JMIR Publications

ABSTRACT

Background:

Just-in-time adaptive interventions (JITAIs) are developed to provide support when it is most needed and when users are most receptive. However, the notion of a just-in-time (JIT) state has remained mostly theoretical; there has been little work to gather empirical evidence for what makes a state “just-in-time.” The Just Walk JITAI project uses system identification methods from engineering to systematically vary provision of support across time and context to empirically assess how need, opportunity, and receptivity—three key aspects of the theoretical formulation of JIT states—influence the effectiveness of intervention provision.

Objective:

The purpose of this research is to investigate JIT states empirically and to enable the empirical optimization of a JITAI intended to increase physical activity (steps/day).

Methods:

We recruited English-speaking adults aged 25+ who are physically inactive and own smartphones. Participants wore a Fitbit Versa 3 and used the study app for 270 days. Building on our prior work, including our previously developed mobile health apps (Just Walk and HeartSteps), we conducted a system identification experiment to investigate and optimize two intervention components: 1) notifications delivered up to 4 times per day designed to increase a person’s steps within the next 3 hours via either increased awareness of the urge to walk or via bout planning; and 2) adaptive daily step goals. Specifically, notifications to walk within the next 3 hours were experimentally provided (or not) across varied operationalizations of JIT states accounting for: need (i.e., whether daily step goals were previously met or not), opportunity (i.e., if the next three hours is a time window when a person previously walked based on a previously published machine learning algorithm), and receptivity (i.e., the person received 6 or fewer messages in the last 72 hours or previously walked after receiving a recent notification). The second intervention component is daily steps goals. Suggested daily step goals varied systematically within a range referenced to a person’s baseline level of steps/day (e.g., 4,000 steps) until they meet clinically meaningful targets (e.g., averaging 8,000 steps/day across a cycle, as the lower threshold).

Results:

The study began enrolling participants in June 2022, with a final enrollment of 48 participants. Data collection concluded in April 2023. Upon completion of the analyses, the results of this study are expected to be submitted for publication in Fall/Winter 2023.

Conclusions:

This study will be the first empirical investigation of JIT states that use system identification methods to inform optimization of a scalable JITAI for physical activity. Clinical Trial: ClinicalTrials.gov NCT05273437