Recent equity review guidance encourages reviewers to consider whether it is likely that their findings may impact on health inequalities. Much of the guidance assumes that health inequalities have either already been identified as the focus of the review, or that reviewers are able to recognize if and how health inequalities matter. However, our experience is that this is not necessarily true. Furthermore, theorizing if and how health inequalities matter is not normally integrated into the HTA review process. This presentation describes a novel approach to the development of a theory-led meta-framework to inform health inequality considerations in systematic reviews. The meta-framework aims to increase the usefulness of systematic reviews in informing and implementing changes to practice.

Following the best-fit framework synthesis approach, a meta-framework was generated by ‘deconstituting’ concepts from theories relating to complex interventions and socio-economic health inequalities into a single framework. Feedback was sought from health inequality experts and reviewers.

Complex intervention theories identify four domains and key factors that may influence effectiveness; intervention design, implementation, context and participant response. Applying an equity lens, socio-economic health inequality theories identify key factors and mechanisms associated with these domains that may lead to differential effects across disadvantaged populations.

The meta-framework has the potential to i) facilitate the identification and understanding of when, why and how interventions may impact on socio-economic health inequalities, ii) promote a theory-led approach to incorporating health inequalities in systematic reviews iii) help reviewers identify data to extract and inform a priori analysis on what factors are associated with differential effects, iv) help reviewers to decide whether it is likely that their review findings may have the potential for an intervention to indirectly widen or narrow socio-economic health inequalities, even when evidence of an impact in the primary research is lacking.