Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner

Feature Story

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Good communication is a key factor in avoiding malpractice suits

Primary care physicians who listen to their patients and use a more friendly manner during visits may reduce the risk that they will be sued for malpractice, according to a recent study funded by the Agency for Health Care Policy and Research (HS07289).

The research team was led by Wendy Levinson, M.D., formerly of Oregon Health Sciences University and Legacy Good Samaritan Hospital and Medical Center, Portland, OR, and now Chief of the Section of Internal Medicine at the University of Chicago. They audiotaped at least 10 routine patient visits each with 59 primary care physicians and 65 general and orthopedic surgeons in Colorado and Oregon. Subjects were randomly selected and were divided into two groups depending on their malpractice claims history. The 1,265 audiotapes were analyzed by coders who did not know whether the physicians fit into the "claims" or "no-claims" group. The study found significant differences in communication styles between primary care physicians who had been filed against and those who had not. The differences for surgeons were not as clear.

Primary care physicians in the no-claims group spent more time with patients during routine office visits than those in the claims group, an average of 18.3 minutes compared with 15 minutes. In addition to the length of the visit, researchers found several other differences in the interaction between physicians and patients. Primary care physicians in the no-claims group were more likely than physicians with claims to orient patients to the process of the visit by telling them exactly what was going to happen during the office visit. These physicians, who had no malpractice claims filed against them, also asked patients for their opinions and elicited questions and were more likely to use humor and to laugh during an office visit.

Length of visits with surgeons did not have the same association as it did for primary care physicians. Furthermore, the study did not identify differences in communication styles that could predict the malpractice risk for surgeons. Researchers attribute this to several possibilities. First, routine communication may not be as important as how surgeons communicate when things go wrong. Second, communication with their surgeon may not be as important to patients initiating a malpractice claim as other factors, including true negligence, surgical complications, or financial incentives. Third, patients may not look to surgeons for interpersonal communication, but rather technical expertise presented in a more businesslike manner.

Researchers are conducting further quantitative analyses to gain more understanding of the communication patterns of surgeons. In the meantime, they suggest that physicians who improve their communication skills and incorporate these behaviors into routine practice can decrease their malpractice risks.

For more information, see "Physician-Patient Communication: The relationship with malpractice claims among primary care physicians and surgeons," in the February 19, 1997, issue of the Journal of the American Medical Association 277(7), pp. 553-559.

Return to Contents

Patient Outcomes/Effectiveness Research

New medication reduces serious problems and repeat operations following coronary angioplasty

A powerful new antiplatelet agent can substantially reduce major complications and the need for repeat revascularization procedures following initial coronary angioplasty. And despite the cost of additional bleeding problems caused by the medication, it nevertheless recoups most of the cost of the therapy and has the potential to pay for itself, concludes a study supported in part by the Agency for Health Care Policy and Research (HS05636 and HS06503). The agent, c7E3 Fab or abciximab, prevents the clumping together of platelets that can lead to life-threatening blood clots.

A multicenter trial recently reported that use of c7E3 Fab in first-time angioplasty patients resulted in a 35 percent reduction in deaths, heart attacks, and the need for repeat revascularization within 30 days, as well as other clinical events related to abrupt closure of a coronary artery following angioplasty. The medication also reduced by 23 percent subsequent major ischemic (deficient blood flow to the heart) events, such as heart attack or death or the need for revascularization (another angioplasty or bypass surgery) up to 6 months after initial angioplasty.

This economic substudy examined the economic tradeoff between reduced ischemic events and increased major bleeding during initial hospitalization with use of c7E3 Fab. Results showed that a potential cost savings of $622 per patient during the initial hospitalization from reduced acute ischemic events with c7E3 Fab was offset by an equivalent rise of $521 due to increased bleeding episodes. Baseline medical costs for c7E3 Fab therapy averaged $13,577 (exclusive of drug cost) compared with $13,434 for placebo.

During the 6-month followup, c7E3 Fab decreased repeat hospitalization rates by 23 percent and repeat revascularization by 22 percent, producing a mean savings of $1,270 per patient (exclusive of drug cost). With a cost of $1,407 for the c7E3 Fab regimen (a bolus and12-hour infusion), the cumulative net 6-month cost to switch from standard care to routine c7E3 Fab averaged $293 per patient.

