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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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

Ammar, K.A., Makwana, R., Jacobsen, S.J., and others (2007, January). "Impaired functional status and echocardiographic abnormalities signifying global dysfunction enhance the prognostic significance of previously unrecognized myocardial infarction detected by electrocardiography." (AHRQ grant HS10239). Annals of Noninvasive Electrocardiology 12(1), pp. 27-37.

Unrecognized myocardial infarction (UMI, heart tissue damage) is diagnosed by electrocardiogram (ECG) in persons without clinically recognized MI (RMI). Common clinical practice discounts ECG-UMI as a false positive, if echocardiography does not demonstrate regional wall motion abnormalities (RWMA). However, a new study found that ECG-UMI patients had an increased risk of mortality independent of RWMA. Patients with ECG-UMI alone had a 3.7 times higher risk of dying, but 7.2 times higher risk of dying with UMI and abnormal functional status, and 9.5 times higher risk of dying with ECG-UMI, abnormal functional status, and any echocardiographic abnormality, not just RWMA. ECG-UMI was not associated with increased mortality if the echocardiogram was completely normal.

These findings were based on a population-based random sample of 2,042 residents of 1 county who were 45 years and older. The researchers examined questionnaire responses, reviewed medical charts, ECGs, echocardiograms, scores on a functional activity scale, and 5-year mortality rates. They conclude that patients with UMI on an ECG should be further evaluated for functional status and any echocardiographic abnormality.

Barry, L.C., Lichtman, J.H., Spertus, J.A., and others (2007, March/April). "Patient satisfaction with treatment after acute myocardial infarction: Role of psychosocial factors." (AHRQ grant HS11282). Psychosomatic Medicine 69, pp. 115-123.

Researchers found that assessing a patient's level of social support, inclination to be optimistic, and depression severity before hospital discharge may indicate which patients are likely to be more satisfied with posthospital cardiac care 1 month after a heart attack. The research team examined the association of psychosocial variables (social support, disposition to optimism, and depression) in 1,847 heart attack patients with their responses to the Treatment Satisfaction Scale of the Seattle Angina Questionnaire a month later.

The patients were predominantly white men in their sixties, who participated in a multicenter study evaluating heart attack events and recovery. Satisfaction with posthospitalization treatment following heart attack increased as social support and disposition to optimism increased. Depressed participants were significantly less satisfied with posthospital care than nondepressed participants. The authors conclude that an available supportive network of people who can take on tasks such as daily chores, and/or be available to talk with, may decrease the patient's distress and ultimately translate into better treatment satisfaction.

Carney, P.A., Yi, J.P., Abraham, L.A., and others (2007, February). "Reactions to uncertainty and the accuracy of diagnostic mammography." (AHRQ grant HS10591). Journal of General Internal Medicine 22, pp. 234-241.

Radiologists who have more discomfort with the uncertainty inherent in clinical medicine are more likely to recall women for additional evaluations after interpreting a recent screening mammogram for a potential abnormality, according to a new study. The researchers surveyed 132 radiologists from 3 States about their reactions to uncertainty. About half of the radiologists (52 percent) reported a prior medical malpractice lawsuit, but only 14 percent reported a previous mammography-related lawsuit.

Radiologists with more years interpreting mammography and higher mammography volume had slightly lower uncertainty scores. Radiologists reporting any prior medico-legal experience had slightly higher uncertainty scores (30.4 vs. 28.7 out of 50), although this was not significant. A total of 131,482 diagnostic mammograms were included in the analysis. Radiologists less comfortable with uncertainty were more likely to interpret abnormalities as cancer that was later diagnosed (higher sensitivity), but were also more likely to interpret cancer for abnormalities later diagnosed as noncancerous (lower specificity).

Clancy, D.E., Huang, P., Okonofua, E., and others (2007, March). "Group visits: Promoting adherence to diabetes guidelines." (AHRQ grant HS10871). Journal of General Internal Medicine 22, pp. 620-624.

