Still Looking for Mechanisms: A Realist Review of Batterer Intervention Programs

Abstract

Introduction: Complex interventions, including batterer intervention or treatment programs (BIPs), require that program strategies interact with contextual factors to trigger often unseen mechanisms in individuals or communities. Although hundreds of evaluations of BIPs have been published, few identify mechanisms or explain for whom, under what conditions, and why programs are effective. The goal of this realist review is to identify evidence of the mechanisms that contribute to successful immediate outcomes in BIPs. Methods: In accordance with published realist review standards, we defined the review questions and rational, defined outcomes and formulating initial theories, searched the literature, applied inclusion and exclusion criteria, and analyzed and synthesizing data. An initial search yielded 5,149 citations, and after a systematic process using realist principles, six articles with sufficient information were included. Results: Few evaluations contain the detail necessary to discern clear generative explanations of program processes. Evidence suggested that under some contextual conditions, strategies that trigger a self-reflexive process in participants may lead to changes in attitudes about violence, which may lead to the development of empathy for their partner. Additionally, programs that help participants differentiate between shame and guilt appear to increase both acceptance of responsibility and empathy. Discussion: This realist synthesis illustrates several gaps in the evaluative literature. Few evaluations captured sufficient detail to examine the influence of contextual factors. Likewise, most evaluations only describe if specific outcomes were achieved, not how, or through what mechanisms, those outcomes were accomplished. Recommendations to strengthen program theory and further examine self-reflection as a mechanism are discussed.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 31 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.