These findings are from one part of a prospective study of the multicenter EPIC (Evaluation of c7E3 for the Prevention of Ischemic Complications) trial, which was conducted between 1991 and 1992 and involved 2,038 patients at 56 U.S. hospitals who were at high risk for ischemic complications following angioplasty. The researchers conclude that c7E3 Fab is the first medication demonstrated to improve the clinical outcomes of angioplasty patients, while recouping most of the cost of therapy required to produce these benefits.

For more information, see "Economic assessment of platelet glycoprotein IIb/IIIa inhibition for prevention of ischemic complications of high-risk coronary angioplasty," by Daniel B. Mark, M.D., M.P.H., J. David Talley, M.D., Eric J. Topol, M.D., and others, in Circulation 94(4), pp. 629-635, 1996.

Studies examine recent angioplasty experience by geographical region and patients' sex

Two recent studies, conducted by David J. Malenka, M.D., and other members of the Northern New England Cardiovascular Disease Study Group, compare angioplasty experiences in Northern New England with other regions of the country and look at differences in outcomes between men and women undergoing angioplasty. The two studies, described below, were supported in part by the Agency for Health Care Policy and Research through the Ischemic Heart Disease Patient Outcomes Research Team (PORT, HS06503) project, which is led by Elizabeth R. DeLong, Ph.D., of Duke University.

Malenka, D.J., and the Northern New England Cardiovascular Disease Study Group. (1996). "Indications, practice, and procedural outcomes of percutaneous transluminal coronary angioplasty in Northern New England in the early 1990s." American Journal of Cardiology 78, pp. 260-265.

Ninety percent of patients undergoing coronary angioplasty in Northern New England will have success, that is, the procedure will open up the blocked vessel that is restricting blood flow to the heart. For many patients with coronary artery disease, the procedure is a preferred alternative to medical management or coronary artery bypass graft (CABG) surgery. Less than 1 percent of patients die, and less than 5 percent will need CABG for a failed angioplasty. These results are similar to those of patients in New York and California and add practical clinical data to that of controlled studies, according to the Northern New England Cardiovascular Disease Study Group.

The researchers conducted a prospective study from 1989 through 1993 of the current indications, practices, and procedural outcomes of percutaneous transluminal coronary angioplasty (PTCA) in Northern New England. This procedure involves threading a catheter into the blocked heart vessel to inflate a balloon and flatten plaque against the vessel wall to open up the artery. Thirty-five cardiologists contributed data on 12,232 PTCA patients at hospitals in New Hampshire and Maine, plus one hospital in Massachusetts. Unstable angina (46 percent), stable angina (23 percent), and postinfarction angina (21 percent) were common indications for the procedure.

Overall, 87 percent of patients had one-vessel PTCA, including 66 percent of those with multivessel disease. Ninety percent of the procedures were successful; 88 percent of patients had one or more lesions successfully dilated with no adverse results. The risk of death, nonfatal heart attack, or coronary artery bypass grafting was 6 percent. These results were similar to reports from registries in other regions but different from the National Heart, Lung, and Blood Institute PTCA Registry, which included patients with prior PTCA and more patients with multiple vessel disease.

Malenka, D.J., O'Connor, G.T., Quinton, H., and others (1996). "Differences in outcomes between women and men associated with percutaneous transluminal coronary angioplasty." Circulation 94(Suppl. II), pp. 99-104.

Heart disease is the leading cause of death among women in the United States. Two invasive revascularization procedures, CABG surgery and coronary angioplasty, are often used to open up blocked coronary arteries. It is well known that women are more likely than men to die in the hospital after CABG; a recent study shows that women also are more likely to die following angioplasty.

Even though angioplasty was nearly 90 percent successful in opening up blocked arteries for both men and women, women had 1.6 times greater risk of death following the procedure than men. Also, more women (5.3 percent) than men (4.5 percent) had to undergo bypass surgery or suffered a heart attack following angioplasty.

Women undergoing angioplasty tend to be older and are more likely to be hypertensive and diabetic. These factors account for their excess risk for nonfatal problems after angioplasty, according to the study. But women's elevated risk of death compared with men's risk following angioplasty remained, even after adjusting for these factors.

