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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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

Albrecht, S.J., Fishman, N.O., Kitchen, J., and others (2006, June). "Reemergence of gram-negative health care-associated bloodstream infections." (AHRQ grant HS10399). Archives of Internal Medicine 166, pp. 1289-1294.

Gram-positive organisms have been the predominant cause of primary health care-associated bloodstream infections (PHA-BSIs) since the 1970s. Recently, gram-negative organisms have reemerged as a cause of PHA-BSIs. Researchers identified all PHA-BSIs among adults hospitalized at the University of Pennsylvania hospital from 1996 through 2003. They calculated the annual proportion of PHA-BSIs accounted for by each of the 10 most common bacterial and fungal organisms, and the following specific organisms or organism groups: Staphylococcus aureus, coagulase-negative staphylococci (CNS), enterococci, gram-negative organisms, and Candida species. They identified a total of 3,662 PHA-BSIs caused by 4,349 bacterial and fungal isolates. From 1999 to 2003, the proportion of PHA-BSIs due to gram-negative organisms (mostly Klebsiella pneumoniae, E. coli, Pseudomonas aeruginosa, and Enterobacter cloacae) increased from 16 to 24 percent. This increase was accompanied by a decline in the proportion of PHA-BSIs from CNS (from 33.5 to 29.9 percent) and S. aureus (from 18.8 to 11.8 percent).

Apter, A.J., Kinman, J.L., Bilker, W.B., and others (2006, April). "Is there cross-reactivity between penicillins and cephalosporins?" (AHRQ grant HS10399). American Journal of Medicine 119(4), pp. 354-363.

The risk of anaphylaxis, a severe allergic reaction that can lead to shock, is low for patients with penicillin allergy who take cephalosporins. Researchers retrospectively studied the United Kingdom General Practice Research Database to identify patients who received a prescription for a penicillin followed by a prescription for a cephalosporin and compared this group with patients who received a prescription for penicillin followed by a prescription for a sulfonamide antibiotic, which can also cause allergic reactions in those who are allergic to penicillin. A total of 3,375 patients received a penicillin. The 15 percent of these patients who received a subsequent cephalosporin had 10 times the risk of developing an allergic reaction compared with those who had no prior allergic reaction to penicillin, unadjusted for other factors. However, cross-reactivity was not an adequate explanation for the increased risk of subsequent allergic reactions, and the absolute risk of anaphylaxis after a cephalosporin was less than 0.0001 percent. The unadjusted risk of an allergic reaction to a subsequent sulfonamide antibiotic after a prior allergic reaction to penicillin was 7.2 times that of those with no prior allergic reaction to penicillin.

Baker, D.W., Feinglass, J., Durazo-Arvizu, R., and others (2006, November). "Changes in health for the uninsured after reaching age-eligibility for Medicare." (AHRQ grant HS10283). Journal of General Internal Medicine 21, pp. 1144-1149.

Researchers analyzed data on 3,419 middle-aged people who transitioned from private insurance or being uninsured to having Medicare coverage at the 1996, 1998, 2000, or 2002 interviews of the Health and Retirement Study. Those who were uninsured during the 2 years prior to gaining Medicare insurance were more likely than those who had been privately insured to have a decline in their self-reported overall health (17 vs. 9 percent), to have their mobility worsen (29 vs. 20 percent), and to develop new agility difficulties (36 vs. 27.5 percent) during that period. The health of the previously uninsured, who had been on Medicare for less than 1 or 2 years at the time of their followup interview, was not significantly different from that prior to Medicare coverage. However, 2 years after obtaining Medicare coverage, this group no longer had a greater risk of deteriorating health compared with their previously insured counterparts. These findings underscore the importance of expanding insurance to those in late middle age, especially those with diabetes, hypertension, and heart disease, suggest the researchers.

Bolcic-Jankovic, D., Clarridge, B.R., Fowler Jr., F.J., and Weissman, J.S. (2007, January). "Do characteristics of HIPAA consent forms affect the response rate?" (AHRQ grant HS11928). Medical Care 45(1), pp. 100-103.

Researchers are required to obtain written authorization from patients to gain access to protected health information. However, what the forms ask may affect the likelihood of patient consent, according to this study. Forms requesting patients' permission for access to their medical records and other personal health information should be clear about the institution from which they come, easy to understand, and should not ask for social security number or other highly sensitive information unrelated to health care, suggest the study authors. They surveyed previously hospitalized patients from 16 Massachusetts hospitals, who were asked to provide authorization for review of their medical records and were sent a copy of the hospital's form to sign and return. Only half the patients returned signed forms. The likelihood of return was significantly affected by the requirement of the social security number and the clarity of the hospital name.

Clancy, C.M. (2007, February). "Emergency departments in crisis: Opportunities for research." HSR: Health Services Research 42(1), pp. xiii-xx.

