The objective of this study is to determine the feasibility of using qualitative comparative analysis (QCA) within systematic reviews of complex health interventions to enrich review results through identification of key intervention features associated with effectiveness. Systematic reviews are the foundation for evidence- based health care, and form the basis for clinical guidelines, quality measures, and health policy. However, reviews of complex interventions are challenging because of the large degree of clinical and methodological heterogeneity that is often present among included studies. Review investigators need new methods to complement existing methods of review for complex interventions. We expect QCA to be used to analyze study-level covariates to better explore heterogeneity, and may offer an advantage over existing methods because of its origins as a method to study complex phenomena. QCA is a case-oriented, nonprobabilistic technique based on set-theoretic causal relationships (i.e., set, superset, and subset) between independent and dependent variables. It shares features of both quantitative and qualitative methods and produces results through a process of logical minimization using Boolean algebra. We will apply QCA to an already completed Agency for Healthcare Research and Quality-funded comparative effectiveness review of interventions to improve medication adherence. This review exemplifies a complex intervention review because it included studies with multiple intervention targets and agents, vastly different intervention components, and differences in duration and intensity of intervention. The proposed study is the first such application of QCA to systematic reviews.
The specific aims are to (1) use QCA to determine which intervention features alone or in combination are related to improved medication adherence among outpatients who are prescribed long-term medication therapy for chronic conditions, and (2) describe our process for applying QCA within the context of the systematic review of interventions to increase medication adherence. We will treat each included study within the medication adherence review as a unit of analysis and will use QCA to evaluate study covariates for their relationship to the treatment effect. In this context, we would express our findings in set-theoretic statements of necessity and sufficiency. For example, "Complex health interventions to improve medication adherence require component X and either component A or D, or the combination of B and C." We expect that this work will determine the initial feasibility f using QCA within a systematic review context. If feasible, QCA will significantly enhance review methods for complex interventions because it applies a fundamentally different paradigm (the use of set relationships) for evaluating causal relationships in addition to taking advantage of existing heterogeneity within complex health intervention reviews.
Systematic reviews are the foundation for evidence-based health care and public health and often form the basis for clinical guidelines and health policy. However, reviews of complex interventions are challenging because of the large degree of variation among studies that are included within a review. The proposed study will assess the feasibility of using a method called qualitative comparative analysis to disentangle some of the complexity within reviews to allow patients, providers, program managers, and policy makers to have richer information upon which to base decisions.