You Are Here: AHRQ Home > Research Findings > Research Activities > December 1998

Announcements

AHCPR invites research proposals focusing on vulnerable populations

The Agency for Health Care Policy and Research has issued a Request for Applications (RFA) seeking research proposals to develop and test measures of quality of care for vulnerable populations. As defined in the Final Report of the President's Advisory Commission on Consumer Protection and Quality in the Health Care Industry, vulnerable populations are those groups of people "made vulnerable by their financial circumstances or place of residence; health, age, or functional or developmental status; or ability to communicate effectively... [and] personal characteristics, such as race, ethnicity, and sex." AHCPR's RFA uses this definition, which encompasses populations whose vulnerability is due to chronic or terminal disease or disability.

The RFA is part of a series of initiatives that respond to the report by the President's Commission, which calls for a significant investment in the further development of research, tools, and information for patients, purchasers, and payers.

AHCPR will award up to $3 million in FY 1999 to support the first year of research for up to 10 projects. The funds stem from the recent $20 million increase in AHCPR's budget for investigator-initiated research. AHCPR's FY 1999 budget is approximately $171 million.

If you are interested in applying for funding under this RFA, you are requested (but not required) to send a letter of intent to AHCPR by January 29, 1999; funding applications must be received by March 15, 1999.

AHCPR is particularly interested in receiving applications from investigators not currently working on AHCPR-funded quality measures. We welcome applications from minority researchers and institutions.

To access the Request for Applications for Measures of Quality of Care for Vulnerable Populations for details and application instructions, consult the NIH Guide for Grants and Contracts, which published this RFA during the week of December 21-25, 1998, at http://grants.nih.gov/grants/guide/index.html.

AHCPR, SAMHSA, and NIMH award $2.1 million in grants to study mental health conditions in children

Grants totaling $2.1 million have been awarded to researchers at three universities and a hospital to study treatments for depression, attention deficit disorder, and oppositional defiant disorder in children and adolescents. This cluster of research projects is designed to improve mental health interventions in primary care settings such as clinics, schools, and doctors' offices. The research also will develop ways to improve the efficiency and cost effectiveness of treatments for this population.

The grants are funded jointly by three Department of Health and Human Services components: the Agency for Health Care Policy and Research, the Substance Abuse and Mental Health Services Administration (SAMHSA), and the National Institute of Mental Health (NIMH). The new grants are as follows:

Youth partners in care. (AHCPR grant HS09908). Principal investigator: Joan R. Asarnow, Ph.D., University of California, Los Angeles, CA. Project period 1998-2003; first-year funding $622,646. This study will evaluate the impact of a quality-improvement intervention in a managed care setting. The intervention focuses on educating the patient and the primary care provider about of depression and the best use of clinic resources. It will measure outcomes such as satisfaction with care, clinical symptoms, daily functioning, service use and costs, indirect costs, and parental psychological distress.
Treating oppositional defiant disorder in primary care. (AHCPR/NIMH Grant MH59462). Principal investigator: John V. Lavigne, Ph.D., Children's Memorial Hospital, Chicago, IL. Project period; 1998-2002; first-year funding $542,263. Oppositional defiant disorder (ODD) is the most common psychiatric disruptive behavior disorder among preschool-aged children, with long-term social consequences ranging from delinquency and substance abuse to high-risk sexual behavior in adolescence. This study will examine the effectiveness of using a psychological intervention in primary care pediatric settings to help identify and treat preschool children with ODD. It also will evaluate how well a 10-week training program on parenting skills reduces the incidence of ODD and how well the intervention reduces the use of ambulatory and emergency room care.
Enhancing ADHD treatment effectiveness by pediatricians and schools. (NIMH Grant MH59461). Principal investigator: Mark L. Wolraich, M.D., Vanderbilt University, Nashville, TN. Project period 1998-2002; first-year funding $465,903. This study is designed to implement a model program for children with attention deficit hyperactivity disorder (ADHD) in schools and physicians' practices. The model program will enhance communications between medical and educational professionals regarding the evaluation and management of these children. ADHD is considered to be one of the most important problems among school-age children and a pressing problem for schools.
Effectiveness of interpersonal psychotherapy (IPT-A) in school-based health clinics. (SAMHSA Grant SM52671). Principal investigator: Laura H. Mufson, Ph.D., Columbia University, New York, NY. Project period 1998-2001; first-year funding $460,747. This research will test the clinical and cost effectiveness of providing interpersonal psychotherapy for adolescents (IPT-A) for economically disadvantaged urban youths with depression. IPT-A is a guideline-based, mental health treatment focusing on improved social function, interpersonal problem-solving skills, and symptom reduction. Mental health professionals working from three Manhattan public school-based health clinics will be trained to deliver IPT-A as a short-term intervention. The research will measure such outcomes as clinical status, social functioning, school performance and attendance, and use of other medical and psychiatric services.

