JMIR Publications

ABSTRACT

Background:

Digitally phenotyping individuals most likely to be non-adherent to personal physical goals could have significant value in the global battle against obesity. Sedentary behavior is increasingly linked to premature death in the developed world. The annual impact of physical inactivity and sedentary behaviors is significant, causing an estimated 2 million deaths and a leading causes of mortality and morbidity. Nevertheless, each year considerable funding and countless public health initiatives promote physical fitness with little impact on sustained behavioral change. Predictive models developed from multimodal methodologies combining data from decision-making tasks with contextual insights and objective physical activity data can be used to identify those most likely to abandon their fitness goals. This information has the potential to develop more targeted initatives to ensure indiviuals fitness goals are achieved.

Objective:

This research aims to determine the feasibility of using predictive models built from computational models of decision-making to identify those most likely to abandon their fitness goals. Objective behavioral data from the participant's fitness tracker will be combined with data captured from assessments delivered via a mobile application (decision-making tasks such as the Iowa Gambling Task (IGT) and personality traits questionnaires) will be used to ascertain if a predictive algorithm can identify digital personae's most likely to be non-adherent to self-determine exercise goals. Furthermore, if it is possible to phenotype these individuals, it may be possible to tailor initiatives to support them to stay the course.

Methods:

This is a longitudinal observational study set in the real world. Combining objective sensor data with decision-making games and contextual personality traits to identify patterns in exercise decay. The data generated will build computational models to predict digital personas, most likely to abandon exercise goals. This study design is entirely virtual and based on a "Bring your own device" (BYOD) model where participants use their own fitness trackers and download the study app to their smartphones. Two hundred healthy adults who are novice exercisers and own a Fitbit will be recruited via social media. Participants will e-consent via the study app, download from the Google/Apple play store. They will also consent to share their Fitbit data. Necessary demographic information concerning age and gender will be collected as part of the recruitment process. Over 12 months, scheduled study assessments will be pushed to the participants to complete. The IGT will be administered via a web application.

Results:

Ethics approval was received in December 2020 from Dublin City University. At manuscript submission, study recruitment is pending. Expected results will be published in 2022.

Conclusions:

It is hoped the study results will support the development of a predictive model and the study design inform future research approaches. Clinical Trial: Registration: Clinical Trials.Gov: NCT04783298, https://clinicaltrials.gov/ct2/show/NCT04783298