Use Patterns of Smartphone Apps and Wearable Devices Supporting Physical Activity and Exercise: a large-scale cross-sectional survey (Preprint)

Abstract

Physical inactivity is a global health issue, and mobile health (mHealth) applications (apps) are expected to play an important role in promoting physical activity. Empirical studies have demonstrated the efficacy and efficiency of app-based interventions, and an increasing number of apps with more functions and richer content have been released in markets. Regardless of the success of mHealth apps, there are important evidence gaps in the literature; that is, it is largely unknown who uses what app functions and which functions are associated with physical activity

To investigate the usage patterns of apps and wearables supporting physical activity and exercise in a Japanese-speaking community sample.

We recruited N = 20,573 online panels who completed questionnaires concerning demographics, regular physical activity levels, and use of apps and wearables supporting physical activity. Participants who indicated that they were using a PA app or wearable were presented with a list of app functions (e.g., sensor information, goal setting, journaling, and reward), among which they selected any functions in use.

Approximately a quarter of the sample (n = 4,465) were identified as app users who showed similar demographic characteristics documented in the literature; that is, compared with non-app users, app users are younger (OR = 0.64, 95%CI [0.56, 0.73]), more likely to be men (OR = 0.83, 95%CI [0.76, 0.89]), have larger BMI scores (OR = 1.02, 95%CI [1.01, 1.03]), have higher levels of education (university or above; OR = 1.49, 95%CI [1.16, 1.94]), more likely to have a child (OR = 1.15, 95%CI [1.04, 1.26]) and job (OR = 1.26, 95%CI [1.16, 1.38]), and have a higher household income (OR = 1.40, 95%CI [1.21, 1.61]). Our results revealed unique associations between demographic variables and specific app functions. For example, sensor information, journaling, and GPS were more frequently used by men than women (ORs < 0.84); reward was more preferred by women (OR = 1.26, 95%CI [1.02, 1.56]). Another important finding is that people typically use two different functions within an app (IQR: 1-4 functions), and the most common pattern was to use sensor information (i.e., self-monitoring) and one other function such as goal setting or reminders.

Regardless of the current trend of app development toward multifunctionality, our findings highlight the importance of app simplicity. A set of two functions (more precisely, self-monitoring and one other function) might be the minimum that can be accepted by most users. In addition, the identified individual differences will help developers and stakeholders pave the way for the personalization of app functions.