Applications now being accepted for director of AHCPR's primary care research program

The Agency for Health Care Policy and Research is accepting applications for the position of Director, Center for Primary Care Research (CPCR). Once on board, this individual will direct and support a program of health services research on primary care, including evaluations of the quality, costs, and effectiveness of primary care services, rural health care services and systems, and special populations.

Candidates must possess an M.D. or equivalent degree with board certification or eligibility in a primary care speciality such as general internal medicine, family practice, or pediatrics; or a degree or diploma from a professional nursing program accompanied by doctoral-level education with either a concentration in a field of nursing or in a closely related non-nursing field (e.g., health services research, health policy) directly applicable to the requirements of the position. In addition, to be considered for the position, applicants must possess a comprehensive knowledge of the current issues in primary care research and major public policy initiatives affecting primary care.

The salary range for this position is $80,658 to $136,700 per year.

AHCPR releases audiotapes from User Liaison Program workshops

The Agency for Health Care Policy and Research's User Liaison Program (ULP) coordinates and hosts workshops for State and local health officials. These workshops are designed to provide policymakers and other officials at the State and local levels with timely information on emerging and critical health care topics.

Audiotapes from a number of recent ULP workshops are listed below and are now available from the AHCPR Publications Clearinghouse. Please be sure to use the AV number when ordering.

CHIP: Implementing Effective Programs and Understanding Their Impacts, September 1998, Portland, OR (AHCPR 98-AV09).
Structuring Health Insurance Markets, May 1998, Lake Buena Vista, FL (AHCPR 98-AV08).
Strengthening the Rural Health Infrastructure, November 1997, Asheville, NC (AHCPR 98-AV07).
Managed Care and Persons with Disabilities and Chronic Illnesses, November 1997, Chandler, AZ (AHCPR 98-AV06).
Long-Term Care, October 1997, Washington, DC (AHCPR 98-AV05).
Workers' Compensation and Managed Care: Challenges and Opportunities, July 1997, Chicago, IL (AHCPR 98-AV04).
Promoting Public Health in an Era of Change, July 1997, Denver, CO (AHCPR 98-AV03).
Providing Services to Children with Special Health Care Needs, September 1997, San Diego, CA (AHCPR 98-AV02).
What Do We Do About the Uninsured? September 1997, Charlottesville, VA (AHCPR 98-AV01).
Integrated State Health Information Systems, July 1997, San Diego, CA (AHCPR 97-AV06).
Ensuring Quality Health Care: The Challenges of Measuring Performance and Consumer Satisfaction, June 1997, North Charleston, SC (AHCPR 97-AV05).
Market Forces and State Health Policy: Goals, Tools and Limitations, May 1997, Albuquerque, NM (AHCPR 97-AV04).
Integrated Delivery Systems in Managed Care: March 1997, Boston, MA (AHCPR 97-AV02).
Local Health Departments in a Managed Care Environment: Challenges and Opportunities, December 1996, St. Louis, MO (AHCPR 97-AV01).

