Consumer-Based Wearable Activity Trackers Increase Physical Activity Participation: Systematic Review and Meta-Analysis

Abstract

Background: The range of benefits associated with regular physical activity participation is irrefutable. Despite the well-known benefits, physical inactivity remains one of the major contributing factors to ill-health throughout industrialized countries. Traditional lifestyle interventions such as group education or telephone counseling are effective at increasing physical activity participation; however, physical activity levels tend to decline over time. Consumer-based wearable activity trackers that allow users to objectively monitor activity levels are now widely available and may offer an alternative method for assisting individuals to remain physically active.

Objective: This review aimed to determine the effects of interventions utilizing consumer-based wearable activity trackers on physical activity participation and sedentary behavior when compared with interventions that do not utilize activity tracker feedback.

Methods: A systematic review was performed searching the following databases for studies that included the use of a consumer-based wearable activity tracker to improve physical activity participation: Cochrane Controlled Register of Trials, MEDLINE, PubMed, Scopus, Web of Science, Cumulative Index of Nursing and Allied Health Literature, SPORTDiscus, and Health Technology Assessments. Controlled trials of adults comparing the use of a consumer-based wearable activity tracker with other nonactivity tracker-based interventions were included. The main outcome measures were physical activity participation and sedentary behavior. All studies were assessed for risk of bias, and the Grades of Recommendation, Assessment, Development, and Evaluation system was used to rank the quality of evidence. The guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement were followed. A random-effects meta-analysis was completed on the included outcome measures to estimate the treatment effect of interventions that included an activity tracker compared with a control group.

Results: There was a significant increase in daily step count (standardized mean difference [SMD] 0.24; 95% CI 0.16 to 0.33; P<.001), moderate and vigorous physical activity (SMD 0.27; 95% CI 0.15 to 0.39; P<.001), and energy expenditure (SMD 0.28; 95% CI 0.03 to 0.54; P=.03) and a nonsignificant decrease in sedentary behavior (SMD -0.20; 95% CI -0.43 to 0.03; P=.08) following the intervention versus control comparator across all studies in the meta-analyses. In general, included studies were at low risk of bias, except for performance bias. Heterogeneity varied across the included meta-analyses ranging from low (I²=3%) for daily step count through to high (I²=67%) for sedentary behavior.

Conclusions: Utilizing a consumer-based wearable activity tracker as either the primary component of an intervention or as part of a broader physical activity intervention has the potential to increase physical activity participation. As the effects of physical activity interventions are often short term, the inclusion of a consumer-based wearable activity tracker may provide an effective tool to assist health professionals to provide ongoing monitoring and support.

Keywords: exercise; fitness trackers; meta-analysis; telemedicine.

Figure 1

Flowchart of study selection. Studies with results published and authors that provided unpublished results were included in the systematic review.

Figure 2

Risk of bias summary: review authors’ judgments about each risk of bias item for each included study. Green symbols represent a low risk of bias, yellow symbols represent an unclear risk of bias, and red symbols represent a high risk of bias.

Figure 3

Forest plot of standardized mean difference of steps per day in studies comparing an intervention that included a consumer-based wearable activity tracker with a control group that did not utilize a consumer-based wearable activity tracker. Subgroup analysis was completed on studies that included wearable-based interventions compared with control and multifaceted interventions compared with control. Green square indicates the standardized mean difference for each individual study. Black square indicates the overall standardized mean difference for all studies.

Figure 4

Forest plot of standardized mean difference of time spent in moderate and vigorous physical activity per week in studies comparing an intervention that included a consumer-based wearable activity tracker with a control group that did not utilize a consumer-based wearable activity tracker. Subgroup analysis was completed on studies that included wearable-based interventions compared with control and multifaceted interventions compared with control. Green square indicates the standardized mean difference for each individual study. Black square indicates the overall standardized mean difference for all studies.

Figure 5

Forest plot of standardized mean difference of energy expenditure in studies comparing an intervention that included a consumer-based wearable activity tracker with a control group that did not utilize a consumer-based wearable activity tracker. Green square indicates the standardized mean difference for each individual study. Black square indicates the overall standardized mean difference for all studies.

Figure 6

Forest plot of standardized mean difference of time spent in sedentary behaviors in studies comparing an intervention that included a consumer-based wearable activity tracker with a control group that did not utilize a consumer-based wearable activity tracker. Subgroup analysis was completed on studies that included wearable-based interventions compared with control and multifaceted interventions compared with control. Green square indicates the standardized mean difference for each individual study. Black square indicates the overall standardized mean difference for all studies.