org/0000-0003-0149-8464
Figures
Abstract
Background
Melanoma survivors are at increased risk of developing a second primary melanoma; however, some report sub-optimal sun behaviors and sunburns. We tested the effectiveness of a wearable device with ultraviolet radiation (UVR)-sensing technology to improve sun behaviors and reduce sunburns in cutaneous melanoma survivors.
Materials and methods
We conducted a randomized controlled trial using Shade 2, a commercially available wrist device that measures UVR. The intervention group received the device and mobile application notifications about their exposure and prompts to use sunscreen. The control group received the device and a separate research mobile application without information about their exposure or notifications. Participants wore the device for 12 weeks and self-reported sun behaviors before, during, and after the intervention. The primary outcome was a composite score of sun protection behaviors at week 12.
Results
386 participants were randomized (186 control, 182 intervention). Most were female and 5+ years past their first melanoma diagnosis. The average age was 56 years. Most (93%) completed the study, though 40% experienced device issues. No meaningful differences were observed in self-reported sun protection behaviors at week 12 (controls 3.0±0.5 vs. intervention 2.9±0.5, p = 0.06), any sunburn during the intervention period (controls 14.4% vs. intervention 12.7%, p = 0.75), or average daily objective UVR exposure (controls median 87 vs. intervention 83 J/m2, p = 0.43).
Citation: Vogel RI, Luo X, Brown K, Jewett P, Dona AC, Nagler RH, et al. (2023) A UVR-sensor wearable device intervention to reduce sun exposure in melanoma survivors: Results from a randomized controlled trial. PLoS ONE 18(2): e0281480. https://doi.org/10.1371/journal.pone.0281480
Editor: Paula Boaventura, IPATIMUP/i3S, PORTUGAL
Received: July 25, 2022; Accepted: December 12, 2022; Published: February 10, 2023
Copyright: © 2023 Vogel et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The participant consent form did not include a statement about sharing the data, de-identified or otherwise, and given that the data contain potentially identifying or sensitive health information, we are unable to share the data due to the policies of the University of Minnesota Institutional Review Board. The University of Minnesota Institutional Review Board can be reached by email at irb@umn.edu.
Funding: Funding: This study was funded by the Melanoma Research Alliance Wayne Stinchcomb Big Orange Foundation – MRA Young Investigator Award (RIV, https://doi.org/10.48050/pc.gr.80546) and an American Cancer Society Research Scholar Award (RIV, 133512-RSG-19-014-01-CPPB). Preliminary studies were supported by the Masonic Cancer Center of the University of Minnesota (NIH P30 CA77598). Use of data collection and management tools were supported by the National Institutes of Health’s National Center for Advancing Translational Sciences, grant UL1TR002494. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Approximately 100,000 people in the United States (US) will be diagnosed with melanoma in 2022, and by 2030, there will be a projected 1.8 million melanoma survivors [1]. Melanoma survivors are at significantly greater risk of developing a second primary melanoma [2]. Ultraviolet radiation (UVR) exposure is the strongest risk factors for melanoma [3–5]. While melanoma survivors generally report less UVR exposure and more protective sun behaviors [6]—such as utilizing shade, protective clothing, and sunscreen—than those without melanoma, a nontrivial number of survivors report suboptimal sun exposure [7, 8]. Therefore, interventions to promote healthy sun behaviors among melanoma survivors are needed to prevent subsequent melanomas in this population.
Few interventions have targeted sun protective behaviors in melanoma survivors specifically, despite their high risk for future diagnoses. Most behavioral interventions to date have been designed for those without a previous melanoma diagnosis, most often school-aged children or young adults, high-risk occupational groups, or individuals otherwise at elevated risk for melanoma, such as family members or solid-organ transplant recipients (e.g. [9, 10]). The few prior melanoma survivor interventions have primarily focused on skin self-examinations [11–16]. Two studies, both web-based, also evaluated sun behaviors in melanoma survivors, with moderate improvements for some behaviors at various time points [11, 14].
Many preventive health interventions have drawn from the Health Belief Model, emphasizing risk perceptions and cues to action for promoting behavior change [17, 18]. Technology can be useful to implement these drivers [19]. Wearable technologies, such as smart watches, have become widely used and provide a novel means for health intervention. Some new wearable technologies incorporate UVR sensors to track exposure and prompts to reduce exposure, which may be especially important to reduce unintentional sun exposure.
The objective of this trial was to test the effectiveness of a wearable device with UVR-sensing technology coupled with a mobile application to promote healthy sun behaviors in melanoma survivors. We hypothesized that participants randomized to the intervention, compared to the control participants, would report greater sun protection and avoidance practices and fewer sunburns following a 12-week intervention.
