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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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.

Better assessment of chest pain patients in the ER could reduce unnecessary CCU admissions

Chest pain patients who could be suffering a heart attack (acute myocardial infarction, AMI) are usually sent from the emergency department to the coronary care unit (CCU), even though only about 30 percent of these patients are ultimately found to have suffered a heart attack. Unnecessary and costly CCU admissions could be reduced if these units were reserved for patients with a moderate probability (21 percent or more, depending on the patient's age) of heart attack, unless patients need intensive care for other reasons, concludes a study supported in part by the Agency for Health Care Policy and Research (HS06452). Patients at moderate risk of an AMI typically are those with electrocardiographic changes indicative of coronary blood deficiency (ischemia) and tissue damage (infarction) that appear to be recent, explain the researchers from Dartmouth Medical School, the University of California, San Francisco, Brigham and Women's Hospital, and Harvard Medical School. They used clinical data from over 12,000 emergency department patients with acute chest pain at seven hospitals and resource use data from 900 patients in a decision-analytic model to identify cost-effective recommendations for initial admission to a CCU versus an intermediate care (or stepdown) unit, based on the probability of AMI.

Assuming a 15 percent relative increase in death when patients with AMI were admitted to the intermediate unit instead of the CCU, the costs per year-of-life-saved for triage to the CCU varied markedly, depending on the age of the patient and the probability of AMI. For 55- to 64-year-old patients with a 1 percent probability of heart attack determined in the emergency department, the cost per year-of-life-saved was $1.4 million. When the probability of heart attack was 99 percent, the cost per year-of-life-saved was $15,000. The CCU had a cost-effectiveness comparable to other accepted medical interventions (less than $50,000 per year-of-life-saved) when the initial probability of AMI was greater than 57 percent among patients 30 to 44 years of age but only 21 percent among patients 65 to 74 years of age. The researchers found that initial triage to the CCU is generally more cost effective for older patients who have higher age-specific mortality rates and for patients with ECG changes of ischemia who have a moderate to high probability of AMI.

Details are in "Cost-effectiveness of a coronary care unit versus an intermediate care unit for emergency department patients with chest pain," by Anna N.A. Tosteson, Sc.D., Lee Goldman, M.D., I. Steven Udvarhelyi, M.D., and Thomas H. Lee, M.D., M.Sc., in the July 15, 1996, issue of Circulation 94, pp. 143-150.

Return to Contents

Healthcare Marketplace Research

Marketplace demand for specialists declines

Anecdotal reports suggest that medical school graduates in certain specialties are having difficulty finding employment, while their primary care colleagues are being recruited with lucrative incentive packages. Also, a recent study found that in 1990, there were four advertisements recruiting specialist physicians for every one for generalist physicians, but by 1995, the market for specialists had declined steeply to two advertisements for specialists for every one generalist ad.

Despite recommendations to train more generalists, the demand for internists and pediatricians appears to be flat or declining, perhaps due to increased use of physician assistants and nurse practitioners. On the other hand, the demand for family physicians is rising, according to the study (supported in part by the Agency for Health Care Policy and Research, T32 HS00026).

Based on analysis of the number of physician positions advertised in the September issue (peak month for physicians to conduct job searches and for recruitment advertising) of key specialty and general medical journals in select years from 1984 through 1995, researchers in California and Washington found steep declines in the number of advertised positions for specialist physicians over the past 5 years, with the exception of pediatric specialists. The most dramatic changes occurred in the number of internal medicine specialist positions, which declined by 75 percent from 1990 to 1995, compared with a 9 percent decline in general internist positions. During the same period, anesthesiology positions declined 65 percent, pulmonary positions 50 percent, and orthopedic surgery positions 50 percent. For physicians as a whole, there were four specialist positions for every generalist position advertised in 1990; by 1995, this ratio dropped to 1.8.

Compared with the other generalist fields, only family medicine showed sustained growth, with advertised positions more than doubling over the study period. The researchers conclude that these trends in recruitment advertisements reflect the shift to increasingly managed and capitated systems of care that use fewer specialists.

More details are in "Changes in marketplace demand for physicians," by Sarena D. Seifer, M.D., Barbara Troupin, M.D., M.B.A., and Gordon D. Rubenfeld, M.D., in the September 4, 1996, Journal of the American Medical Association 276(9), pp. 695-699.

