JMIR Publications

ABSTRACT

Background:

Young adults (YAs) engage in behaviors that place them at risk for skin cancer. Dissemination of digital health promotion interventions via social media are potentially promising strategies to modify skin cancer risk behaviors by increasing ultraviolet radiation (UVR) protection and skin cancer examinations.

Objective:

The goal of the study was to compare three digital interventions designed to modify UVR exposure, sun protection, and skin cancer detection behaviors among YAs at moderate to high risk of skin cancer.

Methods:

This study was a Hybrid Type II Effectiveness-Implementation randomized controlled trial of two active interventions, a digital skin cancer risk reduction intervention (UV4.Me [Basic]) compared to an enhanced version (UV4.Me2 [Enhanced]), and an e-pamphlet. Intervention effects were assessed over the course of a year among 1369 US YAs recruited primarily via Facebook and Instagram. Enhancements to encourage intervention engagement and behavior change included more comprehensive goal-setting activities, ongoing proactive messaging related to previously established mediators (e.g., self-efficacy) of UVR exposure and protection, embedded incentives for module completion, and ongoing news and video updates. Primary outcome effects assessed via linear regression were UVR exposure and sun protection and protection habits. Secondary outcome effects assessed via logistic regression were skin self-exams (SSE), physician skin-exams, sunscreen use, indoor tanning, and sunburn. Intervention costs were also explored.

Results:

The active interventions increased sun protection (Basic P=.02, Enhanced P<.001) and protection habits (Basic P=.04, Enhanced P=.01) compared to the e-pamphlet. The Enhanced Intervention increased sun protection more than the Basic one. Each active intervention increased sunscreen use at three-month follow-up (Basic P=0.03, Enhanced P=.01) and skin self-exam at one year (Basic P=0.4, Enhanced P=.004), compared to the e-pamphlet. Other intervention effects were non-significant. After initial development, the active interventions were similar in costs per person at larger sample sizes. They also incurred reasonable costs to achieve an additional SSE, compared with the e-pamphlet.

Conclusions:

The active interventions were effective in improving several skin cancer risk and skin cancer prevention behaviors. Compared with the Basic Intervention, the Enhanced Intervention added to the improvement in sun protection, but not other behaviors. Although all interventions were costly to create, they were less costly to maintain. Future analyses will explore intervention engagement (e.g., proportion of content reviewed) and focus on interventions that are less costly to build. Clinical Trial: ClinicalTrials.gov ID NCT03313492