Psychometric Evaluation of the Moral Comfort Questionnaire

Abstract

Background and Purpose:The ethical challenges of the COVID-19 pandemic have contributed to an increase in moral distress (MD) in nurses—a long-standing, ubiquitous issue in nursing—highlighting its detrimental impact on nurses, patients, and healthcare organizations. MD is the negative outcome of a moral situation or dilemma in which a nurse is prevented from taking action on ethical decisions usually related to organizational constraints. On the opposite end of the spectrum, moral comfort (MC), an emerging concept in nursing, is an individual’s feelings of ease with decisions and actions related to a moral dilemma or the positive outcome of a moral situation or dilemma. However, nursing literature on MC is sparse. While several instruments to measure MD exist, an instrument to measure MC was not found. The MC Questionnaire (MCQ) was theoretically developed. The purpose of this study was to conduct a psychometric evaluation of the new 35-item MCQ.Methods:Psychometric evaluation design was used. Reliability testing consisted of the examination of internal consistency (Cronbach’s alpha values) and stability (test–retest reliability: Spearman’s correlation coefficient, weighted Kappa, and Bland–Altman [B&A] analysis). Validity was examined using content validity (content expert evaluation) and discriminant validity (r< .30). Model fit of a proposed five-factor model was tested using confirmatory factor analysis (CFA).Results:Hospital-based direct-care (HB-DC) RNs (n= 466) participated from February 2019 to September 2019 in this Institutional Review Board–approved study. Participants completed demographic information, the MCQ, and the Moral Distress Scale—Revised (MDSR). The psychometric evaluation included a priori content validation and multiple statistical analyses: Cronbach’s alpha, Spearman’s correlation coefficient, weighted Kappa, B&A, discriminant validity, and CFA. Cronbach’s alpha was .951, suggesting internal consistency. Spearman’s correlation coefficient was 0.605 (p< .001), suggesting a strong correlation between Time 1 and Time 2. Weighted Kappa values for each item (range = .139–.559) suggested slight to moderate agreement between responses over time. The B&A plot suggested agreement of responses over time. Discriminant validity results suggested no correlation between the MCQ and MDSR (r= −.219), which was expected. CFA results suggest a poor model fit of the proposed five-factor model. A post hoc hierarchical cluster analysis showed the presence of two clusters. A subsequent two-factor exploratory factor analysis showed items loading onto one of the two factors (internal and external).Conclusions:MC in nurses is essential to promoting positive outcomes for nurses. An instrument to measure MC in nurses is needed to gain a further understanding of the concept. The MCQ was theoretically developed and psychometrically evaluated. Results suggest the validity and reliability of the MCQ with further testing of a two-factor model. Knowledge acquired from studies using the MCQ could potentially be used to develop strategies to promote MC in nurses, thereby promoting positive outcomes for nurses, patients, and healthcare organizations.