Do Minority Patients Use Lower Quality Hospitals?
Gaskin, Darrell J., Spencer, Christine S., Richard, Patrick, Anderson, Gerard, Powe, Neil R., LaVeist, Thomas A., Inquiry
Employing three years of inpatient discharge data from 11 states and inpatient and patient safety quality indicators from the Agency for Healthcare Research and Quality (AHRQ), this paper explored whether minority (black, Hispanic', and Asian) patients used lower quality hospitals. We found that the association between the share of minority patients and hospital quality depended on how quality was measured and varied by race and ethnicity. Hospitals serving Hispanics performed well on most patient safety measures. Higher percentages of all three minority patient groups corresponded to lower quality for only one measure, postoperative sepsis. Our analysis indicates that it is" incorrect to generalize that minorities use lower quality hospitals. Analysts and policymakers should be cautious when making generalizations about the overall service quality of hospitals that treat minority patients.
Several studies have presented evidence suggesting that race disparities in the quality of hospital care are due more to variation in quality of care across hospitals than variation in the quality of care within hospitals (Barnato et al. 2005; Blustein and Weitzman 1995; Bradley et al. 2004; Hasnain-Wynia et al. 2007; Jha et al. 2007; Skinner et al. 2005). Stated simply, minority patients receive lower quality care because they are more likely to be treated in lower quality hospitals.
For example, a recent study on adult cardiac and pneumonia patients treated in major teaching hospitals found that disparities were explained by differences in quality across hospitals (Hasnain-Wynia et al. 2007). Two studies suggest that the concentration of black patients in a relatively small number of hospitals contribute to disparities in quality care (Skinner et al. 2005; Jha et al. 2007). Black-serving hospitals compared to other hospitals were found to have higher rates for some patient safety indicators for Medicare patients (Ly et al. 2010). Another study found that minorities and white patients treated in the same hospital did not have different rates of inpatient mortality and adverse events (Gaskin et al. 2008). Observed disparities in quality of care in pneumonia can be explained by the difference in case mix and variations in care across hospitals (Mayr et al. 2010). One study found that differences across hospitals accounted for a large proportion of observed disparities in the Hospital Quality Alliance measures (Hasnain-Wynia et al. 2010). The policy conclusion drawn from these articles is that the most effective way to address race disparities in hospital care is to improve the care in low-quality hospitals.
Previous studies that have shown disparities in quality of care for the most part have focused on Medicare patients and/or specific diseases (most commonly acute myocardial infarction [AMI] and other cardiac diagnoses), process quality measures, utilization rates, and black-white differences (Barnato et al. 2005; Skinner et al. 2005; Jha et al. 2007; Mukamel, Murthy, and Weiner 2000; Castellanos, Normand, and Ayanian 2009; Epstein, Gray, and Schlesinger 2010; Popescu et al. 2010). Findings from these studies may not be generalizable to a broad range of diseases, nonelderly patients or other minority groups. Also regional variation in hospital care may be important. A previous study suggested that race disparities in hospital services for Medicare patients were confounded by regional variations in hospital care (Baicker et al. 2004). Black Medicare patients with AMI were more likely to use high mortality hospitals in markets with high residential segregation (Vaughan-Sarrazin, Campbell, and Rosenthal 2009). In this study, we addressed the following question: Is hospital quality inversely associated with the percentage of minority patients served by the hospital?
Study Design, Population, and Data
We conducted a cross-sectional analysis using acute care state inpatient discharge (SID) data from 11 U. …