JMIR Publications

ABSTRACT

Background:

Mobile health applications can serve as a critical tool in supporting the overall health of un- and underinsured individuals and groups who have been historically marginalized by the medical community and may be hesitant to seek formal healthcare. However, data on uptake and engagement with distinct app features are often lacking, limiting our ability to understand what is used by whom and how often.

Objective:

We aimed to characterize socio-demographic differences in uptake and engagement with a smartphone application (uMAT-R) to support substance misuse recovery in a sample of individuals with substance use disorders (SUDs) in the Greater St. Louis area.

Methods:

We enrolled individuals into the uMAT-R service program from facilities providing recovery support for individuals with SUDs in the Greater St. Louis area between January 2020-April 2022. Study participants were recruited from service project enrollees. We describe the number of logins and eCoach messages sent in the first 30 days following enrollment using medians and interquartile ranges (IQRs) and counts and proportions of those who ever (vs. never) logged in and sent an eCoach message. We compared these estimates across socio-demographic sub-groups and for those who did and did not participate in the study using Wilcoxon rank-sum tests and Pearson’s chi-squared tests.

Results:

Of 665 included individuals, 437 (65.7%) logged into uMAT-R at least once during the 30 days following enrollment (median logins: 2; IQR: 0-8). Approximately half of those who ever logged in (51.3%) used the eCoach messaging feature (median messages: 1; IQR: 0-3). Individuals who participated in the research component of the project (n=498) were more likely to log in and use the eCoach messaging feature compared to those who did not (n=167). Younger individuals, those with higher formal education, and White, non-Hispanic individuals were more likely to log in compared to their counterparts. Median number of logins was higher among females, younger individuals, employed individuals and those who were not on Medicaid compared to their counterparts. Among those who logged in at least once, younger individuals and those with lower formal education were more likely to send at least one eCoach message compared to their counterparts.

Conclusions:

Mobile applications are a viable tool for supporting individuals in recovery for SUDs. However, older individuals, racial/ethnic minorities, and those with less formal education may need additional login support, or benefit from alternative mechanisms of SUD recovery support. Moreover, apps may need to be tailored to achieve sustained engagement among men, and individuals who are older, unemployed, or on Medicaid. Older individuals and those with higher formal education who may be less likely to use eCoach messaging features could benefit from features tailored more specifically to their preferences.