Return to Contents

Women's Health

Uterine fibroids occur more often among black women than white women undergoing hysterectomy

From one-fourth to one-half of women of childbearing age will suffer from uterine fibroids at some point. Among women undergoing hysterectomy, uterine fibroids (leiomyomas) develop at an earlier age in black women than they do in white women, are larger and more numerous, and cause more severe symptoms, according to a study supported in part by the Agency for Health Care Policy and Research (HS06865). The study also suggests that being overweight may increase a woman's risk of developing fibroids, since both black and white leiomyoma patients in this study were overweight compared with women in the general population.

Uterine fibroids can cause abnormal bleeding, pelvic pain, limited activity, anemia, fatigue, urinary and bowel problems, miscarriage, and infertility. These benign tumors of the uterine smooth muscle are the leading indication for hysterectomy in the United States, notes Kristen H. Kjerulff, Ph.D., of the University of Maryland School of Medicine, the study's lead author.

The researchers found that, of 409 black women and 836 white women who underwent hysterectomy for noncancerous conditions at 28 Maryland hospitals, 89 percent of black women and 59 percent of white women were found to have leiomyomas. Interviews shortly before surgery with patients who had a presurgical diagnosis of leiomyoma, hospital records, and pathology reports obtained after discharge showed that black women were more likely than white women to have seven or more leiomyomas (57 percent vs. 36 percent), to be anemic (56 percent vs. 38 percent), and to have severe pelvic pain (59 percent vs. 41 percent). Black women were diagnosed with leiomyomas on average at an earlier age (38 years vs. 42 years) and underwent hysterectomy at a younger average age (42 years vs. 45 years). Among women with leiomyomas undergoing hysterectomy, black women had substantially larger uteri than white women; the average uterine weight was 421 g for black women with leiomyomas and 319 g for white women (average uterine weight for pregnant women at 12 weeks gestation is 280 g).

Body mass index (BMI) was significantly associated with uterine weight. An increase of one point in BMI was associated with an increase in uterine weight of 5 g. Forty-one percent of white women and 67 percent of black women were classified as overweight—with BMI values more than 27.3, equivalent to 120 percent of desirable body weight—compared with 33 percent of white women and 49 percent of black women in the general population. The researchers point out, however, that body weight is only one factor in the development and growth of leiomyomas; various hormones and other factors also play a role in this process.

Details are in "Uterine leiomyomas: Racial differences in severity, symptoms, and age at diagnosis," by Dr. Kjerulff, Patricia Langenberg, Ph.D., Jeffrey D. Seidman, M.D., and others, in the July 1996 Journal of Reproductive Medicine 41, pp. 483-490.

PORT researchers link bacterial infection with many preterm births

Preterm births, on the increase since the mid-1980s, are responsible for nearly 75 percent of newborn deaths and as much as half of long-term neurological damage in children. Uterine infection is the key to many of these preterm births, asserts Robert L. Goldenberg, M.D., of the University of Alabama (Birmingham), in a recent editorial. Dr. Goldenberg is principal investigator of the Low Birthweight Patient Outcomes Research Team (PORT), which is supported by the Agency for Health Care Policy and Research (PORT contract 282-92-0055).

Up to 40 percent of women in spontaneous labor will have bacteria in both the amniotic fluid and the membranes. An additional 20 percent will have organisms in the membranes but not in the amniotic fluid, according to Dr. Goldenberg. Associated with these microorganisms, some of which are transmitted sexually, is an increased production of inflammation-producing cytokines that can be detected in the amniotic fluid. These cytokines participate, both directly and indirectly, in various reactions leading to the onset of uterine contractions, changes in cervical consistency, and rupture of the membranes that initiate labor.

Bacteria most commonly associated with spontaneous delivery include Ureaplasma urealyticum, Mycoplasma hominis, Bacteroides, and Gardnerella vaginalis. For the most part, these strains of bacteria are not very virulent and may exist in the vagina and uterus for a long time without producing symptoms. Previous studies by the Low Birthweight PORT have shown up to a three-fold increase in the odds of preterm birth with the presence of bacterial vaginosis.

This type of infection also would explain why women who have one early spontaneous birth are so prone to have a second. There is no reason to believe that after the infant is delivered, the intrauterine bacterial colonization disappears spontaneously. Chronic colonization of the uterine lining with low virulence microorganisms has the potential to explain most of the observations related to early spontaneous preterm birth, notes Dr. Goldenberg, including why black women have more preterm births than white women. He concludes that treatment strategies aimed at the underlying disease (infection) seem far more promising than those targeted to symptoms of preterm labor or the psychosocial, behavioral, or nutritional characteristics of the mother that have been associated with, but not causally related to, preterm birth. In fact, recently reported randomized antibiotic treatment trials of women at high risk for preterm birth, who also had bacterial vaginosis, showed substantial reduction in spontaneous preterm births.

