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

Healthcare Costs and Financing

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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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

Insurance reimbursement levels greatly affect use of chiropractic services

About 7 percent of persons in the United States visit chiropractors, the third largest group of health professionals after medical doctors and dentists. An increasing number of health insurance plans are covering chiropractic services. But the amount the plan reimburses a patient for these services greatly affects their use, according to a recent study supported in part by the Agency for Health Care Policy and Research (HS06920). It shows that when patients have to share 25 percent or more of the cost, they decrease their chiropractic expenses by half.

This pattern makes chiropractic care more sensitive to cost than overall health and dental care and about as sensitive as outpatient mental health services, concludes Paul G. Shekelle, M.D., Ph.D., of RAND, principal investigator of the study. The researchers analyzed data from the RAND Health Insurance Experiment, a randomized controlled trial on the effects of cost sharing on use of health services. Families in six U.S. sites (Dayton, OH, Seattle, WA, Fitchburg and Franklin County, MA, and Charleston and Georgetown County, SC) were randomized to receive fee-for-service (FFS) care that was free or required one of several levels of cost sharing, or to receive care from a health maintenance organization. Families were followed for 3 or 5 years, and chiropractic use among persons in the HMO and FFS plans was compared.

In the fee-for-service plans, persons who had to pay coinsurance of 25 percent or more decreased chiropractic expenditures by about half. Access to free chiropractic care among HMO enrollees increased chiropractic use nine-fold, compared with a contemporaneous sample of HMO enrollees who faced 95 percent cost sharing for chiropractic care. Among this latter group, access to free medical care in the HMO decreased by over 50 percent their use of chiropractic care relative to persons who faced 95 percent cost sharing for both medical and chiropractic care. This suggests that there is a substitution of medical care for chiropractic care, depending on price.

See "The effect of cost sharing on the use of chiropractic services," by Dr. Shekelle, William H. Rogers, PH.D., and Joseph P. Newhouse, Ph.D., in Medical Care 34(9), pp. 863-872, 1996.

Medicaid providers vary in cost and efficiency in managing chronic illnesses

Medicaid providers vary in how efficiently they manage patients who have hypertension, diabetes, or asthma, all chronic diseases for which quality outpatient management can reduce costly hospitalizations and emergency room visits. In a study supported by the Agency for Health Care Policy and Research (HS06170), researchers from Johns Hopkins University developed and applied a claims-data-based approach to study the performance of Medicaid providers who deliver ambulatory care to chronically ill persons with diabetes, hypertension, and asthma in the State of Maryland. They found that a large degree of misclassification on cost performance can occur unless allowances are made for the severity of illness of a provider's patients (casemix adjustment) and that providers are not equally efficient in managing all chronic illnesses.

According to lead author Neil R. Powe, M.D., M.P.H., M.B.A., and his colleagues, this study uncovered inconsistencies in the association of cost performance with clinical performance—that is, patients whose usual source of care is a high-cost office-based physician are more likely to spend time in the hospital than patients of lower cost physicians. On the other hand, high-cost providers were not consistently more likely than others to provide their patients with better followup care after hospital discharge and to have fewer instances of repetitive emergency room use.

The researchers point out, however, that without casemix adjustment, up to half of providers would be misclassified according to costs. For instance, the severity of illness of a provider's mix of patients explained 22 percent, 33 percent, and 36 percent of the variation in total payments for patients with hypertension, diabetes, and asthma, respectively. After adjusting for physicians' casemix, nearly 20 percent of 211 providers who treated all three illnesses were in the same cost group for all three illnesses and 43 percent of 223 providers who treated two of the three illnesses were in the same cost group for both illnesses. These findings were based on a cross-sectional analysis of Medicaid beneficiaries in Maryland with these three conditions, who were treated on an ambulatory basis by hospital-based outpatient departments, physician office-based providers, and community health centers during 1987 and 1988. These results may be useful to State Medicaid programs in structuring, guiding, and evaluating the care of their recipients with chronic illnesses and to other payers and providers who wish to profile the care of patients with these illnesses, conclude the researchers.