Whether women's poorer outcomes following angioplasty are partly due to their shorter stature or smaller arterial diameter remains to be determined. Even if women have a higher procedural risk of an adverse outcome, the excess risk is small, and it may be that the long-term success of the procedure is good. Thus, the risk of the procedure may be acceptable, concludes the Northern New England Cardiovascular Disease Study Group. The group prospectively analyzed data on 12,232 patients who underwent angioplasty between 1989 and 1993.

Angiography patients fare better when treated in high-volume hospitals

Coronary angiography permits radiographic visualization of the coronary arteries with the use of contrast material that is injected into the heart via a catheter. Its use to diagnose heart problems in patients with acute myocardial infarction (AMI or heart attack) varies widely among hospitals. But patients treated in hospitals with high angiography use rates, especially patients judged by published criteria to need angiography, fare better than patients treated at hospitals with low angiography rates, according to a study supported in part by the Agency for Health Care Policy and Research (HS08362). These findings suggest potential underuse of this effective procedure, according to the researchers.

They assessed the risk of death from heart disease and of any heart disease event such as another heart attack or rehospitalization within 1 to 4 years in 6,851 patients hospitalized with AMI at 16 Kaiser Permanente hospitals from 1990 through 1992. The investigators found that 89 percent of patients who met the criteria for use of angiography underwent the procedure at hospitals with higher angiography use rates versus 66 percent of patients at hospitals with lower rates. Hospital variation in angiography use was related to on-site angiography capability, not patient demographic or clinical differences, according to the researchers.

After adjustment for patient demographic and clinical characteristics, being hospitalized at a hospital with a higher rate of angiography was associated with a 25 percent lower risk of death from heart disease and a 15 percent lower risk of any heart disease event. For patients who did not meet the criteria indicating a need for angiography, the apparent benefits of being treated at hospitals with higher angiography rates were smaller than for those who met the criteria (15 percent vs. 33 percent lower risk of death and 10 percent vs. 29 percent lower risk of any heart disease event).

See "Variation among hospitals in coronary angiography practices and outcomes after myocardial infarction in a large health maintenance organization," by Joe V. Selby, M.D., M.P.H., Bruce H. Fireman, M.A., Robert J. Lundstrom, M.D., and others, in the December 19, 1996, issue of theNew England Journal of Medicine 335, pp. 1888-1896.

Study questions advisability of shifting care of elderly heart attack patients from cardiologists to primary care doctors

Shifting the care of elderly heart attack patients from cardiologists to primary care physicians may be a mistake, caution researchers supported by the Agency for Health Care Policy and Research (HS06503 and HS08805). They found that elderly heart attack patients who were admitted to a hospital by a cardiologist were 12 percent less likely to die within 1 year than those admitted by a primary care physician. This survival advantage was partly due to cardiologists use of more cardiac procedures and medications that are associated with improved survival.

Patients admitted by cardiologists underwent more diagnostic and therapeutic procedures, had longer hospital stays (8 or 9 days instead of 1 week), and received more medications to treat ischemic heart disease than patients treated by physicians in internal medicine, family medicine, or general practice. Compared with patients admitted by primary care physicians, those admitted by cardiologists underwent more coronary angiography (49 percent vs. 18 to 36 percent), revascularization procedures (22 percent vs. 7 to 14 percent), stress testing (14 percent vs. 9 to 11 percent), nuclear imaging (19 percent vs. 7 to 15 percent), and echocardiography (55 percent vs. 40 to 52 percent). Also, 6 percent more of the patients admitted by cardiologists were considered eligible for thrombolytic (clot-busting) therapy.

Health care strategies that shift the care of elderly heart attack patients from cardiologists to primary care physicians may lower resource use and potentially lower costs, but they also may lead to decreased survival, conclude the researchers. The AHCPR-supported Ischemic Heart Disease Patient Outcomes Research Team (PORT), based at Duke University Medical Center, studied detailed clinical data from the Cooperative Cardiovascular Project (CCP) on 8,241 Medicare patients hospitalized for heart attacks from June through December 1992.

For more information, see "Outcome of acute myocardial infarction according to the specialty of the admitting physician," by James G. Jollis, M.D., Elizabeth R. DeLong, Ph.D., Eric D. Peterson, M.D., M.P.H., and others, in the December 19, 1996, issue of the New England Journal of Medicine 335, pp. 1880-1887.