New products available from AHCPR and NTIS

The following new products—including research reports, grant final reports, and software—are now available from the AHCPR Publications Clearinghouse and the National Technical Information Service (NTIS).

Building Bridges 1998 Research Conference Proceedings. The proceedings of the fourth annual Building Bridges Research Conference, held May 7-8, 1998, in Oakland, CA, are now available. The proceedings are divided into two sections-the first focuses on overcoming major challenges in clinical research, and the second is a review of new clinical research findings in the managed care setting. The Building Bridges conference series—jointly sponsored by the American Association of Health Plans, the HMO Research Network, AHCPR, and the Centers for Disease Control and Prevention—is a collaborative initiative to bring together the managed care and health services research communities. A limited number of copies of Improving the Public's Health Through Research Partnerships (Publication No. OM99-0001) are now available from the AHCPR Publications Clearinghouse.

For information about the 1999 Building Bridges Research Conference, to take place April 12-13, 1999, in Chicago, visit AAHP's Web site at http://www.aahp.org.

CCHPR Software. The 1998 Clinical Classifications for Health Policy Research (CCHPR) software is now available online from AHCPR's Web site. The CCHPR software helps researchers translate discharge data into meaningful information for decisionmaking. The CCHPR software program, a "clinical grouper," aggregates ICD-9-CM codes into a smaller number of clinically meaningful, relatively homogenous clusters, which can facilitate focused statistical analyses. CCHPR can be applied to all ICD-9-CM data from 1980 forward, with simple adjustments needed for data prior to 1993. CCHPR can be used in various types of analyses, including: cost-of-illness studies; cost-effectiveness analyses; comparisons of resource use across hospitalizations; and trend analyses. Select for online access to CCHPR software and related information.

To illustrate how CCHPR can be used with discharge data to track changes in national hospital trends, AHCPR applied CCHPR to a national sample of hospital discharge records, and the findings are now available from AHCPR. The aggregated statistics, available in Clinical Classifications for Health Policy Research: Hospital Inpatient Statistics, 1995 (AHCPR Publication No. 98-0049), group hospital discharges into clinical categories and provide information about mean length of stay, mean charges, mean age, sex, and in-hospital mortality. Copies are available from AHCPR Publications Clearinghouse.

Grant final reports. The following grant final reports are now available from NTIS. Each summary includes the title of the project, the principal investigator's name and affiliation, grant number and project period, and a brief description of the project.

Child Health Services Research: Building Research Capacity. Christopher B. Forrest, M.D., Ph.D., Johns Hopkins University, Baltimore, MD. AHCPR grant HS09320, project period 5/1/96 to 4/30/98.

On May 5, 1996, an expert meeting was held in Washington, DC, to discuss issues related to the development of an agenda for child health services research. Participants included academic researchers, AHCPR officials, pediatricians, and representatives of private organizations, all of whom had significant expertise in the field of child health services research. The goals of the meeting consisted of identifying the principal barriers limiting the development of child health research and identifying specific strategies to overcome those barriers.

Abstract and executive summary of the proceedings are available from National Technical Information Service (NTIS accession no. PB99-110439; 22 pp, $23.00 paper, $12.00 microfiche).
Combining Different Data Sources to Assess Treatments. Christopher H. Schmid, Ph.D., New England Medical Center Hospital, Boston, MA. AHCPR grant HS08530, project period 2/1/95 to 1/31/98.

This project developed algorithms and computer software that can construct Bayesian hierarchical models for meta-analyses and applied these methods to the original analysis of data from various randomized trials, combined single-patient studies, and other studies and models.

Abstract, executive summary, and dissertation are available from the National Technical Information Service (NTIS accession no. PB99-110413; 16 pp, $23.00 paper, $12.00 microfiche).
Cost-Effective Hemochromatosis Screening in Primary Care. Pradyumna D. Phatak, M.D., University of Rochester, Rochester, NY. AHCPR grant HS07616, project period 3/1/94 to 2/28/98.

