Medical Effectiveness/Outcomes Research

AMI PORT examines regional variations in heart attack treatments and outcomes and the use of angiography in elderly patients

Each year at least 750,000 Americans have an acute myocardial infarction (AMI). Medicare patients, who account for nearly half of these cases, have a cumulative mortality of 40 percent. The Harvard-based AMI Patient Outcomes Research Team (PORT), led by Barbara J. McNeil, M.D., Ph.D., and supported by the Agency for Health Care Policy and Research (HS06341), recently published the following two studies, which examine regional variations in the use of invasive cardiac procedures and coronary angiography in elderly heart attack patients.

Guadagnoli, E., Hauptman, P.J., and Ayanian, J.Z. (1995). "Variation in the use of cardiac procedures after acute myocardial infarction." The New England Journal of Medicine 333, pp. 573-578.

Physicians in Texas are more likely to perform invasive cardiac procedures on heart attack patients than physicians in New York. However, Texas' high-technology approach does not improve patients' survival or quality of life compared with patients in New York, according to AMI PORT researchers.

They studied elderly Medicare patients with heart attacks who were admitted to 478 hospitals in New York and Texas during 1990 and found that the patterns of treatment for clinically similar patients were quite different in the two States. Coronary angiography, a procedure that images heart blood vessels to detect heart problems, was performed more often in Texas than in New York (45 percent vs. 30 percent). In Texas, the overall rate of coronary artery bypass surgery was 15 percent, and the rate of coronary angioplasty was 15 percent, compared with 13 percent and 7 percent in New York, respectively. Nevertheless, over the 2-year period following a heart attack, the adjusted likelihood of death was lower in New York than in Texas, and 2 years after a heart attack, patients from Texas were 41 percent more likely than New York patients to report angina and 62 percent more likely to be unable to perform activities requiring energy expenditures of 5 or more metabolic equivalents.

Gatsonis, G.A., Epstein, A.M., Newhouse, J.P., and others (1995). "Variations in the utilization of coronary angiography for elderly patents with an acute myocardial infarction." Medical Care 33(6), pp. 625-642.

Regions of the United States vary three-fold in their use of coronary angiography in Medicare patients 65 years of age or older who have suffered a heart attack. Elderly patients who are male, younger (between 65 and 74 years of age), non-black (especially in the Southeast), and live in States with ready availability of angiography, are more likely than other elderly patients to undergo this procedure to detect heart damage following a heart attack, according to AMI PORT researchers.

Coronary angiography uses contrast dye, inserted into the coronary arteries via a catheter, to visualize the arteries. Use of this diagnostic technique was lowest in the Northeastern part of the United States and ranged from 13.8 percent of AMI Medicare patients in Rhode Island to 38.3 percent in Montana. States with more hospitals having on-site angiography capability tended to have higher angiography rates after adjusting for patient characteristics and geographic region. On average, adjusted angiography rates were higher for younger Medicare patients, males, and non-blacks. The odds of blacks receiving angiography compared with whites ranged from a low of 0.41 (less than half the chance) in Mississippi to 0.94 (almost equal chance) in Kansas.

These results show that large variations in the use of procedures can exist, despite uniform insurance coverage and a relatively homogeneous patient population, points out Barbara J. McNeil, M.D., Ph.D., of Harvard Medical School and Brigham and Women's Hospital. Dr. McNeil and her colleagues studied claims and administrative data from the Health Care Financing Administration to identify a group of Medicare patients 65 years of age and older who had a heart attack in 1987. They estimated the probability of angiography during the first 90 days following a heart attack based on patient age, sex, race, and coexisting conditions for patients in 50 States and the District of Columbia.

Recent findings from the Ischemic Heart Disease PORT

Coronary artery disease (CAD) remains the primary cause of death and a major cause of morbidity in the United States. Significant clinical manifestations of CAD are seen in one of every three men and one of every 10 women by the age of 60. CAD causes the majority of cases of ischemic heart disease (IHD) wherein there is an insufficient flow of blood to the heart and consequent high risk of heart attack (infarction). The Ischemic Heart Disease Patient Outcomes Research Team (PORT) is a 5-year study supported by the Agency for Health Care Policy and Research that is designed to assess the effectiveness of various surgical and nonsurgical treatments of ischemic heart disease.

