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

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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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

Research Briefs

Ay, H., Benner, T., Arsava, E.M., and others (2007). "A computerized algorithm for etiologic classifications of ischemic stroke. The causative classification of stroke system." (AHRQ grant HS11392). Stroke 38, pp. 2979-2984.

Etiologic classification of ischemic stroke is important in both patient care and stroke research. Since stroke is a heterogeneous disorder with multiple potential causes, the reliable classification of a stroke's cause is a complex task. In order to facilitate classification in multicenter settings, the researchers developed a computerized Causative Classification System (CCS), derived from their evidence-based classification algorithm (SSS-TOAST). The CCS uses clinical, epidemiological, and diagnostic data to determine stroke type in five major categories (large artery atherosclerosis, cardio-aortic embolism, small artery occlusion, other causes, and undetermined causes).

Five neurologists evaluated its reliability through assessment of 50 ischemic stroke patients. The neurologists' expert opinions on stroke subtype were different from the CCS assignment in only 6 of 250 ratings (2.4 percent). The high interexaminer and expert-CCS agreement rates strongly suggest a potential use for automated CCS for classifying the cause of stroke in multicenter settings.

Bao, Y., Fox, S.A., and Escarce, J.J. (2007, June). "Socioeconomic and racial/ethnic differences in the discussion of cancer screening: 'Between-' versus 'within-' physician differences." (AHRQ grant HS10770). HSR: Health Services Research 42(3), pp. 950-970.

Researchers found that patients with less than a high school education were far less likely than college graduates to have discussed screening for colon, breast, and prostate cancer during medical visits with the same physicians. Patients' race/ethnicity also influenced discussion of cancer screening. Asians were much less likely than whites to discuss fecal occult blood testing for colon cancer, and prostate-specific antigen testing for prostate cancer. Black women were more likely than white women to discuss mammogram screening for breast cancer. The findings were based on patient and physician surveys of 2 community trials of 5,978 patients aged 50 to 80 treated by 191 primary care physicians who practiced in Southern California.

Clancy, C. M., Isaacson, S., and Henriksen, K. (2007). "Design of the physical environment for changing healthcare needs." Health Environments Research & Design Journal 1(1), pp.10-11.

In an editorial in the first issue of Health Environments Research & Design Journal, Carolyn M. Clancy, M.D., Director of the Agency for Healthcare Research and Quality, and colleagues comment on improving the design of the health care environment. The physical environment of health care—including light, noise, air quality, toxic exposures, temperature, humidity, wayfinding, and aesthetics—as well as other workplace design features and physical layouts have an impact on staff communication and job satisfaction, as well as patient outcomes.

Evidence-based design principles can help to tailor health care environments to the needs of multiple users. For example, decreasing distractions and noise leads to health care workers making fewer mistakes. However, the optimal methods for designing spaces that decrease noise and distractions remain to be discovered. More research is needed on this and related questions, and it is important that there be a central resource for architects, CEOs, and researchers to retrieve needed information.

Reprints (AHRQ Publication No. 08-R027) are available from the AHRQ Publications Clearinghouse.

Clancy, C.M., Kiley, J.P., and Weiss, K.B. (2007). "Eliminating asthma disparities through multistakeholder partnerships." CHEST 132, pp. 1422-1424.

The Agency for Healthcare Research and Quality (AHRQ) has long been concerned about the burden of asthma and has sponsored research on the quality of care for children in the Medicaid programs and in the State Children's Health Insurance Program, according to AHRQ Director Carolyn M. Clancy, M.D. In addition, AHRQ encourages quality improvement and disparity reduction interventions at the health plan and policy levels. To this end, AHRQ has worked with six State asthma coalitions, five of them located in State government offices. Other Federal agencies are also involved, including the Centers for Disease Control and Prevention and the National Heart, Lung, and Blood Institute. A national workshop documented in a supplement to this issue of CHEST underscores the necessity of a sustained effort by multiple stakeholders: public and private, national and regional, and local to reduce asthma disparities.

Reprints (AHRQ Publication No. 08-R045) are available from the AHRQ Publications Clearinghouse.

Clancy, C.M. and Slutsky, J.R. (2007). "Guidelines for guidelines: We've come a long way." CHEST 132, pp. 746-747.

