Psychiatry · Medical Literature
An mHealth Intervention Promoting Physical Activity to Reduce Psychological Distress Among Workers: Randomized Controlled Trial.
Abstract
Although mobile health (mHealth) interventions serve as potential solutions for addressing mental health problems, evidence on whether mHealth interventions targeting physical activity can reduce psychological distress among generally healthy workers is limited. This study aimed to investigate the effectiveness of a stand-alone smartphone app, which passively monitors physical activity and psychological distress, in reducing psychological distress among workers. This open-label, 2-arm, parallel-group randomized clinical trial was conducted in Japan over 6 months, with assessments at baseline, 3 months, and 6 months. Eligible participants were workers aged 18 years or older who were able to complete the Japanese questionnaires and owned compatible smartphones. Individuals currently absent from work or with a history of absence due to sickness within the past 12 months were excluded. Participants were randomized (1:1) to the intervention or control group. Participants in the intervention group used a smartphone app that passively monitored physical activity and mental health for 3 months. The control group received a booklet on job stress. The primary outcome was psychological distress measured using the 6-item Kessler Psychological Distress Scale. Secondary outcomes included self-reported physical activity and digitally recorded activity duration in the intervention group. Analyses followed the intention-to-treat principle. For the main analysis, linear mixed modeling for repeated measures was used to estimate between-group differences in changes in psychological distress across the 3 time points. A total of 793 workers were randomized (intervention: n=397; control: n=396). Psychological distress decreased significantly in the intervention group at 3 months (mean difference -0.56, 95% CI -1.11 to -0.02; P=.04; Cohen d -0.17), although this effect was not sustained at 6 months. Self-reported physical activity levels declined in both groups without significant between-group differences. In contrast, in the intervention group, digitally recorded activity increased during the intervention period (7.92 min increase, 95% CI 1.36-14.49; P=.02); however, these data were not comparable with those in the control group. Subgroup analysis showed a larger effect among participants with no or minimal baseline distress (mean difference -0.63, 95% CI -1.16 to -0.11; P=.02; Cohen d -0.19). The stand-alone smartphone-based app that enables workers to monitor physical activity and psychological distress modestly reduced psychological distress among workers. However, it remains unclear whether the app reduces psychological distress through biological and psychosocial mechanisms, including increased physical activity. Furthermore, the benefit may be short-lived rather than durable and may have limited clinical and practical significance. Although mHealth interventions may contribute to preventing mental health problems in workplace settings, booster engagement strategies are necessary to sustain benefits over time. Subgroup analyses suggested that the app may be more effective among generally healthy populations.
Clinical relevance
Does a stand-alone smartphone app (ASHARE) that passively monitors physical activity and psychological distress reduce psychological distress among generally healthy workers compared with an educational booklet on job stress over 6 months? __SNAPSHOT__ {"studyDesign":"Open-label, 2-arm, parallel-group randomized controlled trial","setting":"Japan; occupational/workplace setting; multi-site recruitment of employed adults","sampleSize":"n=793 (intervention: n=397, control: n=396)","population":"Workers aged ≥18 years, able to complete Japanese-language questionnaires, and owning a compatible smartphone; excluded if currently on sick leave or with any sickness absence in the prior 12 months — representing a generally healthy, currently working population","intervention":"3-month use of the ASHARE smartphone app, which passively monitors physical activity (via Apple Health/Google Fit) and uses a deep learning model to generate daily predicted K6 psychological distress scores displayed as a 'mental health forecast,' with feedback delivered via animated weather-style interface","comparator":"Educational booklet on job stress and stress management","primaryOutcome":"Psychological distress measured by the 6-item Kessler Psychological Distress Scale (K6) at 3 and 6 months, analyzed by intention-to-treat linear mixed modeling for repeated measures","followUp":"6 months (assessments at baseline, 3 months, and 6 months; active intervention period was 3 months)","fundingSource":"Government (Japan Society for the Promotion of Science and Japan Agency for Medical Research and Development, based on prior related studies by the same group; specific RCT funding not confirmed in abstract — not fully reported in available sources)"}
Key findings
Among 793 randomized workers, the ASHARE app produced a statistically significant but small reduction in K6 psychological distress at the 3-month (active intervention) timepoint compared with the educational booklet control: mean difference = −0.56 points (95% CI: −1.11 to −0.02; P = .04; Cohen's d = −0.17). This effect was not sustained at the 6-month follow-up assessment, 3 months after the active intervention period ended. A pre-specified subgroup analysis among participants with no or minimal baseline distress showed a modestly larger but still small effect at 3 months: mean difference = −0.63 (95% CI: −1.16 to −0.11; P = .02; Cohen's d = −0.19). Secondary outcomes included self-reported physical activity and digitally recorded activity duration in the intervention group; specific results for these endpoints are not reported in available abstract-level sources. Clinically important adverse events were not reported in available sources. The K6 scale ranges from 0–24; the observed mean difference of 0.56 points is well below commonly cited minimal clinically important difference thresholds for this instrument, indicating the effect — while statistically significant — is of uncertain clinical importance. NNT is not calculable from the reported continuous outcome data.
