Thrombolytic therapy found to be as effective as angioplasty in reducing deaths from heart attacks

A recent study shows that heart attack victims treated at community hospitals with primary angioplasty (procedure is done within the first few hours after the attack) are no less likely to die than patients treated with powerful clot-busting (thrombolytic) medications. What's more, angioplasty (surgical insertion of a balloon catheter, which inflates to flatten plaque and open up clogged arteries) ends up costing significantly more over time. Although a few small-scale clinical trials suggest that primary angioplasty is more beneficial than thrombolytic therapy (TT), the findings from these studies have never been replicated in the average community hospital, which is where most Americans are taken when they have a heart attack.

The study was conducted by Nathan Every, M.D., M.P.H., of the University of Washington as part of the Cardiac Arrhythmia Patient Outcomes Research Team (PORT), which is supported by the Agency for Health Care Policy and Research (HS08362) and led by Mark A. Hlatky, M.D., of Stanford University. Dr. Every and his colleagues analyzed data on more than 3,000 heart attack patients treated at 19 Seattle, WA, hospitals between 1988 and 1994 and followed for up to 3 years after discharge.

The research team found no significant differences in short- or long-term mortality between the two groups (5.6 percent for TT and 5.5 percent for primary angioplasty). They did find, however, that TT led to fewer tests and other procedures than did angioplasty. For example, patients treated with thrombolytics had 33 percent fewer angiograms (images of the heart aided by injected dye), 20 percent fewer followup coronary angioplasties, and 14 percent lower costs after 3 years, for a savings of $3,000 per patient. The researchers conclude that nationwide cost savings could be very significant if applied to the nearly 200,000 patients eligible for thrombolysis each year.

For more information, see "A comparison of thrombolytic therapy with primary coronary angioplasty for acute myocardial infarction," by Nathan R. Every, M.D., M.P.H., Lori S. Parsons, Dr. Hlatky, and others, in the October 24, 1996 New England Journal of Medicine, 335 (10), pp.1253-1260.

Return to Contents

Family Health

PORT researchers examine the effects of neonatal intensive care and maternal streptococcal screening on newborns

The Patient Outcomes Research Team (PORT) on Low Birthweight in Minority and High-Risk Women, supported by the Agency for Health Care Policy and Research (contract 290-92-0055), examines ways to prevent low birthweight and improve the outcomes of low-birthweight infants. Led by Robert L. Goldenberg, M.D., of the University of Alabama, PORT researchers recently published findings on the importance of level III neonatal intensive care units (NICUs) for reducing deaths among high-risk infants and on the impact of streptococcal screening of pregnant women on pediatricians' treatment of potentially exposed infants.

Phibbs, C.S., Bronstein, J.M., Buxton, E., and Phibbs, R.H. (1996, October). "The effects of patient volume and level of care at the hospital of birth on neonatal mortality," Journal of the American Medical Association 276(13), pp. 1054-1059.

High-risk infants born in hospitals with advanced neonatal intensive care units (level III NICUs) have a 38 percent lower death rate than similar infants born in hospitals without NICUs, concludes this study by Low Birthweight PORT researchers and colleagues. The high volume of newborns treated at these NICUs and the full range of specialized neonatal care provided, including surgery, are the keys to improved infant survival, according to the researchers. The study examined the effects of NICU patient volume and level of NICU care available at the hospital of birth on neonatal mortality by examining the outcomes of 53,229 infants classified as likely NICU admissions who were born in non-Federal hospitals in California in 1990.

Investigators found that after adjusting for birthweight, complicating diagnosis, and demographic factors, the risk of death was 38 percent lower for infants born in hospitals with a level III NICU which treated at least 15 newborns a day than for infants born in hospitals without NICUs. However, infants born in hospitals with intermediate level NICUs (II and II+), which provide either limited or full ventilatory support but not the other services provided in a level III NICU, had a comparable risk of dying as infants born in hospitals without NICUs. Also, it was no more costly to care for these at-risk infants in level III NICUs than in level II or level II+ NICUs.

The researchers recommend that urban areas do not add intermediate level NICUs but instead concentrate high-risk deliveries in a small number of hospitals that provide level III NICU care as a way to improve neonatal outcomes without increasing costs. The PORT researchers plan to extend their study to look for associations between obstetric condition and neonatal outcomes to identify those types of obstetric cases that are at highest risk for adverse neonatal outcomes.

Peralta-Carcelen, M., Fargason, C.A., Cliver, S.P., and others (1996, August). "Impact of maternal group B streptococcal screening on pediatric management in full-term newborns," Archives of Pediatric and Adolescent Medicine 150, pp. 802-808.

