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

Outcomes/Effectiveness Research

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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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

Being overweight or underweight does not preclude elective noncardiac surgery for most patients

Relatively healthy overweight and underweight patients are not at any higher risk than normal weight patients for complications or longer hospital stays following elective, noncardiac surgery. The only exception is elective abdominal or gynecologic surgery, after which overweight patients have double the wound infection rates of normal weight patients, according to a recent study supported in part by the Agency for Health Care Policy and Research (HS06573).

The most overweight and underweight patients do, however, have higher costs. This may indicate that more resources are expended on these patients to prevent complications, explains Lee Goldman, M.D., of the University of California at San Francisco. Dr. Goldman and his colleagues conclude that as long as the person is not severely malnourished, being overweight or underweight is not a reason to deny elective noncardiac surgery.

The researchers correlated the body mass index (BMI, weight in kilograms divided by height in meters squared) of 2,964 patients 50 years and older undergoing elective noncardiac surgery with complications, length of hospital stay, and costs. Complications were no different among four BMI groups, after taking into account other factors affecting the likelihood of complications. BMI groups were those underweight (BMI less than 20), normal (BMI 20 to 29), overweight (BMI 30 to 34), and the most overweight (BMI more than 34). For instance, a 5'4" woman with a BMI of 40.5 would weigh 238 pounds, 82 percent above ideal body weight, and a 5'9" man with a BMI of 38.1 would weigh 244 pounds, 61 percent above ideal body weight. A 5'4" woman with a BMI of 18 would weigh 103 pounds and a 5'9" man with a BMI of 18 would weigh 119 pounds, or 21 percent below ideal body weight for both.

Overall, the researchers found no significant increases in complication rates for the most overweight group compared with normal weight patients. However, patients with a BMI of 30 and over who underwent abdominal or gynecologic procedures had wound infection rates of 11 percent compared with nearly 5 percent for normal weight patients and 0 percent for underweight patients. The most overweight patients and underweight patients had insignificantly longer stays (0.8 and 0.9 days longer, respectively), but they incurred significantly higher costs ($834 higher and $3,150 higher, respectively) than patients of normal weight.

Details are in "Body mass index as a correlate of postoperative complications and resource utilization," by Eric J. Thomas, M.D., M.P.H., Dr. Goldman, Carol M. Mangione, M.D., M.S.P.H., Thomas H. Lee, M.D., M.S.C., and others, in the March 1997 issue of the American Journal of Medicine 102, pp. 277-283.

Heart attack patients in the United States fare slightly better in the short-term than Canadian patients

A recent study compared care provided to elderly heart attack patients in the United States and Canada and found that, within 30 days of hospital admission, the U.S. patients underwent coronary angiography five times more often (34.9 percent vs. 6.7 percent), coronary angioplasty almost eight times more often (11.7 percent vs. 1.5 percent), and coronary artery bypass surgery nearly eight times more often (10.6 percent vs 1.4 percent) than the Canadian patients. A slight but significantly lower proportion of U.S. patients than Canadian patients died within 1 month (21.4 percent vs. 22.3 percent); about 34 percent of patients in both countries had died 1 year later. These results appear to favor the more conservative Canadian approach to revascularization, according to the authors of the study, which was supported by the Agency for Health Care Policy and Research (HS08071).

The researchers compared the use of invasive cardiac procedures and mortality rates among 224,258 elderly Medicare beneficiaries in the United States and 9,444 elderly patients in Ontario, Canada, each of whom suffered a heart attack in 1991. The better short-term outcomes in the U.S. patients may reflect the intensity and timeliness of hospital care in this country. For example, a higher proportion of U.S. than Canadian patients were initially admitted to hospitals that were able to perform both cardiac catheterization and revascularization procedures (34.5 percent vs. 14 percent), including hospitals that specialized in cardiac catheterization (22.8 percent vs. 4.1 percent).

The absence of a sustained survival benefit 1 year later probably reflects factors other than the differences in use of revascularization procedures. Many medical therapies, for example, beta-blockers and aspirin, are known to improve long-term survival after heart attack, and it is possible that they were used more frequently in the Canadian group of patients. Also, better long-term outcomes in Canada may reflect greater access to primary care, prescription drugs, and long-term care, which are universally provided to the elderly with minimal or no copayments under the Canadian health care system. The researchers point out, however, that mortality is not the only important outcome; functional status should be considered as well.

