Sources of self-efficacy for physical activity in older adults with multiple chronic conditions

<p><b>Objectives:</b> To examine whether the four sources of self-efficacy (SE) deducted from Bandura’s self-efficacy theory can predict SE for physical activity (PA) in older adults with multiple chronic conditions, and furthermore, to test whether SE mediates the relationship between these sources and subsequent PA.</p> <p><b>Methods:</b> A sample of 309 older adults with multiple chronic conditions was assessed within the project PREFER at 3 time points for self-reported PA (T3), SE (T2), and the four sources of SE (T1): (a) past PA experience; (b) modelling (friends/ family); (c) persuasion (friends/family); (d) somatic and emotional states – positive affect (PANAS), subjective health, and objective health (peak expiratory flow). Results: After controlling for age and gender, past experience, modelling, and subjective health had significant indirect effects via SE on PA (<i>p</i> < 0.05). Objective health had a direct non-mediated effect on PA (<i>p</i> < 0.01).</p> <b>Conclusions:</b> In accordance with SE theory, past PA and modelling are the strongest predictors of SE for PA in older adults with multiple conditions. Contrary to theory, persuasive arguments for PA do not predict SE for PA. Subjective health is more predictive of self-efficacy than objective health or positive affect.