Most low-risk patients with pneumonia prefer to be treated at home, but few are given the option

One-third of the 3 million U.S. adults diagnosed with community-acquired pneumonia (CAP) each year are initially hospitalized. However, most patients who are at low risk of death from CAP prefer to be treated at home rather than the hospital. Even though the majority of low-risk patients fare well with outpatient treatment, few clinicians offer them this option, according to a study supported in part by the Agency for Health Care Policy and Research (HS06468). Identifying which patients are candidates for home treatment is one goal of the Pneumonia Patient Outcomes Research Team (PORT), which conducted the study.

Christopher M. Coley, M.D., of Massachusetts General Hospital and Harvard Medical School, PORT leader Wishwa N. Kapoor, M.D., M.P.H., of the University of Pittsburgh, and their colleagues explored preferences for home or hospital treatment among 159 patients diagnosed with low-risk CAP and enrolled in the multiple-site Pneumonia PORT study, for whom outpatient therapy would be a reasonable option. The team evaluated the patients' responses to seven low-risk pneumonia scenarios. These ranged from progressive return to usual health within 6 weeks, whether initially treated at the hospital or not, to slower rate of recovery with no improvement in fever or symptoms after 1 week, to serious complications after initial treatment with usual health returning after 12 weeks.

When asked, 80 percent of the patients preferred home-based care for the present episode of low-risk CAP, including 60 percent of those who were initially hospitalized. Nearly three-quarters of the patients were willing to pay an average of 24 percent of 1 month's household income to be assured of this preference. Nevertheless, 69 percent of patients felt that the attending physician alone made the decision for home or hospital treatment. Only 9 percent of patients felt that they alone, or in concert with family, made the decision. Fifteen percent felt the decision was made jointly with their physicians,and 7 percent said that "other determinants" played a key role.

Only 11 percent of patients recalled being asked if they had a preference for either site of care. The researchers conclude that most persons with low-risk CAP should be informed of their favorable prognosis and encouraged to play an active role in the decision for hospital or home care.

More details are in "Preferences for home vs. hospital care among low-risk patients with community-acquired pneumonia," by Christopher M. Coley, M.D., Yi-Hwei Li, Ph.D., Anne R. Medsger, M.S.Hyg., and others, which appears in the July 22, 1996, Archives of Internal Medicine 156, pp. 1565-1571.

Women and the elderly are less likely to receive life-saving drugs for heart attacks

Evidence-based guidelines issued by the American College of Cardiology and the American Heart Association advise physicians to treat patients suffering from heart attack (acute myocardial infarction, AMI) with medications such as thrombolytics (clot-dissolving drugs), beta blockers, and aspirin, in order to reduce complications and death. The guidelines advise against use of lidocaine hydrochloride to prevent the likelihood of primary ventricular fibrillation because it can lead to an increased risk of death, especially in uncomplicated AMI. Nevertheless, lidocaine is still used inappropriately in many patients with AMI, and appropriate medications are still underused in elderly patients and women, according to a study supported in part by the Agency for Health Care Policy and Research (HS07357).

Physicians need to follow AMI clinical practice guidelines more closely, especially for elderly patients and women, asserts Thomas J. McLaughlin, Sc.D., lead author of a recently published report of an ongoing study from Harvard Medical School and Harvard Pilgrim Health Care. The study was designed to test the dissemination of national evidence-based guidelines through use of local opinion leaders. Stephen B. Soumerai, Sc.D., the study's principal investigator, Dr. McLaughlin, and their colleagues reviewed the medical records of nearly 2,500 patients admitted to 37 Minnesota hospitals for suspected AMI during 1992 and 1993. Among patients eligible for effective, life-saving drugs, they observed moderately high rates of use of aspirin (81 percent), moderate use of thrombolytic agents (72 percent), and low use of beta blockers (53 percent). Twenty percent of patients with no indication for needing lidocaine received this potentially harmful agent.

