Applying Sequence Analysis to Explore Real-World Usage Patterns in the 10,000 Steps Digital Physical Activity Program

Digital health programs can enhance physical activity, but engagement duration influences behavioral impact. We examined general medium/long-term engagement and disengagement patterns of the 10,000 Steps program.

Sequence analysis identified patterns of program use over 7 years (2018-2024) of adults from Queensland, Australia. Subgroup patterns of program engagement were identified. Logistic regression identified demographic and program correlates of subgroup membership.

Most users showed short-term program engagement. Two user groups engaged over a longer period compared with most users. One group (7.1%) showed a pattern of program use between periods of inactivity. These users were more likely to have heard about the program through health professionals/programs (odds ratio [OR] = 4.59; 95% CI, 3.49-6.06) or family/friends (OR = 3.88; 95% CI, 2.94-5.12), and participated in Tournaments (OR = 1.37; 95% CI, 1.30-1.45), or used Fitbit devices (OR = 1.73; 95% CI, 1.42-2.11) or the website (OR = 2.25; 95% CI, 1.82-2.78). A small group consistently used the program with little inactivity (4.5%), characterized by participating in more individual Challenges (OR = 1.32; 95% CI, 1.26-1.38) and Tournaments (OR = 1.76; 95% CI, 1.59-1.94), and using Garmin devices (OR = 319.37; 95% CI, 219.20-465.31) and the Google app (OR = 6.37; 95% CI, 2.50-16.21).

These findings reveal that medium/long-term engagement with the 10,000 Steps program is characterized by varying levels of engagement over time, depending on program feature engagement and device use.