Clinical practice guidelines have increasing potential to influence how medical care is delivered and measured. An article in this issue of CHEST outlines the approach to guideline development taken by the American College of Chest Physicians (ACCP), and describes an evolution in processes and methodologies for developing, assessing, and implementing clinical practice guidelines.

The ACCP has outlined a detailed algorithm for managing many aspects of guideline development, including conflicts of interest. This algorithm provides clear direction for both guideline developers and users. Guidelines, to be credible, must be updated, and the ACCP has also addressed this issue. Guideline development has advanced steadily, embracing more rigorous evidence standards, explicit methods, broad disciplinary participation, and recognition of the importance of understanding patient perspectives. Health information technology, including the Internet, is helping to bring guidelines closer to the point of care. Since patients frequently have multiple illnesses, the next frontier in guideline development will be developing integrated guidance for complex cases.

Reprints (AHRQ Publication No. 08-R042) are available from the AHRQ Publications Clearinghouse.

Cohen, L.A., Harris, S.L., Bonito, A.J., and others (2007, Winter). "Coping with toothache pain: A qualitative study of low-income persons and minorities." Journal of Public Health Dentistry 67(1), pp. 28-35.

A study by Richard J. Manski, D.D.S., Ph.D., M.B.A., of the Agency for Healthcare Research and Quality, analyzed discussions from 8 focus group sessions with 66 low-income white, Hispanic, and black adults. This group had suffered tooth pain in the past year and used self-care or care from a nondentist to relieve their pain. These individuals described their toothache pain as intense, throbbing, miserable, or unbearable. The pain was bad enough to affect their ability to perform normal activities, such as their job, housework, social activities, sleeping, talking, and eating, and made them depressed, anxious, mean, irritable, and cranky. Some resorted to getting arrested to get dental care to relieve the pain. Others pulled their teeth out with pliers, or even rinsed with caustic substances such as gasoline, kerosene, and rubbing alcohol. Nearly two-thirds (63 percent) of the focus group participants said they never visited the dentist or only visited when they had a dental problem. However, they all had a history of multiple toothaches, with 36 percent having suffered 4 to 9 toothaches and 27 percent reporting 10 or more. Most reported the high cost of dental care as a predominant barrier to seeking care from a dentist. Some also cited fears or distrust of dentists and dental practice, problems obtaining transportation, long waiting lists for appointments, and lack of sick leave.

Reprints (AHRQ Publication No. 07-R072) are available from the AHRQ Publications Clearinghouse.

Coopey, M., Nix, M.P., and Clancy, C. (2007). "Evidence-based practice: ARHQ's role in generating and disseminating knowledge." AORN Journal 86(5), pp. 857-860.

Evidence-based practice depends on a research-based body of knowledge. This body of knowledge provides the foundation for clinical practice guidelines and care performance measures. The Agency for Healthcare Research and Quality (AHRQ) assists in providing this foundation by focusing on the development of evidence; synthesis of findings, dissemination, and implementation of those findings; and evaluation of health care quality. For example, since 1989, AHRQ has awarded 174 grants to nurse principal investigators whose research contributes to the evidence base of nursing. In addition, many of the reports developed under AHRQ's Evidence-Based Practice Program apply to nursing practice, including one recently released on nurse staffing and the quality of patient care, a topic suggested by the American Organization of Nurse Executives.

Another AHRQ initiative, the National Guidelines Clearinghouse™, provides detailed information on clinical practice guidelines. The Clearinghouse currently contains 105 guidelines developed by 13 nursing organizations and departments.

Reprints (AHRQ Publication No. 08-R023) are available from the AHRQ Publications Clearinghouse.

Go, A.S., Yang, J., Gurwitz, J.H., and others (2007, August). "Comparative effectiveness of beta-andrenergic antagonists (atenolol, metoprolol tartrate, carvedilol) on the risk of rehospitalization in adults with heart failure." (AHRQ contract 290-05-003). American Journal of Cardiology 100, pp. 690-696.

Over half (54 percent) of elderly patients hospitalized for heart failure were rehospitalized within a year, according to a new study. The rehospitalization rate was similar regardless of which of three types of beta-blockers they were taking: atenolol, short-acting metoprolol tartrate, or carvedilol. However, patients taking other beta-blockers had a 31 percent higher risk of rehospitalization and those taking no beta-blocker had a 12 percent higher risk of rehospitalization, after adjusting for factors such as cardiovascular history, other coexisting illnesses, and age. Even among a subgroup of 664 patients with known left ventricular systolic heart failure, who were taking both digoxin and a beta-blocker, readmission rates were similar for those taking atenolol, metoprolol tartrate, or carvedilol.

