The effect of activity trackers on physical activity and health in older adults

Background Despite a range of efforts to increase physical activity participation in Australia, inactivity levels in older adults remain high and unchanged over the last two decades, contributing to increased rates of chronic health conditions. Lifestyle interventions have shown to be effective at improving physical activity participation and associated health outcomes over the short-term, however ongoing feedback and support is often required to maintain these changes. Investigating how newer technologies such as wearable activity trackers may assist in providing ongoing support to maintain physical activity levels and health outcomes warrants further investigation. Methods Systematic Review and Meta-analysis: Included studies used a consumer-based wearable activity tracker to improve physical activity participation. Outcome measures included physical activity participation, sedentary behaviour and health risk factors. 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. Randomised Controlled Trial: Participants were randomised at the end of the Strength2Strength Exercise Treatment Program (S2S) to receive either a wearable activity tracker (AT), telephone counselling (TC) or usual care (UC) for 12-months. All groups received standard care, including an individualised home-based exercise program and optional referral to a community-based program. The primary outcome measure was daily steps as measured by an ActivPAL‚Äöv묢 accelerometer. Secondary outcome measures included body composition, blood pressure, measure of physical function and health-related quality of life. All outcome measures were obtained at the beginning and end of the S2S program and at 3, 6 and 12-months of the intervention. Randomised participants were included to determine costs associated with each intervention. Base-case analysis of intervention costs were calculated from a healthcare payer perspective. All participants were retrospectively invited to take part in a health services usage analysis. Participants from the AT group were invited to take part in focus groups to explore their experiences of using the Jawbone UP24‚Äöv묢 activity tracker and the associated health professional feedback. Acceptability data relating to activity tracker wear time was examined. Focus group transcripts were coded and thematically analysed using a qualitative descriptive approach. Results Systematic Review and Meta-analysis: A significant increase in daily step count (Standardised mean difference [SMD] 0.24; 95% CI 0.16 to 0.33; p ‚Äöv¢¬ß 0.00), moderate and vigorous physical activity (SMD 0.27; 95% CI 0.15 to 0.39; p ‚Äöv¢¬ß 0.00) and energy expenditure (SMD 0.28; 95% CI 0.03 to 0.54; p = 0.03) and a non-significant decrease in sedentary behaviour (SMD -0.20; 95% CI -0.43 to 0.03; p=0.08) was observed following the intervention versus control comparator across all studies in the meta-analyses. Randomised Controlled Trial: Both the wearable activity tracker and telephone counselling groups maintained daily step counts throughout the 12-month intervention, while those receiving usual care showed a significant reduction in both daily steps and lean mass at 12 months (MD; - 981 steps, 95% CI; -1668 steps to -294 steps, p ‚Äöv¢¬ß 0.01 and MD; -1.13kg, 95% CI; -2.26kg to -0.01kg, p = 0.005 respectively). A significant increase in body fat between baseline and 12-months was observed in the AT and TC groups (MD; 1.56%, 95% CI; 0.43% to 2.70%, p = 0.01 and MD; 1.27%, 95% CI 0.14% to 2.41%, p = 0.03 respectively) while a significant reduction in lean mass was observed in the UC group between baseline and 12-months (MD; -1.13 kg, 95% CI; -2.26 kg to -0.01 kg, p = 0.05). Diastolic blood pressure significantly decreased between baseline and 12-months in the UC group (MD; -4.10mmHg, 95%CI; -7.02mmHg to - 1.18mmHg, p = 0.02), and differed significantly from the AT group at 12-months (MD; 5.62mmHg, 95%CI; 1.30mmHg to 9.94mmHg, p = 0.01). Participants in the UC group significantly improved their time taken to perform a ten time sit to stand test between baseline and 12-months (MD; -1.10 sec, 95% CI; -2.80 sec to 0.70 sec, p = 0.02) and differed significantly from the TC group at 12-months (MD; 2.36 sec, 95% CI; -0.14 sec to 4.87 sec, p = 0.02). No other changes in health outcomes were observed. The cost of delivering a 12-month activity tracker-based intervention was between $145 and $234 per participant depending on if the cost of the activity tracker itself was included. Delivering a 12-month telephone counselling intervention cost was $195 per participant. No differences in health services usage or health state utilites were observed between groups or in the 12-months following S2S compared to the 12-months preceding S2S. The activity tracker was well accepted with the device worn on an average 86% of possible days and participants reported an overall positive experience. Four themes were identified: (1) increased sense of awareness of activity levels is related to motivation; (2) the level of engagement with the activity tracker influences the user experience; (3) the role of feedback from a health professional in providing ongoing support; (4) the role of habits in supporting long-term behaviour change. Conclusions The major conclusions of this thesis are: 1. Wearable activity trackers are effective in a range of adult populations, particularly when combined with established behavioural change interventions. 2. Both telephone counselling and activity trackers were effective at maintaining daily step count and lean mass over a 12-month intervention period, while usual care resulted in a significant decline in daily step count and lean mass. There were no significant differences between groups at 12-months for daily steps and lean mass. 3. The costs associated with delivering a wearable activity tracker-based intervention are comparable to the costs of delivering a telephone counselling intervention. Health service usage and health state utilites did not differ between groups. 4. The Jawbone UP24‚Äöv묢 activity tracker combined with health professional feedback was well accepted by older adults. Themes identified suggested that older adults who were more engaged with the activity tracker had a better overall experience. Support from a health professional was highlighted as an important factor in helping participants continue to engage with the activity tracker. The significance of this research is that it has demonstrated that activity trackers are an effective and well accepted method of providing ongoing support and feedback to maintain physical activity levels in older adults. Additionally, an activity tracker intervention with health professional support has similar delivery costs to established methods of providing ongoing support. The findings of this thesis can assist health professionals to determine an appropriate method of proving ongoing support in a clinical setting and have confidence in recommending activity tracker use in an older adult population.