Validating measures of perceived discrimination in medical care for American Indian women

Background: The majority of research on medical discrimination may not be generalizable to diverse racial and ethnic groups, such as American Indian and Alaska Natives (AI/AN). There are currently no validated instruments to accurately measure medical discrimination in AI/AN populations. Objectives: Our objectives were to validate measures of medical discrimination and report prevalence of medical discrimination in American Indians (AI). Methods: Data were collected using a survey specifically designed for this study among 270 eligible AI women whose primary medical care is from Northwest reservation-based Indian health care facilities. Results: Three medical discrimination scales tested in our study had high internal reliability (alpha=0.85, 0.90, and 0.94). Approximately 37.8-62.8% of perceived (self-report) medical discrimination (PMD) were missed when assessed by single-item questions when compared to a multi-item scale. Prior studies using single-item measures of PMD may under-report PMD in AIs by as much at 63%. Our reported prevalence of general and race-based PMD was 4 to 6.5 times greater than that reported in previously published studies. Conclusions: Our study found that a large proportion of AI women report PMD, and that our measured prevalence greatly exceeds that of earlier published studies. The use of multi-item measures of PMD was found to be reliable and valid. Our findings suggest that single-item measures of PMD may grossly underestimate PMD. Additionally, inclusion of a broad-qualitative question provided important new ideas for future measures of PMD. More research is needed to investigate associations of PMD on AI health care decisions.