Does Relationship Status Affect the Association Between Comorbidity, Physical Activity and Frailty in Older People? Findings from the Southampton Longitudinal Study of Ageing (SaLSA)

Background: Comorbidity is common in older adults and is associated with lower physical activity (PA) and greater frailty risk. We assessed whether relationship status affects these relationships, using the Southampton Longitudinal Study of Ageing. Methods: 448 community-dwelling older participants (median age of 80 years) were analysed. A questionnaire ascertained relationship status, number of comorbidities, PA (Physical Activity Scale for the Elderly) and frailty components (weight loss, self-reported exhaustion, low PA, low self-reported walking speed, difficulty carrying 10 lb). Associations of number of comorbidities with physical activity and number of frailty components were examined using linear and Poisson regression, respectively. Models were adjusted for age, sex, BMI and smoking status and included an interaction term between number of comorbidities and relationship status. Sex-stratified analyses without interactions were also performed. Results: Evidence of interaction effects were weak (p > 0.14 for both tests). Among females, higher comorbidity was related to lower PA and a greater number of frailty components: difference in PA (95% CI) and multiplicative increase in number of frailty components (95% CI) per additional comorbidity was −0.10 (−0.17,−0.03) and 1.25 (1.16,1.34) respectively. Effect sizes for PA and number of frailty components were broadly similar among females when stratified according to relationship status. Associations were weaker among males (p > 0.27 for all associations). Conclusions: Evidence of relationship status modifying these associations was weak. Females with greater comorbidity had lower physical activity and more frailty components. Differences in observed strength of associations between males and females may reflect different comorbidity profiles.