Hereditary hemochromatosis (HH) can lead to progressive accumulation of tissue iron stores with consequent organ dysfunction. Early diagnosis and institution of phlebotomy treatments will prevent disease manifestations and normalize life expectancy, making HH a natural target for the development of a routine screening strategy. To estimate the prevalence of HH in a primary care setting and establish the feasibility of screening, the researchers screened a sample of 16,031 primary care patients. The prevalence of clinical and biopsy-proven HH combined was 4.5 per 1,000 (5.4 per 1,000 in white patients). The prevalence was higher in men than in women. The researchers concluded that HH is a relatively common disorder among white individuals, and primary care physicians should consider routine screening in this population.

Abstract, executive summary, and final report are available from the National Technical Information Service (NTIS accession no. PB98-152820; 36 pp, $25.50 paper, $12.00 microfiche).
Domestic Violence Identification: Outcomes/Effectiveness. Robert S. Thompson, M.D., Group Health Cooperative, Seattle, WA. AHCPR grant HS07568, project period 3/1/95 to 5/17/98.

The goal of this randomized clinical trial was to improve identification and management of domestic violence (DV) by primary care providers. Five clinics at a large HMO were recruited and randomized to intervention (IC) or control (CC) status. The clinics participated in an intensive, mulifactorial program over a 1-year period, in which skill training was emphasized. Changes over time in provider knowledge, attitudes, and beliefs were measured by survey, and process of care measures (asking, case-finding, and quality of management) were assessed by record abstraction. The intervention had very positive effects on providers' self efficacy and three other domains at 9 months, with sustained positive effects on self-efficacy, fear of offense, and safety concerns at 21 months. Preliminary record-based results at 1 year were as follows: recorded asking about DV increased from 3.5 percent to 20 percent, which was 6.5-fold higher; case-finding increased 1.6-fold in the ICs compared with the CCs; and quality of recorded management improved by 20 percent in ICs and did not change in CCs. This is the first rigorously designed trial in primary care to show sustained improvements in provider knowledge, attitudes and beliefs, and process of care measures.

Abstract, executive summary, and final report are available from the National Technical Information Service (NTIS accession no. PB98-151228; 54 pp, $27.00 paper, $12.00 microfiche).
Effect on Pharmacist Work Patterns of Electronic Guideline-Based Treatment Suggestions: Results of a Prospective, Randomized Trial. William M. Tierney, M.D., Indiana University, Indianapolis, IN. AHCPR grant HS07763, project period 7/1/93 to 12/31/97.

The researchers studied a computer-based drug utilization review (DUR) program that provided treatment suggestions to (a) physicians when they wrote outpatient orders through a network of microcomputer workstations and (b) pharmacists filling outpatient prescriptions. They enrolled 706 patients with heart disease, 712 with lung disease, and 706 with hypertension who received care in an inner city academic practice. Each patient was followed for 1 year and had an average of five physician visits during the study. The intervention had no measurable effect on patients' quality of life, compliance with medications, or satisfaction with care. There was no effect on emergency department visits or hospitalizations, blood pressure, or occurrence of vascular complications. There also was no effect on mortality or inpatient, outpatient, or total health care charges. The researchers conclude that this intensive (and expensive) computer-based DUR intervention had no measurable effects in this practice, possibly because of high baseline compliance with the treatment protocols, negative attitudes towards guidelines, or perhaps because guideline-driven DUR, per se, is ineffective.

Executive summary, final report, and appendixes A-D are available from the National Technical Information Service (NTIS accession no. PB98-139751; 470 pp, $71.50 paper, $25.50 microfiche).
Estimating Wages and Returns to Education Within Nursing. Anne E. Sales, M.S.N., University of Minnesota, Minneapolis, MN. AHCPR grant HS08452, project period 9/1/94 to 4/30/96.

Using data from the 1992 National Sample Survey of Registered Nurses, this study examined the effects of changes in the health care sector on wages and return to education for registered nurses (RNs) in the United States. Wages were lowest in the nursing home sector and highest in the hospital sector, both unadjusted and adjusted for human capital. No returns to education were found in the nursing home sector; the highest returns to education were found in the ambulatory sector.