Led by Elizabeth R. DeLong, Ph.D., of Duke University Medical Center, the Ischemic Heart Disease PORT (HS06503) was awarded by AHCPR in July 1990 in an effort to find ways to reduce treatment variations and improve the survival and quality of life of patients suffering from heart disease. The following three studies were published recently by members of the Ischemic Heart Disease PORT.

Burnett, R.L., Blumenthal, J.A., Mark, D.B., and others (1995). "Distinguishing between early and late responders to symptoms of acute myocardial infarction." The American Journal of Cardiology 75, pp. 1019-1022.

It is well known that the earlier persons with heart attack symptoms call for medical help, the better their chances of survival. Why do certain individuals wait hours, while others call within minutes? Certain psychological and situational factors play a role, according to a study by members of the Ischemic Heart Disease PORT. The team analyzed questionnaire responses of 501 hospitalized patients with documented acute myocardial infarction (AMI), who were similar in illness severity and demographic profile. Early responders (204) averaged 20 minutes to request medical assistance, whereas late responders (249) waited an average of nearly 5-1/2 hours before seeking medical help.

Persons who believed their symptoms were serious and involved the heart were more likely to request medical assistance early, whereas those who attributed their symptoms to indigestion, muscle pain, fatigue, or another cause responded later. Early responders also were more anxious or upset when they first noticed symptoms, felt they had little control over the symptoms, and were more comfortable seeking medical assistance than late responders. Late responders were more likely to be unmarried and to experience their symptoms at work rather than elsewhere, suggesting that they may have been more hesitant to call for help from their office than their home.

Public education campaigns to reduce heart attack deaths have focused on how to recognize symptoms and contact the emergency medical system. However, this study suggests that these programs may achieve better results by focusing on emotional factors that may delay calls for help and ways to distinguish heart attack symptoms from indigestion, muscle pain, and other problems.

Mark, D.B., Hlatky, M.A., Califf, R.M., and others (1995). "Cost effectiveness of thrombolytic therapy with tissue plasminogen activator as compared with streptokinase for acute myocardial infarction." The New England Journal of Medicine 332(21), pp. 1418-1424.

AMI is caused by blockage of the coronary arteries and rapid onset of damage to the heart muscle. It is what most people think of as a "heart attack."

The benefit of treating patients with AMI with thrombolytic (clot-dissolving) drugs to open blocked arteries is well established. Accelerated tissue plasminogen activator (t-PA) costs more but results in a higher survival rate than streptokinase, according to a new study by the Ischemic Heart Disease PORT. Accelerated t-PA is given over a period of 1-1/2 hours rather than the conventional 3 hours, and two-thirds of the dose is given in the first 30 minutes.

The research team conducted a cost-effectiveness analysis to compare the value of these two thrombolytic treatments based on information on mortality rates and use of health resources from the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO) study. One year after enrollment, patients who received t-PA had both higher costs ($2,845) and a higher survival rate (an increase of 1.1 percent or 11 per 1,000 patients treated). Based on the projected life expectancy of each treatment group, the incremental cost-effectiveness ratio was $32,678 per year of life saved.

This cost compares favorably with the cost of other therapies judged by society to be worthwhile, according to the PORT researchers. The researchers conclude that the routine substitution of accelerated t-PA for streptokinase in the treatment of the approximately 250,000 eligible patients who have AMI in the United States each year would cost the Nation approximately $500 million each year. However, this treatment would also provide 38,000 additional years of life for patients after AMI.

Hunink, M.G., Wong, J.B., Donaldson, M.C., and others (1995, July). "Revascularization of femoropopliteal disease." Journal of the American Medical Association 274(2), pp. 165-171.

Femoropopliteal disease is a common co-morbid condition in patients suffering from heart disease. It is caused when peripheral arteries in the leg become blocked (occluded), and circulation of blood is significantly reduced. Many patients who suffer from this disorder eventually require amputation of the affected leg.