Researchers used pharmacy and hospital discharge databases to compare the risk of rehospitalization for heart failure associated with receipt of different beta blockers in 7,883 adults hospitalized for heart failure within two large health plans in 2001 and 2002. More than two-thirds (68 percent) of this group received a beta-blocker at hospital discharge and/or during the first 12 months after discharge. The beta-blocker prescribed was usually atenolol, shorter-acting metoprolol tartrate, or carvedilol.

Goldman, L.E., Vittinghoff, E., and Dudley, R.A. (2007, June). "Quality of care in hospitals with a high percent of Medicaid patients." (AHRQ grant HS16117). Medical Care 45(6), pp. 579-583.

Researchers found that acute care hospitals with a high percentage of Medicaid patients had lower adherence than other nonteaching hospitals on 10 quality of care indicators for 3 conditions: heart attack, congestive heart failure, and community-acquired pneumonia (CAP). The researchers examined compliance with 10 processes of care for the 3 conditions among acute care hospitals that participated in the first Hospital Compare public report released in November 2004 and the 2004 American Hospital Association hospital survey. Among the 2,874 nonteaching hospitals, high Medicaid hospitals (1 standard deviation above the national mean, or more than 28 percent Medicaid patients) prescribed beta-blockers and aspirin less often at discharge for heart attack patients and were also less likely to administer antibiotics on time in CAP. These differences persisted after adjusting for potential confounders of this relationship such as hospital ownership, location, size, and region. However, among teaching hospitals there were few differences between high Medicaid and other hospitals.

Hall, B.L., Hirbe, M., Waterman, B., and others (2007). "Comparison of mortality risk adjustment using a clinical data algorithm (American College of Surgeons National Surgical Quality Improvement Program) and an administrative data algorithm (Solucient) at the case level within a single institution." (AHRQ grant HS11913). Journal of the American College of Surgeons 205(6), pp. 767-777.

In evaluating health care outcomes to improve quality of care, risk adjustment methods based on either clinical data or administrative data sets may be used. However, there is some controversy from studies showing that administrative (claims-based) data could be less accurate, making risk adjustment less reliable.

This study compares the American College of Surgeons' National Surgical Quality Improvement Program (NSQIP) risk adjustment algorithm, based on a strictly clinical data approach, to Solucient's risk adjustment algorithm based on a strictly administrative data approach. Both methods were used with a sample of 1,234 surgical patients at an academic teaching hospital. Actual mortality in this group was 3.32 percent; however, the NSQIP predicted mortality (2.99 percent) was lower and Solucient predicted mortality (3.83 percent) was higher than actual mortality. With respect to mortality risk estimation, the clinically-based NSQIP and the claims-based Solucient systems performed comparably.

The authors concluded that because the use of claims-based data is potentially less expensive than clinical data, the Solucient-type algorithm should continue to be investigated.

Handrigan, M. and Slutsky, J. (2007). "Funding opportunities in knowledge translation: Review of the AHRQ's 'Translating Research into Practice' initiatives, competing funding agencies and strategies for success." Academic Emergency Medicine 14, pp. 965-967.

This article provides an overview of the Agency for Healthcare Research and Quality's (AHRQ) knowledge translation activities and relates them to the field of emergency medicine. Translating research into practice is the key to creating sustainable improvements in emergency medicine and other health care settings. However, scientific evidence alone rarely leads to improvements in practice or changes in policy. Beginning in 1999, AHRQ launched a two-phase grant program called Translating Research Into Practice (TRIP I and TRIP II). TRIP I was intended to develop new approaches that are effective in promoting the use of evidence in clinical settings. TRIP II was intended to develop strategies for translating research into practice through the development of partnerships between researchers and health care systems and organizations.

Investigators interested in funding opportunities in this area of concern are encouraged to visit AHRQ's Web site at The role of emergency medicine in health services research and translating research into practice brings a unique perspective and a natural forum for linking evidence to action.

Reprints (AHRQ Publication No. 08-R043) are available from the AHRQ Publications Clearinghouse.

Harrison, M.I., Henriksen, K., and Hughes, R.G. (2007). "Improving the health care work environment: Implications for research, practice, and policy." Joint Commission Journal on Quality and Patient Safety 33 (11 Suppl), pp. 81-84.