Group visits may improve the quality of care for patients with type 2 diabetes because they have the advantage of being longer, more frequent, more organized, and more educational. Researchers evaluated the effect of group visits on clinical outcomes and adherence to American Diabetes Association (ADA) guidelines and cancer screening guidelines among 186 adults at the Medical University of South Carolina primary care center.

The patients had poorly controlled diabetes (HbA1c greater than 8 percent). Patients were randomized either to group visits or to traditional physician-patient visits at the primary care center. A primary care doctor and registered nurse led the group visits of 14 to 17 patients, which met monthly for 2 hours for 1 year. The sessions included socialization, interactive discussion of topics such as foot care or healthy eating strategies, and 60 minutes for one-on-one consultations with the physician. One-on-one visits were available for care needed between group visits or for care not amenable to group visits, such as Pap smears or mammograms. By 1 year, group patients were more likely to have at least eight ADA process of care indicators performed compared with the patients who received usual care.

Clark, D.E., Lucas, F.L., and Ryan, L.M. (2007, March). "Predicting hospital mortality, length of stay, and transfer to long-term care for injured patients." (AHRQ grant HS15656). Journal of Trauma 62(3), pp. 592-600.

A recent time-phased model predicted an overall 56 percent hospital mortality rate and 8.5 day hospital stay, along with a 38 percent probability of discharge to long-term care (LTC) among traumatically injured patients. Prediction of length of stay (LOS) and other outcomes from the model was reasonably good for most patient subgroups, and explained about 17 percent of individual variation in LOS.

Researchers analyzed National Trauma Data Bank records of 369,829 injured patients hospitalized in trauma centers from 1999 to 2003 to develop a multistate model divided into 4 time periods. The rate of death among hospitalized patients was highest on day 1, decreased during days 2 to 5, and gradually increased thereafter. The rate of discharge to LTC rose until days 6 to 11, and then gradually decreased. The rate of discharge home decreased steadily after the first day. Penetrating and vehicle trauma were associated with higher mortality on the first day, but lower mortality thereafter. Increased injury severity and coma at admission were associated with a lower probability of transfer to LTC for the first 11 days, but a higher probability thereafter. Increased age or female sex were associated with decreased rates of discharge home, but increased rate of discharge to an LTC facility.

Curtis, J.R., Patkar, N., Xie, A., and others (2007, April). "Risk of serious bacterial infections among rheumatoid arthritis patients exposed to tumor necrosis factor a antagonists." (AHRQ grant HS10389). Arthritis & Rheumatism 56(4), pp. 1125-1133.

A new study reveals that patients prescribed tumor necrosis factor a (TNFa) antagonists are two to four times more likely to be hospitalized for a serious bacterial infection than those taking the more traditional rheumatoid arthritis (RA) drug, methotrexate (MTX). Researchers retrospectively studied U.S. RA patients enrolled in a large health care organization, who were taking either TNFa antagonists or MTX. They examined hospitalizations with possible bacterial infections over a median of 17 months, which infectious disease specialists confirmed by reviewing medical records.

A total of 187 suspected bacterial infections were identified from medical records of the 2,393 patients in the TNFa antagonist group and the 2,933 patients in the MTX group. During the study period, 2.7 percent of TNFa antagonist patients compared with 2 percent of MTX patients were hospitalized with a serious bacterial infection. However, TNFa antagonist patients had a fourfold higher rate of hospitalization for bacterial infection within the first 6 months of starting TNFa antagonist therapy (2.9 versus 1.4 infections per 100 person-years, after adjusting for other factors).

Dillard, D., Jacobsen, C., Ramsey, S., and Manson, S. (2007, February). "Conduct disorder, war zone stress, and war-related posttraumatic stress disorder symptoms in American Indian Vietnam veterans." (AHRQ grant HS10854). Journal of Traumatic Stress 20(1), pp. 53-62.

Childhood conduct disorder (CD) may underlie greater posttraumatic stress disorder (PTSD) among American Indian (AI) Vietnam War veterans, concludes a new study. Researchers interviewed 591 men, who participated in the American Indian Vietnam Veterans Project, to examine factors related to PTSD in AIs. They used standard diagnostic tests to assess CD, PTSD, depression, anxiety disorder, and other mental health problems. CD was diagnosed with the presence of three or more symptoms before age 15, such as threatening or assaulting persons, being cruel to animals, willfully destroying property, or running away from home several times.