Executive summary of doctoral dissertation are available from the National Technical Information Service (NTIS accession no. PB98-152853; 10 pp, $12.00 paper, $12.00 microfiche).
HMO Impact on Integrated Networks and Services. Lawton R. Burns, M.B.A., Ph.D., University of Pennsylvania, Philadelphia, PA. AHCPR grant HS09237, project period 9/30/95 to 3/31/98.

This study sought to gauge the impact of HMO market structure on the formation of hospital-based integrated delivery systems (IDSs) and gauge the impact of IDSs on hospital costs and HMO premiums. Longitudinal data (1993 to 1995) were taken from the annual census of HMOs conducted by InterStudy, the Annual Surveys of the American Hospital Association, the Medical Group Census conducted by the American Medical Association, and the Area Resource File. The researchers found that contrary to prevailing wisdom, IDS formation is promoted by the HMOs in a market not by HMO penetration, and that IDSs do not reduce hospital costs but actually may increase them by developing tight linkages with physicians. In addition, IDSs do not serve to blunt the downward force in HMO premiums in a market. These findings suggest that IDSs are primarily HMO contracting vehicles, and they cast doubt on the ability of IDSs to improve hospital efficiency or counterbalance the discounting pressures exerted by managed care.

Abstract, executive summary, and final report are available from the National Technical Information Service (NTIS accession no. PB99-104044; 94 pp, $29.50 paper, $12.00 microfiche).
Impact of Institutional and Technical Forces on Internal Organizational Restructuring: An Examination of the Motivators, Content, and Outcomes of Hospital Reengineering Programs. Stephen L. Walston, M.P.A., University of Pennsylvania, Philadelphia, PA. AHCPR grant HS09581, project period 7/1/97 to 6/30/98.

Organizations often have the need to realign their structures, processes, and products to remain competitive. This research combines both institutional and technical perspectives and explores why organizations adopt reengineering; what external and internal factors influence what they implement; and what the competitive outcomes are in U.S. acute care hospitals. A national survey on restructuring and reengineering collected detailed data from approximately 875 hospitals. The survey data were combined with financial, activity, and area data to allow empirical analyses. The findings suggest that institutional and technical factors affect both the likelihood of adoption and content of reengineering. Institutional forces and integrative mechanisms also appear to have significant effects on competitive outcomes. Hospitals that codified their process, implemented earlier, and balanced their change program with the degree of environmental technical change improved their competitive performance. This research demonstrates that both institutional and technical forces are factors in organizational change, and the subsequent outcomes and should be conjointly considered in both theoretical and practical examination of organizational change.

Abstract, executive summary, and dissertation are available from the National Technical Information Service (NTIS accession no. PB99-104069; 250 pp, $51.00 paper, $23.00 microfiche).
Regional Conference on Dissemination and Implementation of Evidence-based Clinical Guidelines. Julie Morath, M.S., Allina Health System, Minneapolis, MN. AHCPR grant HS09547, project period 9/1/97 to 3/31/98.

Identifying ways to influence the people providing health care services to adopt clinical guidelines or other evidence-based tools and work with aligned incentives was the subject of a conference held in Bloomington, MN. The purpose of the conference was to apply a dissemination and implementation framework that would be useful for planning and implementing evidence-based clinical practice guidelines in various clinical settings.

Abstract, executive summary, and final report are available from the National Technical Information Service (NTIS accession no. PB99-104051; 64 pp, $27.00 paper, $12.00 microfiche).
State Antidivestiture Programs and Their Effectiveness: A Pilot Study. John A. Nyman, M.S., Ph.D., University of Minnesota, Minneapolis, MN. AHCPR grant HS08171, project period 9/30/96 to 12/31/97.