This article reviews a group of articles in the same issue that focus on different aspects of improving the health care work environment. The articles address the following subjects: the implementation of work-hour limits, enhancing work flow to reduce crowding, nurse staffing in acute care settings, organizational climate and health care outcomes, the role of the physical environment in Crossing the Quality Chasm, and a sociotechnical systems approach to improving the work environment. An important implication for future research is the need to know more about how, where, and when such improvement practices work. Among the implications for practice are the importance of limiting work hours, improving schedules, and providing sleep hygiene training (a method to correct things a person does on a regular basis that disturb their sleep) to help combat clinician fatigue. Also, teams planning construction or renovation of health care facilities should include people with experience in evidence-based design. Suggestions for policymakers and payers include strict limits on the length of work shifts, greater emphasis on public reporting, and creation of incentives for demonstrated work environment improvements.

Reprints (AHRQ Publication No. 08-R026) are available from the AHRQ Publications Clearinghouse.

Harrison, M.I., Henriksen, K., and Hughes, R.G. (2007). "Improving the health care work environment: A sociotechnical systems approach." Joint Commission Journal on Quality and Patient Safety 33 (11 Suppl), pp. 3-6.

This article introduces a journal supplement containing a series of articles that synthesize findings on key elements in the hospital work environment and suggest operational and policy challenges in each area. These articles were based on research supported by the Agency for Healthcare Research and Quality (AHRQ) and other government and private agencies. The authors, all AHRQ researchers, outline a sociotechnical framework to provide context for the themes examined in this issue.

The sociotechnical system consists of a social subsystem (people, tasks, and organization), a technical subsystem (technologies, equipment, and physical settings), and social and organizational context (regulatory forces, payers, markets, suppliers, along with science, technology, social norms, and values). All but one of the articles focus on the organizational component of the sociotechnical system. The sociotechnical approach emphasizes work environments as products of interactions among system components, stresses mutual relations between technologies and workplace processes and conditions, and highlights potential influences of forces outside of provider organizations such as markets, regulators, and payers.

Reprints (AHRQ Publication No. 08-R022) are available from the AHRQ Publications Clearinghouse.

Hughes, R.G. and Clancy, C.M. (2007). "So what? The challenge of doing 'need to know' versus 'would like to know' research." Applied Nursing Research 20, pp. 210-213.

Researchers need to be able to differentiate between research that addresses high-priority issues and research whose findings may be interesting, but are not necessarily usable or actionable. They need to be able to define who is experiencing positive or adverse effects from the current organization and processes of health care, and who wants the information derived from the research and for what purposes. Funding needs to be directed toward research that will facilitate evidence-based improvements in health care. It is important to distinguish critical research gaps that are preventing decisionmakers from improving the processes and effects of health care, a process that itself necessitates research.

The goal of the Agency for Healthcare Research and Quality (AHRQ) in supporting research on the organization, delivery, and cost of care is to actually use what we know to be safe and of high quality. ARHQ seeks to bridge the gap between the information needs of decisionmakers and the nuances of research findings.

Reprints (AHRQ Publication No. 08-R024) are available from the AHRQ Publications Clearinghouse.

Hughes, R.G. and Clancy, C.M. (2007). "Improving the complex nature of care transitions." Journal of Nursing Care Quality 22(4), pp. 289-292.

The potential for errors due to handovers and patient transfers—collectively known as "care transitions" is enormous. A key strategy for reducing errors is to examine the processes that occur when patients move from one site or clinician to the next. The Agency for Healthcare Research and Quality (AHRQ) has funded numerous studies assessing efforts to improve patient transfers from one site or one type of provider to another, according to AHRQ Director Carolyn Clancy, M.D.

Coordination of care is important since it often involves many individuals—the patient, a friend or family members, and various clinicians and providers. The possibility of medication errors is also a concern during admission, transfer, and discharge. Avoiding communication failures through the accurate two-way communication of clear and up-to-date information is critical. The growing use of processes such as medication reconciliation (reconciling new medications prescribed during care transitions, such as from the emergency department to an intensive care unit, with a patient's current medications) and the Situation-Background-Assessment-Recommendation technique is helpful, but challenges remain.

Reprints (AHRQ Publication No. 08-R014) are available from the AHRQ Publications Clearinghouse.