Materials and methods
Overview and study design
Details regarding the protocol including the study design and schematic, intervention and comparator choice, eligibility and recruitment methods, randomization procedures, measures, and sample size considerations are described elsewhere [19]. Briefly, we conducted the MELD (Melanoma Exposures and Life after Diagnosis) study, a randomized controlled trial among 368 melanoma survivors over two waves (Summer 2020 and 2021). This two-wave summer method was utilized to account for seasonality of UVR exposure in the Midwest and allow for adequate sample size recruitment. Participants were randomly assigned 1:1 to the intervention or control group for 12 weeks and surveyed regarding their sun protection and surveillance behaviors at study entry and every 4 weeks during the intervention period. All participants provided informed consent prior to study enrollment. The study received approval from the University of Minnesota Institutional Review Board (STUDY00006257) and HealthPartners Institute Institutional Review Board (A18-353), and all participants provided written informed consent. The trial was registered on ClinicalTrials.gov (NCT03927742).
Recruitment
Individuals diagnosed with or treated for invasive cutaneous melanoma between 2010 and 2020 were recruited from HealthPartners healthcare system, a large consumer-governed nonprofit healthcare organization based in Minnesota, from May through July 2020 and 2021. Eligible participants were 18–75 years old, previously diagnosed with stage I–IV cutaneous invasive melanoma, able to read/write in English, owners of a smartphone, and able to provide voluntary informed consent.
Potential eligible individuals were identified via electronic health records and were mailed a letter introducing the study and elements of informed consent. Within 2 weeks of sending the letter, potential participants were contacted via phone to determine interest in participating in the study. Interested individuals were sent an introductory email from the University of Minnesota team including a link to online consent and Health Insurance Portability and Accountability Act (HIPAA) forms, which then led directly to the online baseline survey in Research Electronic Data Capture (REDCap) [20]. Potential participants were encouraged to reach out to the study team with questions or concerns prior to and after providing consent.
Randomization
Following written consent and completion of the baseline survey, participants were randomized 1:1 to the intervention or control arm. Randomization was stratified by age (<50, 50+ years old), gender (male, female), disease stage (I, II, III/IV, unknown), and years since first melanoma diagnosis (< 2 years, 2–5 years, 6+ years). Randomization was implemented within REDCap by the study coordinator. Everyone except the study coordinator and statistician were blinded to the randomization. Participants were not informed of their randomized group or provided specific information about what to expect from the mobile app.
Intervention
Participants were provided a commercially available UVR-sensor-enabled wrist wearable device and asked to wear the device and sync it with the mobile app every day for 12 weeks regardless of planned outdoor activities, except while swimming or doing water sports (device is not waterproof). We originally planned to use the Microsoft Band 2 that measured UVR exposure and additional health behaviors, such as steps and heart rate; it also sent notifications about using sunscreen, wearing a hat, seeking shade, and minimizing midday sun exposure. The Microsoft device, however, was discontinued and no longer supported by the time of study initiation. Instead, we used the Shade 2nd generation device, the only other commercially available UVR-sensor-enabled wrist wearable at the time. Shade was significantly more accurate in measuring sun exposure than similar devices, including the Microsoft Band [21]. However, the Shade device does not collect information on other health behaviors beyond UVR exposure, and messaging was not customizable and focuses solely on reminders to apply sunscreen.
While all participants were provided the same device, they used different versions of an associated mobile app based on their randomization. Intervention participants received the commercially available version of the mobile app, which helped users set daily UVR exposure limits, taking into account UVR intensity, time of exposure, and sunscreen application. When approaching their daily UVR limit, smartphone notifications were sent to users as a cue to action. The mobile app version used by participants in the control group only collected UVR exposure, not sunscreen use, and did not provide notifications or information about measured UVR exposure to participants. They only received a message indicating that their UVR exposure data had been reported to the study team. We considered multiple options at the time of study design and ultimately chose the attention-control group for three reasons: one, we can objectively measure UVR exposure in both groups; two, providing the device to the control group will help with recruitment, engagement, and retention; and three, the intervention is low risk.
Participants in the intervention group also received a study brochure with information about the importance of UVR avoidance and protection, as well as tips for making sun-safe behaviors routine and sustainable. This content was provided both as a printed brochure at baseline and via a website, viewable after completing each subsequent survey. Participants in the control group did not receive the brochure nor were they directed to the website.