In this paper, the Director of the Agency for Healthcare Research and Quality (AHRQ) notes that U.S. emergency departments (EDs) are in crisis. They can barely handle the current everyday demand for their services, let alone respond to public health disasters such as Hurricane Katrina. She describes some AHRQ research in this area, such as the Center for Safety in Emergency Care. This consortium addressed the cognitive psychology of human behavior and error, clinical epidemiology of adverse events, use of technology to improve performance and reduce errors, and the application of human factors engineering to improve safety.

AHRQ also supports the State Emergency Department Databases (which researchers can use to identify trends and develop strategies to enhance emergency care services), projects to improve ED patient flow and monitor crowding, and information technology to improve the safety and quality of ED care.

Reprints (AHRQ Publication No. 07-R041) are available from the AHRQ Publications Clearinghouse.

Culler, S.D., Atherly, A., Walczak, S., and others (2006, Summer). "Urban-rural differences in the availability of hospital information technology applications: A survey of Georgia hospitals." (AHRQ grant HS11918). Journal of Rural Health 22(3), pp. 242-247.

A survey of Georgia community hospitals suggests that rural hospitals have not incorporated health information technology (HIT) to the extent that urban hospitals have, especially in certain clinical areas. On average, Georgia hospitals had 59 percent of 97 functional HIT applications and technological devices available. More than 40 percent of rural hospitals and 47 percent of urban hospitals had over 70 percent of all HIT applications available. However, over 40 percent of rural hospitals reported less than 50 percent of HIT applications available compared with only 16 percent of urban hospitals. Also, 26 percent of rural hospitals had adopted less than 30 percent of the HIT applications. Urban hospitals had significantly more HIT applications available than rural hospitals in 4 areas: emergency room services (7 of 10), surgical/operating room (8 of 12), laboratory (7 of 12), and radiology (5 of 11). Some of these urban-rural differences may be due to the narrow scope of laboratory, radiology, emergency room, and surgery services provided by rural hospitals.

de Pablo, P., Losina, E., Mahomed, N., and others (2006, June). "Extent of followup care after elective total hip replacement." (AHRQ grant HS09775). Journal of Rheumatology 33, pp. 1159-1166.

Researchers found that less than half of elderly adults who underwent total hip replacement (THR) had consistent radiographic followup. The researchers analyzed Medicare claims data to identify a group of 622 elderly patients in 3 States (Ohio, Pennsylvania, and Colorado), who received elective primary THR in 1995. They surveyed the patients 3 and 6 years after the surgery. Overall, 15 percent of patients had no followup radiographs, 43 percent had early followup only, and 42 percent had consistent followup radiographs over 6 years. After accounting for other factors, older patients were 24 percent less likely to have radiographic followup than younger patients per each 5-year increase in age. Patients with no college education were 42 percent less likely to have radiographic followup than those with more education, and those with lower income were 50 percent less likely to have followup than those with a higher income.

Dunn, S.L., Corser, W., Stommel, M., and Holmes-Rovner, M. (2006). "Hopelessness and depression in the early recovery period after hospitalization for acute coronary syndrome." (AHRQ grants HS10531 and HS09514). Journal of Cardiopulmonary Rehabilitation 26, pp. 152-159.

Researchers found that a considerable number of people suffer from hopelessness and depression during the early recovery period after hospitalization for acute coronary syndrome (ACS). They administered a depression and hopelessness scale to 525 post-ACS patients at 5 hospitals in Michigan and found that 36 percent of the post-ACS patients studied suffered frequent and moderate to severe depression symptoms. Another 27 percent suffered from frequent and moderate to severe symptoms of hopelessness. Some patient characteristics were more predictive of depression, while others were more predictive of hopelessness. For example, women were more likely to become depressed, but not more hopeless than men, and people with less education were more likely to feel hopeless than their more educated counterparts. Having had coronary artery bypass graft or coronary angioplasty was also predictive of hopelessness.

Fiscella, K., Holt, K., Meldrum, S., and Franks, P. (2006, September) "Disparities in preventive procedures: Comparisons of self-report and Medicare claims data." (AHRQ grant HS13173). BMC Health Services Research 6(122), available at

Researchers found that with the exception of prostate specific antigen (PSA) testing for prostate cancer, racial/ethnic disparities in use of preventive procedures were generally larger when using Medicare claims data than when using elderly patients' self-report. They analyzed self-report and matching claims data from elderly Medicare beneficiaries who participated in the Medicare Current Beneficiary Survey, 1999-2002. Six preventive procedures were included: PSA testing, influenza vaccination, Pap smear testing, cholesterol testing, mammography, and colorectal cancer testing. Minorities were more likely than whites to self-report preventive procedures in the absence of billing claims, after adjusting for several sociodemographic factors. For Pap testing, some were up to twice as likely to report Pap smear testing in the absence of claims. Self-report differences in receipt of preventive care procedures by minority status (majority vs. minority) ranged from -2.4 percent for cholesterol testing to 17.9 percent for influenza vaccination. In contrast, differences based on claims data ranged from 5.1 percent for cholesterol testing to 19.9 percent for influenza vaccination.