For more information, see "Use of cardiac procedures and outcomes in elderly patients with myocardial infarction in the United States and Canada," by Jack V. Tu, M.D., Ph.D., Chris L. Pashos, Ph.D., C. David Naylor, M.D., D.Phil., and others, in the May 22, 1997 issue of the New England Journal of Medicine 336, pp. 1500-1505.

Many physicians have not fully adopted recommended cholesterol management practices

More aggressive screening, counseling, and medication treatment for hyperlipidemia would substantially aid the prevention of cardiovascular disease, concludes a study supported in part by the Agency for Health Care Policy and Research (HS07892). The study found that physicians screen relatively few people for high cholesterol, are less apt to prescribe lipid-lowering medications for obese than non-obese persons, often do not prescribe recommended lipid-lowering medications, and may not account for multiple risk factors for heart disease, such as cigarette smoking, in their cholesterol management practices. Only 1 in 12 patients without known hyperlipidemia (high levels of cholesterol in the blood) is screened annually for cholesterol; patients with hyperlipidemia are counseled about cholesterol reduction about once every 3 years. Randall S. Stafford, M.D., Ph.D., of Harvard Medical School, and his colleagues examined reports of 2,332 office-based physicians on cholesterol-related screening, counseling, or medications used during 56,215 office visits during 1991 and 1992. For the estimated 1.03 billion U.S. visits by patients without reported hyperlipidemia, cholesterol screening (2.8 percent of visits) and counseling (1.2 percent) were infrequent. In the 85 million visits by patients with hyperlipidemia, cholesterol testing was reported in 23 percent, cholesterol counseling in 34 percent, and lipid-lowering medications in 23 percent.

Physicians were less likely to test cholesterol levels or prescribe lipid-lowering medications to obese than non-obese patients, perhaps relying on weight loss as primary management. Also, physicians prescribed statins or fibrates for 83 percent of visits by patients taking lipid-lowering medications, even though guidelines at the time favored resins and niacin as drugs of first choice. What's more, whether a patient smoked cigarettes or not did not affect cholesterol management practices.

These findings indicate that physicians have not fully adopted the cholesterol management practices first recommended in 1988 by the National Cholesterol Education Program and in specific guidelines that have been issued for cholesterol testing, dietary counseling, and pharmacologic treatment, conclude the researchers.

Details are in "Variations in cholesterol management practices of U.S. physicians," by Dr. Stafford, David Blumenthal, M.D., and Richard C. Pasternak, M.D., in the January 1997 Journal of the American College of Cardiology 29, pp. 139-146.

Better communication increases satisfaction among men facing a decision about surgery for BPH

Benign prostatic hyperplasia (BPH, enlarged prostate) affects roughly 25 percent of older men. These men must choose between watchful waiting to see if symptoms progress and active treatment, which often means surgical removal of the prostate (prostatectomy). Despite a nationwide trend toward less invasive management of BPH, even among patients with moderate to severe urinary tract symptoms, physicians themselves vary in their recommendations.

A shared decisionmaking program (SDP) that synthesizes video, audio, and computer graphics to present a program tailored to the age, health status, and symptom severity of the patient apparently helps men feel more comfortable as they work through the decisionmaking process. There were no significant differences between the treatment choices made by men exposed to the program and the choices made by men who received only an informational brochure (control group). But the men exposed to the SDP were nevertheless more knowledgeable about their condition and more satisfied with the decisionmaking process, and they showed less deterioration in physical functioning and perception of their general health 1 year later. Men in the SDP group were not significantly different from the brochure-only group in satisfaction with the treatment decisions, BPH symptom severity, social functioning, or preferences for participation in decisionmaking.

These findings are the result of a study supported by the Agency for Health Care Policy and Research (HS06540 and HS08397) and led by Michael J. Barry, M.D., of Massachusetts General Hospital. Dr. Barry and his colleagues conclude that traditional methods of obtaining consent may be suboptimal, and to the extent that true informed consent requires patients to be well-informed about their condition, the SDP appears to be a valuable tool for informing men with BPH. They randomized 227 men with BPH who were being treated at one of three urologic practices to the SDP group (104 men) or the brochure-only group (123 men) and followed them for 1 year.