Data collection and measures
Due to the COVID-19 pandemic, all study procedures were conducted remotely. Study measures were collected via online surveys and the Shade device. Participants were mailed their device, training materials, and mobile app access at the time of randomization. Participants were asked to wear the device and sync it with the mobile app every day for 12 weeks. The study coordinator monitored device use of all participants and contacted participants who did not use their device or sync the app data on more than 3 days in any given week to determine barriers and encourage use. We used a weekly $100 raffle-drawing incentive to promote adherence. Almost half (40%) of participants unexpectedly experienced broken straps or device malfunction during the study. Damaged or defective devices were promptly replaced; nevertheless, many participants were without a functioning device for approximately 5–10 days during the study.
Self-administered questionnaire data were collected at baseline, week 4, week 8, and week 12 (end of intervention), as well as week 64 (end of the summer, one year after intervention completion) to assess potential continued behavior change.
The primary outcome was a composite measure of sun protection behaviors (sun habit score) at the end of the intervention period (12-week survey) [22]. A total sun habit score for each individual was created by taking the average of 6 protective behaviors (wearing a shirt with sleeves, wearing sunglasses, staying in the shade, using sunscreen, spending time in the sun to get a tan [reverse coded], and wearing a hat) measured on a 5-point ordinal scale ranging from 1 (never) to 5 (always), recoded to 1 (never or rarely) to 4 (always) to align with previously published studies.
Secondary outcomes included a) sunburns (the proportions of individuals who self-reported at least one red or painful sunburn in the past 3 months at the 12-week survey) [23] and b) UVR exposure, which was both self-reported (number of hours spent outside per day between 10am and 4pm in the summer separately for weekdays and weekend days) and automatically measured by the Shade device (instantaneous UVR intensity and accumulated UVR dose over time). The device collects exposure over the course of the day and reports accumulated joules per meter squared (J/m2) every 6 minutes. If a participant in the intervention group entered into the mobile app that they applied sunscreen, the Shade commercial mobile app mathematically reduced the cumulative UVR exposure that was reported to the participant for the duration of 90 minutes. Therefore, for the following 90 minutes, these individuals did not receive reports about the true incoming UVR, but a modified (reduced) UVR exposure calculated as the objective UVR exposure divided by the SPF of the reported sunscreen. Furthermore, these modified data were provided by Shade as the raw data through the research portal for participants in the intervention group. For our analyses, we back-transformed this modified exposure estimate if individuals had UVR exposure and reported sunscreen use, using the time-stamped sunscreen application and SPF data. In this manuscript, we refer to unmodified objective UVR exposure (among those who did not report using sunscreen) and to the back-transformed UVR exposure (among those who used sunscreen at the time) as ‘objective UVR exposure’. We refer to the UVR exposure that was reduced by the SPF factor as ‘SPF modified UVR exposure’. All UVR data in the control group were objective UVR exposures.
We measured self-reported physical activity, depression, and anxiety to examine potential unintended negative consequences of the intervention. Physical activity was measured using the Godin Leisure-Time Exercise Questionnaire [24], which asks about times and minutes per week of engaging in strenuous (vigorous), moderate, and mild physical activity on average during the preceding month. Depression and anxiety were measured using the 14-item Hospital Anxiety and Depression Scale (HADS) instrument [6]. We measured satisfaction and usability of the device at week 12 using items from the System Usability Scale [25].
Statistical considerations
We calculated the sample size to achieve high power for the primary outcome (sun habit score) while also maintaining moderate power for experience of a sunburn during the intervention period, a key secondary outcome. A sample size of 314 randomized achieves 95% power to detect an effect size of 0.4 between intervention and control groups using a two-sided two-sample t-test assuming a significance level of 0.05. This sample size also achieves 70% power to detect 20% of the control participants reporting a sunburn and 10% of the intervention participants reporting a sunburn as statistically significant at a significance level of 0.05. We estimated that 85% of those randomized would complete the 12-week follow-up survey, resulting in a final calculated sample size of 368.
Participant demographic characteristics and sun behaviors reported at baseline were summarized using descriptive statistics. The main analysis of the primary outcome, total sun habit scores, was analyzed using two-sided two-sample t-test comparing scores at the end of the intervention period (12 weeks) between intervention and control groups. A secondary analysis explored changes in the total sun habit scores over the intervention period (baseline, weeks 4, 8 and 12) using a linear mixed effects regression model with fixed effects of time, group and time-group interactions, assuming that data were missing at random (MAR) [26]. We further analyzed the data at week 64. Because the groups were balanced across demographic characteristics, adjustment was unnecessary. Pre-planned subgroup analyses for gender (female, male), age (<50, 50+ years old), time since first diagnosis (0–5 years, 5+ years), stage (I, II/III, IV), and study wave (2020, 2021) were conducted for the primary outcome at week 12 by fitting linear regression models with the treatment indicator, subgroup indicator(s), and their interactions. All analyses were conducted following intention-to-treat procedures: all subjects were followed up, evaluated, and analyzed as members of their randomized group regardless of their compliance with the assigned intervention.
