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

Research Briefs

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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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

Akincigil, A., Bowblis, J.R., Levin, C., and others (2007, April). "Adherence to antidepressant treatment among privately insured patients diagnosed with depression." (AHRQ grants HS16097 and HS01182). Medical Care 45(4), pp. 363-369.

According to this study, only half of patients with depression adhered to antidepressant therapy for the first 4 months of treatment, and only 42 percent of patients kept taking their antidepressants from 17 to 33 weeks after starting treatment. Patients who received followup care from a psychiatrist (28 percent of patients) were more likely to continue taking their antidepressant medication. Those who took the newer antidepressants (selective serotonin reuptake inhibitors), which have fewer side effects and are easier to tolerate than older drugs, were also more likely to do so. Younger age, alcohol or other substance abuse, coexisting cardiovascular or metabolic conditions, use of older generation antidepressants (tricyclics and monoamine oxidase inhibitors), and residence in lower-income neighborhoods were associated with lower medication adherence during the first 4 months. Members of HMO plans were less likely than those in non-HMO plans to keep taking their medication as long as 5 to 8 months.

Bailit, J.L. and Love, T.E. (2008). "The role of race in cesarean delivery rate case mix adjustment." (AHRQ HS14352). American Journal of Obstetrics and Gynecology 69, pp. e1-e5.

Perinatal outcomes such as infant and maternal death, prematurity, and cesarean delivery are used as a measure of the quality of obstetric care. These poorer perinatal outcomes are also known to be higher in the black population than in the white population. Risk-adjusted primary cesarean delivery rates can now be used to identify hospitals with poorer perinatal outcomes. Building on previous studies that had found a lower rate of primary cesarean delivery to be correlated with higher rates of poor maternal and neonatal outcomes, the researchers decided to test two risk-adjustment models for primary cesarean rates. Their objective was to see if adding race and ethnicity to an otherwise identical model would help to determine whether race and ethnicity improved the predictive impact of the model. The study found that the two models did not differ substantially in predictive discrimination or in model calibration. This suggests that race and ethnicity can be safely left out of cesarean rate risk-adjustment models.

Buckley, J., Coffin, S. E., Lautenbach, E., and others (2007). "Outcome of Escherichia coli and/or Klebsiella bloodstream infection in children with central venous catheters." (AHRQ grant HS10399). Infection Control and Hospital Epidemiology 28(11), pp. 1308-1310.

The purpose of this study was to identify risk factors for poor outcome in children with catheter-associated bloodstream infections (BSI) due to Escherichia coli and/or Klebsiella. Of the 118 eligible patients (ages 0-9) with this type of BSI who were identified from a hospital clinical microbiology database, 16 (14 percent) had a poor outcome. Poor outcomes were defined as either recurrences of infection (8 cases) or death (8 cases). Receiving mechanical ventilation and receiving total parenteral nutrition were independently associated with death or recurrence of infection. Sixty-four of the children had their catheters removed with a median removal time of 2 days following the initial positive blood culture result. However, catheter removal did not affect either the infection recurrence rate or the death rate. A significant proportion of children with catheter-associated BSI were treated successfully without catheter removal.

Carroll, J., Epstein, R., Fiscella, K., and others (2007, June). "Caring for Somali women: Implications for clinician-patient communication." (AHRQ grant HS14105). Patient Education and Counseling 66, pp. 337-345.

Between 1991 and 2000, up to 100,000 Somali refugees arrived in the United States due to civil war in Somalia. In-depth interviews with resettled Somali women in Rochester, New York, who had diverse spoken languages, stages of acculturation, and literacy level revealed elements of U.S. preventive care that are important for them to feel respected and cared for. These included ease and familiarity of accessing the health care system; availability of interpreters; a trusting relationship with clinicians; preference for female interpreters and female clinicians, especially for gynecologic concerns; and Somali-organized community-based health education programs.

Somali women wanted clinicians to show kindness and patience and to recognize and value their life experience. They also expected clinicians to be sensitive to cultural practices such as female circumcision. They wanted clinicians to understand that female circumcision was common and recommended that clinicians respectfully inquire about it in medical history-taking when appropriate. Because gender roles and responsibilities are distinct in traditional Somali culture, cross-gender interactions can be more awkward. This is one reason Somali women prefer female clinicians. Interpreters may not only improve communication, but may also work to reduce disparities by serving as cultural liaisons and advocates.

