New publications now available from AHCPR and NTIS

The following publications and final reports are now available from either the AHCPR Publications Clearinghouse or the National Technical Information Service (NTIS). Select hot links for ordering information.

Conference Summary Report—Moving Toward International Standards in Primary Care Informatics: Clinical Vocabulary. AHCPR Publication No. 96-0069. October 1996.

In November 1995, the Agency for Health Care Policy and Research and the American Medical Informatics Association cosponsored an international conference in New Orleans, LA, which was the beginning of a process to move toward international standards in the clinical vocabulary used in primary care. Standards for coding primary care data and a clinical vocabulary are needed to increase the research usefulness of documentation from primary care practitioners. Conference participants determined that no existing vocabulary is sufficient for the many needs of primary care, health statistics, billing, and health services research. In this report, the strengths and weaknesses of the current primary care vocabularies are identified, and targets are set for future research. Conference participants agreed on three vocabularies—the International Classification of Primary Care (ICPC), the Read Codes, and the Systematized Nomenclature of Human and Veterinary Medicine (SNOMED)—to use as building blocks for the standard. In addition, the group recommended that all primary care vocabularies be added to the Unified Medical Language System (UMLS). The group developed a framework to further efforts toward a standardized vocabulary and a plan to integrate clinical vocabularies into other ongoing public- and private-sector efforts.

Prospective Evaluation of Enigmatic Meningitis. AHCPR grant HS07682. 4/1/94 to 3/31/96. Vincent J. Quagliarello, M.D., Yale University, New Haven, CT.

This purpose of this pilot study was to determine whether clinical features at presentation can identify the high-risk subgroup of patients with enigmatic meningitis (fever, headache, and cerebrospinal fluid [CSF] inflammation, with a negative CSF Gram stain) who warrant hospitalization and urgent antibiotic therapy and which patients are at low risk and can be safely managed as outpatients. The researchers identified and followed 118 patients presenting with enigmatic meningitis to the two emergency rooms serving New Haven, CT. They found that most patients were young and healthy, and almost half presented during nonsummer months. The majority of patients (76 percent) were hospitalized and had a mean length of hospital stay of over 7 days, underwent cranial imaging (53 percent) and were treated with empiric intravenous antibiotics for presumptive bacterial meningitis (67 percent). According to the researchers, only 36 percent of patients benefited from their hospitalization, and only 10 percent benefited from empiric antibiotic therapy. In fact, only 15 percent of these patients had an established cause of their meningitis that was treatable. Patients most likely to benefit from admission and empiric antibiotic therapy presented with a focal neurological exam, abnormal mental status, and extreme abnormalities in the initial CSF formula (i.e., CSF granulocyte count greater than 50 percent, CSF glucose level less than 50 mg/dl).

To order abstract, executive summary, and final report, contact the National Technical Information Service. NTIS accession no. PB96-198171; 8 pp, $6.50 paper; $10.00 microfiche.

Public Policy and Prevention of AIDS at the State Level. AHCPR grant HS07981, 7/1/94 to 6/30/95. James N. Schubert, Ph.D., Northern Illinois University, Dekalb.

This study was designed to explore the consequences and origins of competing policy models of AIDS prevention at the State level. The researchers focused on States because of key role States play in the formulation and implementation of public preventive policies. Three broad theoretical objectives were addressed: (1) the systematic description of State AIDS policies, (2) explanation of State policies, and (3) assessment of their preventive effects on the AIDS epidemic within States. The original texts of 750 pieces of State legislation enacted during the period 1983 to 1992 were acquired and coded. The researchers assessed whether legislation reflected the appearance of policy models of contain-and-control or cooperation-and-inclusion for responding to the epidemic in the United States. These models were found to describe the policy choices made by the States. Policies of the cooperation-and-inclusion variety were significantly and substantially affected by economic and government capabilities. Cooperation-and-inclusion policy, containment policy, and prevention spending had statistically significant effects associated with lower than predicted incidence of AIDS among States with smaller caseloads. Inclusion policy showed much stronger effects than containment policy. There were no significant preventive effects associated with AIDS policy among States with high caseloads.

To order abstract, executive summary, and final report, contact the National Technical Information Service. NTIS accession no. PB96-176888, 196 pp; $38.00 paper, $14.00 microfiche.

Regional Conference on Local Adoption of Guidelines. AHCPR grant HS07967, 9/1/94 to 5/31/96. Christel Mottur-Pilson, Ph.D.