About 15 to 20 percent of pregnant women carry group B streptococcal (GBS) bacteria, the most common cause of sepsis in newborns in the United States. Strategies for screening pregnant women for GBS have been proposed by the American Academy of Pediatrics (AAP) and the American College of Obstetrics and Gynecology (ACOG). Recently, the Centers for Disease Control and Prevention adopted these strategies with minor modifications. ACOG recommends giving intrapartum antibiotic treatment to all women who have prolonged rupture of membranes, maternal fever, or premature labor (often brought about by intrauterine infection). Women without these risk factors for GBS would not receive intrapartum antibiotics, according to the ACOG recommendations. In contrast, the AAP policy advocates screening all mothers for GBS at 28 weeks gestation. Mothers who have both risk factors and positive screening cultures are given intrapartum antibiotics. In general, screening-positive mothers with no risk factors and screening-negative mothers with risk factors do not receive antibiotics under the AAP approach.

Low Birthweight PORT investigators examined the results of survey data from a national random sample of pediatricians designed to evaluate the impact of the two screening strategies on the treatment decisions of pediatricians caring for asymptomatic, full-term newborns. These screening strategies, especially the one recommended by the AAP, prompted pediatricians participating in the survey to order laboratory tests more frequently and administer antibiotics to treat potential GBS infections in asymptomatic full-term newborns born to mothers with positive GBS screening tests. Pediatricians were more apt to increase their use of antibiotics in response to a positive maternal GBS screen than if maternal GBS status were unknown (62 percent vs. 37 percent, respectively, in risk-factor-positive mothers). Use of intrapartum antibiotics increased the number of pediatricians who reported that they would prescribe antibiotic therapy for the newborn in risk-factor-positive mothers with unknown results of GBS screening. When maternal risk factors were absent, maternal intrapartum treatment had little impact on pediatric practice.

Disadvantaged mothers and children benefit from linking primary care practice and public health nursing

Socially disadvantaged children typically face poverty, stress, and inadequate stimulation, as well as limited access to health care, all leading to increased health risks. Linking primary care and public health outreach may improve the quality and effectiveness of care for these children, according to a study supported by the Agency for Health Care Policy and Research (HS07106). It shows that poor families with pregnant women and newborn infants, who were encouraged to seek care at primary care practices and received visits from public health nurses, were more likely than other similar families to have had a prenatal visit with a pediatrician, a primary care office as their regular source of sick care, and less waiting time for care. They were also more apt to have received patient education materials.

Linking primary care and public health efforts can increase the availability and continuity of care, conclude Peter A. Margolis, M.D., Ph.D., and his colleagues at the University of North Carolina at Chapel Hill. They conducted a randomized trial in two poverty-stricken North Carolina counties in 1993. The study tested the feasibility of linking home visits by public health nurses with assistance to primary care offices in the delivery of preventive services. It included 93 Medicaid-eligible, first-time pregnant women in their third trimester and their subsequently born infants, who were followed until they were 6 months of age.

The researchers compared the impact of linked home visits with office assistance alone or usual care. Home visits every 1 to 2 weeks by public health nurses provided parental education and social support and linked families with needed community resources. Women in the "usual care" group were not given information about where to seek health care for their child, an omission that was not unusual in the poor North Carolina counties studied.

For details, see "Linking clinical and public health approaches to improve access to health care for socially disadvantaged mothers and children," by Dr. Margolis, Carole M. Lannon, M.D., M.P.H., Rachel Stevens, R.N., Ed.D., and others, in the August 1996 Archives of Pediatric and Adolescent Medicine 150, pp. 815-821.

Return to Contents

Patient Outcomes/Effectiveness Research

Uterine and prostate cancer survival rates vary substantially by region

A person's chance of survival following treatment for localized cancer often depends on the region of the country where he or she receives treatment. This is especially true for cancers of the prostate and uterus, according to a study supported by the Agency for Health Care Policy and Research (HS06879). For example, 5-year survival following diagnosis of prostate cancer is nearly 33 percent higher in Hawaii compared with Connecticut. Yet there is almost no variation from one region to another in 5-year survival following diagnosis of cancers of the stomach, lung, bladder, and ovary, and only modest regional differences in survival for patients with cancer of the colon, rectum, or breast. Surgery is the usual treatment for all of these localized cancers, explains the study's principal investigator Jonathan M. Samet, M.D., M.S., of The Johns Hopkins University.

