HIV/AIDS News & Information

Results
We mailed 1,030 questionnaires and received responses from 710 (70%) of 1,009 eligible subjects. We excluded 21 subjects who moved without leaving a forwarding address, died, or were no longer in practice. Respondents were predominantly white males with busy practices (Table 2). Respondents had experience with managed care, fee-for-service reimbursement (Table 3), capitation, utilization review, preauthorization by telephone, HMO review, and review by a committee of colleagues. Groups of physicians randomized to receive different versions of the questionnaire did not differ significantly in characteristics.

TABLE 2—Characteristics of Respondents According to Financial Incentive and Utilization Review Depicted in Questionnaire*

* There were no statistically signficant differences between groups,
! Values given as the mean ± SD.

TABLE 3— Percent of Physicians who Experience Each Method of Reimbursement and Type of Utilization Review in Their Practice

The percentage of physicians stating that they would order an indicated test or referral varied from scenario to scenario (Table 4). The more restrictive the financial incentive, the less likely physicians were to state that they would order the indicated test or referral. In the case of the patient with low back pain, 86% of physicians randomized to full capitation scenarios said that they would order the test, compared to 94% of physicians randomized to fee-for-service scenarios. Among physicians randomized to the full capitation scenario, 51% said that they would order a referral to a gastroenterologist for the patient with dyspepsia, compared to 61% of physicians randomized to fee-for-service scenarios. We found similar, but not statistically significant, trends toward fewer services in the other scenarios by physicians randomized to capitation.

Managed care utilization review did not significantly decrease the percentage of physicians who said they would order tests or referrals (Table 5), although a trend in that direction was found. We found no association between physicians stating that they would order the indicated test or referral and their actual experience with utilization review.

TABLE 5— Percent of Physicians Indicating They Would Order the Test or Referral According to Utilization Review

*P=.06

We analyzed responses to the scenarios by the combination of incentives to detect any interaction between them (Table 6). For the scenarios in which the trends were statistically significant, only 84% of physicians randomized to the most restrictive managed care scenarios of full capitation and preauthorization by telephone said they would order the indicated x-ray for low back pain, compared to 94% of physicians randomized to fee-for-service scenarios. Only 50% of physicians randomized to the most restrictive scenario said they would refer the patient with dyspepsia for endoscopy compared to 61% of physicians randomized to the least restrictive scenarios. female pink viagra

TABLE 6.—Percent of Physicians indicating They Would Order the Test or Referral According to Combination of Financial Incentive and

*Р < .005
! P = .03

Discussion
The medical community has shown concern that incentives in managed care will lead physicians to withhold medically indicated services, thereby compromising the quality of care and eroding patient trust.

Our survey results suggest this concern is warranted, but should not be overstated. Compared to fee-for-service, capitation and utilization review caused a small but consistent decrease of 3% to 11% in physicians stating that they would order indicated tests and referrals. Because of our study’s randomized design, we conclude that this difference, though modest, resulted from the financial incentives and utilization review depicted on our questionnaire, and not from factors such as demographics, practice characteristics, or experience with managed care; all factors were balanced in the randomized groups.

We also found, under various types of financial incentives and utilization review, that many physicians would not order indicated tests and referrals. Even among those physicians randomized to fee-for-service scenarios, 6% to 39% failed to say that they would order the indicated services. cialis super active

These results provide modest support for the concerns that conflicts of interest may lead physicians to fail to act in their patients’ best interests. The incentives used in our study led to only a small amount of under-utilization, not gross lapses in the quality of care. Even this level of underutilization, however, matters for patients. For instance, a physician who fails to order an indicated exercise treadmill test for a woman with atypical angina could miss the diagnosis of significant coronary artery disease and fail to prevent a myocardial infarction or death.

Studies of actual practice suggest that our findings may underestimate the true level of underutilization. In a study that closely parallels our scenarios, salaried physicians in a staff-model HMO underutilized x-rays for patients with low back pain by 71%. Adherence to preventive guidelines ranges from 50% for the National Cholesterol Education Program guidelines to 15% to 26% for mammography in an inner-city population.
Canadian Pharmacy

Several reasons can explain why our findings may not reflect true rates of underutilization and the influence of incentives. Our subjects, ACP members of whom most are board certified in internal medicine, may be more knowledgeable about guidelines than a general sample of primary care physicians. Also, physicians are more likely to say that they would order the test or referral on a questionnaire than in real practice if they realize the “correct” answer. A physician’s decision regarding a given intervention may be affected by the source of the guideline, the invasiveness of the intervention, the need to involve another physician, and the implications of a missed diagnosis. Although our study was not designed to examine this issue directly, our results suggest that the quality of care might be improved by focusing as much, if not more, attention on encouraging physicians to adhere to reliable guidelines, rather than simply working to eliminate the effects of incentives in managed care. In fact, the level of underutilization we found for fee-for-service scenarios suggests that unless we improve adherence to reliable guidelines in all systems of care, the quality of care may suffer even in systems without any incentives to withhold services.

Our study has several limitations. Our results may not be generalizable to physicians who work in staff-model HMOs or government institutions, or to physicians who practice in areas of low, managed care penetration. We chose areas of the country with at least 30% penetration of managed care to insure that physicians would have at least some exposure and experience with the incentives depicted. Also, using different physicians as a pretest population, different guidelines or different scenarios may have affected our findings. Omnicef 125 5

Conclusions
A major ethical concern about incentives in managed care is that incentives may lead physicians to fail to act in the patient’s best interest. Within the limitations of our study, we found some evidence that such incentives may have a small but real influence, leading physicians to withhold indicated services. This excess underutilization, though modest, may undermine patient trust in physicians. We also found that, regardless of incentives, a significant number of physicians failed to state that they would order indicated services. Poor adherence to guidelines also threatens the quality of care. Our findings underscore the need for better methods for detecting underutilization. Programs are also needed to ensure that physicians provide indicated services to their patients. These steps could mitigate ethical concerns regarding incentives, improve the quality of care, and bolster patient trust.