This project supported a 1-day research conference, held April 8, 1995 in Chantilly, VA, on local guideline adoption. Three hypotheses were tested: (1) guideline use and adoption are subject to regional variation; (2) interactive learning promises behavior change; and (3) guideline use is subject to institutional constraints. In general, information learned at the conference supported the above hypotheses. However, there were a number of unexpected results. For example, local variation in guideline adoption is always institution-specific. Thus, guideline adoption is a function of the particular institutional culture in a given region. Although institutions may share general structural similarities, there still may be fundamental differences between them. These differences have a strong impact on guideline adaptation and adoption. Professional responsibilities and functions predispose individuals to adapt guidelines. These professional roles cut across medical disciplines and age groups. This research suggests that guideline dissemination should not be separated out from adaptation and adoption. Furthermore, unless an institutional fiat decrees guideline use, a certain degree of guideline familiarity and acceptance must be present for a guideline to be adapted/adopted. Finally, institutional policy is the primary determinant of guideline use and adaptation. To order abstract, conference proceedings, and final report, contact the National Technical Information Service. NTIS accession no. PB96-203468; 145 pp, $31.00 paper, $14.00 microfiche.

Retention of Physicians in Community Health Centers. AHCPR grant HS07053, 2/1/92 to 10/31/94. Stephen M. Davidson, M.S.W., Ph.D., John Snow, Inc., Research and Training Institute, Boston, MA.

The inability of community health centers (CHCs) to retain physicians has been a perennial problem and was the focus of this three-phase study. The components of the study were to: (1) determine how long primary care physicians stay in CHCs; (2) pinpoint when in their tenure physicians are most likely to leave CHCs; and (3) identify factors that affect a physician's likelihood of remaining in a CHC, emphasizing those items which are under the control of CHC management. In phase one, the researchers found that physicians with an obligation to the National Health Service Corps (NHSC) are less likely to leave during the first 2 years on the job but are at much greater risk of leaving after that time than physicians without an NHSC obligation. At the end of 5 years, 16.8 percent of physicians with an NHSC obligation remained at a CHC, compared with 35.6 percent of other physicians. In phase two, a national survey was conducted of two representative samples of primary care physicians who had practiced in CHCs: group one was still there, while group two had left. Survey results suggest that physicians who stay tend to be more satisfied with a wide range of factors associated with the CHCs than physicians who leave, and the strongest and most consistent predictor of physicians' satisfaction was their perception of the CHC's management. As might be expected, a more positive perception was associated with greater satisfaction and retention. In the third phase of the study, site visits were made to eight CHCs with different physician retention rates. Corresponding differences in both the behavior of CHC executives and the attitudes of their physicians were found. According to the researchers, their findings—detailed in the final report—provide specific guidance to center managers who are seeking actions they can take to extend the tenure of their physicians.

To order abstract, executive summary, and final report, contact the National Technical Information Service. NTIS accession no. PB96-203450; 77 pp, $25.00 paper, $10.00 microfiche)

Return to Contents

Research Briefs

Albertsen, P.C., Fryback, D.G., Storer, B.E., and others (1996, July). " The impact of co-morbidity on life expectancy among men with localized prostate cancer." (AHCPR grant HS06770). The Journal of Urology 156, pp. 127-132.

This paper explores the extent to which coexisting medical conditions (comorbidity) decrease survival among men with prostate cancer. The researchers evaluated three indexes of comorbidity, not originally designed for use in prostate cancer patients, to determine whether they could predict patient death more accurately than those based on patient age alone. The indexes were used to retrospectively predict length of survival among elderly men identified by the Connecticut Tumor Registry who were diagnosed with prostate cancer sometime during 1971 to 1976. The Kaplan-Feinstein index identifies 12 categories of comorbid illnesses, the index developed by Charlson and colleagues defines two groups of diseases with varying weights for severity of the condition, and the index of coexistent disease evaluates 14 medical conditions that are graded into four severity levels. Nearly all men with severe comorbidities according to these indexes had died within 10 years. Since life expectancy greater than 10 years is frequently cited as a criterion for recommending aggressive intervention, the indexes should assist clinicians who advise patients 65 to 75 years of age at diagnosis concerning appropriate treatment options for newly diagnosed, localized prostate cancer.

Clancy, C.M., and Kamerow, D.B. (1996, July). "Evidence-based medicine meets cost-effectiveness analysis." Journal of the American Medical Association 276(4), pp. 329-330.