Grant, R.W., Wald, J.S., Poon, E.G., and others (2006, October). "Design and implementation of a Web-based patient portal linked to an ambulatory care electronic health record: Patient gateway for diabetes collaborative care." (AHRQ grant HS13660). Diabetes Technology & Therapeutics 8(5), pp. 576-586.

This paper describes the conceptual framework, design, implementation, and analysis plan for a Web-based diabetes patient portal linked directly to the electronic health record (EHR) of a medical center via secure Internet access. The Web portal, called Patient Gateway, allows patients to interact directly with their EHR. The portal maximizes patient engagement by importing the patients' current clinical data in an educational format, providing patient-tailored decision support, and enabling the patient to author a diabetes care plan. The study authors are assessing the impact of this advanced informatics tool for collaborative diabetes care in a study involving 14 primary care practices.

Gresenz, C.R., Rogowski, J., and Escarce, J.J. (2007, February). "Health care markets, the safety net, and utilization of care among the uninsured." (AHRQ grant HS10770). HSR: Health Services Research 42(1), pp. 239-264.

Facilitating transport to safety net providers and increasing the number of such providers are likely to increase use of care among the rural uninsured. Use of care could also be improved among uninsured persons living in areas with substantial managed care presence, especially where managed care competition is limited, concludes this study. The researchers analyzed data from 1996 to 2000 of the Medical Expenditure Panel Survey and other sources to analyze medical expenditures among uninsured adults living in urban and rural areas. Distances rural uninsured persons had to travel to safety net providers were significantly associated with care use. In urban areas, higher percentages of uninsured individuals in the area, pervasiveness and competitiveness of managed care, limited primary care physician supply, and limited safety net capacity were significantly related to less health care use.

Harrold, L.R., Saag, K.G., Yood, R.A., and others (2007, February). "Validity of gout diagnoses in administrative data." (AHRQ grants HS10391 and HS10389). Arthritis & Rheumatism 57(1), pp. 103-108.

Use of administrative data alone in epidemiologic and health services research on gout may lead to misclassification of patient diagnosis. Medical record reviews for validation of claims data may provide an inadequate gold standard to confirm gout diagnoses, according to this study. The researchers identified patients from four managed care plans who had at least two ambulatory claims for a diagnosis of gout between 1999 through 2003. Trained medical record reviewers and two rheumatologists reviewed the medical records of a random sample of 200 patients. Based on record reviews, patients were also classified according to standard gout criteria. There was low agreement between physician assessments and established gout criteria.

Hartz, A., Kent, S., James, P., and others (2006). "Factors that influence improvement for patients with poorly controlled type 2 diabetes." (AHRQ grant HS13581 and HS14410). Diabetes Research and Clinical Practice 74, pp. 227-232.

Improved blood-sugar control among people with poorly controlled type 2 diabetes depends largely on patient self-care behaviors, concludes this study. The researchers examined 69 patients with type 2 diabetes from 7 practices who had 2 glycosylated hemoglobin (HbA1c) levels of at least 8 percent in the past 6 months (uncontrolled blood sugar levels). After at least 1 year of followup, 26 patients became well controlled, 14 had intermediate control, and 29 remained in poor control. Achieving blood-sugar control was positively associated with patients' understanding of diabetes, adherence to recommended meal plans, and glucose monitoring. It was not significantly associated with patient gender, age, duration of diabetes, body mass index, or HbA1c levels prior to baseline.

Hughes C.M. and Lapane, K.L. (2006, August). "Nurses' and nursing assistants' perceptions of patient safety culture in nursing homes." (AHRQ grant HS11835). International Journal for Quality in Health Care 18(4), pp. 281-286.

A survey of 367 nurses and 636 nursing assistants employed at 26 nursing homes in Ohio indicates that about 40 percent of nursing staff find it difficult to make changes to improve patient safety most or all of the time. Although 40 percent of nursing staff reported that management seriously considered staff suggestions to improve resident safety, only half of nurses reported management discussions with staff to prevent recurrence of mistakes. One in five reported feeling punished for reporting medical errors, and two in five noted that reporting errors was often seen as a "personal attack" against a staff member or management. U.S. nursing homes that do not meet the regulations may be subject to a number of penalties, the most severe being closure of the facility. Thus, staff may feel reluctant to report safety issues that may draw attention to individuals and to the nursing home. Despite these consequences, at least 60 percent of nursing assistants and 80 percent of nurses reported a safety problem at least once in the previous month. Yet, only two in five staffers reported that they were told what happens as a result of incident reports most or all of the time.

Keenan, G., Yakel, E., and Marriott, D. (2006). "HANDS: A revitalized technology supported care planning method to improve nursing handoffs." (AHRQ grant HS15054). Consumer-Centered Computer-Supported Care for Healthy People: IOS Press, 2006.

Each day, nurses create and file care plans in medical records. However, current forms of care plans do little to either enhance the flow of clinical information or communicate shared patient goals. This paper introduces the theoretical model underpinning the HANDS care planning method. It also presents findings on the first year of a 3-year multisite study in which this method and a new health information technology application supporting the process were introduced. The HANDS model focuses on the handoff as a focal point for information transfer and reinforcing shared meaning and goals. Early findings show the method has the potential for revolutionizing nursing practice.