The objective of this pilot study was to test whether it is possible to gather accurate data on divestiture laws that have been enacted and implemented in eight States. The study was intended to lay the groundwork for a full-fledged investigation the effectiveness of the various antidivestiture laws in reducing Medicaid nursing facility patients, nursing facility patient-days, and expenditures on nursing facility care, using all 50 States. The project resulted in information deemed sufficient to conduct a full-fledged study involving all States.

Abstract, executive summary, final report, and appendixes A and B are available from the National Technical Information Service (NTIS accession no. PB99-104010; 240 pp, $51.00 paper, $23.00 microfiche).
Trauma Resource Allocation. Charles C. Branas, M.S., Johns Hopkins University, Baltimore, MD. AHCPR grant HS09326, project period 9/30/96 to 9/29/97.

The study objective was to develop a mathematical model for the location of trauma care resources. A Trauma Resource Allocation Model for Ambulances and Hospitals (TRAMAH) was formulated. Severely injured patients were queried from Maryland data and a spatial injury profile was created by parsing these patients into ZIP codes. In order to maximize coverage of severely injured patients, trauma centers and aeromedical depots were simultaneously sited using TRAMAH. From 1992 to 1994, 26,774 severe injuries were considered for coverage. Across Maryland, 94.8 percent of severely injured residents had access to trauma system resources within 30 minutes, and 70.3 percent had access within 15 minutes. For the same number of resources as the existing Maryland Trauma System, TRAMAH achieved a coverage objective of 99.97 percent within 30 minutes. This translated into an additional 1,384 severely injured people covered over the 3-year study period. Holding in place the trauma centers of the existing system, approximately the same percent coverage as the existing system was achieved within 15 minutes by optimally locating six fewer aeromedical depots. TRAMAH will allow trauma systems planners to better locate their resources with respect to spatial needs and response times.

Executive summary of dissertation are available from the National Technical Information Service (NTIS accession no. PB98-152861; 16 pp, $23.00 paper, $12.00 microfiche).

Medical Expenditure Panel Survey (MEPS). MEPS is the third in a series of nationally representative surveys of medical care use and expenditures sponsored by AHCPR. MEPS is cosponsored by the National Center for Health Statistics (NCHS). The first of these surveys, the National Medical Care Expenditure Survey (NMCES), was conducted in 1977, and the second, the National Medical Expenditure Survey (NMES), in 1987. MEPS collects detailed information on health care use and expenses, sources of payment, and insurance coverage of individuals and families in the United States. The following MEPS Highlights on health insurance coverage and job-related health insurance are now available.

Characteristics of Nursing Home Residents-1996. MEPS Research Findings No. 5 (AHCPR Publication No. 99-0006). This report profiles the health and demographic characteristics of the approximately 1.56 million people living in nursing homes in the United States on January 1, 1996. The data were obtained from a nationally representative sample of nursing homes and nursing home residents from the 1996 Nursing Home Component (NHC) of MEPS. Nursing home residents were highly dependent on assistance with activities of daily living (ADLs). Nearly all residents (97.2 percent) required assistance with at least one ADL, and 83.3 percent required assistance with three or more ADLs. People under age 65 were as dependent for assistance as older residents. In general, cognitive problems were more prevalent than behavior problems among residents, and nearly half experienced communication problems. More than half of residents also experienced bladder or bowel incontinence or both. The majority of nursing home residents (65.8 percent) were in private for-profit facilities. Almost 90 percent were white, and half (49.3 percent) were age 85 or over. More than two-thirds (71.6 percent) of residents were women. Although the data presented in the report do not reflect the insurance source actually paying for nursing home care, most residents were enrolled in Medicare (93.2 percent), and two-thirds were enrolled in both Medicare and Medicaid (63.5 percent). Copies are available from the AHCPR Publications Clearinghouse.
Uninsured Workers-Demographic Characteristics, 1996. MEPS Highlights No. 7 (AHCPR Publication No. 99-0007). This publication shows estimates for U.S. workers ages 16 to 64 selected from MEPS Research Findings No. 2, Health Insurance Status of Workers and Their Families: 1996 (AHCPR Publication No. 97-0065; available from the AHCPR Publications Clearinghouse). Major highlights include: workers with the following demographic characteristics were most likely to be uninsured: young adults ages 19 to 24, Hispanic males, and those with poor or fair health status. More than one-third of young adult workers ages 19 to 24 were uninsured. Workers age 35 and over had a lower risk of being uninsured than all workers in general. Black and Hispanic workers were much more likely than white workers to lack any type of health insurance coverage. Hispanic males were far more likely to be uninsured than any other group, with 44 percent uninsured. Workers in poor or fair health were more likely to be uninsured than those in excellent or very good health. Female workers were more likely than male workers to be insured. Copies are available from the AHCPR Publications Clearinghouse.
Uninsured Workers-Job Characteristics, 1996. MEPS Highlights No. 8 (AHCPR Publication No. 99-0008). This publication presents estimates for U.S. workers ages 16 to 64 selected from MEPS Research Findings No. 2 (see information in previous entry). In this study, self-employed individuals, those working in small businesses, those earning low wages, and part-time workers were the most likely to be uninsured. Self-employed workers were almost twice as likely as wage earners to be uninsured, and the risk of being uninsured was related to the size of the business. Wage earners were more likely to be uninsured if they worked for establishments with less than 25 employees. Workers earning less than $10.00 per hour were at substantially greater risk of lacking health insurance than those earning $10.00 or more per hour. Over three-fourths of full-time workers had job-related insurance. Copies are available from the AHCPR Publications Clearinghouse.