For all four drugs, patients 75 years or older were significantly less likely to be treated than those aged 64 years or younger, regardless of whether the agent was likely to be beneficial or ineffective (lidocaine). This level of undertreatment of elderly patients was most pronounced for thrombolytic agents. Use of all study drugs was consistently lower for women than for men. After adjusting for age and hospital type, women were less likely than men to be treated with aspirin and thrombolytic agents. A similar trend was observed for the use of beta blockers and lidocaine. These results correspond with other studies that show less use of invasive cardiac procedures for women and the elderly compared with men and younger patients.

See "Adherence to national guidelines for drug treatment of suspected acute myocardial infarction," by Drs. McLaughlin and Soumerai, Donald J. Willison, Sc.D., and others, in the Archives of Internal Medicine 156, pp. 799-805, 1996.

Return to Contents

Hospital Use/Quality of Care

Neonatal ICU admission of normal weight infants could be avoided if intensive monitoring were available elsewhere

Half of infants admitted to neonatal intensive care units (NICUs) are of normal weight (5.5 pounds or heavier). Yet many of them could avoid the NICU if they could receive intensive monitoring in alternative care settings, conclude James E. Gray, M.D., M.S., and Marie C. McCormick, M.D., Sc.D., of the Joint Program in Neonatology of the Harvard Medical School, and their colleagues. They found that of normal-birthweight (NBW) infants admitted to ne NICU, only 59 percent received active therapy, such as intravenous infusions, respiratory support, or tube feeding. The remainder received only intensive monitoring and probably did not need to be admitted to an NICU if they could have been monitored in a less costly setting, explain the researchers.

They examined the medical records of 521 NBW infants admitted to a single NICU during 1989 and 1990, mostly for respiratory problems, jaundice, and congenital anomalies, and determined the infants' health status 6 months later via telephone interviews. The researchers found that the median length of hospitalization for NBW infants was 7.7 days (ranging from 1 to 110 days), and median hospital charges were $5,222 (ranging from $565 to $317,820). Overall, about 10 percent of NBW infants required rehospitalization, but 30 percent of those with congenital anomalies were readmitted to the hospital. Infants with congenital anomalies had significantly higher charges and longer hospital stays than other NBW infants.

The major determinants of NICU admission among NBW infants were congenital anomalies (21.6 percent), prematurity despite NBW (22 percent), and acute complications of the neonatal period. Congenital malformations accounted for 22 percent of NBW admissions, 35 percent of hospital days, and 45 percent of total charges in the newborn period.

Details are in "Normal birth weight intensive care unit survivors: Outcome assessment," by James E. Gray, M.D., M.S., Dr. McCormick, Douglas K. Richardson, M.D., M.B.A., and Steven Ringer, M.D., Ph.D., in the June 1996 issue of Pediatrics 97(6), pp. 832-838.

Race and sex influence use of hospital-based procedures in California

Women and minorities in California receive fewer hospital-based procedures than men and whites in that State, according to a study supported by the Agency for Health Care Policy and Research (HS07558). It found that men had much better odds than women of undergoing seven of nine high-technology medical procedures. Whites had greater odds for undergoing five of the nine procedures than blacks, three of the nine procedures compared with Latinos, and two of the nine compared with Asians, even after controlling for patients' insurance status, age, diagnosis, and number of medical conditions.

Analysis of California hospital discharges in 1989 and 1990 showed that the odds ratio favored men over women for seven of the nine procedures studied, which included hip replacement, pacemaker implant, endarterectomy (that is, removal of plaque deposits from coronary arteries), angioplasty, defibrillator implant, heart transplantation, and coronary bypass surgery. Whites showed higher odds than blacks of undergoing angioplasty, endarterectomy, coronary bypass surgery, defibrillator implant, and kidney transplantation. Whites also had greater odds than Latinos of undergoing coronary bypass surgery, kidney transplantation, and angioplasty. Finally, whites showed higher odds than Asians for receipt of angioplasty and endarterectomy. However, Asians showed favorable odds compared with whites for hip replacement.