See "Intrauterine infection and why preterm prevention programs have failed," an editorial by Dr. Goldenberg and William W. Andrews, M.D., Ph.D., in the American Journal of Public Health 86(6), pp. 781-782, 1996.

Women who have regular checkups appear to live longer

Debate continues about which aspects of primary care are most critical for improving patient health. Is it having a usual doctor or clinic, regular checkups, or better access to care? A recent study points out the importance of regular checkups for women's survival but also underscores the difficulty of identifying which aspects of health care access or use measured in population surveys influence patient outcomes. The study was carried out by Peter Franks, M.D., and Marthe R. Gold, M.D., of the University of Rochester, and Carolyn M. Clancy, M.D., 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.

The researchers found that women who have regular checkups have better survival rates than women who do not. However, no association was found between checkups and men's survival rates; instead, checkups among men are more likely to be a marker for higher income. No relationship was found between having a usual source of health care (physician or clinic) or forgoing care for perceived medical problems and subsequent mortality for either men or women.

These findings probably underestimate the relationship between access to and use of health care and survival, cautions Dr. Clancy. Better understanding of the factors underlying differences between the sexes in perceived health and health care use and health outcomes could enhance the development of more useful measures. For example, women are more inclined to get a checkup for preventive health reasons, such as screening mammograms and pap smears. Men are more apt to obtain a checkup if it is required for work or insurance. Also, women generally have a more positive attitude toward health and health care, and they have been socialized to take more responsibility for family health. So while they may be more sensitive to symptoms of illness, they also are less apt to adopt the sick role when ill. This may explain why in this study twice as many men as women did not have a usual source of care (17 percent vs. 9 percent), but slightly more women had chosen not to obtain care (16 percent vs. 13 percent).

These findings were based on data from the National Health and Nutrition Examination Survey Epidemiologic Follow-Up Study, which followed a representative group of working-age U.S. adults for up to 16 years. Dr. Clancy and colleagues measured the impact on survival of availability of a usual source of care, care not received for perceived medical problems (forgone care), and receipt of a general checkup (other than for illness).

For more information, see "Use of care and subsequent mortality: The importance of gender," by Drs. Franks, Gold, and Clancy, in the August 1996 HSR: Health Services Research 31(3), pp. 347-363. Reprints (AHCPR Publication No. 97-R006) are available from the AHCPR Publication Clearinghouse.

Return to Contents

Patient Outcomes/Effectiveness Research

Benefits and costs weighed for specialty versus primary care for stroke patients

Stroke patients who are treated by neurologists may pay more, but they often experience better outcomes, according to a recent study by the Stroke Patient Outcomes Research Team (Stroke PORT). Led by David Matchar, M.D., of Duke University Medical Center, and supported by the Agency for Health Care Policy and Research (contract 290-91-0028), the Stroke PORT researchers analyzed claims data for a random 20 percent sample of Medicare patients admitted to the hospital with nonhemorrhagic stroke between January 2 and September 30, 1991. Three months following the stroke, study patients treated by neurologists had a 31 percent lower death rate than study patients treated by internists and a 36 percent lower death rate than patients treated by family practitioners. Patients cared for by neurologists also appeared to be more functional after their hospital stay. They were more likely to be sent home or discharged to inpatient rehabilitation facilities rather than to a nursing or rest home.

Neurologists were 34 percent more expensive than family practitioners and 22 percent more expensive than internists or other specialists. If both a neurologist and a primary care physician treated a stroke patient, the costs increased another 5 percent.

Lead author Janet Mitchell, Ph.D., of Health Economics Research in Waltham, MA, suggests that several dimensions of stroke management by neurologists may be relevant both to the increased costs and the improved outcomes. Neurologists are significantly more likely to order diagnostic cerebrovascular tests, especially brain scans, which often identify the pathophysiologic mechanism of the stroke and may affect the course of treatment. Also, they are more likely to prescribe the anticoagulant warfarin and to begin early rehabilitation.

These findings could warrant a broader dissemination to primary care physicians of stroke treatment guidelines that incorporate the methods used by neurologists, Dr. Mitchell concludes. However, further research, preferably by randomized trial, would be needed first to provide definitive answers.

For more information, see "What role do neurologists play in determining the costs and outcomes of stroke patients?" by Dr. Mitchell, David J. Ballard, M.D., Ph.D., Jack P. Whisnant, M.D., and others, in the November 1996 issue of Stroke 27, pp. 1937-1943.