Two procedures used to restore flow of blood to the leg are percutaneous transluminal angioplasty (PTA) and bypass surgery (BS). Both treatments are generally considered to be effective in treating leg ischemia and occluded leg arteries in patients with peripheral arterial disease. Both procedures have been shown to decrease the number of amputations in these patients.

The optimum revascularization procedure, however, remains debatable. A new study by the Ischemic Heart Disease PORT concludes that angioplasty is the preferred initial treatment in patients with disabling claudication (limp or walk with pain) due to stenosis (narrowing) or occlusion of the femoropopliteal artery in the thigh, and in those with chronic critical ischemia and stenosis. Bypass surgery is the preferred initial treatment in patients with critical ischemia and a femoropopliteal occlusion.

Using data combined from 26 studies on PTA and BS, the investigators constructed a model to simulate a hypothetical cohort of patients receiving these procedures. The model "followed" these patients over time to estimate long-term outcomes of each treatment in terms of mortality, morbidity, patency (degree to which an artery is open), and costs.

PORT researchers estimated life expectancy for 65-year-old men with peripheral arterial disease to range from 2.7 to 7.4 quality-adjusted life years. Analysis showed that for 65-year-old men with (1) femoropopliteal stenosis presenting with either disabling claudication or critical ischemia or (2) femoropopliteal occlusion presenting with disabling claudication, initial angioplasty increased quality-adjusted life expectancy by 2 to 13 months and resulted in decreased lifetime expenditures compared with bypass surgery. For patients with critical ischemia and a femoropopliteal occlusion, initial bypass surgery increased quality-adjusted life expectancy by 1 to 4 months and resulted in decreased lifetime expenditures compared with angioplasty.

Three-rescuer CPR shown to be more effective than standard two-rescuer CPR

Many urban emergency medical systems dispatch the nearest fire engine company to begin "first-response" cardiopulmonary resuscitation (CPR) until paramedics can reach the scene. Three-rescuer CPR should be adopted as the method of choice by first-response engine companies with three or more trained rescuers, concludes a study supported by the Agency for Health Care Policy and Research (HS06094). The study, led by Arthur L. Kellermann, M.D., M.P.H., of Emory University, shows that three firefighters can perform CPR more effectively than two firefighters, when they use a bag-valve-mask device, and the three-person technique is easily learned and remembered.

The researchers compared the performance of Memphis firefighters who were trained in the three-person technique with that of firefighters who received refresher training in standard two-rescuer CPR. Most firefighters use a bag-valve-mask or oxygen-powered resuscitator to provide rescue ventilations in the field, partly due to fear of disease transmission during mouth-to-mouth or mouth-to-mask ventilation. However, unassisted ventilation with a bag-valve-mask is difficult to perform and can result in the victim receiving insufficient oxygen.

The three-rescuer technique overcomes this problem: one rescuer opens the victim's airway and forms a tight mask-to-face seal with both hands. The second rescuer ventilates the victim with the bag-valve device, while the third rescuer performs standard chest compressions. Three-rescuer teams had more success than two-rescuer teams in achieving every action considered crucial for adequate rescue ventilation: proper head tilt (53 percent vs. 25 percent), adequate mask-to-mouth seal (75 percent vs. 37 percent), and delivery of full breaths (67 percent vs. 35 percent). They were able to deliver more breaths per minute and more oxygen per breath than two-rescuer teams, which resulted in an average of 60 percent more oxygen per minute to the victim.

More details are in "Three-rescuer CPR: The method of choice for firefighter CPR?" by Bela B. Hackman, M.D., Dr. Kellermann, Patty Everitt, R.N., and Linda Carpenter, M.D., in the Annals of Emergency Medicine 26(1), pp. 25-30, 1995.

Stroke PORT publishes latest findings

Stroke is a major cause of disability and the third leading cause of death in the United States. Ischemic stroke and transient ischemic attack (TIA, a mini stroke, usually lasting a few minutes without residual symptoms) account for 67 percent of cerebrovascular disease. Current practices to manage acute stroke are limited in their ability to reduce associated disabilities, which range from paralysis to depression and impaired thinking. Therefore, much effort is directed at stroke prevention, the goal of the Secondary and Tertiary Prevention of Stroke Patient Outcomes Research Team (PORT), supported by the Agency for Health Care Policy and Research (Contract 282-91-0028) and led by David Matchar, M.D., of Duke University Medical Center.

