Because the baby boomers are already at their retirement age, there is need to find feasible ways to support maintenance of health and functioning of this large ageing population. Compelling evidence shows that participation in regular physical activity reduces the risk of many chronic diseases (2018 Physical Activity Guidelines Advisory Committee, 2018) and mobility limitations in old age (Pahor et al., 2014). However, majority of older adults do not meet the physical activity recommendations (Evenson et al., 2012; Wullems et al., 2016). In addition, older adults are the most sedentary age group (Healy et al., 2011), further underlining the fact that effective interventions to promote physical activity among older adults are needed. Traditional intervention methods, such as face-to-face group/individual counselling or training sessions have shown to have the potential to be effective at increasing physical activity of older adults (Zubala et al., 2017). However, these methods usually require time and resources hampering their implementation as preventive programs at large scale. The more scalable and low-cost e-health intervention strategies utilizing devices and programs have been emerged (Muellmann et al., 2018). Consequently, more high quality studies of the use of technology-based methods to promote physical activity among older population are needed (Stockwell et al., 2019).

Modern fitness and activity trackers have increased their popularity among consumers but also as an intervention method because wearable technology can be harnessed with evidence-based behavioral change techniques, such as self-monitoring, goal setting and feedback, without face-to-face contact (Chia et al., 2019; Laranjo et al., 2020). According to the recent studies among older adults, short-term physical activity interventions utilizing activity trackers can increase physical activity (Coughlin and Stewart, 2016; Jonkman et al., 2018; Stockwell et al., 2019), and decrease body weight (Wijsman et al., 2013; Ashe et al., 2015; Cadmus-Bertram et al., 2015; Lyons et al., 2017; Vandelanotte et al., 2018; Ballin et al., 2020) and systolic blood pressure levels (Wijsman et al., 2013; Ashe et al., 2015). In addition, Wijsman et al. found that the use of activity monitors and digital coaching (DirectLife program, Philips) for 12 weeks increased daily moderate-to-vigorous physical activity by 11 min among inactive 60–70 years old adults and further improved their metabolic health in terms of decreased waist circumference and fat percentage, and improved glucose metabolism when the intervention group was compared to the control group not using the program (Wijsman et al., 2013). Consumer-wearable activity trackers have also been found to increase physical activity and to be associated with improvements in cardiometabolic health, including waist circumference, systolic blood pressure, and low-density lipoprotein cholesterol concentration in middle-aged and older patients with chronic diseases (Franssen et al., 2020). A 2-years pedometer-based intervention prevented decrease in daily steps but did not improve cardiometabolic health in over 60 years old patients with prediabetes or type 2 diabetes (Rossen et al., 2021). However, the long-term effects of activity tracker based physical activity interventions on cardiometabolic health outcomes among older adults are still unknown (Coughlin and Stewart, 2016; Stockwell et al., 2019).

We conducted a 12-months-long physical activity intervention that primarily aimed to increase total physical activity among recent retirees with the daily use of a commercial activity tracker (REACT, NCT03320746). The intervention was based on observational findings suggesting that older adults are prone to change physical activity behavior especially in the retirement transition (Barnett et al., 2012; Stenholm et al., 2016). Results regarding the primary outcome, daily physical activity, has been reported earlier (Leskinen et al., 2021). We did not find long-term intervention effect on daily total physical activity although a tendency for increased levels in total and light physical activity was seen during the first 6 months of the intervention. The changes were not however different from that of controls (Leskinen et al., 2021). Because the previous literature suggest that the retirement transition is also related to negative changes in health, such as weight gain, and increased blood pressure and lipid levels (Pedron et al., 2020) as well as increased sedentary behavior (Leskinen et al., 2018; Suorsa et al., 2020), this time-window is of importance for the prevention of cardiovascular disease risk factors (Xue et al., 2020), and thus should be further evaluated. Therefore, in the present study, we examined the effects of the REACT physical activity intervention on body composition and cardiometabolic health, the secondary outcomes of the trial.

Baseline characteristics of the intervention and control group participants are presented in Table 1.

