Coincidence Analysis: A New Method for Causal Inference in Implementation Science

Abstract

Background: Implementation of multifaceted interventions typically involves many diverse elements working together in interrelated ways, including intervention components, implementation strategies, and features of local context. Given this real-world complexity, implementation researchers may be interested in a new mathematical, cross-case method called Coincidence Analysis (CNA) that has been designed explicitly to support causal inference, answer research questions about combinations of conditions that are minimally necessary or sufficient for an outcome, and identify the possible presence of multiple causal paths to an outcome. CNA can be applied as a standalone method or in conjunction with other approaches, and can reveal new empirical findings related to implementation that might otherwise have gone undetected. Methods: We applied CNA to a publicly available dataset from Sweden with county-level data on human papillomavirus (HPV) vaccination campaigns and vaccination uptake in 2012 and 2014 and then compared CNA results to the published regression findings. Results: The original regression analysis found vaccination uptake was positively associated only with the availability of vaccines in schools. CNA produced different findings and uncovered an additional solution path: high vaccination rates were achieved by either (1) offering the vaccine in all schools or (2) a combination of offering the vaccine in some schools and media coverage. Conclusions. CNA offers a new comparative approach for researchers seeking to understand how implementation conditions work together and link to outcomes.

Core implementation strategies for improving cirrhosis care in the Veterans Health Administration

“…[21,34] CCMs draw on Boolean algebra and set theory to identify combinations of strategies (configurations) linked to high performance. [18,19,35] Factors analyzed included facility and respondent characteristics and implementation strategies. Following a previously published method, we used the "minimally sufficient condition" function within cna to analyze the complete data set and exhaustively search all 1-, 2-, and 3-condition configurations instantiated in the data set, retaining configurations meeting pre-established consistency (i.e., cases with the outcome and the solution divided by all cases with the solution) and coverage (i.e., cases with the outcome and combination divided by all cases with the outcome) thresholds, and then identify particular combinations of strategies with the strongest connections to the outcome.…”

“…As more emphasis is placed on advancing methods of causal inference, configurational comparative methods (CCMs) offer potential solutions. [18] CCMs provide a systematic way to assess all factors and cases in a data set at once, analyze how specific combinations of conditions relate empirically to an outcome of interest, and then identify the key set of difference-makers that distinguish patients receiving high-quality versus low-quality care. [18][19][20][21] Because CCMs differ from traditional probabilistic regression analytic methods, combining these methods with qualitative data enables convergent validation and mechanistic understanding.…”

“…[18] CCMs provide a systematic way to assess all factors and cases in a data set at once, analyze how specific combinations of conditions relate empirically to an outcome of interest, and then identify the key set of difference-makers that distinguish patients receiving high-quality versus low-quality care. [18][19][20][21] Because CCMs differ from traditional probabilistic regression analytic methods, combining these methods with qualitative data enables convergent validation and mechanistic understanding. [19] This evaluation aimed to (1) identify implementation strategies associated with implementation of EBPs for veterans with cirrhosis, and (2) specify these core strategies for replication in facilities with opportunities to improve cirrhosis care.…”