PORT researchers identify risk factors for retinal detachment after cataract surgery and confirm the reliability of the VF-14 index of visual function

Cataracts are the second leading cause of blindness in the United States. About 18 percent of persons 65-74 years of age and almost half of those aged 75-84 years have cataracts that impair their everyday activities and ability to live independently. The Cataract Patient Outcomes Research Team (PORT) was funded by the Agency for Health Care Policy and Research (HS06280) to study variations in cataract management, patient outcomes, and the economic aspects of cataract treatment. PORT researchers, led by Earl P. Steinberg, M.D., M.P.P., of The Johns Hopkins University, recently published the results of two studies, which are discussed here.

Tielsch, J.M., Legro, M.W., Cassard, S.D., and others (1996). "Risk factors for retinal detachment following cataract surgery: A population-based case-control study." Ophthalmology 103(10), pp. 1537-1545.

Performance of Nd:YAG laser posterior capsulotomy (incision of the lens capsule with a laser) after cataract surgery increases the risk of retinal detachment nearly four-fold (3.8), according to this study by the Cataract PORT investigators. This finding was based on an analysis that adjusted for other factors which can increase the risk of retinal detachment, such as prior history of retinal detachment and refractive error. Retinal detachment occurs in 0.2 percent to 3.6 percent of persons following cataract surgery. However, given the large number of cataract surgeries performed, the number of these vision-threatening complications becomes substantial. This suggests the need for strong clinical and functional justification for performance of Nd:YAG laser capsulotomy, according to the researchers. Using Medicare claims data, they identified a group of Medicare beneficiaries who underwent this surgery during 1988-1990 and then compared 291 cases who had claims for retinal detachment following surgery with 870 matched controls who did not. Data regarding clinical risk factors, as well as clinical events (e.g., performance of capsulotomy or repair of retinal detachment), were obtained directly from the ophthalmologists who cared for the patients included in the analysis—thus overcoming limitations related to the Medicare claims data.

Cassard, S.D., Patrick, D.L., Damiano, A.M., and others (1995, December). "Reproducibility and responsiveness of the VF-14." Archives of Ophthalmology 113, pp. 1508-1513.

The VF-14 is an index of functional impairment in patients with cataracts. It was developed by the cataract PORT researchers for use in routine clinical practice and in research studies. The Cataract PORT investigators have found that the VF-14 is three times more responsive to small but clinically important changes in vision following cataract surgery than the Sickness Impact Profile (SIP), a general health status measure.

The VF-14 is used to measure a cataract patient's ability to perform 14 vision-dependent activities such as reading small print, doing fine handwork, and night and day driving. This study also shows that the VF-14 elicits highly reproducible responses over an 8-month period in clinically stable patients.

Short-term improvement in carefully selected patients with sciatica and spinal stenosis appears better with surgery than other treatments

When carefully selected, persons suffering from sciatica and spinal stenosis may improve substantially more with surgery than patients treated nonsurgically, according to a study conducted by the Maine Medical Assessment Foundation (MMAF), directed by Robert B. Keller, M.D. The study is a component of the Back Pain Patient Outcomes Research Team (PORT) led by Richard A. Deyo, M.D., M.P.H., of the University of Washington.

Sciatica generally results from herniation or protrusion of an intervertebral disc, which presses on spinal nerve roots. It often results in leg pain, numbness or tingling in the leg and/or foot, weakness in the leg or foot, and back pain. The pain and loss of mobility from sciatica may severely reduce a person's ability to function at home and at work. Spinal stenosis (constriction) occurs predominantly among the elderly and is being diagnosed more frequently due to widespread use of sophisticated noninvasive imaging techniques. It often causes leg pain, numbness, and weakness and pain while walking. Physicians continue to disagree about the best approach to these problems, with surgery rates varying up to 15-fold among U.S. regions.

The Maine study showed that among 389 patients with sciatica and 148 patients with spinal stenosis, outcomes of surgical patients were substantially better at 1 year than for the patients who underwent nonsurgical treatment (for example, back exercises, physical therapy, and spinal manipulation). Patients were recruited from orthopedic, neurosurgical, and occupational medicine practices across the state of Maine and, at the time of study entry, had undergone at least 2 weeks of conservative therapy within the previous 2 months without satisfactory improvement.

Seventy-five percent of surgical sciatica patients had much less or no back or leg pain compared with 55 percent of nonsurgical patients, while 31 percent and 11 percent of patients, respectively, reported that all symptoms were completely gone. Overall quality of life was moderately improved in 80 percent of surgical patients and 58 percent of nonsurgical patients, while 60 percent and 40 percent, respectively, reported they would be satisfied to spend the rest of their lives in their current state. In fact, 86 percent of surgical patients stated that they would elect surgery again if they had to remake their decision. The benefits of surgery were only modest for the least symptomatic patients, resulting in symptoms, quality of life, and satisfaction outcomes similar to medically treated patients.