Lambert, M.C., Essau, C.A., Schmitt, N. and others (2007). "Dimensionality and psychometric invariance of the Youth Self-Report Form of the Child Behavior Checklist in cross-national settings." (AHRQ grant HS08385). Assessment 14(3), pp. 231-245.

The Youth Self-Report (YSR) form and other forms that are part of the Child Behavior Checklist are considered well-normed measures of children's behavioral and emotional problems within the United States. Yet there is little evidence that the YSR form is as psychometrically constant in other national settings in which it is widely used. The researchers thus decided to test this form with adolescent children who were not referred for emotional and behavioral problems in German and Jamaican schools. After administering the form to hundreds of these schoolchildren in both countries, they used item-response theory (IRT) procedures to reveal whether items on a focus dimension psychometrically varied across groups, a phenomenon known as "bias."

The study found that the current YSR factor model might not be appropriately specified for nonreferred adolescents' ratings in Germany and Jamaica. In addition, most items seemed incapable of measuring levels of functioning across the two nations on an identical metric. More accurate and economical assessment might be achieved if psychometric variance and redundancy were addressed and children's psychological severity levels were matched with items measuring this severity.

Leppert, P.C., Legro, R.S., and Kjerulff, K.H. (2007, September). "Hysterectomy and loss of fertility: Implications for women's mental health." (AHRQ grant HS06865). Journal of Psychosomatic Research 63(3), pp. 269-74.

Young women who wanted to have children but needed a hysterectomy were more likely to put off the surgery and seek mental health counseling, a new study finds. Researchers interviewed 1,140 Maryland women before they underwent hysterectomies in 1992 and 1993 and followed up with them for 2 years after surgery. Of the women interviewed, 120 (10.5 percent) said they would have liked to have children, 40 (3.5 percent) said maybe they would have liked to have children, and 980 (86 percent) said they were fine with their baby days being over. The women who wanted children tended to put off their surgeries for 4.5 years, despite severe pelvic pain. Women who wanted children or were ambiguous about the decision tended to be young; had never given birth; had uterine fibroids and indications of endometriosis; and had high test scores for depression, anxiety, anger, and confusion. The women who wanted children were twice as likely to have sought mental health counseling in the 3 months before surgery. Further, 2 years after surgery, these women were more likely to be depressed when compared with women who did not want children.

Liang, H., Chen, D., Wang, Y., and others (2007). "Different risk factor patterns for metabolic syndrome in men with spinal cord injury compared with able-bodied men despite similar prevalence rates." (AHRQ grant HS11277). Archives of Physical Medicine Rehabilitation 88, pp.1198-1204.

Since men with spinal cord injuries (SCI) have increased mortality from type 2 diabetes and cardiovascular disease, it was thought that they had a higher rate of metabolic syndrome than able-bodied men. Metabolic syndrome is a cluster of risk factors for diabetes and cardiovascular disease that includes abdominal obesity, elevated triglycerides, low high-density lipoproteins (HDL), elevated blood pressure, and elevated glucose. Researchers found, however, that men with SCI do not appear to have higher rates of metabolic syndrome than able-bodied men. They compared risk factors associated with metabolic syndrome and other risk factors among 185 men with SCI (between 20 and 59 years of age) who were at least one year postinjury with a control group of similar aged able-bodied men. Men with SCI were at higher risk for abdominal obesity and low HDL, but had a lower risk for elevated levels of glucose, total cholesterol, and triglycerides. There were no differences between the two groups for a history of hypertension, diabetes, or medication use for these disorders. The researchers conclude that for men with SCI and metabolic syndrome, future interventions should focus on abdominal obesity, low HDL, and other nontraditional risk factors.

Longo, D.R., Hewett, J.E., Ge, B., and Schubert, S. (2007, Summer). "Rural hospital patient safety systems implementation in two states." (ARHQ grant HS11885). Journal of Rural Health 23(3), pp. 189-197.

A study of two States reveals that rural hospitals are less likely than urban hospitals to implement patient safety systems such as computerized physician order entry (CPOE), alert systems, medication administration procedures, and error reporting systems. The researchers surveyed acute care hospitals in Missouri and Utah in 2002 and 2004. Rural hospitals had a lower mean score than urban hospitals for CPOE systems, computerized test results, and assessment of adverse events (33.63 vs. 46.39); use of specific patient safety policies (29.58 vs. 32.22); and use of data in patient safety programs (34.18 vs. 40.21). They also had lower mean scores for drug storage, administration, and safety procedures (69.26 vs. 75.19); manner of handling adverse event/error reporting (13.43 vs. 15.89); prevention policies (7.86 vs. 7.72); and root cause analysis (14.39 vs. 16.37).