The Stroke PORT recently published two studies. The first study presents theories about why, even though African-Americans experience more severe strokes resulting in greater disability than whites, they are less likely than whites to undergo carotid endarterectomy, a surgical procedure that is effective in preventing ischemic stroke. The second study points out variation in the availability and use of specialized stroke prevention services.

Horner, R.D., Oddone, E.Z, and Matchar, D.B. (1995). "Theories explaining racial differences in the utilization of diagnostic and therapeutic procedures for cerebrovascular disease." The Milbank Quarterly 73(3), pp. 443-462.

Carotid endarterectomy, which involves the removal of plaque deposits from the carotid artery, is the only surgical procedure known to prevent stroke. It is usually reserved for patients with 70 percent or greater blockage in an operable location within the carotid artery. Despite being at higher risk for ischemic stroke, African-American patients are substantially less likely than white patients to undergo this procedure. In this report, the researchers discuss three widely recognized alternative explanations to racial bias that could account for disparities in the use of carotid endarterectomy. These include (1) racial differences in important clinical factors related to the appropriateness of carotid endarterectomy, such as the extent and location of carotid artery stenosis, comorbid conditions, and differences in operative risk; (2) the ability to pay, regardless of race, may explain variations in the use of expensive diagnostic and therapeutic procedures; and (3) differences along racial lines in patients' decisions to undergo invasive procedures.

Dramatic racial differences have been observed in the use of effective invasive therapies for cerebrovascular disease. The researchers note that available evidence is insufficient to exclude racial discrimination as an explanation for these differences. They conclude that additional studies are needed to clarify the importance of alternative explanations for the observed variations.

Goldstein, L.B., Bonito, A.J., Matchar, D.B., and others (1995, September). "US national survey of physician practices for the secondary and tertiary prevention of ischemic stroke." Stroke 26, pp. 1607-1615.

The stroke PORT conducted a national survey of stroke prevention practices among a stratified random sample of 2,000 physicians. The survey included questions regarding availability of stroke prevention services and use of diagnostic and preventive strategies for patients at elevated risk of stroke. Survey results showed that physician-perceived availability of stroke prevention services varied with type of physician (specialist, such as neurologist, vs. primary care physician) and practice setting (nonmetropolitan vs. small metropolitan or large metropolitan area). Perceived availability of carotid endarterectomy also varied by region of the country. For example, the odds of carotid endarterectomy being reported as readily available were approximately 2.5 to 3.5 times greater for physicians practicing in regions other than the South, which ironically has been dubbed the "Stroke Belt" due to the high number of these events occurring in this part of the country.

Diagnostic studies considered readily available by at least 90 percent of physicians included carotid ultrasonography, transthor-acic echocardiography, Holter monitoring, computerized tomo-graphy, and magnetic resonance imaging (MRI) scans. MR angiography was perceived as being readily available by 68 percent and transesophageal echocardiography by 74 percent of physicians; 12 percent reported cerebral arteriography and 10 percent reported carotid endarterectomy as not being readily available.

Physicians also varied in their use of stroke prevention practices. About 61 percent of physicians reported prescribing daily doses of 325 mg of aspirin for stroke prevention, 33 percent recommended less than 325 mg, and 4 percent used doses of 650 mg or more. Eighty percent of physicians monitored patients on anticoagulant (blood-thinning) therapy at least once a month, whereas 14 percent of physicians monitored these patients less often.

Researchers examine outcomes of back pain treatments provided by chiropractors, surgeons, and general physicians

Patients with acute low back pain have similar recoveries, whether they are treated by primary care practitioners, chiropractors, or orthopedic surgeons. But, they pay substantially more for treatments provided by chiropractors and surgeons and are most satisfied with the care chiropractors provide, according to a study supported by the Agency for Health Care Policy and Research (HS06664). University of North Carolina researchers, led by Timothy S. Carey, M.D., compared the outcomes of treatment provided by 208 practitioners (including chiropractors, orthopedic surgeons, and primary care physicians) in North Carolina for more than 1,500 people who sought care for acute low back pain.