Canadian experience bolsters evidence for regionalization of cardiac procedures in the United States

Mortality rates for patients in the United States who undergo coronary artery bypass graft (CABG) surgery vary widely, depending on the surgeon performing the procedure and the hospital and region where it is done. This is less true in Ontario, Canada, where the government restricts cardiac surgery to several high-volume hospitals. This difference is probably due to Canada's regionalization of CABG surgery so that experienced hospitals and cardiac surgeons—who perform a large number of annual CABG surgeries—are the only ones performing the surgery, concludes a study supported in part by the Agency for Health Care Policy and Research (HS08464).

Jack V. Tu, M.D., Ph.D., F.R.C.P.C., of the Institute for Clinical Evaluative Sciences in Ontario, and his colleagues on the Steering Committee of the Provincial Adult Cardiac Care Network of Ontario, examined clinical risk factors and surgical outcomes of 15,608 patients undergoing CABG surgery between 1991 and 1993 at the nine hospitals performing adult cardiac surgery in Ontario. The overall in-hospital mortality rate after CABG surgery was low at 3 percent. The amount of interhospital variation in risk-adjusted mortality rates was no greater than that expected by chance alone. All Ontario hospitals performed more than 300 CABG procedures in 1992 and 1993 and only 2 of 42 cardiac surgeons performed less than 50 CABG procedures in 1993.

An American College of Cardiology/American Heart Association task force recommends that all cardiac surgery hospitals perform a yearly minimum of 200 to 300 open heart operations and that all cardiac surgeons perform a yearly minimum of 100 to 150 open heart operations—the majority of which should be CABG operations—because frequency of operations performed has been correlated with patient outcomes. This study provides further evidence that regionalization of CABG surgery and a very low prevalence of low-volume cardiac surgeons have a positive impact on surgical outcomes. The researchers conclude that this approach may avoid the problem of low-volume hospitals and surgeons with higher-than-expected mortality rates found in U.S. regions that have a mixture of low- and high-volume hospitals.

For more details, see "Coronary artery bypass mortality rates in Ontario: A Canadian approach to quality assurance in cardiac surgery," by Dr. Tu, C. David Naylor, M.D., D.Phil., F.R.C.P.C., and others, in the November 15, 1996 issue of Circulation 94(10), pp. 2429-2433.

Low Birthweight PORT publishes recent findings

Nearly 70 percent of all infant deaths and about one-third of all handicapping conditions are associated with low birthweight. Determining the causes of low birthweight and treatments that can prevent it are two goals of the Patient Outcomes Research Team (PORT) on Low Birthweight in Minority and High-Risk Women, which is supported by the Agency for Health Care Policy and Research (PORT contract 290-92-0055) and led by Robert L. Goldenberg, M.D., of the University of Alabama at Birmingham.

Five recent articles by members of the Low Birthweight PORT are summarized here. The first two present findings on pregnancy loss following amniocentesis and factors influencing low birthweight.

Three other articles discuss the following key findings: outcomes of newborns weighing less than 3 pounds improve when the mothers are given a combination of antenatal corticosteroids and tocolytics; elective cesarean delivery of very large babies (over 9 pounds) is medically and economically unsound for nondiabetic women; and mounting a randomized clinical trial to test the effectiveness of maternally administered magnesium sulfate to prevent cerebral palsy in preterm infants would require a huge, multicenter effort.

Wenstrom, K.D., Andrews, W.W., Tamura, T., and others (1996). "Elevated amniotic fluid interleukin-6 levels at genetic amniocentesis predict subsequent pregnancy loss," American Journal of Obstetrics and Gynecology 154(4), pp. 830-833.

The current rate of pregnancy loss following amniocentesis is 0.5 percent. However, women who have high levels of interleukin-6 (IL-6) in the amniotic fluid during the second trimester are likely to be at high risk for postprocedure pregnancy loss that is actually related to preexisting intrauterine inflammation rather than the procedure itself, according to this study. IL-6 is a type of protein secreted in response to infection or inflammation. The researchers measured IL-6 levels in stored second-trimester amniotic fluid and maternal serum samples obtained from women undergoing genetic amniocentesis at the University of Alabama at Birmingham from 1988 to 1995, who experienced spontaneous postprocedure loss within 30 days of amniocentesis, and from normal control women who delivered at term. They found that four times as many women who suffered postprocedure pregnancy loss had high amniotic fluid IL-6 levels (2.5 ng/ml or more) than women who delivered at term (12 percent vs. 3 percent). The researchers conclude that up to 12 percent of pregnancy losses following amniocentesis may result from preexisting subclinical intrauterine inflammation. This inflammation is most likely localized and may not be identified by maternal preprocedure blood levels of IL-6.