Malek, M.A., Curns, A.T., Holman, R.C., and others (2006, June). "Diarrhea- and rotavirus-associated hospitalizations among children less than 5 years of age: United States, 1997 and 2000." Pediatrics 117(6), pp. 1887-1892.

Severe rotavirus disease is estimated to account for 4 to 5 percent of all hospitalizations and about 30 percent of hospitalizations for diarrhea among U.S. children under 5 years of age, according to a new study. A newly approved rotavirus vaccine, RotaTeq, should substantially reduce this burden. Claudia A. Steiner, M.D., of the Agency for Healthcare Research and Quality, and colleagues at the Centers for Disease Control and Prevention analyzed data from a national sample of pediatric hospital discharges from the Kids' Inpatient Database. They estimated the number and rate of diarrhea- and rotavirus-associated hospitalizations among U.S. children under 5 years of age in 1997 and 2000. They calculated that diarrhea was the cause of 13 percent of childhood hospitalizations in 1997 and 2000 by age 5. One-third (35 percent) of these diarrhea-related hospitalizations were coded as viral. Rotavirus was the most common pathogen recorded for 18 percent of diarrhea-associated hospitalizations in 1997 and 19 percent in 2000. The researchers estimated that rotavirus was associated with 4 to 5 percent of all childhood hospitalizations, and that 1 in 67 to 1 in 85 children will be hospitalized with rotavirus infection by 5 years of age.

Reprints (AHRQ Publication No. 07-R004) are available from the AHRQ Publications Clearinghouse.

McConnochie, K.M., Conners, G.P., Brayer, A.F., and others (2006, July). "Differences in diagnosis and treatment using telemedicine versus in-person evaluation of acute illness." (AHRQ grant HS10753). Ambulatory Pediatrics 6(4), pp. 187-195.

Children seen by their usual physician for an acute problem at a medical center were randomly assigned to be seen either by an in-person study physician or a telemedicine study physician. An ear-nose-throat endoscope/camera and all-purpose digital camera captured tympanic membrane, eye, nose, throat, and skin images. An electronic stethoscope captured heart and lung sounds. The study telemedicine physician guided the telemedicine assistant in obtaining digital images and audio files, and eliciting information that required tactile sensation, such as palpable attributes of a skin rash or presence of tender lymph nodes. Telemedicine and in-person study physicians differed in diagnosis from usual physicians for 13.8 percent and 8.3 percent, respectively, of the 492 visits studied. The differences in rate of disagreement with usual physicians about prescriptions were similar (32.2 vs. 27.4 percent). However, telemedicine evaluation for children with upper respiratory tract infection (URI)-ear symptoms involved unique technical requirements and clinical judgments. Thus, for the 202 URI-ear visits, telemedicine physicians disagreed with usual physicians on diagnosis more than twice as often as the in-person study physicians (16.6 vs. 6.3 percent). However, for the remaining 290 visits, telemedicine and in-person study physicians disagreed with usual physicians on diagnosis about equally (11.5 vs. 9.9 percent).

McConnochie, K.M., Conners, G.P., Brayer, A.F., and others (2006). "Effectiveness of telemedicine in replacing in-person evaluation of acute childhood illness in office settings." (AHRQ grant HS10753). Telemedicine and e-Health 12(3), pp. 308-316.

This study randomly assigned 253 children to in-person evaluation by study physicians and 239 children to evaluation by study physicians via telemedicine. Children were seen in a pediatric primary care practice or pediatric emergency department of a university-affiliated medical center. Using a base telemedicine model, telemedicine study physicians completed 74.1 percent of visits (that is, made a diagnosis) compared with 76.7 percent for in-person study physicians and 76 percent for usual physicians. The simple telemedicine model (use of simple office laboratory tests and albuterol administration) increased completion rates substantially. Using this model, telemedicine study physicians completed 84.9 percent of visits compared with 86.6 percent for in-person study physicians and 85.2 percent for usual physicians. Using the extended telemedicine model (use of a complete set of tests and procedures), study physicians completed 97.1 percent of visits compared with 96.8 percent for in-person study physicians and 100 percent for usual physicians.

Mold, J.W., Woolley, J.H., and Nagykaldi, Z. (September, 2006). "Associations between night sweats and other sleep disturbances: An OKPRN study." (AHRQ grant HS13557). Annals of Family Medicine 4(5), pp. 423-426.

Night sweats are more prevalent than previously thought, according to a recent study of adults seen at 10 network primary care offices. One-third (34 percent) of 363 patients interviewed reported night sweats. Symptoms such as daytime tiredness, legs jerking during sleep, and awakening with pain in the night were associated with nearly twice the likelihood of having night sweats. Snoring, snore frequency or loudness, having fallen asleep while driving in the past year, or body mass index were not associated with night sweats.