Comparisons of secondary outcomes—including experience of sunburn, self-report sun behaviors, skin cancer knowledge, emotional health, and physical activity—by randomization group were conducted using Chi-squared and t-tests as appropriate. Individual objective UVR data were summarized as minutes of UVR exposure per day, exposure per weekday, exposure per weekend day, and per day between 10am and 4pm (peak hours). Because the UVR data were skewed to the right with a few large exposures, median exposure levels were calculated and compared by randomization group using Wilcoxon rank sum tests and within the intervention group using Wilcoxon signed rank test. Descriptive statistics were used to summarize satisfaction and usability. Surveys had minimal missing data, and with little loss-to-follow-up, we analyzed the complete data only. All tests were two-sided. Data were analyzed using SAS 9.4 and p-values <0.05 were considered statistically significant.
Results
A total of 2,655 individuals were invited to participate; 458 provided verbal consent for their names and email to be passed to the University of Minnesota team, 372 provided written consent, and 370 completed the baseline survey and were randomized (Fig 1). Following randomization, it was determined that two participants did not have a phone compatible with the mobile app and were therefore ineligible, resulting in the planned 368 participants, 182 randomized to intervention and 186 randomized to control. In wave 1 (May-July 2020), 25% of participants were recruited and randomized (n = 92), with the remainder recruited in wave 2 (May-July 2021).
Participants were on average 56 years old (24–75 years), and most (60%) were female, highly educated, and married or partnered (Table 1). Most reported stage I disease, though approximately 1/3 were not aware of their stage. More than half were >5 years from their first melanoma diagnosis. Approximately 20% reported at least one additional (or subsequent) primary melanoma diagnosis.
Study retention was high, with 92.4% of participants (91.2% intervention, 93.5% control) completing the 12-week survey and 87.5% of participants (85.2% intervention, 89.8% controls) completing the 64-week survey. Participants at baseline reported overall high sun protection scores, with average protection scores near 3.0 indicating that participants “often” engaged in most of the protection behaviors. At 12 weeks, no statistically or clinically meaningful differences were found between groups for the primary outcome, the sun habit score (control 3.0±0.5 vs. intervention 2.9±0.5, t(337) = 1.90, p = 0.06; effect size = 0.21). Scores remained stable over the intervention period, both overall and for each separate behavior, and the repeated measures analysis showed no significant difference in overall or individual scores between the two arms during the intervention period (Table 2). At 64 weeks, the results remained similar; no differences were observed between groups for the sun habit score (control 3.2±0.5 vs. intervention 3.2±0.5, t(311) = 0.42, p = 0.67; effect size = 0.05).
Pre-planned subgroup analyses (test for interaction effects) did not reveal differences in the primary outcome (Table 3). We also stratified by baseline self-reported weekday and weekend sun exposure and did not observe differences by subgroup.
At baseline, both groups reported moderate sun exposure, with 80% reporting more than 2 hours of sun exposure on average during peak hours on the weekend and approximately 30% reporting a sunburn in the previous year. During the intervention period (assessed at week 12), 25 (14.4%) participants in the control group and 21 (12.7%) in the intervention group reported at least one sunburn (p = 0.75; Table 4). No statistically significant differences in self-reported weekday or weekend sun exposure during peak hours were observed at any time point, with the exception of fewer participants in the intervention group reporting being outside for 2 hours or more at week 64 on weekend than those in the control group (64.0% vs. 74.5%, p = 0.04).
Most participants downloaded the mobile app, used the device, and provided at least some automatically collected UVR exposure data (92.1% overall; 90.0% intervention and 94.1% controls). Crossover was minimal; four control participants downloaded and used the commercially available (intervention) mobile application; no one in the intervention group downloaded the control mobile app. The median number of days with objective UVR data was 59 days (range: 0–126) for the intervention group and 53 days (range: 0–97; p = 0.06) for the control group. Approximately 75% of participants reported wearing the device all of the time at least 5 days a week, and this was similar between study arms. The total objective UVR dose measured by the device over the intervention period was similar between groups, both during peak hours (10am-4pm) and during full days (Table 5). In addition, participants in both groups had a similar median number of minutes with any exposure per day and average objective daily dose.