See "A randomized trial of a multimedia shared decision-making program for men facing a treatment decision for benign prostatic hyperplasia," by Dr. Barry, Daniel C. Cherkin, Ph.D., YuChiao Chang, Ph.D., and others, in the January/February 1997 Disease Management and Clinical Outcomes 1(1), pp. 5-14.

Biliary Tract PORT studies focus on training physicians for new procedures and diagnosis of common bile duct stones

Laparoscopic cholecystectomy was introduced in the United States in 1988 as a less invasive procedure for gallbladder removal. In this technique, several small incisions are made in the patients abdomen through which surgical instruments are inserted, along with a tiny camera to visualize the operative area.

Laparoscopic cholecystectomy is associated with fewer deaths and complications, less postoperative pain, shorter hospital stays, and earlier return to usual activities than traditional open gallbladder surgery. But to achieve these outcomes, surgeons must be properly trained. The technique's one disadvantage is its higher incidence of retained common bile duct stones, most likely because the physician cannot directly observe and palpate the duct as in open surgery. There also may be a higher rate of common bile duct injuries.

Two newly published studies by the Biliary Tract Disease Patient Outcomes Research Team (PORT) are summarized here. They discuss the training required for surgeons to perform the laparoscopic surgery, as well as how gastroenterologists and surgeons vary in their diagnosis and management of common bile duct stones. The PORT was led by J. Sanford Schwartz, M.D., of the University of Pennsylvania and supported by the Agency for Health Care Policy and Research (HS06481).

Escarce, J.J., Shea, J.A., and Schwartz, J.S. "How practicing surgeons trained for laparoscopic cholecystectomy." Medical Care 35(3), pp. 291-296, 1997.

Rapid adoption of laparoscopic cholecystectomy raised concerns that some surgeons were performing the procedure without adequate training. And from 1989 until the end of 1991, half of them may have been, according to this study. At that point only half of surgeons who had adopted the technique followed the 1992 Society of American Gastrointestinal Endoscopic Surgeons (SAGES) training criteria for hospitals to grant privileges to physicians performing the procedure that eventually evolved. These criteria include prior practice on animals, serving as first assistant or camera operator for experienced surgeons, and supervision or assistance by a more experienced surgeon during initial laparoscopic procedures until proficiency is observed.

The research team surveyed 1,240 non-Federal general surgeons who had adopted the procedure, which they derived from a 15 percent random sample of physicians drawn from the American Medical Associations Physician Masterfile. Results showed that nearly 50 percent of adopters used self-instructional materials, 49 percent were taught by another surgeon with more experience, and 93 percent took a formal course in the procedure. Only 1.3 percent neither were taught by another surgeon nor took a formal course.

Overall, 93 percent of surgeons had practiced on animals, and 76 percent had served as first assistant or camera operator for colleagues before performing their first laparoscopic cholecystectomy. In addition, 64 percent had their first laparoscopic procedures supervised or assisted by a surgeon with more experience. Although the proportion of physicians who satisfied the guidelines more than doubled during the study period, only two-thirds had done so by the end of 1991. These researchers recommend that hospital credentialing bodies, aided by professional organizations, quickly identify new procedures which involve use of new instruments or techniques unfamiliar to practicing physicians so that specific privileging criteria can be developed.

Shea, J.A., Asch, D.A., Johnson, R.F., and others. "What predicts gastroenterologists and surgeons diagnosis and management of common bile duct stones?" Gastrointestinal Endoscopy 46(1), pp. 40-47, 1997.

Because surgeons performing laparoscopic cholecystectomy cannot directly observe and feel for common bile duct stones as in open cholecystectomy, physicians must rely on laboratory and diagnostic test results and elements of the history and physical examination. Common indicators include patient age, history of jaundice or pancreatitis, common bile duct diameter on ultrasound, and levels of serum alanine, aminotransferase, alkaline phosphatase, amylase, and total bilirubin.

The researchers surveyed a random sample of 1,500 gastroenterologists and 1,500 surgeons on the importance they gave to each potential indicator of common bile duct stones, the likelihood that stones were present for each of nine clinical vignettes, and whether they would order a preoperative endoscopic retrograde cholangiography (ERCP, x-ray of the common bile duct following infusion into the duct of a contrast dye).

