Incorporating husband influences into a model of physical activity among older women

Health behaviour models focus primarily on intrapersonal constructs (e.g., self-efficacy) which are good predictors of intention but less so of actual behaviour. Influences from the social environment, namely from close others, may improve prediction of engagement in ongoing behaviour.

This study sought to broadly assess husband influence on physical activity and to determine whether a multidimensional assessment of husband influence would improve prediction of wives' physical activity.

A two-wave (separated by 4 months) observational study of married women was employed.

Women (n = 160, Mage  = 63 years) reported intention, planning, and self-efficacy for physical activity as well as previous month physical activity through Wave 1 mailed surveys. In Wave 2 telephone interviews conducted 4 months later, women reported on their perception of husband influences and their own physical activity in the previous month.

Psychometric analyses supported five components of social support along with social control, injunctive norms, and descriptive norms as a broad assessment of husband social influence on physical activity. Husband influences significantly improved prediction of wives' physical activity, over and above intrapersonal predictors (ΔR(2)  = .09). Social control related negatively to behaviour; moderation analyses revealed a strong negative association when women's intentions were low.

This work highlights the interpersonal social context and post-intentional influences on daily health behaviour and offers guidance on the expansion of models of health behaviour as well as potential targets for intervention. Statement of contribution What is already known on this subject? Behavioural intentions are the best predictors of behaviour; however, much variability in actual behaviour remains to be explained. Romantic partners are frequent providers of health-related social support and social control. What does this study add? Provides multidimensional assessment of husband influence. Improves prediction of behaviour using husband influences. Demonstrates moderator of social control on behaviour.