Details of the group comparisons for all body composition and cardiometabolic outcomes are presented in Supplementary Tables S1, S2. Figures 1, 2 illustrate the results, respectively. Over the 12 months, there were no significant differences in the mean changes in body composition and cardiometabolic health indicators between the groups (group*time interaction >0.20 for all outcomes). The body weight decreased in the intervention group over the 12 months (−0.7 kg, 95% CI −.3 to −0.2); but no significant change was observed in the control group (−0.7 kg, 95% CI −0.8 to 0.3) (Figure 1A). Both groups decreased waist circumference (−2.7 cm, 95% CI −3.4 to −2.1 for the intervention, and −2.6 cm, 95% CI −3.3 to −2.0 for the control group) (Figure 1B) and fat free mass (−1.0 kg, 95% CI −1.4 to −0.5 for the intervention, and −0.8 kg, 95% CI −1.2 to −0.3 for the control group) (Figure 1C), but only control group increased body fat percent (0.6 kg, 95% CI −0.1 to 1.3 for the intervention, and 1.0 kg, 95% CI 0.3 to 1.7 for the control group) (Figure 1D). Systolic blood pressure decreased significantly in the intervention (−7.8 mmHg, 95% CI −10.7 to −4.9) and control (−6.5 mmHg, 95% CI −9.4 to −3.6) groups (Figure 2A). Similarly, diastolic blood pressure decreased in both groups (−3.0 mmHg, 95% CI −4.3 to −1.6 for the intervention, and −2.0 mmHg, 95% CI −3.4 to −0.7 for the controls) (Figure 2B). There was also enhancement in the lipid profile among both groups over the 12 months (Figures 2C,D). The change in daily total physical activity (primary outcome) over the 12 months showed mild, but significant correlations with the change in systolic (r = -0.16, p = 0.0148) and diastolic blood pressure (r = 0.22, p = 0.0014), but not with any other secondary health related outcomes (results not shown).

The REACT trial examined the effect of a 12-months commercial activity tracker based intervention primarily on physical activity behavior and secondarily on other health-related outcomes among recently retired older adults. The present secondary analyses showed no differences in the changes in body composition and cardiometabolic health indicators between the groups over time.

Previous short-term multicomponent physical activity interventions utilizing activity trackers as part of their interventions have succeeded to increase physical activity among individuals having low levels of physical activity or overweight at baseline (Coughlin and Stewart, 2016; Jonkman et al., 2018; Stockwell et al., 2019). Our long-term intervention among general population of recent retirees showed that the use of commercial activity tracker as a stand-alone intervention method did not induce significant changes in daily total physical activity over the 12 months (Leskinen et al., 2021). This result likely explains why no intervention effect was observed in any of the body composition and cardiometabolic health outcomes included in the secondary analysis. The previous digital exercise interventions targeted to sedentary older adults having overweight or chronic diseases have increased mostly moderate-to-vigorous physical activity and resulted in improved metabolic health in the short-term (Wijsman et al., 2013; Ashe et al., 2015; Ballin et al., 2020; Franssen et al., 2020). However, apart from our long-term RCT design, we were unable to examine the short-term effects, because we did not measure body composition nor cardiometabolic outcomes during the intervention. Due to lack of changes in physical activity behavior and especially in the levels of moderate-to-vigorous physical activity over 12 months, our findings are in line with 2-years pedometer-based intervention in which no significant increase in daily steps nor improved metabolic health was observed (Rossen et al., 2021). It is likely that changes in moderate-to-vigorous physical activity are needed to have more marked changes in metabolic health. Therefore, our findings may present aging-related changes or they may be explained by changes in other health behaviors, such as changes in dietary intake, which were not controlled in the study. The controls were not instructed (nor restricted) to engage in physical activity but for example the baseline health measurements may have motivated participants to change their physical activity behavior. The participants of this study were mainly healthy and active older women, which may limit the comparison as well as the generalizability of our findings to other populations.

In conclusion, among general population of recent retirees, the 1-year use of commercial activity trackers with daily activity goal did not induce different cardiometabolic health effects compared to the non-user controls. Alternative methods are therefore needed to promote especially moderate-to-vigorous physical activity and health among retired older adults.