The (cost-)effectiveness of preventive, integrated care for community-dwelling frail older people: A systematic review.

Integrated care is increasingly promoted as an effective and cost-effective way to organise care for community-dwelling frail older people with complex problems but the question remains whether high expectations are justified. Our study aims to systematically review the empirical evidence for the effectiveness and cost-effectiveness of preventive, integrated care for community-dwelling frail older people and close attention is paid to the elements and levels of integration of the interventions. We searched nine databases for eligible studies until May 2016 with a comparison group and reporting at least one outcome regarding effectiveness or cost-effectiveness. We identified 2,998 unique records and, after exclusions, selected 46 studies on 29 interventions. We assessed the quality of the included studies with the Effective Practice and Organization of Care risk-of-bias tool. The interventions were described following Rainbow Model of Integrated Care framework by Valentijn. Our systematic review reveals that the majority of the reported outcomes in the studies on preventive, integrated care show no effects. In terms of health outcomes, effectiveness is demonstrated most often for seldom-reported outcomes such as well-being. Outcomes regarding informal caregivers and professionals are rarely considered and negligible. Most promising are the care process outcomes that did improve for preventive, integrated care interventions as compared to usual care. Healthcare utilisation was the most reported outcome but we found mixed results. Evidence for cost-effectiveness is limited. High expectations should be tempered given this limited and fragmented evidence for the effectiveness and cost-effectiveness of preventive, integrated care for frail older people. Future research should focus on unravelling the heterogeneity of frailty and on exploring what outcomes among frail older people may realistically be expected.