JMIR Publications

ABSTRACT

Background:

Smartphone apps and wearable activity trackers are expected to play an important role in promoting physical activity (PA). Although studies suggest that use of commercial mobile health tools is associated with increased PA levels, most of the evidence is of a cross-sectional design, and thus longitudinal evidence is lacking.

Objective:

Therefore, we aimed to reveal the app-use patterns that are prospectively associated with increases in and maintenance of PA. We were specifically interested in (a) whether continued app use is associated with adherence to the recommended levels of PA (i.e., 23 METs-h/w for adults or 10 METs-h/w for people aged>65 years) during a follow-up assessment and (b) whether any functions and features of PA apps predict the changes in PA level.

Methods:

A two-wave longitudinal survey was conducted with the baseline and follow-up assessments separated by six months. A total of 20,573 Japanese-speaking online respondents participated in the baseline, among which 16,286 (8,289 women; mean age=54.7, standard deviation=16.8) completed the follow-up. On each assessment occasion, participants reported their current PA levels and whether they were using any PA apps and wearables. Each participant was then classified into following four categories: continued users (those who were using apps at both the baseline and follow-up; n=2,150, 13.2%), new users (those who started using apps before the follow-up; n=1,462, 9.0%), discontinued users (those who had been using apps at the baseline but not anymore at the follow-up; n=1,899, 11.7%), and continued nonusers (those who had never used apps; n=10,775, 66.2%).

Results:

Most continued users (n=1,538, 71.5%) improved or maintained their PA to the recommended levels over six months; however, discontinued users showed the largest reduction in PA (-7.95 METs-h/w on average), and most failed to meet the recommended levels at the follow-up. Analyses of individual app functions showed that energy analysis (e.g., app calculation of daily energy expenditure) and journaling (e.g., users manually entering notes and maintaining an exercise diary) were significantly associated with increases in PA (odds ratio [OR]=1.67, 95% confidence interval [CI] [1.05, 2.64] and OR=1.76, 95%CI [1.12, 2.76], respectively). On the other hand, those who maintained the recommended PA levels at the follow-up typically used the goal setting (OR=1.73, 95%CI [1.21, 2.48]), sleep information (OR=1.66, 95%CI [1.03, 2.68]), and blood-pressure recording (OR=2.05, 95%CI [1.10, 3.83]) functions.

Conclusions:

The results support the importance of continued app use in increasing and maintaining PA levels. Different app functions may be suited to increasing PA (e.g., goal setting and journaling) vs. maintaining high levels of PA (their health in general, covering sleep and blood pressure, not limited to PA).