Matheny, M.E., Gandhi, T.K., Orav, E.J., and others (2007, November 12). "Impact of an automated test results management system on patients' satisfaction about test result communication." (AHRQ grant HS11046). Archives of Internal Medicine 167(20), pp. 2233-2239.

This study found that patients liked receiving test results via an automated management system that provides centralized test result tracking and facilitates contact with patients. Patients also liked receiving information on the conditions and treatments related to the chemistry, hematology, pathology, radiology, or microbiology tests they took. The automated system was built into the electronic health record system of a network of 26 primary care practices in Massachusetts.

The researchers examined 570 patient encounters in both intervention and control practices. Doctors in 13 intervention practices were trained and given access to the physician test results management tool with imbedded patient notification functions. Physicians at the 13 control clinics had to track outpatient test results manually, which often required repeated checks of a patient's record. They were notified of new test results through paper printouts generated by the testing facilities. After adjusting for other factors affecting patient satisfaction such as age and insurance type, patients were nearly 2.4 times more satisfied with test results communication with the automated test management system than without it. They were also nearly 3.5 times more satisfied with the information given them for medical treatments and conditions regarding their test results.

Mosdol, A., Witte, D.R., Frost, G., and others (2007). "Dietary glycemic index and glycemic load are associated with high-density-lipoprotein cholesterol at baseline but not with increased risk of diabetes in the Whitehall II study." (AHRQ grant HS06516). American Journal of Clinical Nutrition 86, pp. 988-994.

This study did not confirm the protective effect of low dietary glycemic index (GI) and glycemic load (GL) diets on the risk of developing type 2 diabetes in a healthy group of middle-aged men and women. A Whitehall II research team examined the associations of dietary GI and GL in 1991-1993 (baseline) among 7,321 middle-aged, nondiabetic British civil servants and development of diabetes 13 years later. The civil servants completed questionnaires about food intake for 13 years. Higher dietary GI and GL were not associated with increased risk of diabetes. On the contrary, higher dietary GL was associated with decreased risk of diabetes. This protective effect of GL remained after adjustments for employment grade, smoking, alcohol intake, and physical activity, and was only moderately weakened by adjustment for total carbohydrate and fiber intake. These findings are consistent with a possible protective role of dietary factors on disease risk. However, they do not support the idea that high dietary GI and GL may be a risk factor for type 2 diabetes.

O'Leary, J.F., Sloss, E.M., and Melnick, G. (2007, Summer). "Disabled Medicare beneficiaries by dual eligible status: California, 1996-2001." (AHRQ grant HS10256). Health Care Financing Review 28(4), pp. 57-67.

Nonelderly persons with disabilities are expected to comprise nearly 17 percent of the Medicare-insured population by 2010. Over one-third of this group is dually eligible, that is, receives health insurance from Medicare and Medicaid (Federal-State program for low-income persons). A California study from 1996 to 2001 revealed differences in profile and hospital use among dually eligible and Medicare-only beneficiaries in that State.

Dually eligible persons were younger (42 percent aged 18 to 44) than Medicare-only persons (48 percent aged 55 to 63), and fewer were white (66 vs. 76 percent) and enrolled in an HMO (8 vs. 34 percent). Throughout the study period, dually eligible beneficiaries had more hospitalizations than Medicare-only beneficiaries, with the difference in any given year ranging from 56 to 112 more discharges per 1,000 beneficiaries. Dual eligibles, especially blacks, also had longer average hospital stays than Medicare-only patients throughout the period (57 percent more days for those in fee-for-service plans and 30 percent more days for those in HMO plans), with a difference of about 1 day per stay between the two groups.

O'Malley, A.J., Marcantonio, E.R., Murkofsky, R.L., and others (2007). "Deriving a model of the necessity to hospitalize nursing home residents." (AHRQ HS10645). Research on Aging 29, pp. 606-625.