AI veterans with CD had more war-related PTSD symptoms than those without CD (mean symptom score of 99.4 vs. 90.6), even after adjusting for level of war zone stress, premilitary traumatic experiences, and other factors potentially affecting PTSD. Both groups had similar age, education, employment status, and mental health before, at entry, and during the military, except that AI veterans with CD were more likely to be alcohol and/or drug dependent prior to military service.

Editor's Note: Another AHRQ-supported study (HS10854) by the same researchers found that American Indians are more likely to participate in community-based research, if it is being conducted by a tribal college/university or national organization, an American Indian is leading the study, or the study is addressing health problems of concern to the community. More details are in: Noe, T.D., Manson, S.M., Croy, C., and others (2007, Winter). "The influence of community-based participatory research principles on the likelihood of participation in health research in American Indian communities." Ethnicity & Disease 17 (Suppl.1), pp. S6-S14.

Graham, J., Bennett, I.M., Holmes, W.C., and Gross, R. (2007, May). "Medication beliefs as mediators of the health literacy-antiretroviral adherence relationship in HIV-infected individuals." (AHRQ grant HS10399). AIDS and Behavior 11, pp. 385-392.

Low literacy and mistaken beliefs about HIV medication affect adherence to antiretroviral drug regimens, according to a new study. Researchers found that 64 percent of those with at least a ninth-grade reading level complied with their complex, multidrug highly active antiretroviral therapy (HAART) regimen compared with 40 percent of those with less than a ninth-grade reading level.

The study also suggests that erroneous beliefs about the use of HIV medications and their side effects may contribute to lower adherence rates among individuals with low literacy. Investigators at the University of Pennsylvania Center for Education and Research on Therapeutics and colleagues studied 87 predominantly black, low-income, HIV-infected men on HAART, who were cared for in 1 of the university's 2 HIV clinics in 2003.

Patients completed a questionnaire about HAART medication beliefs. Researchers used pharmacy refills to measure medication adherence to a single index drug over the prior 3 months. Participants with 95 percent or greater adherence were significantly more likely to have undetectable viral loads than those with less than 95 percent adherence (73 vs. 45 percent). About 74 percent of participants had at least one mistaken belief about the medication.

Grossman, J.M., Gerland, A., Reed, M.C., and Fahlman, C. (2007, April). "Physicians' experiences using commercial e-prescribing systems." (AHRQ Contract No. 290-05-0007). Health Affairs 26(3), w393-w404.

This study indicates that substantial gaps may exist between advocates' vision of e-prescribing (electronic prescribing) systems and how physicians use them in practice. The findings were based on interviews with administrators and physicians at 15 e-prescribing practices and 6 practices without e-prescribing, as well as a few health plan, vendor, and pharmacy representatives in 2005 and 2006. All but one of the practices' e-prescribing systems offered some clinical decision support in the form of drug-drug interaction alerts. However, access to more advanced decision support was limited. Only half of practices reported being able to check for drug-allergy interactions, and only 20 percent could check for drug-condition contraindications. Finally, physicians in slightly more than half of the practices did not have electronic access to formulary data (indicating the specific drugs that are covered by health plans) when they wrote prescriptions. Either the systems did not have the feature or the practice had chosen not to enable it. Also, e-prescribing systems sometimes omitted the formularies of major health insurers, including Medicaid.

Gupta, R., Plantinga, L.C., Fink, N.E., and others (2007, April). "Statin use and hospitalization for sepsis in patients with chronic kidney disease." (AHRQ grant HS08365). Journal of the American Medical Association 297(13), pp. 1455-1464.

