W La Revue canadienne de psychiatrie, vol 54, no 3, mars 2009 160 Objective: While previous research has established that regular involvement in physical activity (PA) is associated with better mental health in old age, the socio-cognitive factors that mediate the association have only been partially tested. We examined whether changes in PA are associated with changes in distress during a 6-year period, and whether this association is mediated by changes in global self-esteem, mastery, and physical health status.
Method:A residualized regression technique was used to examine changes over time in a national longitudinal survey of adults aged 65 years and older (n = 1327).Results: There is a significant association between change in PA and change in distress. Separately, physical health status accounted for 30% of the explained variance of the longitudinal relation between PA and distress, while global self-esteem and mastery accounted for 39%. Combined, they accounted for 50% of the explained variance of PA on distress.
Conclusion:These findings highlight the importance of psychosocial factors in the relation between PA and distress. Results suggest that PA interventions focused on improving mastery or self-worth, as well as physical fitness, may yield the greatest benefit in alleviating psychological distress.Can J Psychiatry. 2009;54(3):160-169.
Clinical Implications· PA may play a protective role in distress among older adults. · Beyond physical health benefits, the positive impact of PA on distress may be through improved sense of self-worth and greater perceived mastery. · PA interventions that target perceptions of self-worth and perceived control may have the greatest impact on psychological distress in older adults living in the community.
Limitations· Our measure of PA is based on recreational activities only and does not reflect other activities, such as vacuuming, raking leaves, and other household chores, which can be physically demanding. · We were only able to assess change during a single, 6-year interval. Further work with repeated measures for a longer duration are needed to confirm our results. · Other potential mediating factors, especially biological markers of stress, are excluded from surveys such as the National Population Health Survey.