In this editorial, Carolyn Clancy, M.D., Director of the Center for Primary Care Research and Acting Director of the Center for Outcomes and Effectiveness Research, and Douglas B. Kamerow, M.D., M.P.H., Director of the Office of the Forum for Quality and Effectiveness in Health Care, Agency for Health Care Policy and Research, note current efforts to narrow the gap between research evidence and medical practice. These efforts include clinical practice guidelines and other evidence-based recommendations that should reduce inappropriate variation in practice and associated expense. However, the ultimate impact of these efforts for practicing physicians is not clear. Physicians who consider these guidelines and adjust them to fit their particular practice setting refute charges offered by skeptics of "cookbook medicine." But whether physicians' interpretations of practice guidelines lead to reduced practice variation, better patient outcomes, and cost savings is an open question. Dr. Clancy points out that evidence-based recommendations and cost-effectiveness analysis are implicitly intended for different target audience: clinicians and policymakers, respectively. She concludes that, if the goal is to move practice toward evidence-based medicine, it is necessary to study and clarify the information needs of practitioners and determine how relevant information is best delivered to them.

Freedman, V.A., and Kemper, P. (1996). "Designing home care benefits: The range of options and experience." Journal of Aging & Social Policy 7(3/4), pp. 129-148.

Public funding for home care services in the United States has been increasing over the past decade. The authors, Vicki Freedman, Ph.D., of RAND, and Peter Kemper, Ph.D., formerly of the Agency for Health Care Policy and Research and now with the Center for Studying Health System Change in Washington, DC, present a framework for identifying important home care benefit design decisions and review the experience of 55 home care programs where existing designs have been adopted into practice. Fundamental home care benefit design questions include what type of benefits are provided, how the benefit level is determined, what services are covered and who can provide the services, who has the authority over the mix of services used, and who chooses the provider. The authors describe and provide examples of three basic benefit designs that predominate: service entitlements, managed-service benefits, and cash disability allowances. However, their review of U.S. and foreign government programs and private long-term care insurance policies also identifies examples of a wide variety of designs being used in practice.

Reprints (AHCPR Publication No. 96-R128) are available from the AHCPR Publications Clearinghouse.

Humphreys, B.L., Hole, W.T., McCray, A.T., and Fitzmaurice, J.M. (1996, August). "Planned NLM/AHCPR large-scale vocabulary test: Using UMLS technology to determine the extent to which controlled vocabularies cover terminology needed for health care and public health." Journal of the American Medical Informatics Association 3(4), pp. 281-287.

This article, by J. Michael Fitzmaurice, Ph.D., Director of the Center for Information Technology, Agency for Health Care Policy and Research, and his colleagues from the National Library of Medicine (NLM), describes the joint efforts of NLM and AHCPR to sponsor a large-scale vocabulary test. The purpose of the test is to determine the extent to which a combination of existing health-related classifications and vocabularies covers vocabulary needed in information systems supporting health care, public health, and health services research. The test vocabularies are the 30 that are fully or partially represented in the 1996 edition of the Unified Medical Language System (UMLS) Metathesaurus, plus three planned additions: the portions of SNOMED International not in the 1996 Metathesaurus, the Read Clinical Classification, and the Logical Observations Identifiers, Names, and Codes (LOINC) system. These vocabularies are available to testers through a special interface to the Internet-based UMLS Knowledge Source Server. The test will determine the ability of the test vocabularies to serve as a source of controlled vocabulary for health data systems and applications.

Reprints (AHCPR Publication No. 96-R126) are available from the AHCPR Publications Clearinghouse.

McDowell, I., and Newell, C. (1996). Measuring Health: A Guide to Rating Scales and Questionnaires. (AHCPR grant HS06206). New York: Oxford University Press.

This reference book provides a critical overview of the field of health measurement, with a technical introduction and discussion of the history and future directions of health measurements. It covers measurements of physical disability, social health, psychological well-being, depression, mental status, pain, general health status, and quality of life. The book is intended for researchers from the medical and social sciences and for health professionals wishing to evaluate the progress of their patients. Its principal aim is to guide readers in choosing among rival health measurement methods and to score the instrument chosen. The authors give full descriptions and copies of over 80 health measurement methods, summarize the reliability and validity of each, and provide the information readers need to select the most appropriate measurement for their purposes and then to apply and score the method chosen.

Pathman, D., Konrad, T.R., Freed, G.L., and others (1996, September). "The awareness-to-adherence model of the steps to clinical guideline compliance: The case of pediatric vaccine recommendations." (AHCPR grant HS07286). Medical Care 34(9), pp. 873-889.

