This analysis used data from the PREDIMED study to examine whether habitual moderate-to-vigorous physical activity during leisure time (MVLTPA) is associated with long-term adiposity measures and obesity incidence in older adults at high cardiovascular risk. The investigators sought associations between MVLTPA and trajectories of BMI and waist-to-height ratio (WHtR), and the incidence of obesity (BMI ≥30 kg/m2) and central adiposity (WHtR ≥0.6).
Participants were men and women aged 60–80 years enrolled in PREDIMED. The analytic samples included n=5,273 participants for BMI analyses and n=4,940 for WHtR analyses; 61% of participants were women. Usual MVLTPA was estimated using the Minnesota leisure-time physical activity questionnaire and expressed as MET-minutes per day. A cumulative mean of MVLTPA measurements was used to reflect habitual exposure.
MVLTPA was dichotomized for primary comparisons: adequate MVLTPA (≥100 MET-min/day) versus insufficient MVLTPA (<100 MET-min/day). Outcomes were repeated measures of BMI and WHtR over follow-up and incident cases of obesity and central adiposity based on standard thresholds.
To model longitudinal adiposity, the authors used mixed-effects models with cubic splines to estimate trajectories of BMI and WHtR according to MVLTPA categories. Time-to-event analyses for incident obesity and central adiposity used Cox proportional hazards models. Nonlinear dose–response relationships between MVLTPA and adiposity outcomes were also explored.
Compared with participants reporting <100 MET-min/day, those reporting ≥100 MET-min/day had lower average adiposity measures across trajectories. Reported differences in BMI ranged from −1.27 to −1.48 kg/m2 favoring the more active group. Corresponding differences in WHtR ranged from −0.021 to −0.026. The more active group also showed a slower decline in BMI over time compared with less active participants.
In time-to-event analyses, adequate MVLTPA was associated with a substantially lower incidence of central adiposity. The hazard ratio for central adiposity among participants meeting the ≥100 MET-min/day threshold was 0.66 (95% CI 0.56–0.77), representing a 34% lower risk compared with those below the threshold.
For general obesity (BMI ≥30 kg/m2), associations were evaluated overall and by sex (see next section). The primary reported protective association for incident obesity was specific to women in stratified analyses.
Nonlinear analyses indicated that the reduction in central adiposity risk increased with higher MVLTPA up to approximately 400 MET-min/day, after which the association appeared to plateau. The source reports that risk reduction for central adiposity did not continue to decline beyond that level.
A sex-specific finding was reported for incident obesity. Adequate MVLTPA (≥100 MET-min/day) was associated with a 24% lower risk of developing obesity in women (HR 0.76; 95% CI 0.57–1.00). The interaction p-value for sex and MVLTPA was reported as 0.084. The study did not report a statistically robust protective association for incident obesity in men within the provided summary.
Among older adults at high cardiovascular risk, higher habitual moderate-to-vigorous leisure-time physical activity was associated with more favorable adiposity trajectories (lower BMI and WHtR) and a lower incidence of central adiposity. The authors highlight a dose–response pattern for central adiposity that levels off at about 400 MET-min/day. A sex-specific protective association for incident obesity was observed in women but not conclusively in men.
These findings support the potential role of regular MVLTPA in attenuating age-dependent increases in adiposity and reducing central fat accumulation in an older, cardiometabolic high-risk population.
The study was conducted according to the Declaration of Helsinki and received approval from the Ethics Committee of Parc de Salut Mar (reference 2024/11670/I; November 18, 2024). All participants provided written informed consent.
Individual-level PREDIMED data are not publicly available due to national data-protection legislation and consent restrictions; access may be granted to qualified investigators on reasonable request to the PREDIMED Steering Committee. The code used for database construction and statistical analyses is publicly available at the authors' GitHub repository (link reported in the source).
Several authors disclosed relationships with food industry groups and other organizations (personal fees, grants, board memberships). All other authors declared no conflicts of interest.
Note: This rewrite adheres to facts and figures reported in the source article. No additional study details, numerical estimates, or interpretations beyond those reported in the source were added.