In a post-hoc exploratory analysis, we examined the objective and SPF modified UVR exposure data among those in the intervention group. Most (75%) participants in the intervention group used the sunscreen feature at least once; those who did not were more likely to have stage 3 or 4 disease (24.4% vs. 8.1%, p = 0.007) but were otherwise similar in demographics. Both those who did and did not use the sunscreen feature reported similar minutes per day of exposure (median 160 vs. 151). Comparisons of the objective UVR exposure, however, found that those who reported using sunscreen within the app had greater objective UVR exposure than those who did not, which remained statistically significant even after adjusting for stage (e.g. average daily dose median 87 vs. 56 J/m2; Table 6). In contrast, the SPF modified UVR exposure dose seen by participants who reported using sunscreen was on average 25–30% less than the objective UVR exposure dose.
Concerning physical activity and emotional health, we identified no differences between the control and intervention groups (e.g. strenuous minutes per week 39.8±20.1 vs. 41.1±34.4, p = 0.70), rates of clinically relevant symptoms of anxiety (9.2% vs. 12.7%, p = 0.54) or depression (1.8% vs. 1.2%, p = 0.92) at 12 weeks, respectively (Table 7).
About half of participants reported being satisfied or very satisfied with the device (Table 8). Those in the intervention group reported paying greater attention to the device and the mobile application, and more often reported that the device affected their sun behaviors compared to the control group (all p<0.0001). While most agreed the device was easy to use, less than half reported they would like to use the system frequently and would continue using the device, even among those in the intervention group. Those assigned to the intervention group, however, were more likely to recommend this device and mobile app to a friend or relative than those in the control group (57.6% vs. 39.5%, p = 0.004). Participant comments about the device centered on concerns about durability and quality, recommendations to integrate the technology into a device they were already wearing with more features, and the need for a waterproof device.
Discussion
Contrary to our hypotheses, we found no differences in self-reported sun protection behaviors, UVR exposure, or sunburns following the 12-week intervention, including no behavioral differences in subgroup analyses. We further found no long-term behavioral differences with the one exception of slightly fewer participants in the intervention group reporting spending at least 2 hours outside on weekends at 64 weeks. Below we discuss possible reasons for our overall null findings and lessons learned for future research in this area.
We chose a previously validated summary measure of self-reported sun protection behaviors as the primary outcome for this study to enable comparisons with other similar intervention studies [22]. However, the Shade device only included recommendations for sunscreen use and no other protective measures (wearing protective clothing and/or hat, seeking shade, avoiding peak sun), which may have contributed to a lack of change in the summary measure and in these individual behaviors. In a 2021 prospective cohort study among young adults using a UV-dosimeter and messaging and goal-setting app, reductions in chest and shoulder sunburns and a trend towards fewer scalp sunburns over a 28-day period were observed, and wearing protective clothing was the most common goal reported [27]. It is possible that wearing t-shirts and hats are easily achievable sun protection measures among young cancer-free adults who may be likely to wear little protective clothing otherwise. We were unable to provide messaging about protective clothing or other sun protection behaviors. Additionally, in our study population of melanoma survivors, we recorded high levels of sun protective behaviors in the study population already at baseline (both in the intervention and control arms) with overall little room for sun protection improvements. As a self-reported outcome, social desirability to use sun protection may also have affected reported measures, leading to a ceiling effect.
A key secondary outcome for this study was self-report of at least one red or painful sunburn during the intervention period. The National Institutes of Health has deemed sunburns following an intervention to be a clinically relevant endpoint and suitable outcome for trials aiming to prevent skin cancer [28]. Although sun protection behaviors were high at baseline, 30% of participants reported a sunburn in the previous year at baseline, more than previously estimated in a similar population [7], and indicating a need for behavior change for at least some participants. Post-intervention sunburn rates were lower (<15%), which may indicate potential improvements in both groups, or we may have missed sunburns in late spring before the study.
We chose an attention-control group over no intervention so that we could objectively measure UVR exposure in both groups and increase engagement in the control group over the 12-week intervention. A 2021 study that used the Shade device and control mobile application found that unintentional tanning and time spent outside decreased among children (aged 8–17) who wore the device for a two-week period; however, no changes in UVR exposure or sunburn were observed [29]. Notably, the study also observed that sunscreen use decreased over the study period, with the authors concluding this change could dampen the detection of intervention effects. While it is possible that our control group was affected by the device, sun behavior scores were stable over the intervention period, indicating this was likely not the case. In retrospect, however, we acknowledge that a second control group who did not receive the device would have strengthened the study.