Carroll, K.N., Gebretsadik, T., Griffin, M.R., and others (2007, June). "Maternal asthma and maternal smoking are associated with increased risk of bronchiolitis during infancy." (AHRQ grant HS10384). Pediatrics 119(6), pp. 1104-1112.

Researchers at the Vanderbilt Center for Education and Research on Therapeutics and colleagues studied hospitalizations for bronchiolitis among infants of 100,000 women enrolled in the Tennessee Medicaid program during 1995-2003. Infants of mothers who smoked and had asthma were twice as likely to end up in the ED with bronchiolitis as infants whose mothers had neither problem. Infants whose mothers had only one of these risk factors had a lower, but still significantly elevated risk for ED visits, clinics, and/or hospitalizations than infants whose mothers had neither risk factor. However, maternal asthma was the more important of these two risk factors. Similarly, infants were 50 percent, 23 percent, and 19 percent, respectively, more likely to be hospitalized longer than 3 days if their mothers had asthma, also smoked, or just smoked.

Chi, C., Street, W.N., and Ward, M.M. (2007). "Building a hospital referral expert system with a prediction and optimization-based decision support system algorithm." (AHRQ grant HS15009). Journal of Biomedical Informatics.

The researchers aim to develop a tool that patients can use to help them in selecting a hospital that is most likely to yield the desired outcome for their condition. To this end, they are proposing a new method, the Prediction and Optimization-Based Decision Support System (PODSS) algorithm. This algorithm is a decision tool that can provide suggestions by utilizing captured knowledge and optimizing the effectiveness of the chosen action. The algorithm incorporates both institutional characteristics of the hospital and patient characteristics such as age, admission type, comorbidities, and the maximum tolerated distance that the patient is willing to travel to get to the hospital. The article discusses how to capture knowledge, transform it into an objective function, apply the captured knowledge, and use the algorithm in a hospital referral problem. Experimental results of both single and multiobjective optimization examples are discussed and an indirect evaluation method for determining the effectiveness of the method is introduced. The proposed process can also be adapted to recommendations of treatment in clinical care as well as recommendations of healthy lifestyle choices.

Clancy, C.M. (2008). "Designing for safety: Evidence-based design and hospitals." American Journal of Medical Quality 23(1), pp.66-69.

A growing body of literature on evidence-based design describes the links between a hospital's physical design and its quality and safety outcomes. The current boom in hospital construction provides an unusual opportunity to make design decisions that will impact the quality of care over the next 20 to 40 years. Design principles in the areas of noise control, lighting and visual performance, falls, and infection control can lead to higher quality care, according to Carolyn M. Clancy, M.D., director of the Agency for Healthcare Research and Quality (AHRQ). For example, the use of sound-absorbing materials in ceiling tiles and carpeting and single-patient rooms are effective strategies for reducing noise. High illumination of work surfaces reduces medication error rates among pharmacists. Locating bathrooms behind headwalls and providing handrails in patient rooms reduces the likelihood of falls. High-efficiency air particulate filtration systems reduce the spread of airborne pathogens. An AHRQ-produced DVD on "Transforming Hospitals: Designing for Safety and Quality" highlights the achievements of three hospitals in using evidence-based design principles in the construction of new environments of care that contribute to patient safety and the healing process.

Reprints (AHRQ Publication No. 08-R035) are available from the AHRQ Publications Clearinghouse.

Cohn, F., Goodman-Crews, P., Rudman, W., and others (2007, Summer). "Proactive ethics consultation in the ICU: A comparison of value perceived by healthcare professionals and recipients." (AHRQ grant HS10251). Journal of Clinical Ethics 18(2), pp. 140-147.

Calling in an ethics consultant can help resolve end-of-life care discussions and improve family and clinician satisfaction, suggests a study of 122 adult patients in an intensive care unit (ICU). Consultations were offered to the families of these patients in response to latent or manifest conflicts, rather than specific requests for ethics consultations. The majority of clinicians and family members found the ethics consultations helpful (92 and 87 percent), informative (81 and 88 percent), fair (93 and 84 percent), and respectful of personal values (92 and 85 percent). In addition, nearly three-fourths of families and clinicians did not find the consultations to be stressful. Both clinicians and family members found the ethics consultation helped identify (88 and 87 percent), analyze (88 and 87 percent), and resolve (74 and 71 percent) ethical issues. Also, the majority of clinicians and family members agreed with the decision reached in the ethics consultation (81 and 72 percent) and would seek out further ethics consultations in similar situations (95 and 80 percent).