O'Malley, A.J., Landon, B.E., and Guadagnoli, E. (2007, February). "Analyzing multiple informant data from an evaluation of the health disparities collaboratives." (AHRQ grant HS13653). HSR: Health Services Research 42(1), pp. 146-164.

The authors of this study used multiple informant data to evaluate Health Disparities Collaboratives. They surveyed executive directors, medical directors, and providers from 65 community health centers (176 informants) who participated in an evaluation of the Health Disparities Collaboratives. Multiple informants' analysis of both Collaborative participation and quality improvement efforts found significant effects and differences between informants that traditional methods failed to find. Executive directors and medical providers were the most discrepant in their analyses. Different informants may have different insights or experiences. Thus, it is important that differences among informants be measured, and ultimately understood by health services researchers.

Phillips, R.L., Dovey, S.M., Graham, D., and others (2006, September). "Learning from different lenses: Reports of medical errors in primary care by clinicians, staff, and patients." (AHRQ grant HS11584). Journal of Patient Safety 2(3), pp. 140-146.

A new study found that physicians, clinic staff, nurse practitioners (NPs), physician assistants (PAs), and resident physicians in family physician offices will all report medical errors, their consequences, and their potential remedies. The different perspectives provided by various staff of family medicine practices improve understanding of the factors contributing to errors, error cascades, the broader ways that errors affect people, and potential solutions, note the study authors. Clinicians, staff, and patients of 10 family medicine clinics reported errors they observed through anonymous reports submitted via a Web site, paper forms, and a voice-activated phone system. A total of 401 clinicians and staff reported 935 errors within 717 events. Staff completed 53 percent of reports, clinicians completed 47 percent of reports, resident physicians completed 6 percent, and NPs and PAs 3 percent. Patients submitted 126 reports, 18 of which included errors. Most (96 percent) errors reported were process errors, not related to knowledge or skill. Most of the health consequences of these errors placed the patient at elevated risk for harm (49 percent) or made the patients, their families, or their clinicians upset (33 percent).

Platt, R. (2007). "Speed bumps, potholes, and tollbooths on the road to panacea: Making best use of data." (AHRQ grant HS10391). Health Affairs 26(2), pp. w153-w155.

Electronic health databases promise to transform both the assessment of health care delivery and our understanding of treatments' safety and effectiveness, assert the authors of this paper. However, to achieve these goals, it will be necessary to recognize limits on inferring causality, protect confidentiality while allowing important societal gain, and link health data back to the individual patient. It will also be necessary to obtain additional information from medical records, understand ways in which electronic data can misrepresent reality, and create the infrastructure, expertise, and resources to use the data. However, realizing databases' potential will require long-term commitment and investment beyond the maintenance of the databases themselves.

Poon, E.G., Cina, J.L., Churchill, W., and others (2006, September). "Medication dispensing errors and potential adverse drug events before and after implementing bar code technology in the pharmacy." (AHRQ grant HS14053). Annals of Internal Medicine 145, pp. 426-434.

Implementing bar code technology can substantially decrease dispensing errors and potential harm to patients due to adverse drug events (ADEs). Researchers examined medication dispensing errors and potential ADEs before and after implementing bar code technology in the hospital pharmacy of an academic medical center. The bar code-assisting dispensing system was implemented in three configurations. In two configurations, all doses were scanned once during the dispensing process. In the third configuration, only one dose was scanned if several doses of the same medication were being dispensed.

In the pre- and post-bar code implementation periods, the researchers observed 115,164 and 253,984 dispensed medication doses, respectively. Overall, the rates of target potential ADEs (dispensing errors that often harm patients, such as wrong medication, wrong dose, or wrong formulation errors) and all potential ADEs decreased by 74 and 63 percent, respectively. The two configurations that required staff to scan all doses had a 93 to 96 percent relative reduction in incidence of target dispensing errors and 86 to 97 percent relative reduction in the incidence of potential ADEs. However, the configuration that did not require scanning of every dose had only a 60 percent relative reduction in target dispensing errors and a 2.4-fold increased incidence of target potential ADEs.

Poulose, B.K., Speroff, T., and Holzman, M.D. (2007, January). "Optimizing choledocholithiasis management: A cost-effectiveness analysis." (AHRQ grant HS13833). Archives of Surgery 142, pp. 43-48.

Endoscopic retrograde cholangiopancreatography (ERCP, radiographic examination of the bile ducts and pancreas) is both less costly and more effective than laparoscopic common bile duct exploration (LCBDE) for gall stone management, concludes this study. Factors important to choosing the best strategy for gall stone (choledocholithiasis, CDL) management included the cost of a potential case lost due to LCBDE performance and cost of ERCP hospitalization, note the authors. They performed a cost-effectiveness analysis to compare ERCP with LCBDE using a decision model. The base case patient evaluated was a woman 18 years of age or older with symptomatic cholelithiasis (gall stones in the gallbladder or bile duct) and incidental CDL discovered at the time of intraoperative cholangiogram.

Rask, K., Hawley, J., Davis, A., and others (2006, September). "Impact of a statewide reporting system on medication error reduction." (AHRQ grant HS11918). Journal of Patient Safety 2(3), pp. 116-123.