Return to Contents

Research Briefs

Larme, A.C., Meyer, J.S., and Pugh, J.A. (1998, July). "Use of qualitative methods to evaluate diabetes education programs." (AHCPR grant HS07397). The Diabetes Educator 24(4), pp. 499-508.

A process evaluation using qualitative methods complements and strengthens the outcome evaluation of a diabetes education program, concludes this study. Quantifiable evaluation methods typically track pre- and postprogram physiological, cognitive, behavioral, and attitudinal changes to determine the effectiveness of the program in achieving desired goals. Process evaluation helps to monitor program functioning, identify areas that require modifications to improve effectiveness, and interpret outcomes data through analysis of why patients may be responding in a particular way to an outcome measure.

Riley, A.W., Forrest, C.B., Starfield, B., and others. "Reliability and validity of the adolescent health profile-types" and Riley, A.W., Green, B.F., Forrest, C.B., and others, "A taxonomy of adolescent health: Development of the adolescent health profile-types." (1998). (AHCPR grant HS07045). Medical Care 36(8), pp. 1228-1236 and pp. 1237-1248.

These researchers used four domains of health (satisfaction, discomfort, risks, and resilience) to group adolescents into 13 distinct profile types. They characterized the profile types primarily by the number of domains including poor health and identifying the unique combinations of problems that characterize different subgroups of adolescents. The researchers used cluster analysis and clinically based conceptual methods to identify these patterns in samples of adolescents from schools and from clinics that serve adolescents. They assigned individuals with similar patterns of scores across multiple domains to the same profile type. A second study by the researchers demonstrated the reliability and validity of this taxonomy of health profile types in four ethnically diverse population samples of urban and rural youths aged 11 to 17 in public schools.

Schmid, C.H., Lau, J., McIntosh, M.W., and Cappeller, J.C. (1998). "An empirical study of the effect of the control rate as a predictor of treatment efficacy in meta-analysis of clinical trials." (AHCPR grants HS07782 and HS08532). Statistics in Medicine 17, pp. 1923-1942.

Key risk factors that vary across patients and can be measured only as aggregate values, such as age and sex, are difficult to address adequately by meta-regression. The control rate (a summary measure) proposed by some can serve as a surrogate for these patient risk factors. This study found a significant correlation between the control rate (CR) and the treatment effect (TE). This suggests that, rather than merely pooling the TE in a single summary estimate, investigators should search for the causes of heterogeneity related to patient characteristics and treatment protocols to determine when treatment is most beneficial. The researchers examined 115 meta-analyses covering a wide range of medical applications for evidence of a linear relationship between the CR and three TE measures: the risk difference (RD), the log relative risk (RR), and the log odds ratio (OR). The control rate was two times more likely to be significantly related to the RD (31 percent) than to the RR (13 percent) or the OR (14 percent).