Of the 148 patients with lumbar spinal stenosis, 81 were treated surgically and 67 were treated nonsurgically. On average, patients undergoing surgery had more severe imaging findings, symptoms, and worse functional status than nonsurgical patients. Twice as many surgically treated patients reported near or complete relief of leg or back pain 1 year after surgery as nonsurgically treated patients (55 percent vs. 28 percent). Even patients with moderate symptoms improved compared with nonsurgically treated patients. Days of disability during the past month declined in both groups, but more so in surgical patients. Overall quality of life was at least moderately improved in 81 percent of surgical and 49 percent of nonsurgical patients. Overall results were viewed as very good or excellent in 69 percent of surgical and 36 percent of medical patients, and 88 percent of surgical patients said they would make the same decision for surgery again.

Although 1-year results appear better in surgical patients with spinal stenosis, few nonsurgical patients worsened or required subsequent surgery, and fully 20 percent of surgical patients reported no improvement in back or leg pain. And, while surgery provides a greater chance for rapid relief of sciatica symptoms, nonsurgically treated patients appear likely to gradually improve. Ultimately, the decision to undergo surgery for either condition is an individual one,conclude the PORT researchers. Determining whether the benefits of surgery for stenosis or sciatica persist over time requires longer followup, which is underway.

Additional support for portions of the Maine study came from another AHCPR-funded project, "Outcomes dissemination: The Maine Study Group Model." This project has focused on methods of dissemination and evaluation of the study group methodology developed by the MMAF.

More details are in the following three articles: "The Maine lumbar spine study: I. Background and concepts," by Robert B. Keller, M.D., Steven J. Atlas, M.D., M.P.H., Daniel E. Singer, M.D., and others; "The Maine lumbar spine study: II. 1-year outcomes of surgical and nonsurgical treatment of sciatica," by Dr. Atlas, Dr. Deyo, Dr. Keller, and others; and "The Maine lumbar spine study: III. 1-year outcomes of surgical and nonsurgical treatment of lumbar spinal stenosis," also by Dr. Atlas, Dr. Deyo, Dr. Keller, and others. All three articles appear in the August 1996 issue of Spine, 21, pp. 1769-1795.

Educating patients about prostate screening and treatment influences their medical decisions

When a videotape shown in clinical practice informs men about the natural course of early-stage prostate cancer and the medical uncertainty surrounding the routine screening and treatment of this condition, they are less likely to prefer screening and treatment than men who don't see the tape, concludes a study by the Prostate Patient Outcomes Research Team (PORT). However, even informed men varied in their decisions about prostate screening. These findings underscore the importance of incorporating individual preferences into PSA screening decisions, conclude the Prostate PORT researchers, whose work was supported by the Agency for Health Care Policy and Research (HS06336 and HS08397).

Routine prostate-specific antigen (PSA) screening for prostate cancer is controversial because of frequent false-positive results (the test shows cancer that does not exist) and uncertainty surrounding the benefits of treating early-stage prostate cancer. Even if diagnosed early, the majority of men with prostate cancer will not experience significant symptoms and will die from another cause. Moreover, it is questionable whether treatment, usually with radical prostatectomy or radiation therapy, improves survival over watchful waiting or decreases disease-related symptoms. Also, the side effects of treatment range from impotence, incontinence, and rectal injury to operative death.

The research team presented an educational videotape designed by the team to inform men about the uncertainty surrounding PSA screening and the treatment of early-stage prostate cancer to two groups of male patients 50 years of age or older who had no history of prostate cancer who were being seen at the Dartmouth-Hitchcock Medical Center in Hanover, NH. Men seeking a free screening were preassigned to view the videotape (184 men) or another videotape (188). Men scheduled to visit a general internal medicine clinic viewed either the educational videotape (103) or no videotape (93).

Men shown the educational videotape were more inclined to say that they would choose watchful waiting over active treatment if cancer were to be found (63 and 86 percent vs. 26 and 40 percent) in two different clinic settings. They were also half as likely as men who did not view the videotape to have a PSA test (12 percent vs. 23 percent of general internal medicine clinic patients).

For more details, see "The importance of patient preference in the decision to screen for prostate cancer," by Ann Barry Flood, Ph.D., John E. Wennberg, M.D., M.P.H., Robert F. Nease, Jr., Ph.D., and others, in the Journal of General Internal Medicine 11, pp. 342-349, 1996.

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