The public's perception of quality hospitals II: Implications for patient surveys

Because there is growing consensus that monitoring quality care should be based, at least in part, on patients' perceptions (Davies and Ware 1988), this article expands earlier findings on how the public perceives the quality of hospitals and hospital care (Boscarino 1988b). In this study, the public's overall quality ratings of 155 short-term medical and surgical hospitals are analyzed by the type, size, staff ratio, mortality rate, case mix, and location. The hospitals in the study represented a national cross-section of institutions, with the results based on 20,000 adults surveyed in 40 U.S. market areas. Initial analysis shows that the public rates nonrural, larger, tertiary care, teaching, higher-patient census, better-staffed, and lower-mortality facilities higher in overall quality. Hospitals that are located in the Midwest or West, have higher average employee salaries, and that are more costly are also perceived to be of a higher quality. A multiple regression analysis reveals that combined these variables account for 50 percent of the public's quality perception, with the most important being tertiary care level, patient-census level, average employee salary, and teaching status (all positively related to higher quality). Using these variables in a discriminant function analysis, hospitals with high-perceived quality can be correctly identified 80 percent of the time. It is suggested that these findings have major significance for monitoring the quality of care, based on patients' perceptions. A practical model for doing this, one that minimizes patient biases and incorporates medical outcomes, is described in detail.