Colmenero, F., Sullivan, S.D., Palmer, J.A., and others (2007, July). "Quality of clinical and economic evidence in dossier formulary submissions." (AHRQ contract 290-05-0006). American Journal of Managed Care 13(7), pp. 401-407.

The Academy of Managed Care Pharmacy (AMCP) recommends a format for drug companies to submit a dossier of data on drugs that they want health plans to consider including in their drug formularies. Researchers reviewed the quality of dossiers submitted to a large health plan, a leader in dossier review, between 2002 and 2005, and found that the information provided in these dossiers could be improved, especially data on off-label drug indications and economic analysis. Only 46 percent of drug dossiers submitted to a large health plan included analyses of the drug's economic value relative to alternative therapies. Also, proportionately more studies supporting on-label claims (medical indications approved by the U.S. Food and Drug Administration) than off-label claims had random treatment allocation (50 vs. 33 percent), were double-blinded (43 vs. 27 percent), or were meta-analyses (8 vs. 1 percent). Adherence to analytic standards was generally higher in economic analysis of high-cost medications (30-day treatment cost over $1,000) and for "innovative" instead of "me-too" products (no clear-cut advantage over competitors).

Coopey, M., James, M.D., Lawrence, W., and Clancy, C.M. (2008). "The challenge of comparative effectiveness: Getting the right information to the right people at the right time." Journal of Nursing Care Quality 23(1), pp. 1-5.

The need to develop better evidence about the benefits and risks of different treatments is paramount. The Agency for Healthcare Research and Quality's (AHRQ) Effective Health Care Program is dedicated to fulfilling this need by performing research and getting results to those who make health care decisions. This program, created by the Medicare Modernization Act of 2003, conducts and supports evidence syntheses and research on topics of highest priority to Federal and State health care programs. It builds on prior investment in outcomes research that has shown, for example, that many patients who could benefit are not receiving treatments known to be effective. Under the Effective Health Care Program, reports on choosing medications for osteoarthritis have already been issued. More reports are forthcoming on treatments for conditions such as low bone density and type 2 diabetes. This type of information is especially important to nurses and other front-line caregivers who have a great stake in the collection and timely dissemination of clinical knowledge.

Reprints (AHRQ Publication No. 08-R031) are available from the AHRQ Publications Clearinghouse.

Dansinger, M.L., Tatsioni, A., Wong, J.B., and others (2007, July). "Meta-analysis: The effect of dietary counseling for weight loss." (AHRQ contract 290-02-0022). Annals of Internal Medicine 147, pp. 41-50.

A meta-analysis of dietary counseling studies found that, when compared with usual care, dietary counseling can produce a modest 6 percent weight loss within a year among overweight adults. However, about half of the initial weight loss is typically regained after 3 years, and the weight difference narrows considerably over 4 years. Dietary advice was conveyed primarily by group meetings in 18 studies, individual meetings in 13 studies, and group and individual meetings in 11 studies. Three trials used the Internet and three trials did not specify how they conveyed advice. The active phase of the behavioral weight loss programs (more intensive and more frequent interactions with participants) ranged from 2.5 to 48 months. The maximum effect of dietary counseling was a loss of 1.9 body mass index (BMI) units, that is, 6 percent of body weight.

Edwards, R.R., Klick, B., Buenaver, L., and others (2007, July). "Symptoms of distress as prospective predictors of pain-related sciatica treatment outcomes." (AHRQ grants HS06344, HS08194 and HS09804). Pain 130, pp. 47-55.

