Health Care Quality and Utilization

Physician claims can provide clues to causes of postoperative complications

Physician claims submitted to Medicare could be used by professional review organizations (PROs) and others to identify hospitals or surgeons who have a high complication rate for surgeries such as carotid endarterectomy, according to a recent study by the Stroke Prevention Patient Outcome Research Team (Stroke PORT).

Led by David B. Matchar, M.D. of Duke University, the PORT investigators extracted physician claims data from a random 20 percent sample of Medicare patients undergoing carotid endarterectomy (surgical removal of plaque deposits from the carotid artery) between January 1, 1991 and November 31, 1991. The study, which involved claims for 8,345 patients, was supported by the Agency for Health Care Policy and Research (contract 282-91-0028).

The Stroke PORT researchers found that by following claims for certain services indicative of complications, they could identify those patients who had problems following carotid endarterectomy—approximately 10 percent of all patients undergoing the procedure. Characteristics of these patients, including their costs and outcomes, were compared with the characteristics of patients whose records implied that they underwent the surgery without complications.

Results indicated that patients who developed complications were not only more likely to die within a month but also were eight times more likely to be discharged to a rehabilitation hospital or nursing home than to their homes (25 percent vs. 3 percent) and three times more likely to be readmitted to an acute care hospital (27 percent vs. 8 percent). Few differences in complication rates were attributable to the patient's age, sex, race, or medical condition; the rural or urban location of the hospital; or the surgeon's specialty.

These findings suggest that the identity of the hospital and the operating surgeon may be key factors in predicting the patient's chance of recovering from carotid endarterectomy without complications. Lead author Janet Mitchell, Ph.D., cautions, however, that the use of claims data as a screening tool for complications should be clinically validated before its use becomes widespread.

For more information, see "Using physician claims to identify postoperative complications of carotid endarterectomy," by Dr. Mitchell, David J. Ballard, M.D., Ph.D., Jack P. Whisnant, M.D., and others, in the June 1996 issue of Health Services Research 31(2), pp.141-152.

Current health, race, and age predict future prescription drug use

While people typically use more medications as they age, a person's current health and race are also important predictors of later medication use, according to a new study supported in part by the Agency for Health Care Policy and Research (HS07819). It shows further that elderly blacks are less likely than elderly whites to take prescription drugs, even though previous studies have reported that blacks are more apt to be given a drug prescription. This may reflect poorer communications between typically white physicians and their black patients, greater reliance of black patients on nontraditional healing methods (for example, faith healing and folk remedies), noncompliance, or inappropriate treatment of disease, notes Joseph T. Hanlon, Pharm.D., of Duke University Medical Center.

Information was gathered through Duke University's Established Populations for Epidemiologic Surveys of the Elderly (EPESE) on 1,778 elderly black and 1,446 elderly white North Carolina community residents in 1986-1987 and 1989-1990. Dr. Hanlon and colleagues then designed a model of health service use to analyze which factors would predict changes in medication use 3 years later. About 74 percent of these elderly persons were taking prescription medications at the outset, compared with about 79 percent 3 years later, a statistically significant increase. The average number of prescription drugs taken increased from 2.12 to 2.49. Older persons were more likely than younger ones to become users of prescription drugs and to increase the number of drugs used, but race and health were better predictors of future prescription drug use than age.

Blacks were less likely than whites to become prescription drug users or to increase the number of these drugs they used. Initial severity of chronic disease and health care visits during the year prior to the first interview also predicted use of prescription drugs 3 years later. The model used was unable to explain why the proportion of nonprescription drug users declined from 73.4 percent to 71 percent, or why the average number of nonprescription drugs decreased from 1.31 to 1.24 during the 3-year period.

For more information, see "Factors predicting change in prescription and nonprescription drug use in a community-residing black and white elderly population," by Gerda G. Fillenbaum, Ph.D., Ronnie D. Horner, Ph.D., Dr. Hanlon, and others, in the Journal of Clinical Epidemiology 49(5), pp. 587-593, 1996.