Some of the Shade device features may have contributed to our null findings. Although Shade is highly sensitive and accurate, defects in the hardware resulted in many participants needing device repair or replacement during the study. Further, the Shade built-in algorithm, which reported SPF-reduced UVR exposure to participants after sunscreen use, is potentially problematic and misleading, as evidenced by our exploratory finding that those who reported using sunscreen compared with those who did not had higher objective UVR exposure. The goal of using sunscreen is to reduce sunburn, but can mislead sunscreen users to believe they can prolong their direct sun exposure without additional protective measures. Because of this and the general lack of proper use of sunscreen (amount and reapplication), the World Health Organization recommends that other protection methods such as clothing and shade be prioritized over sunscreen for primary prevention of skin cancer [30]. Observational studies have previously suggested that rates of sunburn are higher among those who report using sunscreen [31–33]. While this may be because individuals at greater risk for sunburn are more likely to use sunscreen, our data and others’ data suggest that people may also use sunscreen to prolong otherwise unprotected time in the sun. Shade is not waterproof and therefore cannot be worn during aquatic activities. We and others have reported that individuals are often near water or swimming at the time of sunburn [31, 34], suggesting the importance of having a device capable of monitoring sun exposure during aquatic activities. Finally, we did not observe differences in the use of the device or the effect of the intervention by age group, suggesting that the technology-based intervention was not more effective in younger participants. While there are gaps in the ownership of smart phones and use of other technology between young and older adults, these results are not surprising given one, the narrowing of this gap over the last decade [35] and two, the requirement of a smart phone for study entry.
The underlying idea of the intervention was that making people aware of their UVR exposure in real-time would motivate them to reduce that exposure. While unintentional UVR exposure more frequently leads to sunburn than intentional tanning, it is possible that promoting awareness is not sufficient to change action. Barriers to sun behavior changes may be higher in Northern climates where summers are short—for example, in Minnesota, where the study was conducted. Minnesota is second only to Utah in sunburn prevalence [36] and fourth highest for UVR-attributable incidence of melanoma [37]. These high rates can likely be traced to weather-related intermittent sun exposure and greater likelihood of sunburn, which increases melanoma risk, along with outdoor-focused summer and fitness lifestyles.
Finally, behavior change is difficult. Two other studies have aimed to improve sun protection behaviors (in addition to skin self-examination) among melanoma survivors, with varying levels of success. After an interactive online intervention, Manne et al. detected small improvements in sun-protective behaviors at 24 weeks [11]. Bowen et al. reported that following a web-based family communication intervention to increase self-skin examinations and sun-protective behaviors, some behaviors changed—wearing clothing and/or sunglasses, seeking shade, avoiding peak sun hours—but not sunscreen use [14]. Notably, this study reported low engagement in the intervention.
Strengths of the current study include a strong study design with rigorous implementation (blinded randomized clinical trial), successful participant recruitment despite study implementation during the COVID-19 pandemic, high participant retention, and collection of objective UVR data. Beyond the already described flaws of the device, including the need to back-transform data, limitations included the use of self-report outcome measures, which may be prone to social desirability bias in this study population. Further, with melanoma survivors as the participants, we acknowledge that their prior melanoma diagnosis and knowledge of the risk factors may affect their sun exposure behaviors and these results may have been different among participants without a history of skin cancer. While the study was conducted remotely, allowing it to proceed largely unimpeded through the COVID-19 pandemic, it is not clear how or to what extent people’s sun exposure may have been affected by the pandemic. We have no a priori reason to think that such effects would have been differential between the study arms, but this may limit generalizability of our findings in a post-pandemic era.
Conclusion
We found no evidence that a UVR-exposure measurement intervention resulted in differences in sun protective or exposure behaviors, which may have been due, in part, to unexpected problems with the device. Based on our experience, and before further research is conducted with UVR-measuring devices similar to the one we used, the technology needs considerable improvement—for example, by integrating it into reliable devices people already use and by waterproofing the device. Additional changes to the mobile app messages are needed to promote more comprehensive sun protection behaviors, primarily sun avoidance and clothing [38], with sunscreen promotion only as a secondary measure to be used in conjunction with these more effective sun protection behaviors.
Supporting information
S1 Checklist. CONSORT 2010 checklist of information to include when reporting a randomised trial*.
https://doi.org/10.1371/journal.pone.0281480.s001
(DOC)
References
- 1.
American Cancer Society. Cancer Treatment & Survivorship Facts & Figures 2019–2021. Atlanta, GA: American Cancer Society, 2019.
- 2. Bradford PT, Freedman DM, Goldstein AM, Tucker MA. Increased risk of second primary cancers after a diagnosis of melanoma. Arch Dermatol. 2010;146(3):265–72. pmid:20231496.
- 3. Gilchrest BA, Eller MS, Geller AC, Yaar M. The pathogenesis of melanoma induced by ultraviolet radiation. The New England journal of medicine. 1999;340(17):1341–8. pmid:10219070.