Bloodstream infection (sepsis) is a major cause of problems and death in patients with chronic kidney disease who are on dialysis. However, when these patients use statins, they are less likely to be hospitalized for sepsis, according to this new study. The ability of statins to modulate the immune system may play a protective role, suggest the researchers. They prospectively studied 1,041 dialysis patients at 81 U.S. not-for-profit outpatient dialysis clinics from 1995 to 1998, with followup in January 2005. They compared rates of hospitalization for sepsis between statin users and nonstatin users, after adjusting for other factors. Overall, there were 303 hospitalizations for sepsis.

Rates of sepsis-related hospitalizations were significantly lower in patients receiving statins than in those not receiving statins (41 vs. 110/1,000 patient-years). With adjustment for demographics and dialysis modality, statin users were 59 percent less likely to be subsequently hospitalized for sepsis. After further adjustment for coexisting medical conditions and laboratory test results, statin users were 62 percent less likely to be hospitalized for sepsis.

Han, H.-R., Kang, J., Kim, K.B., and others (2007). "Barriers to and strategies for recruiting Korean Americans for community-partnered health promotion research." (AHRQ grants HS13160 and HS13779). Journal of Immigrant Health 9, pp. 137-146.

Cultural, language, and community barriers make it difficult to recruit Korean Americans into health promotion research studies, conclude Johns Hopkins University researchers. Researchers analyzed barriers and facilitators to recruiting Korean Americans for 14 studies conducted between 1998 and 2005 that addressed prevalent problems of Korean Americans, such as high blood pressure. Aside from language barriers, the patriarchal Korean culture forbids women from participating in certain social activities, including research studies, or seeking preventive health care on their own. Many middle-aged Koreans work long hours and have little time to participate in research. Community barriers include low health care coverage (34 percent of this group lack insurance), lack of awareness of research studies, suspicions about consent procedures, and concern about privacy and confidentiality issues. Community gatekeepers, such as Korean physicians, church leaders, and grocery managers, can play a critical role in connecting researchers to potential participants, note the researchers.

Harris, C.B., Krauss, M.J., Coopersmith, C.M., and others (2007, April). "Patient safety event reporting in critical care: A study of three intensive care units." (AHRQ grant HS11898). Critical Care Medicine 35(4), pp. 1068-1076.

Reporting of intensive care unit (ICU) events that have the potential to jeopardize patient safety doubled when a new voluntary card-reporting system was implemented, according to a new study. ICU events reported on SAFE (Safety, Action, Focus, Everyone) 5x8 inch two-sided cards included a combination of risky situations, near misses, and no-harm events. During a 14-month period, nurses, physicians, and other staff at 3 ICUs reported 714 patient safety events using the new card-based reporting system. They reported 41.7 events per 1,000 patient days compared with 20.4 events per 1,000 patient days with the previous online Web-based reporting system.

This twofold rise in reporting using SAFE may be due to the simplicity, brevity, and ease of using the card system. Physicians were more likely than other ICU staff to boost their reporting of patient safety events with the new card system. Their reporting of such events increased 43-fold compared with 1.7-fold for nurses and 4.3-fold for other staff, relative to the prior period with Web-based reporting. Overall reporting of safety events varied by type of ICU and health care worker.

Labarere, J., Stone, R.A., Obrosky, D.S., and others (2007, February). "Comparison of outcomes for low-risk outpatients and inpatients with pneumonia: A propensity-adjusted analysis." (AHRQ grant HS10049). Chest 131(2), pp. 480-488.

Physician judgment plays an important role in deciding which low-risk patients with pneumonia are hospitalized, this study concludes. Clinical guidelines generally recommend outpatient treatment for patients who are defined as low-risk based on the pneumonia severity index (PSI). Yet, based on their judgment, physicians sometimes hospitalize low-risk patients with community-acquired pneumonia (CAP).

Researchers used data from a randomized trial conducted in 32 emergency departments to compare 30-day mortality rates, time to return to work and usual activities, and patient satisfaction with care between 944 CAP outpatients and 549 inpatients in PSI low-risk categories (I to III). After adjusting for likelihood of receiving outpatient or inpatient treatment, which eliminated significant differences in baseline characteristics, outpatients were twice as likely to return to work and nonworkers were 40 percent more likely to return to usual activities than inpatients. The overall mortality rate was higher for inpatients than outpatients (2.6 vs. 0.1 percent). Satisfaction with the site-of-treatment decision, with emergency department care, and with overall medical care was no different between outpatients and inpatients.

