Abstract
BACKGROUND: Whether physical activity (PA) levels relate to mortality in postmenopausal women remains not well understood. METHODS: We analyzed 5,880 postmenopausal women from NHANES 2007-2018 (median follow-up:77 months). PA-including total PA (TPA), leisure-time PA (LTPA), and occupational PA (OPA)-was assessed using the Global Physical Activity Questionnaire and expressed as metabolic equivalent of task (MET)-minutes/week. TPA was classified as no, insufficiently active (< 600), or sufficiently active (≥ 600), while LTPA and OPA were classified as no, low (< 600), or high (≥ 600). Weighted proportional hazards Cox models and restricted cubic spline (RCS) analyses examined associations with all-cause, cardiovascular disease (CVD), and non-CVD mortality. RESULTS: During follow-up, 718 deaths occurred (215 CVD, 503 non-CVD). Compared with no PA, insufficiently and sufficiently active TPA were associated with reduced risks of all-cause, CVD, and non-CVD mortality; both low and high LTPA with reduced risks of all-cause and non-CVD mortality; and low OPA with reduced risks of all-cause and non-CVD mortality. RCS analyses showed linear inverse associations for TPA and non-linear inverse associations for LTPA with all-cause and non-CVD mortality, while OPA showed no significant associations. Tests for trend were significant for TPA in relation to all-cause and non-CVD mortality, and for LTPA across all three outcomes. Sensitivity analyses excluding early deaths yielded similar results. CONCLUSIONS: Higher TPA and LTPA were associated with reduced risks of all-cause and non-CVD mortality among postmenopausal women. Promoting adequate PA, particularly moving from inactivity to modest levels, may be an effective strategy to support healthy aging in this population.