Goldenberg, R.L., Cliver, S.P., Mulvihill, F.X., and others (1996). "Medical, psychosocial, and behavioral risk factors do not explain the increased risk for low birth weight among black women." American Journal of Obstetrics and Gynecology 175(5), pp. 1317-1324.

Black infants are more likely to be born smaller and preterm than white infants. However, a recent study shows that low-income white women have more risk factors for low birthweight and premature birth than low-income black women; these risk factors include smoking, low prepregnancy weight, and low educational level. Nevertheless, black infants weighed about 200 gm less and were born an average of 4 days earlier than white infants; 17 percent of black infants were born preterm compared with 11 percent of white infants. These findings were based on a sample of 1,491 pregnant black and white women who were enrolled for prenatal care between 1985 and 1988. PORT researchers compared various demographic, medical, environmental, and psychosocial risk factors with birth outcomes. By many criteria the black women in this study had fewer risk factors than white women—they were better educated, smoked half as much, reported less frequent use of illicit drugs, generally reported better housing, scored better on psychological testing, and were heavier than the white women. These factors, as well as the increased hypertension among black women, only explained about one-third of the black-white differences in birthweight. Perhaps powerful but yet unknown risk factors more prevalent among black than white women, or some predisposition to both preterm and low birthweight infants among black women, account for their poorer birth outcomes, conclude the researchers.

Piper, J.M., Atkinson, M.W., Mitchel, Jr., E.F., and others (1996). "Improved outcomes for very low birth weight infants associated with the use of combined maternal corticosteroids and tocolytics," in The Journal of Reproductive Medicine 41(9), pp. 692-698.

In this study, Low Birthweight PORT researchers found that infants weighing 500 to 1,500 g (about 1 to 3 pounds) at delivery, whose mothers received combined treatment with tocolytic drugs to inhibit preterm labor contractions and corticosteroids to hasten fetal growth, had better outcomes than infants born to women who received no treatment. These infants had one-third the odds of neonatal and infant death and about half the risk of neonatal seizures (an important source of morbidity) as infants whose mothers received no treatment (the most common approach in Tennessee during 1989 and 1990). While steroids alone were associated with improved maternal outcomes, the use of combined therapy was associated with better neonatal outcomes than use of corticosteroids alone or tocolytics alone. For infants whose mothers received both treatments, the brief additional time in utero gained by tocolysis apparently provided a window of opportunity for the corticosteroids to act. These findings were based on a retrospective study conducted on all infants born in Tennessee in 1989 and 1990 who weighed less than 1,500 g at birth and had no serious malformations.

Rouse, D.J., Owen, J., Goldenberg, R.L., and Cliver, S.P. (1996, November). "The effectiveness and costs of elective cesarean delivery for fetal macrosomia diagnosed by ultrasound." Journal of the American Medical Association 276(18), pp. 1480-1486.

About 97 percent of pregnant women are not diabetic. For these women, elective cesarean delivery of large babies weighing about 9 pounds or more (4,000 g or larger) based on ultrasound estimates is medically and economically unsound, conclude the Low Birthweight PORT researchers. Elective cesarean delivery of very large infants has been recommended to avoid problems extracting the shoulders (shoulder dystocia) during vaginal delivery, which can result in fractures and neurologic injuries. The researchers compared three policies: management without ultrasound, ultrasound and elective cesarean delivery for estimated fetal weight of 4,000 g or more, and ultrasound and elective cesarean delivery for estimated fetal weight of 4,500 g (about 10 pounds) or more. They found that for each permanent injury to the brachial plexus (a network of nerves that innervate the muscles and skin of the chest, shoulders, and arms) prevented by the 4,500-g policy, 3,695 cesarean deliveries were performed at an additional cost of $8.7 million compared with 2,345 cesarean deliveries and an additional $4.9 million with the 4,000-g policy. Rates of cesarean delivery, shoulder dystocia, brachial plexus injury, and total costs were higher for diabetic women with all three policies than for nondiabetic women. However, diabetic women had more favorable ratios for both cesarean deliveries and cost per permanent injury avoided: 443 deliveries and $930,000 with the 4,500-g policy, and 489 deliveries and $880,000 with the 4,000-g policy. The researchers conclude that in pregnancies complicated by diabetes, elective cesarean delivery of very large babies appears to be more tenable, although its merits remain debatable.