- 4. Narayanan DL, Saladi RN, Fox JL. Ultraviolet radiation and skin cancer. International journal of dermatology. 2010;49(9):978–86. pmid:20883261.
- 5. Kricker A, Armstrong BK, Goumas C, Litchfield M, Begg CB, Hummer AJ, et al. Ambient UV, personal sun exposure and risk of multiple primary melanomas. Cancer causes & control: CCC. 2007;18(3):295–304. pmid:17206532.
- 6. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta psychiatrica Scandinavica. 1983;67(6):361–70. Epub 1983/06/01. pmid:6880820.
- 7. Vogel RI, Strayer LG, Engelman L, Nelson HH, Blaes AH, Anderson KE, et al. Sun Exposure and Protection Behaviors among Long-term Melanoma Survivors and Population Controls. Cancer Epidemiol Biomarkers Prev. 2017. pmid:28254810.
- 8. Mayer D, Layman A, Carlson J. Sun-protection behaviors of melanoma survivors. J Am Acad Dermatol. 2012;66(1):e9–10. pmid:22177654.
- 9. Reyes-Marcelino G, Wang R, Gultekin S, Humphreys L, Smit AK, Sharman AR, et al. School-based interventions to improve sun-safe knowledge, attitudes and behaviors in childhood and adolescence: A systematic review. Prev Med. 2021;146:106459. Epub 2021/02/21. pmid:33609617.
- 10. Mayer JA, Slymen DJ, Clapp EJ, Pichon LC, Eckhardt L, Eichenfield LF, et al. Promoting sun safety among US Postal Service letter carriers: impact of a 2-year intervention. Am J Public Health. 2007;97(3):559–65. Epub 2007/02/03. pmid:17267715.
- 11. Manne SL, Heckman CJ, Kashy DA, Ritterband LM, Thorndike FP, Lozada C, et al. Randomized controlled trial of the mySmartSkin web-based intervention to promote skin self-examination and sun protection among individuals diagnosed with melanoma. Transl Behav Med. 2021;11(7):1461–72. Epub 2021/04/28. pmid:33904921.
- 12. Robinson JK, Gaber R, Hultgren B, Eilers S, Blatt H, Stapleton J, et al. Skin self-examination education for early detection of melanoma: a randomized controlled trial of Internet, workbook, and in-person interventions. J Med Internet Res. 2014;16(1):e7. Epub 2014/01/15. pmid:24418949.
- 13. Loescher LJ, Hibler E, Hiscox H, Quale L, Harris R. An Internet-delivered video intervention for skin self-examination by patients with melanoma. Arch Dermatol. 2010;146(8):922–3. Epub 2010/08/18. pmid:20713834.
- 14. Bowen DJ, Burke W, Hay JL, Meischke H, Harris JN. Effects of web-based intervention on risk reduction behaviors in melanoma survivors. Journal of cancer survivorship: research and practice. 2015;9(2):279–86. Epub 2014/11/27. pmid:25425205.
- 15. Robinson JK, Turrisi R, Stapleton J. Efficacy of a partner assistance intervention designed to increase skin self-examination performance. Arch Dermatol. 2007;143(1):37–41. Epub 2007/01/17. pmid:17224540.
- 16. Robinson JK, Wayne JD, Martini MC, Hultgren BA, Mallett KA, Turrisi R. Early Detection of New Melanomas by Patients With Melanoma and Their Partners Using a Structured Skin Self-examination Skills Training Intervention: A Randomized Clinical Trial. JAMA dermatology. 2016;152(9):979–85. Epub 2016/07/02. pmid:27367303.
- 17. Becker MH. The Health Belief Model and personal health behavior. Health Education Monographics. 1974;2:324–508.
- 18.
Glanz K, Rimer BK, Viswanath K, editors. Health Behavior: Theory, Research, and Practice. 5th ed. San Francisco CA: Jossey-Bass Public Health; 2015.
- 19. Vogel RI, Nagler RH, Ahmed RL, Brown K, Luo X, Martinson BC, et al. UVR-sensor wearable device intervention to improve sun behaviors and reduce sunburns in melanoma survivors: study protocol of a parallel-group randomized controlled trial. Trials. 2020;21(1):959. Epub 2020/11/25. pmid:33228807.
- 20. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–81. Epub 20080930. pmid:18929686.
- 21.
Banerjee S, Hoch EG, Kaplan PD, Dumont EL, editors. A comparative study of wearable ultraviolet radiometers. 2017 IEEE Life Sciences Conference (LSC); 2017: IEEE.