A hospital patient safety program, established in Georgia in 2001, includes the Safe Medication Use (SMU) program, a voluntary annual self-improvement program in acute care hospitals. Researchers analyzed survey responses of hospitals in 2001, 2002, and 2003 about their program participation, evidence of reduced medication errors, and program effectiveness. More than 90 percent of eligible hospitals in the State participated in the SMU program each year. Omitting a medication dose was the most common type of error cited by the hospitals. Human factors, frequent interruptions, and communications issues were cited as the most common contributors to medication errors. Most hospitals relied on incident reports to identify errors. However, a small but growing number of hospitals had begun using automated or computer-generated reports. Most hospitals reduced their medication errors after 2001, with a mean error reduction of 28 percent in 2002 and 34 percent in 2003. Improvements were seen across all types of hospitals—urban, rural, large, small, or academic. Overall participation in the SMU program was the only significant predictor of reduction in medication errors.

Schootman, M., Fuortes, L., and Aft, R. (2006, September). "Prognosis of metachronous contralateral breast cancer according to stage at diagnosis: The importance of early detection." (AHRQ grant HS14095). Breast Cancer Research and Treatment 99, pp. 91-95.

A new study reveals that early detection of primary cancer in a second breast following a first primary breast cancer (FPBC) also greatly affects survival. Researchers analyzed 1990-2000 data from the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) Program. The researchers examined women at risk for developing metachronous contralateral breast cancer (MCBC), that is, those diagnosed with stage 0-III FPBC during 1990-2000. Among the 170,453 women who had been diagnosed with stage 0-III FPBC, 3,243 women developed a MCBC during the study period. Of the 2,904 women whose stage of MCBC was known, 329 died from breast cancer, 194 died from other causes, and 2,381 were still alive by December 2000, the end of the study period. The 70 percent of women diagnosed with stage 0-I MCBC had 81 percent better survival than women diagnosed with stage II-IV MCBC.

Sharek, P.J., Horbar, J.D., Mason, W., and others (2006, October). "Adverse events in the neonatal intensive care unit: Development, testing, and findings of an NICU-focused trigger tool to identify harm in North American NICUs." (AHRQ grants HS13698 and HS11583). Pediatrics 118(4), pp. 1332-1340.

Researchers tested a neonatal intensive care unit (NICU) trigger tool by reviewing 749 randomly selected patient charts from 15 NICUs. This revealed 2,218 triggers or 2.96 per patient, and 554 adverse events (AEs) or 0.74 per patient. Forty percent of the AEs were serious enough to warrant prolonged hospitalization, and 23 percent resulted in permanent harm, including death. The most common AEs identified were hospital-acquired infections (27.8 percent), catheter infiltrates (15.5 percent), and abnormal cranial imaging (10.5 percent). AE rates were higher for infants less than 28 weeks gestation and who weighed less than 3.3 lbs at birth. Over half (56 percent) of all AEs were deemed preventable; 16 percent could have been identified earlier, and 6 percent could have been mitigated more effectively. Only 8 percent of AEs were identified in existing hospital-based occurrence reports, and only 6 percent of the identified AEs did not have a trigger associated with them.

Singh, H., Thomas, E.J., Khan, M.M., and others (2007, February). "Identifying diagnostic errors in primary care using an electronic screening algorithm." (AHRQ grant HS11544). Archives of Internal Medicine 167, pp. 302-308.

Diagnostic errors are the leading basis for malpractice claims in primary care. However, these errors are underidentified and understudied. Electronic screening has the potential to identify records that may contain diagnostic errors in primary care, according to this study. The researchers used an algorithm to screen the electronic medical records of patients at a single hospital using a Structured Query Language-based program to screen for a primary care visit followed by a hospitalization in the next 10 days (screen 1) or a primary care visit followed by one or more primary care, urgent care, or emergency department visits within 10 days (screen 2). Among screen 1 and 2 positive visits, 16.1 and 9.4 percent, respectively, were associated with a diagnostic error. The most common diagnostic errors were failure or delay in eliciting information and misinterpretation or suboptimal weighing of critical pieces of data from the patient's medical history and physical examination.

Sloane, P.D., MacFarquhar, J.K., Sickbert-Bennett, E., and others (2006, July). "Syndromic surveillance for emerging infections in office practice using billing data." (AHRQ grant HS13521). Annals of Family Medicine 4(4), pp. 351-358.

A pilot study indicates that primary care practices may be able to provide more timely surveillance of emerging infections such as West Nile virus and bird flu at low cost and minimal staff. Researchers compared billing data on infection-related syndromes from medical visits at one rural North Carolina family medicine office with emergency department (ED) records in the same rural area over a 1-year period. The most frequently recorded syndromes were respiratory illness, gastrointestinal illness, and fever. Syndromes that most commonly exceeded the threshold of two standard deviations for the practice were lymphadenitis (inflammation of one or more lymph nodes), rash, and fever. During the 2003-2004 influenza season, the trend line patterns of the ED visits reflected a pattern consistent with that of the State. However, the trend line in primary care practice cases was less consistent, reflecting the variation expected in data from a single clinic. Still, spikes of activity that occurred in the practice before the ED suggest that the practice may have seen patients with influenza earlier.