Regardless of which type of health practitioner treated them, about 70 percent of patients had recovered completely within 6 months. The rest continued to be bothered by back pain, but only 5 percent were unable to function as well as they had before the pain began. Outpatient charges were highest for patients seen by orthopedic surgeons and chiropractors and lowest for patients seen by primary care physicians in health maintenance organizations (HMOs) and other primary care providers.

Patients visiting chiropractors made two to three times as many visits as those being treated by primary care doctors (15 and 10 visits for urban and rural chiropractors, respectively, vs. fewer than 5 visits to primary care doctors). Also, orthopedists and chiropractors used more x-rays, with orthopedists ordering x-rays for 72 percent of their patients. In contrast, patients visiting HMO doctors had a mean of 3.1 visits, and only 19 percent had x-rays.

As a result, the charge for treating an episode of back pain was about $808 for patients of urban chiropractors and $809 for patients who saw orthopedic surgeons. The charge for treatment from family doctors averaged $478 for urban doctors and $540 for rural doctors. Yet 42 percent of patients treated by chiropractors rated their treatment "excellent" compared with less than 27 percent of those treated by medical doctors. Patients particularly liked the way chiropractors took their health histories, examined them, and explained the cause of their pain.

See "The outcomes and costs of care for acute low back pain among patients seen by primary care practitioners, chiropractors, and orthopedic surgeons," by Dr. Carey, Joanne Garrett, Ph.D., Anne Jackman, M.S.W., and others, in the October 5, 1995 issue of The New England Journal of Medicine 333, pp. 913-917.

PORT researchers examine prenatal zinc supplementation and other factors affecting birthweight

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 282-92-0055) 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 prenatal zinc supplementation, use of prenatal corticosteroids, the costs/benefits of screening for bacterial infections in asymptomatic pregnant women, and the role of low self-esteem and other psychosocial factors in low birthweight.

Goldenberg, R.L., Tsunenobu, T., Neggers, Y., and others (1995, August 9). "The effect of zinc supplementation on pregnancy outcome." Journal of the American Medical Association 274(6), pp. 463-468.

The use of zinc supplements in pregnant women whose blood zinc levels are low early in pregnancy is known to have beneficial effects, including increased birthweight for their infants. Supplementation seems to have a greater impact on the infants of women whose body mass index (BMI, ratio of weight to height) is low, according to the Low Birthweight PORT researchers. These findings are based on a study of 580 poor black pregnant women whose plasma zinc levels were below the median when they began prenatal care. They were randomized to receive either a daily dose of 25 mg of zinc or a placebo until delivery. Infants born to women in the zinc-supplemented group weighed 126 grams more at birth and had a 0.4 cm larger head circumference than infants in the placebo group. In women with a low BMI, zinc supplementation was associated with a 248-gram higher infant birthweight and a 0.7 cm larger infant head circumference.

Leviton, L.C., Baker, S., Hassol, A., and Goldenberg, R. (1995, July). "An exploration of opinion and practice patterns affecting low use of antenatal corticosteroids." American Journal of Obstetrics & Gynecology 173(1), pp. 312-316.

Despite the evidence that prenatal administration of corticosteroids reduces death and problems such as respiratory distress syndrome in preterm infants, only 20 percent of mothers who deliver preterm newborns receive corticosteroids. In this study, PORT researchers examined why obstetricians still underuse corticosteroids. Based on individual interviews and focus groups, they found that obstetricians and maternal-fetal specialists vary widely in their view of the risks and benefits of prenatal corticosteroids. Obstetricians tend to focus on feared negative consequences, for example, that corticosteroid therapy might disguise the early signs of infection, especially in the case of premature rupture of membranes. Neonatologists, on the other hand, who see the positive and negative outcomes for all infants, believe the therapy's risks are negligible and far outweighed by demonstrated benefits.

Bronstein, J.M., and Goldenberg, R.L. (1995, July). "Practice variation in the use of corticosteroids: A comparison of eight data sets." American Journal of Obstetrics & Gynecology 173(1), pp. 296-298.

