Background: Quality indicators are used to quantify the quality of care. A large number of quality indicators makes assessment of overall quality difficult, time consuming and impractical. There is consequently an increasing interest for composite measures based on a combination of multiple indicators.
Objective: To examine the use of different approaches to construct composite measures of quality of care and to assess the use of methodological considerations and justifications.
Methods: We conducted a literature search on PubMed and EMBASE databases (latest update 1 December 2020). For each publication, we extracted information on the weighting and aggregation methodology that had been used to construct composite indicator(s).
Results: A total of 2711 publications were identified of which 145 were included after a screening process. Opportunity scoring with equal weights was the most used approach (86/145, 59%) followed by all-or-none scoring (48/145, 33%). Other approaches regarding aggregation or weighting of individual indicators were used in 32 publications (22%). The rationale for selecting a specific type of composite measure was reported in 36 publications (25%), whereas 22 papers (15%) addressed limitations regarding the composite measure.
Conclusion: Opportunity scoring and all-or-none scoring are the most frequently used approaches when constructing composite measures of quality of care. The attention towards the rationale and limitations of the composite measures appears low.
Discussion: Considering the widespread use and the potential implications for decision-making of composite measures, a high level of transparency regarding the construction process of the composite and the functionality of the measures is crucial.