Even though most gastroenterologists and surgeons used the common bile duct diameter on ultrasound and serum total bilirubin as the most important indicators of stones, they varied substantially in the importance they gave to other clinical indicators when deciding to order an ERCP prior to laparoscopic cholecystectomy.

Physicians varied in their estimate of a common bile duct stone by about 30 percent when presented with the clinical vignettes. On average, physicians wanted to obtain preoperative ERCPs only for patients whose likelihood of a common bile duct stone was greater than 37 percent, although this also varied from 2.5 percent to 83 percent. Thus, patients receive varying recommendations for care depending on whom they see, according to this study.

Return to Contents

Health Insurance/Health Care Costs

Little is known about how consumers choose health insurance plans

With the advent and growth of managed care, choosing a health insurance plan involves more than simply selecting a system for financing medical care. Consumers also must select a set of providers and a system for delivering care. Little is known about how consumers actually choose health insurance plans, except that cost is an important factor. A recent literature review on this issue found that price was inversely related to the probability of choosing a health plan and positively related to the probability of switching or disenrolling from a health plan.

Less is known about how price interacts with other potentially important primary variables such as plan quality, convenience, breadth of coverage, and provider choice. Information is also lacking on the interaction of secondary variables—such as health status, income, and educational level—with price and how this interaction causes some groups to select certain types of health plans. According to the authors of the review, researchers are just beginning to look at whether plan report card data have an effect on enrollment decisions.

When traditional fee-for-service coverage dominated and consumers paid very little out-of-pocket for health plan coverage, it was sufficient to know that price was important in making health plan decisions. Now, with the widespread growth and acceptance of managed care and with a greater proportion of consumers income going toward the purchase of health plans, more information is needed, note the researchers. Their work was supported in part by the Agency for Health Care Policy and Research (NRSA training grant T32 HS00053).

See "Consumer health plan choice: Current knowledge and future directions," by Dennis P. Scanlon, Michael Chernew, Ph.D., and Judith R. Lave, Ph.D., in the Annual Review of Public Health 18, pp. 507-528, 1997.

HMOs' comparative advantage in providing preventive services to women may be eroding

Women enrolled in health maintenance organizations (HMOs) in 1987 were more likely to have received Pap smears and breast exams within the last year and to have ever had a mammogram than women with fee-for-service (FFS) coverage. However, by 1992, HMOs had lost this comparative advantage, according to an Agency for Health Care Policy and Research study.

The study found that an increase in the probability of HMO enrollment was associated with more than twice the likelihood of having received Pap smears and breast exams among women aged 18 to 64 in 1987. Overall in 1992, only 65 percent of women reported they had received Pap smears (aged 18 to 64) and breast exams (aged 30 to 64); these levels are similar to those in 1987 (66 percent and 68 percent, respectively). Substantially more women aged 50 to 64 had ever received a mammogram in 1992 than in 1987 (80 percent vs. 50 percent), which is consistent with the tremendous growth in mammogram use during the period between the two surveys.

Women's use of these preventive services differed significantly between HMO and FFS enrollees in 1987, but was not statistically significant in 1992, note Robin M. Weinick, Ph.D., and Karen M. Beauregard, M.H.A., the study's authors. They analyzed data from the 1987 National Medical Expenditure Survey and the 1992 National Health Interview Survey to evaluate changes over time in use of Pap smears, breast exams, and mammograms.

The results of this study also point to characteristics of women who may be at risk for underutilization of preventive screening services, even though they have private health insurance coverage. These include women with less than 12 years of education, those who have never been married, and women who report no usual source of health care or have a negative attitude toward health care.

More details are in "Women's use of preventive screening services: A comparison of HMO versus fee-for-service enrollees," by Dr. Weinick and Ms. Beauregard, in the June 1997 issue of Medical Care Research and Review 54(2), pp. 176-199. Reprints (AHCPR Publication No. 97-R081) are available from the AHCPR Publication Clearinghouse.

State reimbursement guidelines reduce the number of spinal fusion surgeries for back-injured workers

In 1988, the Washington State Department of Labor and Industries, which pays for most worker's compensation costs in the State, established guidelines which required that requests for spinal fusion surgery for workers with back injuries be screened for appropriateness before authorization for reimbursement. This led to a substantial decline in lumbar fusion surgery for injured workers.