This study found that patients with sciatica, whether they are treated with surgery or nonsurgical methods, have worse pain and function following treatment if they also suffer from anxiety and depression. Patients with sciatica completed a baseline assessment (including mood prior to the start of treatment) and followup questionnaires about pain and disability at 3, 6, 12, 24, and 36 months after starting treatment. The questionnaires asked about sciatica symptoms such as leg pain, leg or foot weakness, leg numbness, and pain in the back or leg while sitting. Disability questions addressed physical impairment due to several low back conditions, as well as general pain and physical functioning. Symptoms of depression and anxiety, both at baseline and during the period prior to each assessment, were significant independent predictors of worse pain and function, after controlling for other factors. The patient's mental health score at the 2-year followup assessment improved by 40 percent the predictive accuracy of pain and functioning compared with their score on the sciatica symptom frequency index at that time.

Feifer, C., Nemeth, L., Nietert, P.J., and others (2007, May/June). "Different paths to high-quality care: Three archetypes of top-performing practice sites." (AHRQ grant HS13716). Annals of Family Medicine 5(3), pp. 233-241.

Researchers studied the top 10 practices (out of 101) that opted to participate in a demonstration project to advance adherence to clinical practice guidelines. Primary care practices that want to deliver high-quality care can adopt one of three archetypes, or a combination of three, to achieve their goal. "Technophiles" champion health information technology, while "Motivated Teams" enable their staff to play key roles in improving the practice and reward them for success. The "Care Enterprise" archetype uses a business approach that focuses on customer service and risk management.

Technophile practices relied heavily on electronic medical records (EMR) to guide staff and clinicians to perform and document routine tasks consistently, such as recording blood pressure and assessing alcohol use. Key to the Technophiles' success was an innovative physician who was computer savvy and a champion of change. Practices that fit the Motivated Team archetype met quarterly for half-day workshops to focus on quality improvements. They let staff members choose which quality indicators they wanted to improve and rewarded them financially when they met their goals. The Care Enterprise archetype, a subset of the Motivated Team, let customer needs drive practice change.

Glance, L.G., Osler, T.M., Mukamel, D.B., and Dick, A.W. (2007, June). "Estimating the potential impact of regionalizing health care delivery based on volume standards versus risk-adjusted mortality rate." (AHRQ grant HS13617). International Journal for Quality in Health Care 19(4), pp. 195-202.

Researchers conducting a retrospective study of California hospitals found that selective referral of patients to high-volume hospitals to undergo either abdominal aortic aneurysm surgery (more than 50 cases/year), coronary artery bypass surgery (more than 450 cases/year), or coronary angioplasty (more than 400 cases/year) would result, at best, in only modest improvement in patient outcomes. Selective referral to high-volume centers would only moderately reduce in-hospital mortality by 2 to 20 percent and would be extremely disruptive by reducing by 70 to 99 percent the number of hospitals treating these conditions. Selective referral to high quality centers was estimated to dramatically reduce in-hospital mortality by 50 percent. However, this would not be a realistic policy option given the need to redirect more than 80 percent of the patients to a small number of centers of excellence. Selective avoidance of low-volume hospitals (lowest quartile of procedure volume) would not improve mortality. Yet selective avoidance of low quality hospitals was estimated to result in a small 2 to 6 percent improvement in overall mortality, while causing relatively minor disruptions in patient referral patterns.

Hanmer, J., Hays, R.D., and Fryback, D.G. (2007). "Mode of administration is important in US national estimates of health-related quality of life." (AHRQ grant HS16574). Medical Care 45 (12), pp. 1171-1179.

The researchers investigated whether different modes of survey administration affected the national averages for health-related quality of life (HRQoL). They used data from two surveys administered by telephone (the Joint Canada/United States Survey of Health and the National Health Measurement Study); one by mail (the 2002 Medical Expenditure Panel Survey); and one self-administered with the interviewer present (the US Valuation of the EuroQol EQ-5D Health States Survey). In their comparison of the HRQoL summary scores from each of the four surveys, the researchers found that when the mode of administration was the same, the age- and gender-stratified mean estimates for HRQoL measures were similar. When the mode of administration differed, respondents aged 70 and over reported better health with a telephone-administered questionnaire than with a self-administered questionnaire. Also, older age groups and females reported worse HRQoL generally, regardless of mode of survey administration.

Harrold, L.R., Patterson, M.K., Andrade, S.E., and others (2007, June). "Asthma drug use and the development of Churg-Strauss syndrome (CSS)." (AHRQ grant HS10391). Pharmacoepidemiology and Drug Safety 16, p. 620-626.

