Do blacks, Hispanics, and the poor get poorer medical care?

The intuitive answer is yes, and in fact previous literature (see this paper, for example) seems to support this assertion. But a new article in the New England Journal of Medicine claims to show the opposite. Say the authors of the new article: "Previous studies have focused on a narrow set of quality indicators and conditions in selected populations and have had a limited ability to adjust for the range of factors associated with poorer quality. Few studies have examined quality across the continuum of care for multiple conditions."

The paper by Asch et al., then, seeks to estimate whether the care actually received by the subjects in their study population correspond to a set of indicators of ideal care. The indicator set is available here as a PDF, in a working paper from RAND (the authors' institution).

Surprisingly, the authors estimate that the population overall received 55 percent of recommend care -- that is, we're all not getting the care we should be. Even more surprising was the finding that quality of care varied only slightly among racial and socioeconomic groups.

The authors conclude, "Quality-improvement programs that focus solely on reducing disparities among sociodemographic subgroups may miss larger opportunities to improve care. "

This is probably the case. But their results should not be accepted uncritically, for two important reasons.

1. Abysmal response rate. The authors included in the study "all persons for whom [they] had at least one record (37 percent of the eligible sample). " Of course, the authors performed sensitivity analyses (in lay terms: estimating what difference the response rate might have had on the calculated results) and constructed models which controlled for non-response. They say, for example, "For nonresponse by blacks [in this study] to have produced the magnitude of disparities commonly found in previous studies, nonrespondent blacks would need to have had implausibly low (near zero) quality scores." Whether this is implausible is open to question, I think.

The study also excluded people who don't have a telephone and who saw no health-care provider during the previous two years. Obviously, these are the poorest of the poor.

2. Quality by what measure? The quality indicators used in this study come from a variety of sources: randomized trials together with consensus statements by professional societies. Furthermore, we don't know whether these quality indicators correspond to decreases in morbidity or mortality.

Despite these weaknesses, this is a highly important result. And I can comment, based on my limited personal experience, that it rings true. Errors are widespread. I have seen the poor get the same care as the very rich, and I have seen both failed by the system.