Diverse elements comprising studies of peer support complicate evidence synthesis.

Background: Innovative approaches to care, such as peer support, are needed to address the substantial and frequently unmet needs of people with serious mental illnesses such as schizophrenia. Although peer support services continue to expand in mental healthcare, findings of effectiveness from systematic reviews are mixed. However, the studies evaluated in these reviews consisted of diverse elements which the review methods neglected to consider.

Aims: This review aims to demonstrate the substantial diversity in intervention components and measured outcomes among studies of peer support and lay the groundwork for more focused reviews of individual intervention components.

Methods: As part of a realist review of the literature, here we synthesize evidence in a way that examines the substantial diversity in intervention components and measured outcomes comprising studies of peer support.

Results: Seven categories of outcomes were represented, including recovery, symptoms and functioning, and care utilization. Importantly, seven distinct intervention components were represented in 26 studies: "being there," assistance in self-management, linkage to clinical care and community resources, social and emotional support, ongoing support, explicit utilization of shared lived experience or peer support values, and systems advocacy. Reflecting diversity in approaches, no study reported all intervention components, and no component was found among all studies.

Implications: Peer support services constitute a category of intervention approaches far too varied to evaluate as a single entity. Results suggest intervention components deserving more focused research, including assistance in self-management, "being there," and explicit utilization of shared lived experience or peer support values.

Prisma/prospero: As this article reports results from a realist review of the literature, we did not follow the PRISMA guidance which is suitable for systematic reviews. We did follow the Realist and Meta-narrative Evidence Syntheses: Evolving Standards (RAMESES) guidelines.This review was not registered on PROSPERO as it is not a systematic review.