Singer, J.D., Davidson, S.M., Graham, S., and Davidson, H.S. (1998). "Physician retention in community and migrant health centers: Who stays and for how long?" (Cosponsored by AHCPR and the Bureau of Primary Health Care, Health Resources and Services Administration). Medical Care 36(8), pp. 1198-1213.

Retention of primary care physicians (PCPs) is a difficult challenge for community and migrant health centers (C/MHCs) in medically underserved communities, especially with the severe reduction in the National Health Service Corps (NHSC), the main source of center clinical staff. Once NHSC physicians have fulfilled their contractual obligation—usually 4 years—they are likely to leave. For instance, after 5 years at C/MHCs in medically underserved areas, only 17 percent of NHSC physicians remained compared with 36 percent of physicians who began their work at a center with no NHSC contractual obligation. For this study, the researchers analyzed data from a central information system on all physicians working in federally funded C/MHCs on January 1, 1990 as well as data from each center. They explored the relationship between employment duration and physician and center characteristics and found that for NHSC physicians, the risk of leaving on the 2-year anniversary of their contract was 0.13, the 3-year anniversary, 0.17, and the 4 year anniversary, 0.41. The corresponding risks of leaving for non-NHSC physicians were 0.09, 0.09, and 0.05, respectively. This suggests that the NHSC physicians are at the greatest risk of leaving at the end of their loan payback period. For NHSC physicians, only personal characteristics (such as age at hire) affected their stay-leave decision, whereas for other physicians, characteristics of centers in which they worked (for example, center size and location) also played a role.

Wynia, M.K., Ioannidis, J.P., and Lau, J. (1998). "Analysis of life-long strategies to prevent Pneumocystis carinii pneumonia in patients with variable HIV progression rates." (AHCPR grants HS07782 and NRSA training grant T32 HS00060). AIDS 12(11), pp. 1317-1325.

For patients with AIDS who are expected to live another 3 years, standard or low-dose trimethoprim-sulfamethoxazole (TS) is the treatment of choice for preventing the potentially fatal Pneumocystis carinii pneumonia (PCP). Both regimens are superior to other strategies for preventing PCP (between 9 and 26 fewer episodes of PCP per 100 patients treated compared with aerosolized pentamidine [AP]-based therapies). However TS-based treatments are more toxic, with 11 to 44 episodes of toxicity per 100 patients. For patients with slower rates of disease progression (expected to live another 7 years), low-dose TS results in nearly three fewer episodes of PCP and 32 fewer episodes of toxicity per 100 patients treated, and the lifetime cost is $1,381 less than standard-dose TS. Researchers compared the efficacy and toxicity of high and low doses of TS, dapsone, and/or AP in sequence in a group of AIDS patients with a broad range of disease progression rates. Without any prophylaxis, this group had an overall 40 percent 2-year risk of developing PCP. At typical and rapid rates of disease progression, the standard and low-dose TS strategies were comparable (0.3 and 1.1 episodes of PCP per 100 patients treated) and were substantially more effective than were the high-AP and AP-switch (low-dose AP to low-dose TS) strategies, which more than doubled the lifetime risk of developing PC compared with either of the TS-based strategies. The low-dose TS strategy was more effective, less costly, and less toxic for those with slow disease progression. Although the AP-switch strategy was clinically comparable for those with slow disease progression, it cost $2,099 more per person.

Return to Contents

AHCPR Publication No. 99-0014
Current as of January 1999

Internet Citation:

Research Activities newsletter. December 1998, No. 222. AHCPR Publication No. 99-0014. Agency for Health Care Policy and Research, Rockville, MD. http://www.ahrq.gov/research/dec98/