Slutsky, J.R. (2007, January). "Moving closer to a rapid-learning health care system." Health Affairs 26(2), pp. w122-w124.

Establishing a culture of learning, while providing care, will take collaboration among the participants in the U.S. health system, asserts the author of this paper. She discusses activities that are necessary for developing a rapid-learning health system, such as recognition of the central role that patients play. Understanding the trade-offs of using a less controlled form of research to inform health care decision making and making necessary investments in methodology and translation will help secure the success of continuous-learning research. Major public policy interest in promoting health information technology and in getting more value for health care spending creates a framework for moving ahead.

Reprints (AHRQ Publication No. 07-R040) are available from the AHRQ Publications Clearinghouse.

Smetana, G.W., Landon, B.E., Bindman, A.B., and others (2007, January). "A comparison of outcomes resulting from generalist vs. specialist care for a single discrete medical condition." Archives of Internal Medicine 167, pp. 10-20.

Studies comparing patient outcomes from generalist versus specialist care for a single medical condition have methodological shortcomings, concludes this comprehensive review of studies. Of 49 studies that met inclusion criteria, only 8 studies reported physician integration into health delivery systems, 4 considered physician experience, 3 documented information technology support, and 2 considered the impact of care management programs. Selection bias was adequately addressed in 58 percent of studies that favored specialty care and 71 percent of studies that found no difference or favored generalist care. Studies that favored specialty care were less likely to consider four key, potentially confounding physician or practice characteristics, compared with studies that found no difference or favored generalist care.

Reprints (AHRQ Publication No. 07-R035) are available from the AHRQ Publications Clearinghouse.

Stommel, M., Olomu, A., Holmes-Rovner, M., and others (2006, October). "Changes in practice patterns affecting in-hospital and post-discharge survival among ACS patients." (AHRQ grants HS10531 and HS09414). BMC Health Services Research 6 (140), available online at

Researchers compared the survival of patients with acute coronary syndrome (ACS) at the same five community hospitals during three periods: 1994-1995, 1997, and 2002-2003. A quality improvement (QI) project to implement ACS care guidelines was implemented at these hospitals in 2001, 1 year prior to the last period studied. The guidelines emphasized the use of aspirin, beta-blockers and angiotensin-converting enzyme inhibitors during and after hospital discharge for ACS patients, as well as use of invasive procedures for eligible patients.

The 2003 ACS group had 84 percent less risk of hospital mortality than ACS patients in the same hospitals 10 years earlier, after controlling for demographics, disease severity, and coexisting conditions. This was accomplished with shorter hospital stays than in earlier years. Also, in-hospital mortality improvements were sustained in the first year after hospital discharge. The researchers conclude that the QI initiative substantially improved survival of ACS patients due to increased use of effective treatments.

Stroupe, K.T., Manheim, L.M., Luo, P., and others (2006, October). "Tension-free repair versus watchful waiting for men with asymptomatic or minimally symptomatic inguinal hernias: A cost-effectiveness analysis." (AHRQ grant HS09860). Journal of the American College of Surgeons 203, pp. 458-468.

Researchers conducted a cost-effectiveness analysis of a trial at six medical centers to examine outcomes of men with asymptomatic or minimally symptomatic inguinal hernias. The men were randomized to standard open tension-free repair (TFR) surgery with mesh or to watchful waiting (WW). Although WW is less costly initially, patients who delay hernia repair may have greater need for more costly care later. Thus, the researchers compared the total 2-year health care costs of 317 patients randomized to TFR and 324 patients randomized to WW. At 2 years, TFR patients had $1,831 higher average total costs than WW patients ($7,875 vs. $6,044), with 0.031 higher quality-adjusted life years (QALYs). The cost per additional QALY for TFR patients was $59,065. The probability that TFR was cost-effective at the $50,000 per QALY level (considered the cost-effectiveness standard for medical interventions) was 40 percent. Increased quality of life for those assigned to TFR, although significant, was very modest. Given the marginal cost-effectiveness of TFR, both TFR and WW are reasonable approaches from the viewpoint of cost-effectiveness, conclude the researchers.

Taylor, B.J., Robbins, J.M., Gold, J.I., and others (2006, October). "Assessing postoperative pain in neonates: A multicenter observational study." (AHRQ grant HS13698). Pediatrics 118(4), pp. 992-1000.

Researchers found that while management of postoperative neonatal pain is well accepted, the practice is highly variable, with deficiencies in the assessment and management of postoperative pain in neonates treated at NICUs in 10 hospitals. The neonates studied had a birthweight of 2.4 kg and gestation age of 36 weeks. Participating hospitals used 7 different numeric pain scales; nursing pain assessments were documented for 88 percent of the infants; and physician pain assessments were documented for 9 percent of the infants. Opioids (84 vs. 60 percent) and benzodiazepines (24 vs. 11 percent) were used more commonly after major surgery than minor surgery. Also, a small proportion of infants (7 percent major surgery, 12 percent minor surgery) received no analgesia. Physician pain assessment (not postnatal age or surgery type) was the only significant predictor of postsurgical analgesic use.

