The aim of this study was to evaluate the effectiveness of personalised home-based physical activity programs for older adults by means of a tablet application (Gymcentral), and to explore the relationship with physical, psychological and social wellbeing.
We set up an trial with 40 older adults from Trento, Italy, who were performing a tailored exercise program for a period of two months. Participants were divided in two groups: a study group using the full Gymcentral application, and a control group using a simple application focused only on the training aspect. All participants were provided with a sensor to measure their overal physical activity.
Standardized questionnaires were provided to assess the familiarity with the technology, self-reported physical activity, stage of change, social relationship and well-being. Adherence to the training programs, social interactions and user behavior were evaluated using data collected by the application. Physical performance was measured using validated exercises.
A critical aspect of deploying home-based exercises is understanding the feasibility of the
technical solution in the intended scenario. In its simplest form, this means that users
need to be able to operate the technological support, and to follow the activity program.
We evaluated these aspects considering standard measures such as
With regard to persistence, a total number of 24 sessions were planned during the eight weeks of the study (3 sessions per week). In order for the exercise program to success, participants were asked to carry out at least two of the three sessions that were planned each week. To calculate persistence, the total number of exercise sessions in which each participant took part was divided by 24, the total number of possible training sessions across the 8-weeks period of the study. Below you'll see the persistence of the of the two groups over the training program.
The general persistence rate in the two groups was 76% (SD=22.5%), well above the threshold of 70% that was considered an acceptable criterion for success. More specifically, in the Social group the persistence rate was 85%, while in the Control group it was 65%. Indeed, grouping the participation by week, distributing the users by number of participations (1, 2 and 3), we'll see that participants of the study group not only comply with the Coach instructions, but they did more. This indicates that the various features of the Gymcentral application were more engaging than the simple app.
In terms of usability, the Gymcentral represented no barrier, with users rating the usability of the application at 3 on a Likert scale after their first contact, going to 4.5 at the end of the study. We have seen similar effects on attractiveness, perceived usefulness and satisfaction, with a decrease on anxiety towards the end.
Activity programs are designed to improve specific body functions. By measuring the improvements on the trainees as a result of the training, we're not only assessing the effectiveness of the program per se, but the effectiveness of the Gymcentral platform as a medium for delivering and managing a training class.
Two types of assessment exercises, developed and validated within the Otago Exercise Program (A. J. Campbell et al., 1997; J. C. Campbell & Robertson, 2003), were used to measure participants’ leg muscle strength and walking ability at the beginning and at the end of the study. Below we illustrate the difference between both measures.
With regard to leg muscle strength, a mixed between-within subjects analysis of variance was conducted to compare pre- and post- scores in the 30 second Chair Stand test between participants in the experimental and in the control group. Participants in the Social group performed better that participants in the Control group. This could be due to random variability, possibly related to the small sample size, but still does not affect negatively the results confirming that leg muscle strength significantly improved in both groups after the eight weeks period of training. Moreover, multiple comparison tests with Bonferroni correction showed that both groups significantly improved after eight weeks of training (p < .001 for the Social group and p = .032 for the Control group).
With regard to gait speed, a mixed between-within subjects analysis of variance was conducted to compare pre- and post- gait speed in the Timed Up & Go test between participants in the experimental and in the control group. Although the interaction was not significant, multiple comparison tests with Bonferroni correction showed that gait speed significantly improved for the Social group (p = .002) but not for the Control group (p = .285).
Various aspects of the psychological well-being were measured to understand the effects of a home-based intervention on this important dimension.
Subjective well-being was measured by means of the MPQ (Tellegen & Waller, 2008) before and after the study. Multiple comparison tests with Bonferroni correction showed that subjective well-being significantly improved for the Social (p = .034) as well as for the Control group (p = .002). These results suggests that wellbeing improved for all participants, regardless of the version of the application assigned.
Similar results were obtained when analysing the change in participants behavior towards a more active lifestyle. The extent to which participants used the ten processes of behavior change of the TTM was measured with a questionnaire at the beginning and at the end of the study (Nigg et al., 1999; Prochaska et al., 1988). According to the TTM, processes of change are different kinds of activities and techniques that people use when they try to change their behavior. A series of mixed between-within subjects analyses of variance was conducted to compare pre- and post- scores associated to each of the ten processes of change.
The results obtained suggest an attitude towards a more active lifestyle, independently of the group.
Two questionnaires were used to measures participants' social wellbeing before and after the training program. Social closeness was assessed by means of the MPQ questionnaire (Tellegen & Waller, 2008), while loneliness was measured using the shorter version of the R-UCLA Loneliness Scale developed by Hughes et al. (2004).
Social Closeness. No significant interaction between time and group (p = .357), nor a significant main effect for group (p = .614) or for time (p = .44) were found, suggesting that social closeness did not significantly change from pre- to post- measurement for both groups.
Loneliness. With regard to loneliness, a mixed between-within subjects analysis of variance did not reveal a significant interaction between time and group (p = .18), nor a significant main effect for group (p = .915), but it showed a significant main effect for time (F(1,33) = 8.131, p = .007, partial eta squared = .198). However, multiple comparison tests with Bonferroni correction showed that the perception of loneliness significantly decreased only for participants in the Control group (p = .007), but not for participants in the Social group (p = .281). Although the graph shows a tendency toward a decrease of loneliness in the Social group, this change was not statistically significant. Further studies with a larger sample size and a longer training period should further investigate if there is indeed a non-significant decrease in feelings of loneliness in participants using the full application, or if this result is due to low statistical power.