Taking more daily steps was associated with reduced depressive symptoms and lower risk of depression in adults, according to a systematic review and meta-analysis published in JAMA Network Open.
The analysis of 33 observational studies involving 96,173 adults found that compared with fewer than 5000 steps per day, achieving 10,000 or more steps daily, 7500 to 9999 steps daily, and even 5000 to 7499 steps daily were significantly associated with fewer depressive symptoms in cross-sectional studies.
The protective effect appeared dose-dependent. In prospective cohort studies, participants who took 7000 or more steps per day had a 31% lower risk of developing depression compared with those taking fewer steps. Additionally, each increase of 1000 steps per day was associated with a 9% reduction in depression risk.
The study included 27 cross-sectional and 6 longitudinal studies involving adults aged 18 years or older across 13 countries, with mean ages ranging from 18.6 to 91.2 years. The study population was 54.5% female and 45.5% male, distributed across Asia, Europe, North America, Oceania, and South America.
Daily step counts were measured via accelerometer or pedometer devices over periods ranging from 3 to 365 days, with most studies using a 7-day measurement period. Mean step counts among participants ranged from 2931 to 10,378 steps per day.
For study quality, 6 (22.2%) cross-sectional studies were rated as good quality, 19 (70.4%) as fair, and 2 (7.4%) as poor. Among longitudinal studies, 2 (33.3%) were rated as good quality and 4 (66.7%) as fair. Follow-up periods for longitudinal studies ranged from 2 to 5 years for panel studies and 1.8 to 7.4 years for prospective cohort studies.
The association between steps and reduced depressive symptoms remained significant across multiple subgroup analyses, including different age groups, both sexes, and various measurement devices and methods. The relationship held true for both accelerometer and pedometer measurements and for devices worn at the waist with uniaxial and triaxial accelerations.
The researchers noted substantial between-study heterogeneity in some pooled estimates (I² = 65.4% for cross-sectional studies), which was partially explained by differences in participant characteristics and measurement methods. Results remained significant in sensitivity analyses, including after excluding COVID-19 lockdown studies.
Limitations included possible reverse associations, substantial between-study heterogeneity, and potential residual confounding from unmeasured factors such as physical activity types or chronic stress. The meta-analysis of cross-sectional studies comparing high versus low step counts showed evidence of publication bias.
The study was supported by grants from the University of Castilla-La Mancha, the National Agency for Research and Innovation, the Carlos III Health Institute, and other institutions. The authors reported no conflicts of interest.