Rouse, D.J., Hauth, J.C., Nelson, K.G., and Goldenberg, R.L. (1996). "The feasibility of a randomized clinical perinatal trial: Maternal magnesium sulfate for the prevention of cerebral palsy." American Journal of Obstetrics and Gynecology 175(3), pp. 701-705.

Premature babies have a higher risk than other newborns of developing cerebral palsy, a neurologic disorder affecting motor function that occurs in about 2 per 1,000 live births. Because survival of extremely premature infants has increased dramatically in recent years, the prevalence of cerebral palsy has also risen. A few studies have indicated that intrapartum administration of magnesium sulfate to the mother to delay preterm labor and delivery may reduce subsequent risk of cerebral palsy in surviving preterm offspring. Although these findings are promising, a randomized clinical trial would be most conclusive from the standpoint of establishing a causal relationship between maternal intrapartum magnesium sulfate therapy and a decreased risk of cerebral palsy in surviving preterm neonates. This study found, however, that such a randomized trial would be a formidable undertaking, requiring a concerted multicenter effort. The research team based its conclusion on a literature review used to calculate the cumulative rates of delivery, neonatal survival, and cerebral palsy for progressively longer gestational age intervals beginning at 24 weeks gestation. The researchers calculated that in the 4-week interval from 24 to 28 weeks of gestational age, the minimum number of neonates required to conduct a randomized trial would be 1,189. But with loss to followup of neonates, maternal ineligibility, and failure of enrollment, about 1 million pregnancies would be necessary to generate the necessary number of eligible pregnant women to conduct the study.

Prostate cancer screening debate is analogous to early lung cancer screening debate

Evidence from randomized controlled trials addressing whether screening for prostate cancer actually reduces prostate cancer deaths may not be available for 10 to 15 years. In the meantime, clinicians debate whether to screen now and stop later if prostate cancer screening proves ineffective or does more harm than good, or to not screen now and start later if trials prove that screening does more good than harm. This problem is similar to the early debate surrounding lung cancer screening, note Mary McNaughton Collins, M.D., and Michael J. Barry, M.D., of Massachusetts General Hospital. They reviewed published articles on lung and prostate cancer screening and found many parallels.

Advocates of early detection of prostate cancer with digital rectal examination and prostate-specific antigen (PSA) tests have compared prostate cancer screening with the effective strategy of breast cancer screening, implying that prostate cancer screening should similarly reduce cancer mortality. They also have cited the high burden of prostate cancer, the acceptable accuracy of digital rectal examination and PSA tests, a stage shift to more localized and potentially curable cases among men screened, and the theoretical curability of early-stage disease as sufficient reasons to proceed with screening.

These arguments are reminiscent of early arguments in favor of lung cancer screening with chest x-ray examination and sputum cytology, a practice ultimately proven ineffective in clinical trials. Lung cancer screening advocates pointed enthusiastically to preliminary reports documenting a shift toward localized, potentially curable lung cancer in screened patients relative to controls (about 50 percent vs. 20 percent). But despite the stage shift, subsequent clinical trials did not show a reduction in lung cancer deaths but did report problems and deaths following lung resection for some of those screened. For this reason, the authors caution against widespread dissemination of prostate screening without evidence from randomized controlled trials showing that the screening does more good than harm. This work was supported in part by the Agency for Health Care Policy and Research (Prostate Patient Outcomes Research Team-II, HS08397).

Details are in "Controversies in prostate cancer screening: Analogies to the early lung cancer screening debate," by Drs. Collins and Barry, in the December 25, 1996, Journal of the American Medical Association 276(24), pp. 1976-1979.

Return to Contents
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care