The effects of community interventions on unplanned healthcare use in patients with multimorbidity: a systematic review

Objectives To summarise the impact of community-based interventions for multimorbid patients on unplanned healthcare use. The prevalence of multimorbidity (co-existence of multiple chronic conditions) is rapidly increasing and affects one-third of the global population. Patients with multimorbidity have complex healthcare needs and greater unplanned healthcare usage. Community-based interventions allow for continued care of patients outside hospitals, but few studies have explored the effects of these interventions on unplanned healthcare usage. Design A systematic review was conducted. MEDLINE, EMBASE, PsychINFO and Cochrane Library online databases were searched. Studies were screened and underwent risk of bias assessment. Data were synthesised using narrative synthesis. Setting Community-based interventions. Participants Patients with multimorbidity. Main outcome measures Unplanned healthcare usage. Results Thirteen studies, including a total of 6148 participants, were included. All included studies came from high-income settings and had elderly populations. All studies measured emergency department attendances as their primary outcome. Risk of bias was generally low. Most community interventions were multifaceted with emphasis on education, self-monitoring of symptoms and regular follow-ups. Four studies looked at improved care coordination, advance care planning and palliative care. All 13 studies found a decrease in emergency department visits post-intervention with risk reduction ranging from 0 (95% confidencec interval [CI]: -0.37 to 0.37) to 0.735 (95% CI: 0.688-0.785). Conclusions Community-based interventions have potential to reduce emergency department visits in patients with multimorbidity. Identification of specific successful components of interventions was challenging given the overlaps between interventions. Policymakers should recognise the importance of community interventions and aim to integrate aspects of these into existing healthcare structures. Future research should investigate the impact of such interventions with broader participant characteristics.