Presentation
DH1 - A Meta-Analysis Examining the Effect of Personalization on Fitness Apps
SessionPoster Session 1
DescriptionBackground: There is an abundance of health apps on the market, but the question remains whether or not a subcategory of these apps, fitness apps specifically, are effective in increasing physical activity (PA) in users. Some applications utilize personalization features, such as goal setting and exercise selection, to encourage physical activity and, subsequently, adherence to the app. Additionally, previous work has attempted to evaluate the role of social features on physical activity, as social comparison and social support may moderate physical activity in fitness app users. Prior randomized controlled trials (RCTs) have explored such effects within specific apps, yet findings are mixed.
Objective: The aim of this study was to provide a comprehensive understanding of whether fitness apps with personalization features result in increased physical activity in comparison to apps that lack such features. The secondary aim of this study was to evaluate the interaction between personalization and social features on physical activity levels in fitness app users.
Methods: A meta-analysis of studies that examined the effect of fitness apps, specifically the presence or absence of personalization features and social features, on physical activity was performed. A total of 4,356 articles were identified across five databases. After screening for eligibility, 20 studies (n= 2,159 participants) remained for data extraction and analysis. From these 20 studies, 31 comparisons were run between postintervention differences between the fitness app intervention group and the control group that were included in this meta-analysis.
Results: A random effects model was performed to examine the effect of fitness apps on physical activity levels compared to controls. The number of personalization features, social aspects and the intervention duration was included as a moderator. Results indicated a small to medium effect size across all fitness apps in which the fitness app intervention effectively increased physical activity levels compared to controls (hedge’s g= 0.5033, p=<.001). Analyses for the effect of personalization and social features are currently being conducted.
Conclusions: The results show that both fitness apps with and without personalization features were effective in increasing physical activity. These findings have valuable implications for increasing physical activity in U.S. adults, and subsequently decreasing the risk of NCDs.
Objective: The aim of this study was to provide a comprehensive understanding of whether fitness apps with personalization features result in increased physical activity in comparison to apps that lack such features. The secondary aim of this study was to evaluate the interaction between personalization and social features on physical activity levels in fitness app users.
Methods: A meta-analysis of studies that examined the effect of fitness apps, specifically the presence or absence of personalization features and social features, on physical activity was performed. A total of 4,356 articles were identified across five databases. After screening for eligibility, 20 studies (n= 2,159 participants) remained for data extraction and analysis. From these 20 studies, 31 comparisons were run between postintervention differences between the fitness app intervention group and the control group that were included in this meta-analysis.
Results: A random effects model was performed to examine the effect of fitness apps on physical activity levels compared to controls. The number of personalization features, social aspects and the intervention duration was included as a moderator. Results indicated a small to medium effect size across all fitness apps in which the fitness app intervention effectively increased physical activity levels compared to controls (hedge’s g= 0.5033, p=<.001). Analyses for the effect of personalization and social features are currently being conducted.
Conclusions: The results show that both fitness apps with and without personalization features were effective in increasing physical activity. These findings have valuable implications for increasing physical activity in U.S. adults, and subsequently decreasing the risk of NCDs.
Event Type
Poster Presentation
TimeMonday, March 314:45pm - 6:15pm EDT
LocationFrontenac Foyer
Digital Health (DH)
Simulation and Education (SE)
Hospital Environments (HE)
Medical and Drug Delivery Devices (MDD)
Patient Safety and Research Initiatives (PS)