Who benefits from self-management support? Results from a randomized controlled trial.

Background: Self-management support models adapted to accommodate the needs of each patient are complex interventions that should be evaluated for intervention mechanisms. In a national randomized controlled trial (RCT), we evaluated the efficacy of telephone-based self-management support that demonstrated improved health-related quality of life (HRQoL), no reduction in hospital admissions, and an unexpected increase in primary healthcare services. Objective: The objective of this study is to identify RCT impact mechanisms and explore which participants could benefit the most from the PaHS intervention. Methods: This study evaluates intervention mechanisms through interaction analyses of predefined intervention moderators (sex, age, education, chronic disease, risk of hospital admissions, and coping) and post-hoc intervention mediators (contacts in primary care and anxiety medication). The one co-primary outcome HRQoL was assessed with SF26v2 and analyzed with generalized linear mixed models and the other co-primary hospital admissions was analyzed with poisson regression. Results: PaHS interacted with diabetes, multimorbidity, coping, and anxiety medication on the outcome hospital admissions. PaHS led to a significant reduction in hospital admissions in participants with diabetes or multimorbidity and an increase in hospital admissions in participants with higher baseline coping and participants using anxiety medication. The interaction analyses revealed significant intervention mediation in the outcome HRQoL by sex and diabetes. Conclusions: Participants with diabetes, multimorbidity, and women could benefit the most from telephone-based self-management support, but the intervention involves the risk of over-treatment.