Why Is There a Quality Chasm?

Joseph P. Newhouse


Health Affairs. 2002;21(4) 

In This Article

Inefficiency in Translating Medical Care Services into Health

One of the first indications of inefficiency came from the vast literature on geographic variations in use of treatment, with an implied assumption of little or no variation in outcomes. U.S. variations are now presented in great and colorful detail in the well-known Dartmouth Atlas.[6] And such variations are not confined to the United States.[7]

Also, these variations cannot be accounted for by differences in health status.[8] Although variations in patients' preferences and factor prices have been less studied, it seems implausible that they could differ by enough to explain the magnitude of differences in utilization cross areas. As result, the usual interpretation of this variation is that many, perhaps all, areas are producing health inefficiently.

The early health services research literature tended to assume that the low-rate regions had it right. In the mid-1980s, however, Mark Chassin and others showed that this presumption was incorrect and in so doing brought forth more compelling evidence of inefficiency than had hitherto existed. Chassin and his colleagues defined procedures to be appropriate if the "expected health benefits of procedure exceed its expected negative consequences by sufficiently wide margin that the procedure is worth doing."[9] Conversely, inappropriate procedures had little or no expected benefit -- or even negative benefit. Physician panels assigned appropriateness ratings for patients with varying indications, and then information from sample of charts was used to ascertain each patient's indications.

The initial studies showed that sixth to third of the procedures performed were inappropriate. An additional number were equivocal. These magnitudes certainly suggest substantial problem. And the bad news was not limited to the United States. In the Trent region of the United Kingdom, for example, the rate of inappropriate coronary angiography was 51 percent, and the rate of inappropriate coronary artery bypass graft (CABG) was 42 percent. In four Israeli hospitals the rate of inappropriate or equivocal cholecystectomy was 29 percent.[10] High rates of inappropriateness have not been found in all studies, but they do predominate. Moreover, an economist would find even higher rates of inefficiency, since the economist would consider medical care whose marginal benefit was positive but less than marginal cost to be inefficient, whereas the Chassin definition considers such care to be appropriate.

By setting normative standard rather than simply observing that every region could not have it right, the studies of appropriateness strengthened the inference of inefficiency from the variations literature. Moreover, the studies by Chassin and others tended to find similar rates of inappropriate care among areas with widely varying overall procedure rates, implying underuse in low-rate areas and overuse in high-rate areas.

Later and larger studies produced more evidence of poor quality, namely variation cross states in measures of proper care for given condition. Not only did states at the tenth and ninetieth percentiles exhibit considerable spread, the median state was distressingly far from 100 percent ( Table 1 ). Moreover, these data, as well as the variation shown in the Dartmouth Atlas, almost certainly understate the problem, because they do not account for within-area variation.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.