Wasson, J.H., Ahles, T., Johnson, D., and others (2006, July). "Resource planning for patient-centered collaborative care." (AHRQ grant HS10264). Journal of Ambulatory Care Management 29(3), pp. 207-214.

Researchers suggest three strategies that a practice can use to better customize care management of patients with chronic conditions. The strategy chosen is based on the patient's condition, psychosocial problems, confidence in self-care, and financial status. A low-intensity self-care strategy consists of standard assessment, feedback to the physician from patients so that they are on the "same page," and tailored information from the physician for patients. Most patients who are relatively good at self-care management benefit from this approach. A small percentage of patients need a second strategy that adds an average of three telephone calls from a nurse, who coaches patients in problem solving related to their conditions. However, the telephone approach alone is not sufficient to overcome problems of patients with either low self-confidence for self care, or low financial status and the added burden of pain and psychosocial problems. A third strategy would provide these patients with the addition of a problem-solving coach. This strategy would focus on health literacy and remedying social needs.

Wasson, J.H., Johnson, D.J., Benjamin, R., and others (2006, July). "Patients report positive impacts of collaborative care." (AHRQ grant HS10264). Journal of Ambulatory Care Management 29(3), pp. 199-206.

A survey of 24,609 adult Americans who had common chronic diseases or significant dysfunction revealed that only 1 out of 5 (21 percent) received good collaborative care (both physician information and patient confidence). A total of 36 percent obtained fair collaborative care (either physician information or patient confidence), and 43 percent experienced poor collaborative care (neither one). Good collaborative care was associated with better control of blood pressure, blood glucose level, serum cholesterol level, and treatment effectiveness for pain and emotional problems. For example, 31 percent of patients with diabetes who received good collaborative care reported their blood glucose levels were always in the range of 80-150 (normal) compared to 20 percent of patients receiving fair and 14 percent receiving poor collaborative care. Also, 35 percent of patients receiving good collaborative care had treatments that made pain much better compared with 25 percent of patients receiving fair and 10 percent receiving poor collaborative care. One-third (35 percent) of patients receiving good collaborative care said past treatment had made emotional problems much better compared with 23 percent receiving fair and 13 percent receiving poor collaborative care.

Yang, Z., Olomu, A., Corser, W., and others (2006, October). "Outpatient medication use and health outcomes in post-acute coronary syndrome patients." (AHRQ grants HS10531 and HS09414). The American Journal of Managed Care 12(10), pp. 581-587.

Medications that are recommended to prevent health problems among patients with acute coronary syndrome (ACS) include angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin receptor blockers (ARBs), beta-blockers, lipid-lowering medications, and aspirin. Researchers surveyed 433 patients by telephone on their medication use after hospital discharge 8 months after hospitalization for ACS at 5 Michigan hospitals. Taking at least one type of beta-blocker or ACEI/ARB within 3 months of hospital discharge significantly reduced the probability of hospital readmission 3 months after discharge. Once patients were readmitted during this critical period, they were more likely to be admitted to the hospital a second time or more. After hospitalizations, ACS patients may add, switch, or drop their discharge medications due to transfer of care to another doctor, change in health condition, medication side effects, or changes in insurance coverage. Thus, outpatient physicians could be as important as the hospital physicians who write the discharge prescriptions in promoting use of effective medications for ACS patients, note the researchers. They found that most changes to medication regimens occurred within 3 months after discharge, with fewer changes in the subsequent 5 months.

Zhang, L., Kao, W.H., Berthier-Schaad, Y., and others (2006). "Haplotype of signal transducer and activator of transcription 3 gene predicts cardiovascular disease in dialysis patients." (AHRQ grant HS08365). Journal of the American Society of Nephrology 17, pp. 2285-2292.

Signal transducer and activator of transcription 3 (STAT3) protein has been linked to cardiovascular disease (CVD) through multiple pathways in experimental and animal studies. These researchers examined STAT3 gene variation as a predictor of incident CVD in a group of 529 white dialysis patients. They genotyped 15 single-nucleotide polymorphisms of the STAT3 gene. Compared with common haplotype C-1, C-3 was associated with twice the risk for CVD events. Associations were independent of inflammation markers, interleukin-6 (IL-6), and C-reactive protein (CRP). However, IL-6 levels were 14 percent lower per copy of haplotype A-3 compared with haplotype A-1 in block A after adjustment for CRP and other risk factors. Variation in the STAT3 gene is associated with the risk for CVD among white dialysis patients independent of serum IL-6 and CRP levels, conclude the authors.

Return to Contents

AHRQ Publication No. 07-0042
Current as of June 2007


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


AHRQ Advancing Excellence in Health Care