Leverence, R.R., Williams, R.L., Sussman, A., and others (2007, April). "Obesity counseling and guidelines in primary care: A qualitative study." (AHRQ grant HS13496). American Journal of Preventive Medicine 32(4), pp. 334-339.

Clinicians view their efforts to treat obesity as ineffective in light of family, cultural, social, and community barriers, suggests this new study. Researchers conducted in-depth interviews and 2 focus groups with 20 primary care clinicians caring for predominantly low-income minority patients. Clinicians were frustrated with lack of resources to help their patients lose weight, such as psychologists, health educators, dieticians, and exercise physiologists. They also cited lack of family concern about a child's weight problem or cultural differences in what is considered acceptable weight. Clinicians cited the importance of family-based interventions and addressing the problem of obesity at the community and social level. Patients often mention that their neighborhood is not safe when the clinician recommends 15 minutes of walking. Thus, clinicians had difficulty following obesity guidelines, which recommend diet and exercise counseling, and chose instead to apply their efforts and time to preventive or medical care that they perceived themselves to be more effective at delivering.

Payne, C.H., Smith, C.R., Newkirk, L.E., and Hicks, R.W. (2007, April). "Pediatric medication errors in the postanesthesia care unit: Analysis of MEDMARX data." (AHRQ grant HS10397). AORN Journal 85(4), pp. 731-740.

This new study reveals that medication errors involving infants and children in the postanesthesia care unit (PACU) occur once in 20 medication orders. In addition, these errors are more likely to cause harm than medication errors in the overall population. Researchers analyzed the MEDMARX medication error reporting system for pediatric medication errors originating in the PACU during a 6-year period from 1998 to 2004.

During the study period, 42 hospitals reported 59 medication errors involving children in the PACU, of which 12 (20 percent) were harmful. During the same time period, 354 institutions reported 2,406 overall medication errors in the PACU, of which 6 percent were harmful. The five leading causes of medication errors in the pediatric and overall population were performance deficit, procedure/protocol not followed, communication problems, insufficient knowledge, and calculation error. Medication errors involving calculations, decimal points, and confusion about dosage were reported in higher percentages in the pediatric than overall population.

Powell, C.K., Hill, E.G., and Clancy, D.E. (2007, January). "The relationship between health literacy and diabetes knowledge and readiness to take health actions." (AHRQ grants HS13851 and HS10871). The Diabetes Educator 33(1), pp. 144-151.

Low health literacy is significantly associated with worse glycemic (HbA1c) control and poorer disease knowledge in patients with type 2 diabetes. However, researchers found that it does not seem related to their readiness to take action to manage their disease. The researchers examined demographic and clinical data as well as health care provider visits among 68 adults with type 2 diabetes at a general internal medicine clinic. The researchers also administered the Rapid Estimate of Adult Literacy in Medicine (REALM) literacy instrument; a Diabetes Knowledge Test (DKT); and a Diabetes Health Belief Model (DHBM) scale. The patients were predominantly black middle-aged women with an HbA1c level of 8.25 percent.

Participants with lower health literacy scores had lower DKT scores and had HbA1c levels 1.21 to 1.36 percent higher than those with a REALM literacy level greater than or equal to the ninth grade. However, patients with the lowest literacy levels scored only 2.84 points lower on the DHBM scale than those with the highest literacy levels. This implies that patients with lower literacy are still willing to take action to manage their diabetes.

Clancy, D.E., Yeager, D.E., Huang, P., and Magruder, K.M. (2007, March). "Further evaluating the acceptability of group visits in an uninsured or inadequately insured patient population with uncontrolled type 2 diabetes." (AHRQ grant HS10871). The Diabetes Educator 33(2), pp. 309-314.