Practice implications
This RCT establishes that the ASHARE mHealth app produced a statistically significant but small short-term reduction in K6 psychological distress at 3 months in a generally healthy Japanese working population. Critically, this effect was not sustained at 6 months, 3 months after app use ended, suggesting that any benefit depends on continued active engagement and does not persist beyond the active intervention period. The study did NOT establish durable or clinically meaningful mental health benefit, did NOT demonstrate improvement in physical activity outcomes at the population level (based on prior feasibility data from the same app), and did NOT evaluate workers with clinically significant psychological distress — indeed, those with recent sickness absence were explicitly excluded, making these findings applicable only to currently employed, minimally distressed workers. The subgroup showing a larger effect among those with no or minimal baseline distress is hypothesis-generating but not powered as a primary analysis. The effect size (Cohen's d ≈ −0.17 to −0.19) is small by conventional standards, and clinicians should note that a mean 0.56-point change on a 24-point scale is unlikely to translate into meaningful symptomatic relief for individual patients. This summary is based on abstract-level data from the preprint and published abstract; full-text data including complete secondary outcomes, attrition details, and subgroup analyses were not available for review. __LIMITATIONS__ Open-label design: neither participants nor outcome assessors were blinded, introducing significant risk of performance and detection bias, particularly for self-reported K6 outcomes,Active control imbalance: the comparator (a job stress booklet) is a minimal passive intervention unlikely to be equivalent in attention and engagement, potentially inflating the apparent intervention effect,Effect not sustained: loss of benefit at 6-month follow-up — 3 months after app cessation — raises major questions about durability and real-world clinical utility,Small effect size: Cohen's d of −0.17 is below the threshold for even a small-to-moderate clinically meaningful effect; whether a 0.56-point K6 change impacts patient wellbeing or functioning is not established,Highly selected population: exclusion of workers with any sickness absence in the prior 12 months selects against individuals with the most clinically relevant psychological burden, severely limiting generalizability to higher-risk or symptomatic populations,Surrogate and self-reported primary endpoint: K6 is a screening tool, not a diagnostic instrument; no functional, work performance, or clinical outcomes were assessed as primary endpoints,Single-country, single-language setting: conducted entirely in Japan with Japanese-language tools, limiting generalizability to other cultural and occupational contexts,Prior feasibility data from the same app (ASHARE) in earlier single-arm and non-randomized studies were negative for physical activity and distress endpoints, raising concerns about replication,Mechanism not established: whether app-driven physical activity change is the actual mediator of distress reduction, or whether passive self-monitoring feedback alone drives effect, is not determined,Funding source and potential conflicts of interest for the published RCT are not fully disclosed in available abstract-level sources __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ A stand-alone physical activity monitoring smartphone app (ASHARE) produced a statistically significant but very small reduction in K6 psychological distress at 3 months in healthy Japanese workers (mean difference −0.56 points; Cohen's d −0.17), but this effect was not sustained at 6 months after the intervention ended. Given the open-label design, minimal effect size, non-durable benefit, and restriction to a low-risk working population, this intervention is not ready for clinical adoption as a standalone mental health strategy; further trials with blinded outcomes, durable follow-up, and clinically meaningful endpoints are needed before practice change is warranted.