The model suggests the necessary steps that precede a physician's adoption of a clinical practice guideline: (1) the physician must first become aware of the guideline, (2) intellectually agree with it, (3) decide to adopt it in his or her practice, and (4) succeed in following it at appropriate times, that is, adhere to it. The researchers used data on family physicians' and pediatricians' use of national recommendations for selected pediatric vaccines as a test case for assessing the model they developed. They mailed questionnaires to 3,014 family physicians and pediatricians in nine States. In the case of the recommendation to provide hepatitis B vaccine to all infants, guideline awareness among respondents was 98.4 percent; agreement, 70.4 percent; adoption, 77.7 percent; and adherence, 30.1 percent. Data for 87.9 percent of physicians fit the model at every step. Significant deviation from the model occurred only for the 11 percent of physicians who adopted the hepatitis B recommendation without agreeing with it. According to the researchers, this model may prove useful in identifying ways to improve physicians' adherence to a variety of guidelines by demonstrating where physicians fall along the path to adherence, which physicians are at greatest risk for not attaining each step in the path, and factors associated with a greater likelihood of attaining each step towards guideline adherence.

Weissman, J.S., Levin, K., Chasan-Taber, S., and others (1996). "The validity of self-reported health-care utilization by AIDS patients." (HS06239). AIDS 10, pp. 775-783.

These researchers examined the validity of self-reported health care use by nearly 300 persons with AIDS at three provider sites in Boston, based on personal interviews within 4 months of hospitalization, ambulatory visits, and hours of home care during 1990 and 1991. They identified reporting error by differences between self reports and medical/financial records. Results showed that AIDS patients' overall reports of their use of health care services were accurate. However, patients were somewhat better at reporting major events, such as hospitalizations, than they were at reporting more frequent, and perhaps less emotionally and financially prominent events, such as ambulatory visits or hours of home care. Biases may exist for persons with high use, who underreported all types of services, and for those identified by interviewers as having recall problems. The researchers conclude that smaller studies of patients with AIDS which analyze health care costs may rely on data provided directly by patients.

Wong, H.S. (1996). "Market structure and the role of consumer information in the physician services industry: An empirical test." Journal of Health Economics 15, pp. 139-160.

To draw reliable inferences about the effects of various health care proposals and to explain economic phenomena observed in the physician services industry, it is important to model the physician services market structure properly. Using private and public sources and applying Panzar and Rosse's econometric test of market structure, Herbert S. Wong, Ph.D., of the Agency for Health Care Policy and Research, presents two novel empirical analyses. First, he provides an empirical test that simultaneously evaluates three possible alternatives for the market structure for physician services: monopoly, perfect competition, and monopolistic competition. Second, he provides a more direct test of the hypothesis which suggests that greater physician density raises the search cost of obtaining consumer information and leads to higher prices. For primary care and general and family practice physicians, the monopolistically competitive model prevailed over the competing hypotheses.

Reprints (AHCPR Publication No. 97-R005) are available from the AHCPR Publications Clearinghouse.

Zapka, J.G., Bigelow, C., Hurley, T., and others (1996, July). "Mammography use among sociodemographically diverse women: The accuracy of self-report." (AHCPR grant HS06874). American Journal of Public Health 86(7), pp. 1016-1021.

This study analyzed the accuracy of self-report of mammography among 392 ethnically diverse women aged 50 to 74. Women were randomly selected for telephone or mail surveys. Results showed that 31 percent of women reported accurately the exact month and year of their most recent mammogram, 54 percent reported accurately within 3 months, and 83 percent reported accurately within the year. Greater accuracy was associated with mammogram recency, white race, and non-Hispanic ethnicity but not with age, education, or income. Most women could correctly report the reason for, the findings of, and the payer of their mammograms but knew little about how much they or their insurance paid. The authors conclude that clinical studies requiring more precise information than the year of last mammogram should be cautious when using self-reported data.

Zimmerman, J.E., Wagner, D.P., Seneff, M.G., and others (1996). "Intensive care unit admissions with cirrhosis: Risk stratifying patient groups and predicting individual survival." (AHCPR grant HS05787). Hepatology 23(6), pp. 1393-1401.

This study examines the usefulness of the Acute Physiology, Age, and Chronic Health Evaluation (APACHE III) prognostic system for risk-stratifying groups of intensive care unit (ICU) patients with cirrhosis and in predicting individual survival. The researchers used data for 17,440 ICU admissions at 40 American hospitals to select 117 of the 537 patients with a history of cirrhosis who were ventilated on ICU day 1, a group known to have a high mortality rate. The most important determinants of risk for hospital death on ICU day 1 were the acute physiology score of APACHE III, ICU admission diagnosis, and operative status. Daily changes in the acute physiology score caused a rise or fall in the probability of hospital death and were useful in assessing individual response to therapy. According to the researchers, APACHE III accurately risk-stratifies critically ill patients with cirrhosis because it accounts for many of the factors known to influence prognosis. This capability can be used to assess severity of illness and risk-stratify patients with cirrhosis during clinical trials.

Return to Contents

AHCPR Publication No. 97-0003
Current as of October 1996