A psychometric study of the decisional conflict scale in surrogate decision makers

Objective: To assess the psychometric properties of the 16-item Decisional Conflict Scale (DCS) in surrogate decision makers. Methods: With a sample of 472 surrogates from intensive care units in five academic medical centers across the United States, we performed the analysis in five phases to 1) model the congeneric structure with confirmatory factor analysis and assess 2) internal consistency reliability, 3) the unidimensional or global assessment, 4) factorial invariance across surrogate gender, and 5) individual item influence on the domains. Results: The congeneric model fit the data, with all factor loadings (0.577-0.955) statistically significant at p < 0.05. All subdomains had acceptable internal consistency (0.751-0.981). The bifactor model supported the sub-domains or the global construct as appropriate measurement models. The DCS demonstrated invariance for use across surrogate genders. The most difficult item for surrogates to complete was "This decision is easy for me to make" [MNSQ Infit/Outfit: 2.37/3.27] and the easiest item was "I have enough advice to make a choice" [0.70/0.67]. Conclusions: Overall, the DCS demonstrated good fit, and can be considered a valid and reliable tool to use with the surrogate population. Practice Implications: Measuring surrogate decisional conflict could be especially useful to determine how to improve the decision-making process. (c) 2018 Elsevier B.V. All rights reserved.