- 22. Glanz K, Schoenfeld ER, Steffen A. A randomized trial of tailored skin cancer prevention messages for adults: Project SCAPE. Am J Public Health. 2010;100(4):735–41. Epub 2010/02/20. pmid:20167900.
- 23. Glanz K, Yaroch AL, Dancel M, Saraiya M, Crane LA, Buller DB, et al. Measures of sun exposure and sun protection practices for behavioral and epidemiologic research. Arch Dermatol. 2008;144(2):217–22. Epub 2008/02/20. pmid:18283179.
- 24. Godin G, Shephard RJ. A simple method to assess exercise behavior in the community. Canadian journal of applied sport sciences Journal canadien des sciences appliquees au sport. 1985;10(3):141–6. pmid:4053261.
- 25.
Brooke JB. SUS—a quick and dirty usability scale. In: Jordan P, Thomas B, Weerdmeester B, McLelland I, editors. Usability Evaluation in Industry. London: Taylor and Francis; 1996.
- 26.
Fitzmaurice GM, Laird NM, Ware JH. Applied Longitudinal Analysis. Hoboken, NH: John Wiley & Sons; 2004.
- 27. Robinson JK, Patel S, Heo SY, Gray E, Lim J, Kwon K, et al. Real-Time UV Measurement With a Sun Protection System for Warning Young Adults About Sunburn: Prospective Cohort Study. JMIR Mhealth Uhealth. 2021;9(5):e25895. Epub 20210506. pmid:33955844.
- 28. Perna FM, Dwyer LA, Tesauro G, Taber JM, Norton WE, Hartman AM, et al. Research on Skin Cancer-Related Behaviors and Outcomes in the NIH Grant Portfolio, 2000–2014: Skin Cancer Intervention Across the Cancer Control Continuum (SCI-3C). JAMA dermatology. 2017;153(5):398–405. Epub 2017/03/23. pmid:28329179.
- 29. Parsons BG, Nagelhout ES, Wankier AP, Hu N, Lensink R, Zhu A, et al. Reactivity to UV Radiation Exposure Monitoring Using Personal Exposure Devices for Skin Cancer Prevention: Longitudinal Observational Study. JMIR Mhealth Uhealth. 2021;9(9):e29694. Epub 2021/09/29. pmid:34581683.
- 30.
World Health Organization. Radiation: Protecting against skin cancer: World Health Organization; 2017 [cited 2022 06/11/2022]. https://www.who.int/news-room/questions-and-answers/item/radiation-protecting-against-skin-cancer.
- 31. McLeod GFH, Reeder AI, Gray AR, McGee R. Unintended Sunburn: A Potential Target for Sun Protection Messages. Journal of skin cancer. 2017;2017:6902942. Epub 2017/05/06. pmid:28473925.
- 32. Reinau D, Achermann C, Arnet N, Meier CR, Hatz C, Surber C. Sun protective behaviour of vacationers spending holidays in the tropics and subtropics. The British journal of dermatology. 2014;171(4):868–74. Epub 2014/03/29. pmid:24665976.
- 33. Autier P. Sunscreen abuse for intentional sun exposure. The British journal of dermatology. 2009;161 Suppl 3:40–5. Epub 2009/09/25. pmid:19775356.
- 34. Jewett PI, Henning-Smith C, Lazovich D, Ahmed RL, Vogel RI. Incidental sun exposures as a source of sunburn among rural compared to urban residents in the United States. J Rural Health. 2022. Epub 20220919. pmid:36123966.
- 35.
Faverio M. Share of those 65 and older who are tech users has grown in the past decade: Pew Research Center; 2022 [cited 2022 12/09/2022]. https://www.pewresearch.org/fact-tank/2022/01/13/share-of-those-65-and-older-who-are-tech-users-has-grown-in-the-past-decade/.
- 36. Centers for Disease Control and Prevention. Sunburn Prevalence Among Adults—United States, 1999, 2003, and 2004. MMWR Surveill Summ [Internet]. 2007; 56(21):[524–8 pp.]. Available from: https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5621a2.htm. pmid:17538527
- 37. Islami F, Sauer AG, Miller KD, Fedewa SA, Minihan AK, Geller AC, et al. Cutaneous melanomas attributable to ultraviolet radiation exposure by state. Int J Cancer. 2020;147(5):1385–90. Epub 2020/02/18. pmid:32064604.
- 38. Lazovich D, Vogel RI, Berwick M, Weinstock MA, Warshaw EM, Anderson KE. Melanoma risk in relation to use of sunscreen or other sun protection methods. Cancer Epidemiol Biomarkers Prev. 2011;20(12):2583–93. pmid:22016471.