Case reports suggest that use of leukotriene modifiers, the newest anti-inflammatory drugs for asthma, may be linked to the onset of Churg-Strauss syndrome (CSS). This study did not find an association between the two; however, it is not possible to rule out modest associations between the use of leukotriene modifiers and CSS, given that CSS is so rare and so highly correlated with asthma severity. Researchers analyzed pharmacy and other data on nearly 14 million people enrolled in a U.S. national health plan and 3 U.S. managed care organizations. They identified 47 cases of possible or definite CSS and 4,700 controls who used asthma drugs between 1995 and 2002. Overall, 6 CSS cases and 202 controls used leukotriene modifiers in the 2 to 6 months prior to the onset of CSS. Those who used leukotriene modifiers had four times the likelihood of developing CSS. However, this association disappeared when use of other asthma drugs was taken into account. Those who took leukotriene modifiers were also significantly more likely to use a greater number of other asthma drugs and to use oral steroids.

Kane, R.L., Wilt, T., Suarez-Almazor, M.E., and Fu, S.S. (2007, May). "Disparities in total knee replacements: A review." (AHRQ contract 290-02-0009). Arthritis & Rheumatism 57(4), pp. 562-567.

This systematic review of studies found that women were nearly twice as likely to undergo total knee replacement (TKR) as men. This was primarily because women were more likely than men to be referred by their primary care doctor for the procedure. It is possible that the severity or type of arthritis varied among men and women. Lower rates of TKR among blacks than whites occurred despite a higher prevalence of knee osteoarthritis (KO) among blacks, suggesting that the prevalence of KO was not a mitigating factor. Several studies of primary care doctors, orthopedic surgeons, rheumatologists, and other physicians found varied clinical consensus about indications for TKR or for referring patients for TKR. Race and sex were not listed as either indicators or nonindicators for surgery.

Kent, D.M., Ruthazer, R., Griffith, J.L., and others (2007, May). "Comparison of mortality benefit of immediate thrombolytic therapy versus delayed primary angioplasty for acute myocardial infarction." (AHRQ grant HS10280). American Journal of Cardiology 99, pp. 1384-1388.

This study used patient-level data to develop a model that examined the influence of baseline mortality risk on the degree of primary percutaneous coronary intervention (PPCI) benefit and the risk of treatment delay in heart attack patients. The model showed that as baseline risk rose, the relative benefit of PPCI compared with thrombolytic therapy significantly increased, with high-risk patients benefiting greatly. As baseline risk increased, the risk associated with longer surgery delay also increased. Based on the model, a surgery delay of about 100 minutes would nullify its benefit over immediate thrombolytic therapy for patients who had at least a moderate degree of mortality risk (greater than 4 percent). In contrast, patients at relatively low risk of dying gained little or no incremental mortality benefit from PPCI compared with immediate clot-busting therapy. For this group, consideration of other outcomes such as risks for stroke or reinfarction and logistical concerns should guide treatment decisions more than consideration of mortality trade-offs. Yet studies have shown that thrombolytic medication reestablishes blood flow (reperfusion) in only about 50 percent of patients even 60 minutes after "needle time," whereas over 90 percent of PPCI-treated patients will achieve reperfusion immediately after the balloon inserted into the artery is inflated.

Kroll, T.L., Richardson, M., Sharf, B.F., and Suarez-Almazor, M.E. (2007). "'Keep on trucking' or 'It's got you in this little vacuum': Race-based perceptions in decision-making for total knee arthroplasty." (AHRQ grant HS10876). The Journal of Rheumatology 34, pp. 1069-1075.

Ethnic differences in attitudes and beliefs about total knee replacement (TKR) may contribute to the disparities in use of TKR, according to this study. Differences were most obvious in explanations of illness, perceived changes in lifestyle, physician and health care system trust, and attitudes about paying for surgery. The researchers conducted 6 focus groups of 37 patients with knee osteoarthritis (KO) being treated at the same facility: 2 black groups, 2 Hispanic groups, and 2 white groups. Blacks described internal causes (aging or "bone on bone") for their condition. Hispanics and whites attributed their condition to external causes, or blamed the medical profession for not fully understanding the condition. Blacks and Hispanics described KO as being more debilitating than whites did. Hispanics talked about the way in which their KO limited how they could spend their time. Although whites talked about how osteoarthritis limited their lifestyle, they were more likely to describe ways in which they overcame those limits. Trust in their doctor was critical for the surgery decision among Hispanics. In contrast, blacks and whites were more likely to have their trust based on the reputation of the physician or affiliated hospital. Blacks were willing to pay for the surgery, even it if meant borrowing money, to alleviate their pain. Whites only talked about the out-of-pocket expenses for a highly regarded surgeon. Hispanics were more likely to speak of having the operation and then paying for it over time, after the fact.