For more information, see "Systemwide provider performance in a Medicaid program," by Dr. Powe, Jonathan P. Weiner, Dr.P.H., Barbara Starfield, M.D., M.P.H., and others, in Medical Care 34(8), pp. 798-810.

Return to Contents

HIV/AIDS Research

Expected toxicity should guide choice of treatment for mild to moderate PCP

Despite the advent of medications to prevent Pneumocystis carinii pneumonia (PCP), one of the most serious AIDS-related illnesses, over 15,000 cases of PCP were reported in 1994. A recent study recommends that oral treatment for mild to moderate PCP in AIDS patients should be based on expected toxicity, since the three standard oral treatments for the condition are equally effective. The study, supported in part by the Agency for Health Care Policy and Research (HS07824), examined the toxicity and effectiveness of oral trimethoprim-sulfamethoxazole, dapsone-trimethoprim, and clindamycin-primaquine in 181 patients at 24 U.S. medical centers who were randomly assigned to receive one of the three treatments for 21 days. Albert Wu, M.D., M.P.H., of The Johns Hopkins Schools of Public Health and Medicine, and his colleagues also assessed the effects of these treatments on patient-reported health status measures, such as fatigue, symptoms, physical functioning, and disability.

One week following treatment, patients treated with clindamycin-primaquine reported the most improved health status, particularly in reduced pulmonary symptoms and disability, compared with patients treated with the other two regimens. However, 3 weeks after treatment, all patients reported further improvement in health status, and differences among groups were less evident. These health status outcomes were more sensitive than survival, treatment failure, and toxic effects to differences among treatment groups. They further supported the finding that the most common regimen, trimethoprim sulfamethoxazole, was not superior to the two alternative treatments, notes Dr. Wu.

All three medications were similarly effective, with about 9 percent of each group experiencing failure of their medication to reduce symptoms or correct the condition. However, one-fourth to one-third of patients in all treatment groups had to discontinue therapy because of one or more toxic reactions (36 percent for trimethoprim-sulfamethoxazole, 24 percent for dapsone trimethoprim, and 33 percent for clindamycin-primaquine). Drug toxicities varied, ranging from rash and nausea to liver damage (trimethoprim-sulfamethoxazole) and severe anemia (clindamycin-primaquine). Evidence of a patient's liver insufficiency at the beginning of therapy should prompt the clinician to consider a regimen other than trimethoprim-sulfamethoxazole, and severe bone marrow suppression may suggest a regimen other than clindamycin-primaquine.

Details are in "Comparison of three regimens for treatment of mild to moderate Pneumocystis carinii pneumonia in patients with AIDS," by Sharon Safrin, M.D., Dianne M. Finkelstein, Ph.D., Judith Feinberg, M.D., and others, in the Annals of Internal Medicine 124(9), pp. 792-802, 1996.

Low CD4 count signals the need to screen HIV/AIDS patients for CMV retinitis

Cytomegalovirus retinitis (CMV-R) is a major cause of blindness in persons infected with the human immunodeficiency virus (HIV) that causes AIDS. Nearly one in five persons with end-stage HIV disease (CD4 count below 100, normal is about 1,000) will develop this condition, according to a Multicenter AIDS Cohort Study (MACS), which is supported in part through an interagency agreement between the Agency for Health Care Policy and Research and the National Institute of Allergy and Infectious Diseases.

CMV-R can be diagnosed and treated through early screening and might be prevented through prophylactic medication, explains the study's lead author, Donald R. Hoover, Ph.D., of The Johns Hopkins University School of Public Health. Dr. Hoover and his colleagues at Johns Hopkins and other MACS centers recommend periodic ophthalmologic exams for CMV-R when a patient's CD4 count falls below 100 and intensified screening when the count falls below 50.

