Measuring individual and community capacity factors in people with multimorbidity and exploring associations with health outcomes.

BACKGROUND: People with multimorbidity work to manage their conditions (burden of treatment). Burden of Treatment Theory (BOTT) proposes poorer outcomes when this work outweighs capacity - an individual's ability to successfully undertake the work of self-management. Capacity is influenced by individual and community factors. This study aims to quantify individual and community capacity factors and explore associations, if any, with mortality and hospitalisation in people with multimorbidity. METHODS: Data source is as follows: West of Scotland Twenty-07 cohort (three age-based cohorts - 15, 35 and 55 years at baseline (wave 1), followed up with four additional waves over 20 years). Participants are as follows: people with ≥ 2 chronic conditions. Variables (e.g. car access/self-esteem/neighbourliness) mapped to underlying individual and community BOTT constructs. Directed acyclic graphs (DAGs) informed analysis. Cox regression analysis using time-varying covariates explored mortality associations; multiple logistic regression explored hospitalisation associations. Both analyses were adjusted for age, sex, socioeconomic deprivation (SED), alcohol, exercise, fruit/vegetable intake, BMI, smoking, marital status, number of long-term conditions and blood pressure. Exploratory analysis of potential moderating effect of SED was also undertaken. RESULTS: A total of 2249 people had multimorbidity across the five waves (mean age 51.5 (SD 11.6) at baseline and 61 (14.9) at wave 5; male 40.6% baseline, 41.1% wave 5; smokers 32.7% baseline, 25.3% wave 5). Living in social housing was associated with increased mortality (HR (95% CI) 1.39 (1.14, 1.68)), while registered disability was associated with increased risk of hospitalisation (OR (95% CI) 1.7 (1.27, 2.27)). Feeling fearful about walking in the dark was associated with mortality ("try to avoid" OR (95% CI) 0.74 (0.60, 0.92); "feel uncomfortable" (OR (95% CI) 0.70 (0.55, 0.89); "no worries" 0.69 (0.57, 0.83)). Feeling little control over one's life: disagreeing quite a bit with "care from others helps me to get well" OR (95% CI) 0.53 (0.33, 0.86). Initial exploratory analysis suggests high SED could act as a potential moderator, increasing associations between community factors with mortality and hospitalisations. CONCLUSIONS: Individual and community factors influencing capacity to self-manage multimorbidity are quantifiable and associated with adverse health outcomes. Our work adds to the growing body of evidence that capacity issues may be important when designing future multimorbidity interventions and services.