Use of prenatal corticosteroids for women in premature labor increased dramatically in the United States during the past decade, from a low of 8.1 percent in the early 1980s to 26 percent in the early 1990s. PORT researchers examined eight databases to identify consistencies in this pattern of corticosteroid use. Analysis showed that, although physicians vary widely in their use of prenatal corticosteroids, they do agree on a few appropriate uses. For example, corticosteroid therapy is used more frequently for multiple births and much less frequently in infants less than 28 weeks gestation or older than 31 weeks. The therapy is used more often when delivery is by cesarean section and when tocolytic agents (which inhibit uterine contractions during labor) are used.

Atkinson, M.W., Goldenberg, R.L., Gaudier, F.L., and others. (1995, July). "Maternal corticosteroid and tocolytic treatment and morbidity and mortality in very low birth weight infants." American Journal of Obstetrics & Gynecology 173(1), pp. 299-304.

Very-low-birthweight (VLBW) infants (weighing less than 3 pounds) born to women with any prenatal treatment with corticosteroids have a 50 percent lower risk of neonatal death, 65 percent lower risk of intraventricular hemorrhage, 70 percent lower risk of severe intraventricular hemorrhage, and 60 percent lower risk of seizures than other VLBW infants. Conversely, VLBW infants whose mothers receive tocolytics to inhibit uterine contractions during labor are more than twice as likely as other VLBW infants to experience intraventricular hemorrhage. The numbers of neonatal deaths and seizures are significantly reduced when women are given tocolytics in combination with prenatal corticosteroids. These findings are based on a retrospective study of labor and delivery records from 773 live births at the University of Alabama at Birmingham hospitals from 1979 to 1991. The infants were born between 24 to 28 weeks' gestation and weighed 500 to 1,000 grams.

Rouse, D.J., Andrews, W.W., Goldenberg, R.L., and Owen, J. (1995, July). "Screening and treatment of asymptomatic bacteriuria of pregnancy to prevent pyelonephritis: A cost-effectiveness and cost-benefit analysis." Obstetrics & Gynecology 86(1), pp. 119-123.

Depending on the patient population, 2-10 percent of pregnant women have an asymptomatic bacterial infection at their initial prenatal visit. Without antimicrobial treatment, as many as 30 percent of these women will develop symptomatic urinary tract infections during pregnancy, usually in the form of pyelonephritis (kidney infection sometimes due to obstruction from an enlarged uterus). PORT researchers found that, regardless of whether dipstick or urine culture is used, screening is cost-beneficial compared with a policy of no screening for the prevention of pyelonephritis in pregnancy. No screening resulted in 23.2 cases per 1,000 pregnancies, versus 16.2 cases with dipstick and 11.2 with urine culture. The cost of screening and treatment of asymptomatic bacteriuria per 1,000 pregnancies was $1,968 with dipstick and $19,264 with culture. These results were based on a decision analytic model that compared the two screening strategies with each other and to a policy of no screening.

Hickey, C.A., Cliver, S.P., Goldenberg, R.L., and others (1995, August). "Relationship of psychosocial status to low prenatal weight gain among nonobese black and white women delivering at term." Obstetrics & Gynecology 86(2), pp. 177-183.

PORT researchers examined the relationship between psychosocial well-being (anxiety trait, subjective stress, mastery or sense of control over events, self-esteem, and social support) and low prenatal weight gain among 806 nonobese, high-risk, low-income black and white women who delivered at term from 1985 to 1988. White women in this study who experienced anxiety, depression, and low self-esteem were more apt to gain too little weight during pregnancy, but these psychosocial factors did not affect weight gain among black women in the study, according to the researchers. Women who were depressed gained 2.26 kg less than those who were not depressed, and those with high levels of trait anxiety gained 1.69 kg less than women with low levels. Conversely, women with high self-esteem gained 1.94 kg more than those with low self-esteem, women with a high level of social support gained 2.26 kg more than those with a low level, and women with a high level of mastery gained 2 kg more than those with a low mastery level.

Return to Contents
Proceed to Next Section