The use of spinal fusion in the treatment of low back pain in general, and for occupational lumbar injuries specifically, continues to be controversial. There is no firm consensus about clinical indications for this surgery, and marked geographic variation in use of spinal fusion exists. Also, when compared with nonfusion spinal surgery, spinal fusion is associated with higher rates of complications, longer and more costly hospital stays, and widely variable success rates.

During 1987 through 1992, the lumbar fusion rate for Washington State showed a 26 percent decline compared with a 3 percent decrease for all lumbar operations. After November 1988, when the guidelines went into effect, the rate of spinal fusion surgery declined 33 percent, whereas rates for nonfusion operations stayed about the same, according to a study by the Back Pain Patient Outcomes Assessment Team. The study was supported in part by the Agency for Health Care Policy and Research (HS06344 and HS08194). The researchers used the Comprehensive Hospital Abstract Reporting System, which contains all patient discharge records for non-Federal, acute care hospitals in Washington State, to calculate the number of spinal fusions performed in the State before and after the guidelines went into effect.

The lumbar fusion guidelines were jointly authored by orthopedists, neurosurgeons, and specialists in occupational health and were designed to be nonadversarial and flexible to make them more acceptable to practicing physicians. Also, the rationales for the guidelines were disseminated to clinicians performing spinal fusions during the months preceding their implementation. This approach may have increased the subsequent effects of the guidelines on physician behavior. Also, insurance companies profiling of surgeons for frequency of elective procedures, which began in the 1990s, may have resulted in a more conservative approach by surgeons toward all elective operations, conclude the PORT researchers.

More details are in "Impact of workers compensation practice guideline on lumbar spine fusion in Washington State," by Kenneth Elam, M.D., M.P.H., Victoria Taylor, M.D., M.P.H., Marcia A. Ciol, Ph.D., and others, in Medical Care 35(5), pp. 417-424.

Home care may not be less expensive than institutional care for some ventilator-assisted individuals

Persons who need ventilators on a daily basis in order to breathe are typically adults who have a degenerative neuromuscular disease (46 percent), spinal chord trauma or disease (18 percent), or chronic lung disease (25 percent). Recent initiatives to discharge long-term, ventilator-assisted individuals (VAIs) home to their families instead of to an institution assume that home care will result in a better quality of life for the patient and also be more cost effective.

A recent study, which was supported by the Agency for Health Care Policy and Research (NRSA fellowship F32 HS00054), found that home care may not be less expensive than institutional care for some patients if the long-term economic impact of home care on the VAI's family is taken into consideration. The study was conducted by Mary Ann Sevick, Sc.D., R.N., and Douglas D. Bradham, Dr.P.H., of the Bowman Gray School of Medicine. The authors point out that 43 percent of caregivers surveyed in their study had to adjust their employment situations to accommodate their family members home placements. Of these, half had to stop working, one-third decreased their working hours, 13 percent changed jobs, and 3 percent increased their hours.

Caregivers, usually the women of the household, earned an average of $797 per month less in wages than they would have earned had they not become caregivers. In addition, these estimates do not include lost wages of the VAI or the extent to which the caregiver was financially dependent on the VAI, explain Drs. Sevick and Bradham. In 1995, they mailed a survey to caregivers of 1,404 VAIs in 37 States; they received 277 responses.

The researchers estimated an average monthly cost of $6,411 for formal home care services, such as skilled nursing, physical therapy, occupational therapy, and home health aid tasks. The average cost of home care increased by $960 to $12,483 per month, depending on how the researchers calculated the value of the caregiver's time. This ranged from $4.25 per hour to $10 or $11 per hour for homemaker services or attendant care, up to $38 per hour for hours spent functioning as a private duty nurse, which some caregivers did when they provided tracheostomy care, intravenous infusion, urinary catheter care, and the like. Depending on the estimate of long-term care costs and how the caregiver's time was valued, home care was more expensive than institutional care for between 5 and 37 percent of VAIs.

Details are in "Economic value of caregiver effort in maintaining long-term ventilator-assisted individuals at home," by Drs. Sevick and Bradham, in the March/April 1997 issue of Heart and Lung 26, pp. 148-157.

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