The findings from this study have tremendous policy implications. Providing insurance and admission to high technology hospitals may not ensure equal access to high technology treatments for women and minorities. In order to equalize utilization of these procedures, it may be necessary for policymakers to address not only financing, but also the social and clinical issues that affect allocation of technologies.

It is not yet clear which roles, if any, referral patterns, patient preferences, or evidence of clinical effectiveness play in a provider's selection of candidates for various procedures, comments Mita K. Giacomini, M.P.H., Ph.D., of McMaster University, author of the study. A more disturbing implication is possible discrimination on the part of the provider. A critical question is how well sophisticated technologies—often supported by public funding but developed and tested primarily on whites and men&3151;serve the needs of minority and female patients with the target diseases, concludes Dr. Giacomini.

More details are in "Gender and ethnic differences in hospital-based procedure utilization in California," by Dr. Giacomini, in the June 10, 1996, Archives of Internal Medicine 156, pp. 1217-1224.

Return to Contents

Elderly/Minority Health

Elderly persons' attitudes toward nursing homes predict their future use of these facilities

Health and financial status are two of the factors that determine whether an elderly person will be admitted to a nursing home, but attitude plays an important role as well. Elderly persons who feel that there is lots to do in a nursing home are more likely to later use a nursing home and to stay longer at a home than elderly persons who do not share this opinion, concludes a study supported in part by the Agency for Health Care Policy and Research (T32 HS00032). It shows that those with favorable attitudes toward nursing homes had 1.5 times increased odds of subsequently using nursing home services and 17 percent longer nursing home stays than other elderly persons. Prior nursing home use did not directly affect subsequent attitudes toward nursing homes.

Planners and policymakers need to recognize the importance of these attitudes as a predictor of subsequent nursing home use. They also need to develop opportunities to include patient preferences and attitudes in long-term care treatment plans, suggests Donna J. Rabiner, Ph.D., of the National Center for Health Promotion, VA Medical Center, Durham, NC, author of the study. She analyzed data from the National Long-Term Care Survey of impaired elderly adults living at home in 1982, who were reinterviewed in 1984. This survey provides data on the health and functional status of chronically disabled elderly persons in the United States and their use of hospital, medical, home health, community-based, and institutional care, as well as attitudes regarding this use.

Survey analysis pointed to four "predisposing" factors that predicted the likelihood of any nursing home stay from 1982-1984: black race, being married, advanced age, and the belief that there is lots to do in a nursing home. This attitude increased the odds of nursing home use. Being black or married reduced the odds by 76 and 42 percent, respectively. Each additional year of age increased the odds of having a nursing home stay and the total months spent in a nursing home. Being married reduced total length of stay by 24.3 percent.

Two "enabling" factors predicted the likelihood of any nursing home stay from 1982 to 1984: the number of children and residence in either the north central region or the west. The presence of each child reduced the odds by 11 percent, while living in the north central or western regions of the United States increased the likelihood of nursing home use. Finally, three "need" factors were significant predictors in the model. Each limitation in an activity of daily living (e.g., bathing, feeding oneself, or using the toilet) or instrumental activity of daily living (e.g., using the telephone, driving a car, or preparing a meal) increased the odds of a nursing home stay. In addition, those reporting their health to be good (as opposed to fair) were 17 percent less likely to have any subsequent nursing home stay.

See "Attitudes toward and use of subsequent nursing home services among a national sample of older adults," by Dr. Rabiner, in the August 1996 Journal of Aging and Health 8(3), pp. 417-443.

More than one-third of Mexican Americans lack health insurance

Hispanics have the least health insurance coverage of any ethnic/racial group in the United States. For instance, in 1980, 26 percent of Hispanics (including Mexican Americans) were uninsured compared with 18 percent of blacks, and 9 percent of whites. Currently, more than one-third of Mexican Americans are without health insurance, according to a study supported by the Agency for Health Care Policy and Research (HS07397).

