Quality Differences in Managed Care and Fee-for-Services

11/01/1998
Summary of working paper 6523
Featured in print Digest

Managed care plans on the whole may offer lower quality care than fee-for-service plans.

As authors Sarah Feldman and David A. Scharfstein point out in Managed Care and Provider Volume (NBER Working Paper No. 6523), numerous studies have attempted to measure the quality differences between managed care health plans and more traditional fee-for-service coverage. Yet few, if any, definitive conclusions have been drawn regarding the superiority of one type of health-care plan over the other. The matter is of no small consequence, since the number of Americans covered by managed care has quintupled between 1980 and 1995.

Feldman and Scharfstein take a new approach to the issue. The starting point for their work is the large body of research establishing that patients have better clinical outcomes when they are treated by physicians and hospitals with more experience in treating their diseases. The authors' goal is to examine whether patients in managed care plans tend to be treated by higher or lower volume health care providers. This approach could serve as an indirect way of measuring quality differences between managed care and fee-for-service health insurance.

Feldman and Scharfstein use as the basis of their study data on all inpatient hospital stays in Massachusetts in 1995. These data identify the hospitals and physicians providing the care, and the patients' health insurance plans. The data also include clinical and demographic information on the patients. The researchers focus their study on the surgical treatments of three common cancers: breast cancer, colorectal cancer, and gynecologic cancer. They choose these diseases because all three require surgical interventions; as a result, there is not likely to be a significant difference in whether patients in managed care or fee-for-service plans receive such treatment.

The data indeed indicate that managed care cancer patients tend to be treated at hospitals that perform fewer procedures, while the fee-for-service patients tend to be treated at hospitals that perform more procedures. Likewise, managed care cancer patients tend to be treated by surgeons who perform fewer operations than those operating on fee-for service patients. There is, however, substantial variation across managed care plans in provider volume. Patients in some plans are treated by surgeons with more than 40 percent lower volume than the surgeons of fee-for-service plans and at hospitals with roughly half the volume. Other managed care plans seem to be no different than fee-for-service insurance, and patients in one managed care plan are treated at very high volume hospitals (albeit by lower volume surgeons at those hospitals.)

If indeed, as research indicates, volume and quality of hospitals and physicians are related, then the authors say that two conclusions may be drawn from their study. The first, which Feldman and Scharfstein call unsurprising, is that some managed health care plans may be better than others. The second conclusion is that managed care plans on the whole may offer lower quality care than fee-for-service plans.

Precisely why managed care patients tend to be treated by lower volume providers, the authors say, remains undetermined. One leading possibility, they say, is that managed care plans limit patients to using relatively low-volume providers. These patients might not be referred as frequently to specialists for their procedures, or their access to high-volume specialists may be restricted. Such care plans may in fact have contracts only with low-volume community hospitals, and not high-volume teaching hospitals. This is plausible, say Feldman and Scharfstein, because lower-cost community hospitals tend to charge less for their services than higher-cost teaching hospitals.

Feldman and Scharfstein indicate, however, that there may be an alternative explanation of the findings, namely that managed care plans do not limit patient choice, but rather that members of such health plans are less aggressive health consumers. In this light, the authors speculate, even if they had enrolled in a fee-for-service plan, such patients would choose to be treated by lower-volume providers, either because they care less about the quality of their care or are more concerned about costs -- or both. Indeed, patient characteristics do appear to matter in the choice of provider; younger and higher income patients tend to be treated by higher volume surgeons and hospitals.