Disadvantaged patients with uncontrolled type 2 diabetes find group visits an acceptable way to receive diabetes care, concludes this study. The researchers surveyed a group of predominantly black, poor, or uninsured rural women with poorly controlled type 2 diabetes (HbA1c of 8 percent or more) at one university-affiliated primary care clinic. The researchers randomly assigned 96 patients to group visits and 90 patients to usual care (individual visits with the physician), and surveyed them at baseline, 6 months, and 12 months later.

Group visits were co-led by one of six primary care internal medicine doctors and one of three registered nurses in the clinic. Group visits lasted for two hours. There were no significant differences between the groups at baseline. However, the patients attending group visits felt that their physicians were more knowledgeable about their community and were more culturally competent. Group patients also assessed their care better than the usual care patients, and had attendance rates at least as good as the general clinic population.

Raebel, M.A., McClure, D.L., Simon, S.R., and others (2007, January). "Laboratory monitoring of potassium and creatinine in ambulatory patients receiving angiotensin converting enzyme inhibitors and angiotensin receptor blockers." (AHRQ grant HS11843). Pharmacoepidemiology and Drug Safety 16, pp. 55-64.

This study concludes that nearly one-third of patients dispensed angiotensin converting enzyme inhibitors (ACEIs) or angiotensin receptor blockers (ARBs) do not undergo laboratory monitoring at least once a year. Further, nearly half (46.3 percent) of patients with one test to evaluate serum potassium and creatinine levels did not have a second test. Therefore, the clinical stability of the serum potassium and creatinine levels could not be assessed.

The researchers used administrative data and medical records to assess serum potassium and creatinine monitoring of 52,906 adults in 10 HMOs, who were dispensed ACEIs or ARBs for at least 1 year. Most patients were dispensed an ACEI (89.4 percent), with 7.8 percent dispensed an ARB. Also, 2.8 percent of patients were either codispensed an ACEI plus ARB or were switched from one drug class to the other during the study period. About two-thirds (68.4 percent) of patients received laboratory monitoring. The likelihood of monitoring increased with age; making more than 9 outpatient visits; hospitalization; concomitant therapy with potassium supplements, diuretics, or digoxin; and diagnosis of chronic kidney disease, diabetes, or heart failure.

Scott-Cawiezell, J., Pepper, G.A., Madsen, R.W., and others (2007, February). "Nursing home error and level of staff credentials." (AHRQ grant HS14281). Clinical Nursing Research 16(1), pp. 72-78.

Certified medication technicians or aides (CMT/As) are no more likely to make medication errors than licensed practical nurses (LPNs) or registered nurses (RNs), according to this new study. Medication administration was observed by type of staff over 3 to 4 days at each of five different types and sizes of Missouri nursing homes, both across shifts and among various medication administrators. During the study period, 3,194 doses of medications were ordered to be given, including 3,101 doses observed and 93 omitted doses. Baseline observations involved 8 RNs, 12 LPNs, and 19 CMT/As. The RNs and LPNs more often administered specialized, complex medications such as insulin. CMT/As administered 61.3 percent of medication doses, whereas RNs administered the fewest doses (15.3 percent). RNs had the most interruptions (39.9 percent), which halted medication administration, whereas LPNs had the highest percentage of distractions (41.6 percent), which did not halt medication administration. Although there were no significant differences in medication error by level of credential, RNs had the highest proportion of error without wrong time error included.

Stewart, A.L., Dean, M.L., Gregorich, S.E., and others (2007, March). "Race/ethnicity, socioeconomic status and the health of pregnant women." (AHRQ grant HS10856). Journal of Health Psychology 12(2), pp. 285-300.

A woman's race, education, income, and social status all interact to affect her health during pregnancy, concludes a new study. As part of the Project WISH (Women and Infants Starting Healthy), a research team studied 1,802 ethnically diverse women receiving prenatal care at 6 San Francisco Bay area delivery sites. >The women were fairly healthy with low depression scores. Differences by race/ethnicity were pronounced, with whites and Asians/Pacific Islanders doing better on all measures. Higher percentages of Latinas and black women were in the lower economic and educational strata than whites and Asians/Pacific Islanders.