Elderly private patients are less apt to wait for nursing home admission than those eligible for Medicaid

Medicaid-eligible elderly persons seeking nursing home care are considerably less likely than private payers to gain admission to nursing homes, according to a recent study. It suggests that this difference is due to the low reimbursements nursing homes receive from Medicaid for these patients and State limits on the number of nursing home beds. Based on a statistical model using data from the National Long Term Care Channeling Demonstration, the typical private-pay patient who sought nursing home care was able to gain admission to a nursing home during the 1-year duration of the demonstration. In contrast, the typical Medicaid-eligible person who sought nursing home care had only a 71 percent chance of gaining admission.

The findings suggest that increasing Medicaid reimbursements and relaxing controls on bed supplies would increase access to care for Medicaid-eligible people. Increasing reimbursements to improve access would be expensive, however, and other research suggests that, under some conditions, higher reimbursements may lead to lower quality care, since nursing homes would then have less incentive to compete for higher paying private payers. The study was conducted by James D. Reschovsky, Ph.D., formerly with the Agency for Health Care Policy and Research.

Details are in "Demand for and access to institutional long-term care: The role of Medicaid in nursing home markets," by Dr. Reschovsky, in Inquiry 33, pp. 15-29, 1996. Reprints (AHCPR Publication No. 96-R104) are available from the AHCPR Publications Clearinghouse.

Physician retention continues to be problematic in rural areas

Rural areas typically have fewer health care providers than more urbanized parts of the country and it is difficult for rural communities to keep their physicians. For example, rural areas are rapidly losing obstetric providers. Even retention of National Health Service Corps (NHSC) physicians, after they have paid back their NHSC scholarship obligation, is difficult. Two recent studies supported by the Agency for Health Care Policy and Research (HS06544) demonstrate the influence of rural characteristics on retention of physicians.

The first study shows that black NHSC physicians placed in rural areas were much more dissatisfied with their work and personal lives than other NHSC physicians, primarily because they preferred urban life. Thomas R. Konrad, Ph.D., of the University of North Carolina at Chapel Hill, and coinvestigators surveyed NHSC physicians involved in the program from 1987 through 1990 who had been placed in rural sites. Minority physicians were more apt to prefer access to urban cultural activities and placed less value on living in a small community or an area where they had access to outdoor sports such as fishing and skiing.

Despite similar hours of work, number of nights on call, patient volumes, and incomes, minority physicians had lower work satisfaction scores (2.13 vs. 2.30) and personal life satisfaction scores (2.75 vs. 2.99) than other NHSC physicians. These lower ratings were due almost entirely to the low ratings provided by black physicians. Hispanic physicians did not differ from white physicians in work or personal life satisfaction scores.

In the second study, Dr. Konrad and colleagues examined the migration of obstetrician-gynecologists into and out of rural areas from 1985 to 1990. During the study period, a total of 962 ob-gyns moved out of 531 nonmetropolitan counties and 979 ob-gyns moved into 528 counties. These physicians were more likely to leave rural areas that were near metropolitan counties and that had a low number of hospital beds. A sufficient population to support health care providers and adequate hospital resources were important to retaining physicians.

Of greater potential importance were factors related to State-level policies, such as malpractice premiums, Medicaid reimbursement, or the activity of State offices of rural health. For example, North Carolina, the biggest gainer in the change, made significant changes in its Medicaid program to improve payments for prenatal care and delivery.

For more details on both studies, see "Minority physicians serving in rural National Health Service Corps sites," by Donald E. Pathman, M.D., M.P.H., and Dr. Konrad, and "Migration of obstetrician-gynecologists into and out of rural areas, 1985-1990," by Thomas C. Ricketts, Ph.D., M.P.H., Sarah E. Tropman, M.P.H., Rebecca T. Slifkin, Ph.D., and Dr. Konrad, which appear in Medical Care 34(5), pp. 439-454 and 428-438, respectively.

Return to Contents
Proceed to Next Section