Objective: To examine the influence of long-term exposure and timing of physical activity (PA) on new joint pain/stiffness in mid-age women. Methods: Data were from 5105 participants (born 1946-51) in the Australian Longitudinal Study on Women's Health (ALSWH) who completed survey items on PA (1998, 2001 and 2004) and joint pain/stiffness (2007 and 2010). PA was categorized in five levels at each survey and summed into a cumulative PA score (CPA, range 0-12). Associations were analysed using logistic regression, with separate models for the cumulative model (using CPA), the sensitive periods model (i.e., PA measured at each survey in one regression model) and the critical periods model (i.e., separate regression models for PA at each survey). Results: 951 (18.6%) participants reported new-onset joint pain/stiffness. In the cumulative model, CPA was associated joint pain/stiffness when included as a continuous variable (adjusted odds ratio [OR]=0.97, 95% confidence interval [CI]=0.95-0.99), but not when included as a categorical variable. In both the sensitive periods and critical periods models, low to high levels of PA in 2001 and 2004 had stronger inverse associations with joint pain/stiffness than PA levels in 1998. The model fit was better for the sensitive periods than the cumulative or critical periods models. Conclusions: In mid-age women, PA between the ages 47 and 58 was associated with a lower risk of joint pain/stiffness 9years later. Associations were stronger for PA in the last 6years than for earlier PA.
- Life-course modelling