Lakshminarayan, K., Anderson, D.C., Borbas, C., and others (2007, June), "Blood pressure management in acute ischemic stroke." (AHRQ grant HS11073). Journal of Clinical Hypertension 9(6), pp. 444-453.

This study found that blood pressure management of patients hospitalized for acute stroke varied substantially. Of 1,118 acute stroke patients in 19 Minnesota hospitals, 129 received as-needed antihypertensive medication in direct response to elevated blood pressure in the first 24 hours of admission. Of these 129, 56 percent were overtreated according to American Stroke Association (ASA) guidelines and 24 percent were overtreated according to less stringent European Stroke Initiative (EUSI) guidelines. Of the 1,052 patients not treated, 16 percent were undertreated by ASA guidelines and 3 percent by EUSI guidelines. In contrast, nearly all patients (93 percent) were likely to have their chronic hypertension appropriately treated with medication at discharge. The study authors point out that ongoing clinical trials are trying to resolve some of the differences in current guidelines for managing blood pressure in acute stroke. For example, concern that antihypertensives may further reduce blood flow to the brain during acute stroke prompts some guidelines to caution clinicians about their use. Also, recommended blood pressure levels for initiating antihypertensive medication in acute stroke patients differ. Finally, no guidelines recommend when antihypertensive regimens should be initiated after stroke or whether prestroke regimens should be stopped during an acute phase.

Langfitt, J.T., Westerveld, M., Hamberger, M.J., and others (2007, June). "Worsening quality of life after epilepsy surgery: Effects of seizures and memory decline." (AHRQ grant HS09986). Neurology 68, pp. 1988-1994.

A study of 138 patients who underwent temporal lobe surgery for intractable epileptic seizures found that patients who were seizure-free 2 and 5 years after surgery reported improved quality of life, regardless of memory outcome. Among the 18 percent who still endured persistent seizures after surgery, quality of life remained stable only when memory did not decline (10 percent). Quality of life diminished when memory declined (8 percent). The patients with continued seizures and memory decline had characteristics that predicted poor seizure or memory outcome. They were more likely to have had a dominant hemisphere resection (73 vs. 38 percent), lower baseline IQ (85 vs. 93), later age at epilepsy onset (23.6 vs. 14.1 years), and shorter duration of epilepsy (17.3 vs. 25.3 years). They also tended to be less likely to have hippocampal atrophy on magnetic resonance imaging (45 vs. 72 percent).

Lobach, D.F., Kawamoto, K., Anstrom, K.J., and others (2007). "Development, deployment and usability of a point-of-care decision support system for chronic disease management using the recently-approved HL7 decision support service standard." (AHRQ grant HS15057). MEDINFO 12(Pt 2), pp. 861-865.

Clinical decision support (CDS) systems relying on computers are a very effective way to improve clinician compliance with evidence-based care standards. However, their use has remained limited due to the lack of an efficient method of encapsulating, processing, and delivering medical knowledge for use in clinical software applications. The authors describe the first implementation of a decision support system based on the recently approved HL7 Decision Support Service draft standard that facilitates the implementation of CDS systems using software services. The new CDS system is known as the SEBASTIAN (System for Evidence-Based Advice through Simultaneous Transaction with an Intelligent Agent across a Network) Decision Support Service (DSS). The SEBASTIAN DSS was used to implement a point-of-care chronic disease management system within a health system handling over 60,000 hospitalizations and 1.2 million outpatient encounters a year. A survey of 20 users found that responses for content, accuracy, format, and ease of use were significantly favorable. The authors believe that implementation of this DSS validates the usefulness of the new HL7 draft standard in the context of a large health system.

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