Limits...
Engagement and Nonusage Attrition With a Free Physical Activity Promotion Program: The Case of 10,000 Steps Australia.

Guertler D, Vandelanotte C, Kirwan M, Duncan MJ - J. Med. Internet Res. (2015)

Bottom Line: Cox regression showed that user group predicted nonusage attrition: Web-and-app users (hazard ratio=0.86, 95% CI 0.81-0.93, P<.001) and app-only users (hazard ratio=0.63, 95% CI 0.58-0.68, P<.001) showed a reduced attrition risk compared to Web-only users.Further, having a higher number of individual challenges (hazard ratio=0.62, 95% CI 0.59-0.66, P<.001), workplace challenges (hazard ratio=0.94, 95% CI 0.90-0.97, P<.001), physical activity logging days (hazard ratio=0.921, 95% CI 0.919-0.922, P<.001), and steps logged per day (hazard ratio=0.99999, 95% CI 0.99998-0.99999, P<.001) were associated with reduced nonusage attrition risk as well as older age (hazard ratio=0.992, 95% CI 0.991-0.994, P<.001), being male (hazard ratio=0.85, 95% CI 0.82-0.89, P<.001), and being non-Australian (hazard ratio=0.87, 95% CI 0.82-0.91, P<.001).Better understanding of participant reasons for reducing engagement can assist in clarifying how to best address this issue to maximize behavior change.

View Article: PubMed Central - HTML - PubMed

Affiliation: Institute of Social Medicine and Prevention, University Medicine, Greifswald, Germany. diana.guertler@uni-greifswald.de.

ABSTRACT

Background: Data from controlled trials indicate that Web-based interventions generally suffer from low engagement and high attrition. This is important because the level of exposure to intervention content is linked to intervention effectiveness. However, data from real-life Web-based behavior change interventions are scarce, especially when looking at physical activity promotion.

Objective: The aims of this study were to (1) examine the engagement with the freely available physical activity promotion program 10,000 Steps, (2) examine how the use of a smartphone app may be helpful in increasing engagement with the intervention and in decreasing nonusage attrition, and (3) identify sociodemographic- and engagement-related determinants of nonusage attrition.

Methods: Users (N=16,948) were grouped based on which platform (website, app) they logged their physical activity: Web only, app only, or Web and app. Groups were compared on sociodemographics and engagement parameters (duration of usage, number of individual and workplace challenges started, and number of physical activity log days) using ANOVA and chi-square tests. For a subsample of users that had been members for at least 3 months (n=11,651), Kaplan-Meier survival curves were estimated to plot attrition over the first 3 months after registration. A Cox regression model was used to determine predictors of nonusage attrition.

Results: In the overall sample, user groups differed significantly in all sociodemographics and engagement parameters. Engagement with the program was highest for Web-and-app users. In the subsample, 50.00% (5826/11,651) of users stopped logging physical activity through the program after 30 days. Cox regression showed that user group predicted nonusage attrition: Web-and-app users (hazard ratio=0.86, 95% CI 0.81-0.93, P<.001) and app-only users (hazard ratio=0.63, 95% CI 0.58-0.68, P<.001) showed a reduced attrition risk compared to Web-only users. Further, having a higher number of individual challenges (hazard ratio=0.62, 95% CI 0.59-0.66, P<.001), workplace challenges (hazard ratio=0.94, 95% CI 0.90-0.97, P<.001), physical activity logging days (hazard ratio=0.921, 95% CI 0.919-0.922, P<.001), and steps logged per day (hazard ratio=0.99999, 95% CI 0.99998-0.99999, P<.001) were associated with reduced nonusage attrition risk as well as older age (hazard ratio=0.992, 95% CI 0.991-0.994, P<.001), being male (hazard ratio=0.85, 95% CI 0.82-0.89, P<.001), and being non-Australian (hazard ratio=0.87, 95% CI 0.82-0.91, P<.001).

Conclusions: Compared to other freely accessible Web-based health behavior interventions, the 10,000 Steps program showed high engagement. The use of an app alone or in addition to the website can enhance program engagement and reduce risk of attrition. Better understanding of participant reasons for reducing engagement can assist in clarifying how to best address this issue to maximize behavior change.

No MeSH data available.


Related in: MedlinePlus

Nonusage attrition curves for user groups in the subsample of users who were 10,000 Steps members for at least 3 months (n=11,651).
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4526999&req=5

figure1: Nonusage attrition curves for user groups in the subsample of users who were 10,000 Steps members for at least 3 months (n=11,651).

Mentions: The following results are based on a subsample only including users that had been a 10,000 Steps member for at least 3 months (n=11,651). Figure 1 presents Kaplan-Meier survival curves for the different user groups based on the duration of usage. The log-rank test showed that the survivor functions were significantly different across groups (χ22=161.3, P<.001). Estimated median lifetime usage (time after which 50% stopped logging physical activity) was 30 days for all groups combined (Table 4). For all groups combined, 25.00% (2913/11,651) were still logging steps after 42 days. This was similar to the Web-only and app-only groups, with 41 and 43 days. respectively; however, in the Web-and-app group, 25.0% (220/878) of the sample were still logging steps after 56 days.