The researchers evaluated the prevalence of CMV-R in 367 HIV-infected men who were receiving zidovudine and Pneumocystis carinii pneumonia (PCP) prophylaxis and had CD4 cell counts below 100. They found that during a median 3-year period, 103 of the 367 men developed CMV disease, which caused retinitis in 71 percent, gastrointestinal disorders (mostly colitis) in 20 percent, pneumonitis in 8 percent, esophagitis in 12 percent, and nervous system disorders in 4 percent. Once a person's CD4 count fell below 50, the risk of developing CMV-R increased three-fold.

With a CD4 count of 100 as the starting point, the onset of CMV-R was associated with nearly a two-fold increase in the risk of death. Most progression to CMV-R and/or death occurred within 4 years after CD4 cell counts fell below 100, at which time only 11 percent of men were alive and free of retinitis. During these 4 years, almost 25 percent of the men had been diagnosed as having CMV-R.

The researchers point out that the ability to predict when initial manifestations of CMV-R will appear is important for planning long-range screening programs or entry into treatment studies for these individuals. For example, this study shows that 7 percent of HIV-infected homosexual men with a current CD4 count of 100 will need to be treated within 1.5 to 2 years for an initial clinical manifestation of CMV-R.

More details are in "Occurrence of cytomegalovirus retinitis after human immunodeficiency virus immunosuppression," by Dr. Hoover, Yun Peng, M.S., Alfred Saah, M.D., and others, in the July 1996 Archives of Ophthalmology 114, pp. 821-827.

Problems with HIV-related medications are common, but few patients require hospitalization

Current standard treatment for persons infected with the human immunodeficiency virus (HIV) that causes AIDS consists of antiretroviral drugs to reduce the amount of HIV in the blood and other medication to prevent Pneumocystis carinii pneumonia (PCP), a common and serious HIV-related infection. Adverse reactions to both types of medications are common, but few are severe enough to require hospitalization. HIV-infected persons whose CD4 lymphocyte cell counts decline to less than 100 cells/mm3 are more likely to have problems tolerating these medications than others whose immune systems are stronger. And, sex and race appear to be related to poor reactions to certain drugs, according to a study supported in part by the Agency for Health Care Policy and Research (HS07809). These adverse reactions can lead to dosage reduction, discontinuation of therapy, or the need to treat problems arising from the medication itself, explains Richard D. Moore, M.D., M.H.Sc., the study's principal investigator.

Dr. Moore and colleagues at The Johns Hopkins University School of Medicine calculated rates of adverse drug reactions to zidovudine, didanosine, zalcitabine, cotrimoxazole, and dapsone in 1,450 HIV-infected patients with CD4 counts of 500 cells/mm3 (approximately 1,000 cells/mm3 is normal) being seen at an urban, university-based HIV clinic from 1989 through 1994. Adverse drug reactions were 16 per 100 person years for dapsone, 24 for didanosine, 26 for zidovudine, 26 for cotrimoxazole, and 37 for zalcitabine. In addition, women were nearly three times more likely than men to have an adverse reaction to didanosine and 1.5 times more likely to react poorly to cotrimoxazole. Whites were 1.6 times more likely than blacks to have a bad reaction to cotrimoxazole. Finally, only 6 percent of adverse reactions to these medications required hospitalization, and none resulted in death.

Patient age or injecting drug use did not influence the rate of adverse drug reactions. For all of the drugs studied except dapsone, the risk of adverse reactions increased progressively as the person's CD4 count declined. Depending on the drug, reactions ranged from blood, kidney, and nervous system disorders to gastrointestinal problems, pancreatitis, and rashes.

Details are in "Adverse events from drug therapy for human immunodeficiency virus disease," by Dr. Moore, IIana Fortgang, A.B., Jeanne Keruly, B.S.N., and Richard E. Chaisson, M.D., in The American Journal of Medicine 101, pp. 34-40, 1996.