Uninsured Mexican Americans—who are mostly poor and have low levels of education—are those most in need of health care but are the least likely to receive it, conclude researchers at the Mexican-American Medical Treatment Effectiveness Research Center at the University of Texas Health Science Center.

They interviewed a random sample of 501 Mexican Americans from San Antonio about their sociodemographic characteristics, perceived health status, health insurance coverage, and sources of health care. The survey was conducted in the summer and early fall of 1992.

More than one-third of those surveyed under the age of 65 years were uninsured. Compared with privately insured Mexican Americans, those who were uninsured or publicly insured were less educated, poorer, and more likely to be unemployed. Uninsured or publicly insured individuals were more in need of health care, that is, they were more apt than privately insured persons to consider themselves in fair or poor health. Yet, uninsured Mexican Americans were least able to obtain health care. For instance, 79 percent of publicly insured Mexican Americans reported having personal physicians and making regular visits compared with 36 percent of their uninsured counterparts. Persons with private insurance, regardless of health care need (reporting excellent to poor health), had similar access to personal physicians (63 percent and 69 percent, respectively). Uninsured persons, regardless of health care need, also had similar, but substantially lower, access to personal physicians (33 percent and 40 percent, respectively).

The finding that poor Mexican Americans with health insurance did use the health services available demonstrates that health care access may improve health care outcomes for Mexican Americans. But more comprehensive community-based campaigns to promote health and better use of health services among underprivileged groups should be developed, urge the researchers.

More details are in "Health care access among Mexican Americans with different health insurance coverage," by Robert P. Trevino, M.D., Fernando M. Trevino, Ph.D., M.P.H., Rolando Medina, M.D., M.P.H., and others, in the Journal of Health Care for the Poor and Underserved 7(2), pp. 112-121, 1996.

Return to Contents

HIV/AIDS Research

Symptoms such as fatigue and fever serve as a barometer of physical functioning in AIDS patients

Symptoms such as fatigue, fever, and neurologic problems are a better guide to the physical functioning of AIDS patients than biologic and physiologic factors such as current infections or time since diagnosis. These symptoms explain more than half of how well a patient with AIDS is functioning. Adding biologic and physiologic factors only explains 2 percent more of the differences in patients' physical functioning, according to a study supported by the Agency for Health Care Policy and Research (HS06239).

Biologic and physiologic variables such as weight loss, lower hemoglobin level, current non-pneumonia bacterial infections, Mycobacterium avium complex infection, oral Pneumocystis carinii pneumonia (PCP) prophylaxis, and cigarette smoking were associated with worse physical functioning. Use of zidovudine (ZDV) was associated with better physical functioning. These factors explained 29 percent of variation in physical functioning among patients.

However, symptoms alone explained nearly twice (56 percent) the variation. In each case, better mental health was associated with fewer symptoms. Lower total white blood cell count, longer time since diagnosis, weight loss in the last year, and more episodes of PCP were all associated with more fatigue. Weight loss and asthma were associated with more neurologic symptoms, and ZDV use in the last month was associated with fewer neurologic symptoms. This analysis by Massachusetts researchers was based on a conceptual model which proposed that symptoms mediate the relation between biologic and physiologic variables and physical functioning. The researchers used the model, patient interviews, and medical chart reviews to predict physical functioning in 305 persons with AIDS from three sites in Boston.