Engagement and Nonusage Attrition With a Free Physical Activity Promotion Program: The Case of 10,000 Steps Australia.

Guertler D, Vandelanotte C, Kirwan M, Duncan MJ - J. Med. Internet Res. (2015)

Nonusage attrition curves for user groups in the subsample of users who were 10,000 Steps members for at least 3 months (n=11,651).
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4526999&req=5

figure1: Nonusage attrition curves for user groups in the subsample of users who were 10,000 Steps members for at least 3 months (n=11,651).
Mentions: The following results are based on a subsample only including users that had been a 10,000 Steps member for at least 3 months (n=11,651). Figure 1 presents Kaplan-Meier survival curves for the different user groups based on the duration of usage. The log-rank test showed that the survivor functions were significantly different across groups (χ22=161.3, P<.001). Estimated median lifetime usage (time after which 50% stopped logging physical activity) was 30 days for all groups combined (Table 4). For all groups combined, 25.00% (2913/11,651) were still logging steps after 42 days. This was similar to the Web-only and app-only groups, with 41 and 43 days. respectively; however, in the Web-and-app group, 25.0% (220/878) of the sample were still logging steps after 56 days.

Bottom Line: Cox regression showed that user group predicted nonusage attrition: Web-and-app users (hazard ratio=0.86, 95% CI 0.81-0.93, P<.001) and app-only users (hazard ratio=0.63, 95% CI 0.58-0.68, P<.001) showed a reduced attrition risk compared to Web-only users.Further, having a higher number of individual challenges (hazard ratio=0.62, 95% CI 0.59-0.66, P<.001), workplace challenges (hazard ratio=0.94, 95% CI 0.90-0.97, P<.001), physical activity logging days (hazard ratio=0.921, 95% CI 0.919-0.922, P<.001), and steps logged per day (hazard ratio=0.99999, 95% CI 0.99998-0.99999, P<.001) were associated with reduced nonusage attrition risk as well as older age (hazard ratio=0.992, 95% CI 0.991-0.994, P<.001), being male (hazard ratio=0.85, 95% CI 0.82-0.89, P<.001), and being non-Australian (hazard ratio=0.87, 95% CI 0.82-0.91, P<.001).Better understanding of participant reasons for reducing engagement can assist in clarifying how to best address this issue to maximize behavior change.

View Article: PubMed Central - HTML - PubMed

Affiliation: Institute of Social Medicine and Prevention, University Medicine, Greifswald, Germany. diana.guertler@uni-greifswald.de.

ABSTRACT

Background: Data from controlled trials indicate that Web-based interventions generally suffer from low engagement and high attrition. This is important because the level of exposure to intervention content is linked to intervention effectiveness. However, data from real-life Web-based behavior change interventions are scarce, especially when looking at physical activity promotion.

Objective: The aims of this study were to (1) examine the engagement with the freely available physical activity promotion program 10,000 Steps, (2) examine how the use of a smartphone app may be helpful in increasing engagement with the intervention and in decreasing nonusage attrition, and (3) identify sociodemographic- and engagement-related determinants of nonusage attrition.

Methods: Users (N=16,948) were grouped based on which platform (website, app) they logged their physical activity: Web only, app only, or Web and app. Groups were compared on sociodemographics and engagement parameters (duration of usage, number of individual and workplace challenges started, and number of physical activity log days) using ANOVA and chi-square tests. For a subsample of users that had been members for at least 3 months (n=11,651), Kaplan-Meier survival curves were estimated to plot attrition over the first 3 months after registration. A Cox regression model was used to determine predictors of nonusage attrition.

Results: In the overall sample, user groups differed significantly in all sociodemographics and engagement parameters. Engagement with the program was highest for Web-and-app users. In the subsample, 50.00% (5826/11,651) of users stopped logging physical activity through the program after 30 days. Cox regression showed that user group predicted nonusage attrition: Web-and-app users (hazard ratio=0.86, 95% CI 0.81-0.93, P<.001) and app-only users (hazard ratio=0.63, 95% CI 0.58-0.68, P<.001) showed a reduced attrition risk compared to Web-only users. Further, having a higher number of individual challenges (hazard ratio=0.62, 95% CI 0.59-0.66, P<.001), workplace challenges (hazard ratio=0.94, 95% CI 0.90-0.97, P<.001), physical activity logging days (hazard ratio=0.921, 95% CI 0.919-0.922, P<.001), and steps logged per day (hazard ratio=0.99999, 95% CI 0.99998-0.99999, P<.001) were associated with reduced nonusage attrition risk as well as older age (hazard ratio=0.992, 95% CI 0.991-0.994, P<.001), being male (hazard ratio=0.85, 95% CI 0.82-0.89, P<.001), and being non-Australian (hazard ratio=0.87, 95% CI 0.82-0.91, P<.001).

Conclusions: Compared to other freely accessible Web-based health behavior interventions, the 10,000 Steps program showed high engagement. The use of an app alone or in addition to the website can enhance program engagement and reduce risk of attrition. Better understanding of participant reasons for reducing engagement can assist in clarifying how to best address this issue to maximize behavior change.

No MeSH data available.


Related in: MedlinePlus