Dr. Samet, along with colleagues Diana C. Farrow, Ph.D., and William C. Hunt, M.A., analyzed regional survival rates of persons diagnosed with localized cancer of the stomach, colon, rectum, lung, breast, uterus, ovary, prostate, or bladder during 1983-1991. Their analysis was based on data from nine regional cancer registries of the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program. Five-year survival for prostate cancer varied from 52 percent in Connecticut to 64 percent in Seattle and for uterine cancer, from 73 percent in Connecticut to 84 percent in Hawaii. Yet 5-year survival for stomach cancer ranged from only 10 percent in Utah to 15 percent in New Mexico. For five of the nine cancers investigated, 5-year survival rates were higher in Hawaii than in any other SEER area.

These regional differences in survival rates for all cancers, except rectal cancer, persisted even after adjustments were made for the patient's age, sex, cancer stage, and treatment; race was not a variable, since all patients in this study were white. The proportion of patients treated surgically varied across SEER areas for every cancer site examined.

These regional differences in cancer survival could be due to differences in patient characteristics, such as coexisting medical conditions, socioeconomic status, or treatment compliance; regional differences in availability of health care; or the quality of care delivered, according to the researchers. They conclude that, although regional survival differences are statistically significant, they are nevertheless modest from the standpoint of the practicing clinician.

See "Regional variation in survival following the diagnosis of cancer," by Drs. Farrow, Samet, and Hunt, in the Journal of Clinical Epidemiology 49(8), pp. 843-847, 1996.

Return to Contents

Managed Care/Primary Care

Study links decline in physicians' income to managed care

Managed care may be one factor precipitating the recent drop in physicians' income, suggests a recent study, which shows a 4 percent slip in the average physician's income, from $195,300 in 1993 to $186,600, nearly $9,000 less, in 1994. Fee constraints by Medicare and Medicaid are another factor. This drop in physicians' earnings comes after decades of strong income growth for doctors and affects specialists more than generalists, concludes the study, which was supported by the Agency for Health Care Policy and Research (HS09210).

Managed care systems typically place limits on what they pay doctors, and they closely monitor use of health resources, including referrals to specialists. Managed care seeks to increase the cost-effectiveness of care and, at the same time, control spiraling health care costs, according to the study's authors, Carol J. Simon, Ph.D., Associate Professor at the University of Illinois at Chicago and Patricia H. Born, Ph.D., of the American Medical Association (AMA). The researchers used data from an AMA survey of 4,000 non-Federal physicians across the country. These data show that physicians' average net income has been rising continuously since 1985, except for a 1-year break in 1990. Average earnings rose from $154,800 in 1985 (1994 dollars) to $195,300 in 1993 before dropping in 1994. The 1994 income data may be the first evidence that managed care has had a widespread effect on physicians' earnings in particular and on health spending in general.

The gap between the earnings of primary care physicians and those of specialists has narrowed, although gains by general/family practice physicians are driving these results. Physicians in general internal medicine and pediatrics suffered sharp income losses in 1994, and internists' incomes have lagged significantly behind those of other primary care physicians for most of the 1990s. Also, hospital-based physicians suffered larger relative losses than subspecialists. These data suggest that managed care has shifted the demand for physician services toward primary care providers, while reducing health service use, fees, or both for all physicians, conclude the researchers.

Details are in "Physician earnings in a changing managed care environment," by Drs. Simon and Born, in the Fall 1996 issue of Health Affairs 15(3), pp. 124-133.

Scope of State laws regulating managed care plans is expanding

Today, almost two of every three privately insured Americans are enrolled in managed care plans, and many States have passed laws recently to regulate these plans. Fred J. Hellinger, Ph.D., of the Center for Organization and Delivery Studies, Agency for Health Care Policy and Research, has completed a study that traces the growth of three types of State laws: one, laws that guarantee managed care enrollees direct access to a specialist, notably, access to obstetrician-gynecologists by women; two, laws that prohibit exclusive contracts between managed care plans and providers; and three, laws that mandate a minimum hospital stay for deliveries.

Proponents of direct-access laws argue that women's health specialists are monitoring and coordinating primary care without increasing care costs. Managed care plans argue that laws forbidding exclusive contracts may adversely affect their ability to obtain price concessions from health care providers, but supporters point out that these laws allow open competition among providers. Many States now mandate a minimum 48-hour hospital stay for mother and child if delivery is normal and the child is healthy. This agrees with recommendations made by the American College of Obstetricians and Gynecologists and the American Academy of Pediatrics. Some managed care plans had previously limited this stay to 24 hours, which was believed by many health care professionals to jeopardize the health of mother and child.

