Methods used to conceptualize dimensions of health equity impacts of public health interventions in systematic reviews

• Published in: Journal of clinical epidemiology 2024-05, Vol.169, p.111312, Article 111312
• Main Authors: Hollands, Gareth J., South, Emily, Shemilt, Ian, Oliver, Sandy, Thomas, James, Sowden, Amanda J.
• Format: Article
• Language: English
• Subjects: Algorithms; Datasets; Differential impacts; Equity; Health disparities; Health Equity; Health promotion; Humans; Inequalities; Intervention; Investigations; Machine learning; Public health; Public Health - methods; Reviews; Searching; Sexual orientation; Socioeconomic factors; Systematic review; Systematic reviews; Systematic Reviews as Topic - methods
• ISSN: 0895-4356; 1878-5921; 1878-5921
• DOI: 10.1016/j.jclinepi.2024.111312

Our aims were to, first, identify and summarize the use of methods, frameworks, and tools as a conceptual basis for investigating dimensions of equity impacts of public health interventions in systematic reviews including an equity focus. These include PROGRESS-Plus, which identifies key sociodemographic characteristics that determine health outcomes. Second, we aimed to document challenges and opportunities encountered in the application of such methods, as reported in systematic reviews. We conducted a methodological study, comprising an overview of systematic reviews with a focus on, or that aimed to assess, the equity impacts of public health interventions. We used electronic searches of the Cochrane Database of Systematic Reviews, the Database of Promoting Health Effectiveness Reviews (DoPHER), and the Finding Accessible Inequalities Research in Public Health Database, supplemented with automated searches of the OpenAlex dataset. An active learning algorithm was used to prioritize title-abstract records for manual screening against eligibility criteria. We extracted and analyzed a core dataset from a purposively selected sample of reviews, to summarize key characteristics and approaches to conceptualizing investigations of equity. We assessed 322 full-text reports for eligibility, from which we included 120 reports of systematic reviews. PROGRESS-Plus was the only formalized framework used to conceptualize dimensions of equity impacts. Most reviews were able to apply their intended methods to at least some degree. Where intended methods were unable to be applied fully, this was usually because primary research studies did not report the necessary information. A general rationale for focusing on equity impacts was often included, but few reviews explicitly justified their focus on (or exclusion of) specific dimensions. In addition to practical challenges such as data not being available, authors highlighted significant measurement and conceptual issues with applying these methods which may impair the ability to investigate and interpret differential impacts within and between studies. These issues included investigating constructs that lack standardized operationalization and measurement, and the complex nature of differential impacts, with dimensions that may interact with one another, as well as with particular temporal, personal, social or geographic contexts. PROGRESS-Plus is the predominant framework used in systematic reviews to conceptualize dif