Evaluation of the factor structure and validity of the Group-Based Medical Mistrust Scale (GBMMS) among a racially and ethnically diverse sample of college students

BACKGROUND: A growing number of empirical studies have investigated the contributing role of medical mistrust in observed health disparities among racial/ethnic minorities in the U.S. The Group-Based Medical Mistrust Scale (GBMMS) was developed to measure an individual's beliefs regarding the unequal treatment of minorities by medical providers and institutions. Previous research indicates that the GBMMS consists of three factors including “Suspicion” of doctors and healthcare workers, beliefs about “group-based disparities in healthcare,” and beliefs about the “lack of support” from healthcare providers. The psychometric properties of the GBMMS have not yet been established on a sample of young racial/ethnic minorities. The goals of the current study were to (1) explore the latent factor structure of the GBMMS on a diverse sample of young racial/ethnic minorities and (2) assess the reliability/convergent validity of the GBMMS. METHODS: Exploratory and confirmatory factor analyses were conducted on data from a cross-sectional survey of 956 minority students at a large university in the Southeast. The total sample was randomly divided into two split-half samples to explore the factor structure and to validate a modified model. RESULTS: Evidence for the three-factor structure of the GBMMS was not supported. Parallel analysis of the real and randomly generated eigenvalues, as well as an examination of the scree plot, indicated a two-factor solution as the most empirically plausible. These two-factors accounted for 61% of the variance and encompassed the original “suspicion” and “group-based disparities” factors, which were significantly correlated (r=.59). Internal consistency for the “suspicion” and “group-based disparities” subscales were sufficient (.87 and .82, respectively). A modified two-factor model including these subscales garnered some empirical support for its construct validity (RMSEA=.06, CFI=.98). Evidence of convergent validity was found in significant positive correlations between “suspicion” and beliefs in “group-based disparities” with perceived difficulty in talking to a healthcare provider about health concerns (r = .20 and .22, respectively). CONCLUSION: It is important to understand the causes of medical mistrust as it may explain high levels of disengagement of racial/ethnic minorities from healthcare organizations. Comprehension of factors should be considered in clinical practice to facilitate trust building and improve health care provided to racial/ethnic minorities. The continued development of valid instruments to measure medical mistrust will facilitate a deeper theoretical understanding of health disparities in racial/ethnic minorities. The current study provides some support for the psychometric properties of the GBMMS and offers a modified model to improve its construct validity.