For more information, see "The expanding scope of State legislation," by Dr. Hellinger, in the October 2, 1996, Journal of the American Medical Association 276(13), pp. 1065-1070.

Reprints of the article (AHCPR Publication No. 97-R025) are available from the AHCPR Publications Clearinghouse.

Researchers identify steps for measuring and improving quality in primary care

The increased role of organized health care delivery systems and greater emphasis on cost containment have focused attention on primary care and quality improvement. Both practitioners and organizations are responding to purchasers' demands for better quality and more value for their health care dollar, in part by developing strategies to deliver high-quality, cost-effective care based on scientific evidence.

A recent supplement to the journal Medical Care presents eight papers and a summary detailing the findings from a large, multisite study on quality of care, which was supported by AHCPR (HS03087 and HS05609). The Ambulatory Care Medical Audit Demonstration (ACMAD) Project was a randomized, controlled trial that investigated the impact of quality improvement interventions on primary medical care in 16 group practices (631 practitioners) in Boston. During the period 1978 to 1993, the project measured the cost, feasibility, and effectiveness of ambulatory care quality assurance in these practices for performance related to eight patient-care guidelines.

The researchers analyzed clinical performance data to determine factors affecting performance and the steps that could be taken to improve performance and thus the quality of patient care. Analysis showed that improvements in care occurred for five of the eight guidelines after feedback was provided on performance measures and corrective actions were taken by the clinicians. Physicians generally had a high level of acceptance of externally coordinated quality assurance programs. In fact, fewer than half of the physicians would have preferred quality assurance reviews to be conducted by their own staff. Finally, the study quantifies the sources of variability and bias that affect measures of quality of care and suggests ways to reduce both bias and variability.

Measuring and improving quality in primary care is neither simple nor straightforward, according to Carolyn Clancy, Director of the Center for Primary Care Research and Acting Director of the Center for Outcomes and Effectiveness Research, Agency for Health Care Policy and Research, and James W. McAllister, of AHCPR's Center for Information Technology. Writing in a preface to the journal supplement, Dr. Clancy and Mr. McAllister describe the ACMAD analyses as highly relevant to organizations and practitioners striving to meet external demands for accountability in a competitive environment.

For more details, see "Quality measurement and improvement among primary care practitioners: Results from a multisite, randomized, controlled trial," edited by R. Heather Palmer, M.B., B.Ch., S.M., and Lee J. Hargraves, Ph.D., a supplement to Medical Care 34(9), pp. S1-S113, September 1996.

Return to Contents

Medical Liability

Medical experts commonly disagree about the cause of "medical injury"

Medical malpractice cases frequently hinge on whether or not medical experts agree that the injury was caused by medical error or negligence rather than other factors. However, a recent study shows that medical experts commonly do not agree on whether a patient's problem, whether it is a serious drug reaction or a complication of an illness, is caused by medical care. For example, in 13 percent of medical injury cases reviewed at New York State hospitals, paired physicians vehemently disagreed about the cause of the patient's problem. They were most likely to agree about the cause of wound infections and least likely to agree about events attributed to failure to diagnose or treat the patient's condition. However, more experienced physicians were more apt to agree with one another than less experienced physicians, according to A. Russell Localio, J.D., M.P.H., M.S., of Pennsylvania State University College of Medicine, and his colleagues.

Their findings were based on an analysis by pairs of physicians of a random sample of inpatient medical records at 51 facilities in New York State. On average, 127 paired physicians, working independently, found that a patient's injury was at least "more likely than not" to have been caused by medical management in 18 percent of reviewed cases (2,764 of 15,066). Agreement was greatest for wound infections (0.62), which were covered by specific practice guidelines and were clearly associated with the site and time of surgery. Drug reactions and falls had markedly lower rates of agreement (0.48 and 0.37, respectively). Physicians were least likely to agree that problems were caused by failure to diagnose (0.32) or treat a patient's condition (0.24). These results corroborate previous findings that assessments based on medical records, especially when not guided by practice guidelines or other objective criteria, produce disagreement among physicians on the appropriateness and quality of care.

This research was supported in part by the Agency for Health Care Policy and Research (HS07067). For more information, see "Identifying adverse events caused by medical care: Degree of physician agreement in a retrospective chart review," by Dr. Localio, Susan L. Weaver, M.S., J. Richard Landis, Ph.D., and others, in the September 15, 1996, Annals of Internal Medicine 125(6), pp. 457-464.

Return to Contents
Proceed to Next Section