Modified Decisional Conflict Scale for Primary Caregivers in Long-Term Care Facilities: Psychometric Validation Using Structural Equation Modeling

Abstract

Background: Caregivers of long-term care (LTC) residents often face decisional conflict during unplanned hospitalization decisions. This study aimed to adapt and psychometrically validate a modified Decisional Conflict Scale (DCS) for primary family caregivers. Methods: A cross-sectional survey was conducted among 205 caregivers in 20 LTC facilities. Exploratory and confirmatory factor analyses (EFA/CFA) were performed on randomly split subsamples, and structural equation modeling (SEM) examined associations among Decision Antecedents, Decision-Making Process, and Decisional Conflict. Bollen–Stine bootstrap procedures were applied to provide robust estimates under slight deviations from multivariate normality. Given the cross-sectional design and single-sample nature of this study, the SEM findings should be interpreted as theory-informed associations rather than causal effects. Longitudinal or intervention-based studies are needed to establish temporal ordering and determine whether improvements in caregiver readiness and decision-making processes lead to subsequent reductions in decisional conflict. Results: The modified DCS demonstrated strong internal consistency and a single-factor structure (α = 0.98, factor loadings 0.83–0.90). SEM indicated that Decision Antecedents and Decision-Making Process were significantly associated with Decisional Conflict (R2 = 0.68). The mediation analysis suggested that the Decision-Making Process partially mediated the relationship between Decision Antecedents and Decisional Conflict. Conclusions: The modified DCS appears to be a reliable and valid instrument for assessing decisional conflict among LTC caregivers. Findings highlight the importance of caregiver readiness, support, and communication in shaping decisional experiences. Given the cross-sectional design and single sample, results should be interpreted as associations rather than causal effects. Future research should replicate these findings in larger, diverse samples and explore short-form versions of the scale.