The researchers point out several limitations to their work. First, as with all cross-sectional data, the direction of causal effects could not be determined. Second, only one measure of health-related quality of life—physical functioning—was examined, and these findings do not necessarily apply to other measures. Third, too few patients had current CD4 cell counts, so this important variable could not be included. Fourth, there are limitations related to the size of the study and the constraints of a natural experiment in which data were limited to what physicians decided to order in the course of routine clinical care. And finally, although the researchers enrolled a sociodemographically diverse population, there is no way to know if these findings also apply to other groups of persons with AIDS, such as women. The researchers also point out that although oral PCP prophylaxis was associated with worse physical functioning, when the number of PCP episodes was added to the model, oral PCP prophylaxis was no longer significant. For more information, see "Clinical predictors of functioning in persons with acquired immunodeficiency syndrome," by Ira B. Wilson, M.D., M.Sc., and Paul D. Cleary, Ph.D., in the June 1996 issue of Medical Care 34(6), pp. 610-623.

Return to Contents

AHCPR News and Notes

Ten new members have been appointed to the National Advisory Council

Ten new members have been appointed to the National Advisory Council for Health Care Policy, Research, and Evaluation, and the term of service of the council's Chairman, Walter J. McNerney, M.H.A., has been extended to May 1997. The council, which comprises 17 private-sector health care experts and consumers and 7 top Federal health officials, advises the Secretary of the Department of Health and Human Services and the Administrator of the Agency for Health Care Policy and Research on matters elating to AHCPR's activities to enhance the quality, appropriateness, and effectiveness of health care services and access to such services.

The 10 new members of the council are:

• Richard E. Behrman, M.D., J.D, Managing Director, Center for the Future of Children, The David and Lucile Packard Foundation, Los Altos, CA.
• Nancy Wilson Dickey, M.D., Chair, Board of Trustees, American Medical Association, and Program Director, Family Practice Residency Department of the Brazos Valley, and Associate Professor, Department of Family and Community Medicine, Texas A&M University, College Station, TX.
• Jose Julio Escarce, M.D., Ph.D., Senior Fellow, Leonard Davis Institute and Assistant Professor, School of Medicine, University of Pennsylvania, Philadelphia, PA.
• Sharon C. Kiely, M.D., Director of Health Services, Primary Care Health Services, Inc., Pittsburgh, PA.
• Ada Sue Hinshaw, Dean and Professor, University of Michigan School of Nursing, Ann Arbor, MI.
• Jeffrey P. Koplan, M.D., M.P.H., President, The Prudential Center for Health Care Research, Atlanta, GA.
• Woodrow Meyers, Jr., M.D., M.B.A., Director, Health Care Management, Ford Motor Company, Dearborn, MI.
• Martin Paris, M.D., M.P.H., Vice President, Medical Affairs, Tenet Healthcare Corp., Dallas, TX.
• Stephen M. Shortell, Ph.D., A.C. Buehler Distinguished Professor of Health Services Management, J.L. Kellogg Graduate School of Management, Northwestern University, Evanston, IL.
• W. Leigh Thompson, M.D., Ph.D., Chairman, Profound Quality Resources, Ltd., Charleston, SC.

They will join six existing council members whose 3-year terms will expire in May 1997. In addition to Walter McNerney, the returning members of the panel are:

• Helen Darling, M.A., Manager, Health Strategy and Programs, Corporate Benefits, Xerox Corporation, Stamford, CT.
• Robert M. Krughoff, J.D., President, Center for the Study of Services, Washington, DC.
• W. David Leak, M.D., Medical Director, Pain Control Consultants, Columbus, OH.
• Harold S. Luft, Ph.D., Director, Institute for Policy Studies, School of Medicine, University of California, San Francisco, CA.
• Edward B. Perrin, Ph.D., Professor, Department of Health Services, School of Public Health, University of Washington, Seattle, WA.

The appointment of a 17th council member is expected within the next several months. In addition to the private-sector members, the following Federal officials serve as ex officio members of the council: Director, National Institutes of Health; Commissioner, Food and Drug Administration; Administrator, Substance Abuse and Mental Health Services Administration; Director, Centers for Disease Control and Prevention; Administrator, Health Care Financing Administration; Assistant Secretary for Defense (Health Affairs); and Chief Medical Officer, Department of Veterans Affairs.

---

Return to Contents
Proceed to Next Section