There is widespread variation in nursing home hospitalization rates. Unnecessary hospitalizations should be avoided, since these can be traumatic to the patient and costly for payers. In order to find out the extent to which potentially controllable nonclinical factors contribute to variations in the hospitalization rates, it is first necessary to control for clinical factors. To do this, the authors developed a diagnosis-based model (NeccHosp) for the clinical necessity of hospitalizing a nursing home resident. They used an expert panel of 12 geriatricians to obtain information on the necessity of hospitalizing nursing home residents with common conditions. Using a 9-point scale of the necessity of hospitalization, the experts rated 38 scenarios based on Medicare diagnosis-related groups.

The authors found that the most influential factor in the decision to hospitalize is the primary diagnosis, followed by the Do Not Hospitalize advance directive, and then the secondary diagnoses. The NeccHosp model can be used to find out if nursing homes with very high or low hospitalization rates lacked certain resources or had adopted particular policies regarding resident hospitalization.

Poon, E.G., Wald, J., Schnipper, J.L., and others (2007). "Empowering patients to improve the quality of their care: Design and implementation of a shared health maintenance module in a U.S. integrated healthcare delivery network." (AHRQ grant HS13326). MedInfo 12(Pt. 2), pp. 1002-1006.

Researchers developed an online personal health record that lets patients prepare for discussions with their clinicians and communicate with them before a visit. The online system offers patients a portal for renewing prescriptions, looking at medication lists, accessing a health information library, scheduling appointments, and E-mailing providers. The system also reminds patients of health care services that are due, such as checkups, vaccinations, and screening tests. It allows providers to review patient comments before the visit.

From July 2005 to November 2006, 2,779 primary care patients at 7 study sites participated in a study of the system. Of the 970 patients invited to review their records before scheduled visits, nearly three-fourths (72 percent) of the patients did so and passed information to their clinicians. Clinicians reviewed data from 66 percent of patients before their visits. Eighty-one percent of the 437 patients who participated in the survey found the system easy to use and agreed that providers were able to gather accurate information on the patients.

Rosko, M.D. and Mutter, R.L. (2007). "Stochastic frontier analysis of hospital inefficiency. A review of empirical issues and an assessment of robustness." (AHRQ contract. no. 290-04-0005) Medical Care Research and Review Online First.

Efforts to study and reduce hospital inefficiency have been hindered by a lack of commonly accepted measures. The authors focus on one approach to measuring hospital inefficiency, stochastic frontier analysis (SFA). Frontier methods, developed to overcome drawbacks associated with other methods, attempt to determine "best practice" relationships by measuring inefficiency as the distance between actual performance and a best practice frontier.

The authors sought to discover and summarize best practices for using SFA by reviewing SFA-based studies of U.S. hospitals, and by conducting an empirical analysis of the impact of varying research methods on inefficiency estimates. They conclude that findings from SFA are robust to certain suboptimal strategies (e.g., excluding quality variables because of data constraints). In addition, SFA can be used to identify the correlates of inefficiency, and the inefficiency estimates generated by SFA hold promise for use in schemes that reward efficient providers.

Reprints (AHRQ Publication No. 08-R021) are available from the AHRQ Publications Clearinghouse.

Stewart, K.A., Neumann, P.J., Fletcher, S.W., and Barton, M.B. (2007, August). "The effect of immediate reading of screening mammograms on medical care utilization and costs after false-positive mammograms." Health Services Research; 42(4), pp. 1464-1482.

Immediate reading of mammograms and followup on false positives lessened anxiety among women. A group of 1,140 women aged 40 and older participated in the study at 7 sites in the Boston area between February 1999 and January 2001. Radiologists read the mammograms of 564 women immediately, while the films of 576 women were read in batches after the women left. The study found more false positives with the group whose mammograms were immediately read. However, that setting provided quick resolution of false positives and led to significantly lower anxiety (56 percent knew their mammograms were normal when they left the imaging center) than for the women whose mammogram results would be forthcoming (41 percent knew their mammograms were normal). Immediate reading of mammograms increased health care plans' cost by 10 percent ($4.40) because of less efficiency ($2.22) and extra films being taken ($2.18). About half of the additional cost may be attributed to the higher rate of false positives in the immediate reading setting (22.8 percent) compared with the batch reading setting (18.5 percent). Twelve-month costs for the patient, health plan, and society did not significantly differ between the two groups.

Reprints (AHRQ Publication No. 07-R067) are available from the AHRQ Publications Clearinghouse.

Return to Contents
Proceed to Next Article


The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care