Return to Contents

AHCPR News and Notes

AHCPR's Administrator announces resignation

Clifton R. Gaus, Sc.D., Administrator of the Agency for Health Care Policy and Research, has announced his resignation. He released the following statement on December 2, 1996.

"It has been my privilege to lead AHCPR at a critical juncture in its history and at a time of dramatic change in the American health care system. Now that the goals that I have set forth for AHCPR have been achieved through the hard work and dedication of its staff, it is time for me to return to the private sector.

"I am confident that I leave an AHCPR that is strategically positioned to better help clinicians, health plans, consumers, and policymakers improve the quality of health care services and achieve value in health care spending. To ensure a smooth transition, I will be staying on until HHS Secretary Donna Shalala names a successor."

AHCPR and Kaiser examine consumers' use of quality information

Health care consumers value quality-related information, but currently they do not use it to make key decisions. Instead, they rely on personal recommendations from people they know and trust, according to a national poll co-sponsored by AHCPR and the Kaiser Family Foundation. The results of the poll, which were released at a national conference held in Washington, DC, on October 29-30, are significant because they provide the first national, quantitative snapshot of how quality affects consumer decisions.

For example, in choosing a health plan, quality of care is more of a concern (42 percent) than low costs (18 percent), a wide choice of doctors (17 percent) and a range of benefits (14 percent). However, consumers' ultimate selections actually are based on information they receive from their personal physicians (59 percent) and their family and friends (57 percent).

If they need to find a new doctor, Americans also are more likely to turn to someone they know for a recommendation. Fifty-seven percent would ask their current doctor, and 51 percent would consult friends and family. When selecting a physician, those surveyed said they are more concerned with how well a doctor communicates with patients and if he or she shows a caring attitude (84 percent) and whether the doctor is board certified (71 percent). By contrast, only 25 percent were concerned with whether a doctor has been highly rated by an independent organization.

Consumers' views on selecting a hospital continue these trends. When choosing a hospital, an individual's regular doctor has the most influence (65 percent), followed by friends or family (60 percent), and then patient surveys (43 percent). In choosing between two hospitals, a majority of consumers (72 percent) would choose the hospital that is familiar to them over the hospital rated higher in quality by experts.

Americans' lukewarm reception to quality information produced by independent organizations could reflect their lack of familiarity with such information; only 39 percent said they had seen quality comparisons within the last year. In addition, seven out of ten (69 percent) of those surveyed regard their family and friends as "good" sources of information about health plans because they share common concerns. For the same reasons, patient satisfaction surveys were valued as quality sources of information for those polled. When asked to name a "very believable" source about quality of care, family and friends were considered the most believable (50 percent), followed by patient surveys (34 percent) and individual doctors (29 percent). Respondents consistently considered patient satisfaction surveys to be an influential source of information, after personal doctor and family and friends, in choosing a health plan (45 percent) and a doctor (37 percent).

The vast majority of those surveyed say specific quality-of-care information is "very important" to know when choosing a health plan. The quality-of-care indicators considered very important were: how well a plan cares for members who have health problems (90 percent), how easy it is to get needed care (88 percent), how successfully early disease is detected (87 percent), and how well a health plan keeps its members as healthy as possible (79 percent).

Most Americans (88 percent) believe there is a role for the Government in the quality of health care arena. A majority (52 percent) think the Government should both monitor health providers to ensure a minimum standard of quality and make sure information about quality is available to the public. Another 24 percent think the Government should only make sure information is available so people can make judgments about quality themselves, and 12 percent say the Government's only role should be to monitor for a minimum standard of quality of care.

"Americans as Health Care Consumers: The Role of Quality Information," is a nationally representative telephone survey of 2,006 adults conducted between July 26 and September 5, 1996. The margin of error is plus or minus three percentage points. The survey was designed by the Kaiser Family Foundation, AHCPR, and Princeton Survey Research Associates.

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