What Determines Quality Health Care, and How can we Measure It?

Summary of working paper 7327
Featured in print Digest

After adjusting for other differences between market areas, the mortality measure indicates that for-profit hospitals offer better quality.

Issues of performance and quality in the delivery of health services are critical to assessing the effectiveness of different health care providers and to evaluating different health policies. However, policymakers and care providers have not had adequate measures of quality available because of the inherent problems in gathering timely and relevant data, the fact that medical care has so many dimensions, and because so many factors aside from provider quality affect patient outcomes.

In The Quality of Health Care Providers (NBER Working Paper No. 7327), authors Mark McClellan and Douglas Staiger describe a systematic approach to assessing hospital quality which overcomes many of the key limitations of existing methods. First, it reduces measurement costs by isolating particular combinations of indicators that provide a truer measure of quality at a hospital. Second, it performs better than alternative approaches in reducing the problem of "very noisy" measurements resulting from small patient samples and large numbers of factors outside the health care system affecting patient outcomes. Finally, the authors' approach allows them to make better predictions about the current and future quality of providers than existing methods do.

McClellan and Staiger apply their approach to an analysis of heart disease in elderly Americans, the successful treatment of which is highly dependent on quality care. They find that differences across hospitals in short-term mortality rates following a heart attack, adjusted for patient demographics, are excellent indicators of the quality of care. Those rates vary dramatically across hospitals, with differences that are persistent over time, and are highly correlated with alternative indicators of quality.

Using the same methodology and data on patient outcomes from 1984 to 1994, the authors attempt to compare quality between for-profit and not-for-profit hospitals. In Comparing Hospital Quality at For-Profit and Not-for-profit Hospitals (NBER Working Paper No.7324), they observe that, on average, for-profit hospitals have higher mortality among elderly patients with heart disease, and that this difference has grown over the past decade. Not-for-profit hospitals perform better with elderly patients with heart disease, even after adjustments are made for differences in hospital size, teaching status, whether the hospital is in a city, and patient demographics. Much of this difference is attributed to the location of the for-profit hospitals, though.

For-profits tend to locate in areas with higher costs and worse outcomes. Not controlling for these area effects, it would look like they tend to have somewhat higher costs and worse outcomes -- an error commonly made in the literature. Within the areas where they locate, they tend to do a little better than average on both.

In other words after adjusting for other differences between market areas, the mortality measure indicates that for-profit hospitals offer better quality. More striking are the differences in mortality rates within each hospital ownership type.

The authors examined three market areas with varying for-profit/not-for-profit hospitals within them, resulting in a sample of 3,718 hospitals. Overall, the authors' analysis suggests that factors other than profit status may be the main determinants of quality care in hospitals.

-- Lester A. Picker