Reactivity to UV Radiation Exposure Monitoring Using Personal Exposure Devices for Skin Cancer Prevention: Longitudinal Observational Study

Objectively-Assessed Ultraviolet Radiation Exposure and Sunburn Occurrence

Ultraviolet radiation (UVR) exposure is the primary modifiable risk factor for melanoma. Wearable UVR sensors provide a means of quantifying UVR exposure objectively and with a lower burden than self-report measures used in most research. The purpose of this study was to evaluate the relationship between detected UVR exposure and reported sunburn occurrence. In this study, a UVR monitoring device was worn by 97 parent-child dyads during waking hours for 14 days to measure instantaneous and accumulated UVR exposure. The results showed that the participants' total UVR exposure was associated with reported sunburn after adjusting for Fitzpatrick skin type and geographic location. It was observed that one standard erythemal dose (SED) increase in the participants' daily total UVR exposure was associated with reported sunburn (an odds ratio (OR) of 1.26 with a 95% CI of 1.13 and 1.41, and p < 0.001 for parents and an OR of 1.28 with a 95% CI of 1.12 and 1.47, and p < 0.001 for children). A one-SED increase in the participants' UVR exposure from 10 am to 4 pm was also associated with reported sunburn (an OR of 1.31 with a 95% CI of 1.15 and 1.49, and p < 0.001 for parents and an OR of 1.33 with a 95% CI of 1.12 and 1.59, and p = 0.001 for children). We found that elevated UVR exposure recordings measured by the UVR sensor were associated with reported sunburn occurrence. Future directions for wearable UVR sensors may include their use as an intervention tool to support in-the-moment sunburn prevention.

A UVR-sensor wearable device intervention to reduce sun exposure in melanoma survivors: Results from a randomized controlled trial

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).

CONCLUSION

Wearing a device that measured and alerted melanoma survivors to UVR exposure did not result in different sun behaviors, exposure, or sunburns relative to controls. The technology needs refinement before further attempts to assess the effectiveness of self-monitoring UVR exposure.

CLINICAL TRIALS REGISTRATION

NCT03927742.

Digital Phenotyping in Health Using Machine Learning Approaches: Scoping Review

BACKGROUND

Digital phenotyping is the real-time collection of individual-level active and passive data from users in naturalistic and free-living settings via personal digital devices, such as mobile phones and wearable devices. Given the novelty of research in this field, there is heterogeneity in the clinical use cases, types of data collected, modes of data collection, data analysis methods, and outcomes measured.

OBJECTIVE

The primary aim of this scoping review was to map the published research on digital phenotyping and to outline study characteristics, data collection and analysis methods, machine learning approaches, and future implications.

METHODS

We utilized an a priori approach for the literature search and data extraction and charting process, guided by the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-analyses Extension for Scoping Reviews). We identified relevant studies published in 2020, 2021, and 2022 on PubMed and Google Scholar using search terms related to digital phenotyping. The titles, abstracts, and keywords were screened during the first stage of the screening process, and the second stage involved screening the full texts of the shortlisted articles. We extracted and charted the descriptive characteristics of the final studies, which were countries of origin, study design, clinical areas, active and/or passive data collected, modes of data collection, data analysis approaches, and limitations.

RESULTS

A total of 454 articles on PubMed and Google Scholar were identified through search terms associated with digital phenotyping, and 46 articles were deemed eligible for inclusion in this scoping review. Most studies evaluated wearable data and originated from North America. The most dominant study design was observational, followed by randomized trials, and most studies focused on psychiatric disorders, mental health disorders, and neurological diseases. A total of 7 studies used machine learning approaches for data analysis, with random forest, logistic regression, and support vector machines being the most common.

CONCLUSIONS

Our review provides foundational as well as application-oriented approaches toward digital phenotyping in health. Future work should focus on more prospective, longitudinal studies that include larger data sets from diverse populations, address privacy and ethical concerns around data collection from consumer technologies, and build "digital phenotypes" to personalize digital health interventions and treatment plans.