Although women in all three minority groups reported higher levels of depression and lower self-rated health than white women, only Latinas and black women reported worse physical functioning. After adding socioeconomic status (SES) variables, racial disparities in depression remained for all minority groups, and disparities in self-rated health remained for Asians/Pacific Islanders.

In contrast, disparities in self-rated health between blacks and whites and between Latinas and whites became nonsignificant when any SES measures were included. Subjective social standing was more highly correlated with education and income in whites and Asian/Pacific Islanders than in Latinas and blacks.

Stone, P.W., Mooney-Kane, C., Larson, E.L., and others (2007, June). "Nurse working conditions, organizational climate, and intent to leave in ICUs: An instrumental variable approach." (AHRQ grant HS13114). HSR: Health Services Research 42(3), pp. 1085-1104.

The authors of this study analyzed survey responses from 837 nurses employed in 39 adult ICUs from 23 hospitals located in 20 separate metropolitan areas. They also examined hospital administrative, public use, and Medicare files to investigate causes of nurse intention to leave (ITL), while simultaneously considering organizational climate (OC) in ICUs.

A total of 15 percent of ICU nurses indicated their intention to leave in the coming year, imposing potentially high hiring and training costs on hospitals. OC and the tightness of the labor market had significant roles in determining ITL. For example, improving the OC by one standard deviation was predicted to reduce nurses' probability of intent to leave their position by 13 percent. Furthermore, OC was positively affected by better regionally adjusted ICU wages, hospital profitability, and hospital teaching and Magnet status (Magnet hospitals have organizational characteristics associated with high nurse retention rates). Because higher wages did not reduce ITL, increased pay alone without attention to OC is likely insufficient to reduce ICU nurse turnover.

Stone, P.W., Mooney-Kane, C., Larson, E.L., and others (2007, June). "Nurse working conditions and patient safety outcomes." (AHRQ grant HS13114). Medical Care 45(6), pp. 571-578.

This study is the first to link national data on nosocomial (hospital-acquired) infections and other patient safety problems to nurse working conditions. The authors surveyed 1,096 nurses from 51 adult ICUs in 31 hospitals about ICU working condition factors such as staffing, overtime, wages, and hospital profitability and Magnet accreditation. They then examined the impact of nurse working conditions on several patient safety measures among the 15,846 elderly Medicare patients treated at the ICUs.

The ICUs with higher staffing had a lower incidence of central line associated bloodstream infections (CLBSIs), ventilator-associated pneumonia, 30-day mortality, and decubitus ulcers (pressure sores). Increased nurse overtime was associated with higher rates of catheter-associated urinary tract infections and decubitus ulcers, but slightly lower rates of CLBSIs. Physicians, rather than nurses, typically insert central lines, so CLBSIs are less likely to be affected by nursing care.

Nurses' wages were not associated with any of the patient safety outcomes. The effects of organizational climate and profitability were not consistent. Improving the working conditions of nurses will most likely promote patient safety, conclude the researchers.

Teplin, V., Vittinghoff, E., Lin, F., and others (2007, February). "Oophorectomy in premenopausal women: Health-related quality of life and sexual functioning." (AHRQ grant HS09478). Obstetrics & Gynecology 109(2), pp. 347-354.

This study found that women who underwent bilateral salpingo-oophorectomy (BSO), despite an initial decline in quality of life in the first 6 months after surgery, had no apparent differences in quality of life 2 years after surgery compared with women whose hysterectomy only involved removal of the uterus, not ovaries. For both groups, scores for all health-related quality-of-life outcomes improved after hysterectomy. Six months after surgery, the BSO group had less improvement than the non-BSO group in body image (2 points vs. 14 points), sleep problems (4 vs. 16), and the SF-36 Mental Component Summary (4 vs. 10). There were no differences in sexual functioning, hot flushes, urinary incontinence, or pelvic pain between the two groups. At the 2-year follow-up, both groups had similar scores on all measures of health-related quality of life and sexual functioning, irrespective of estrogen use.

Researchers analyzed data among premenopausal women who underwent hysterectomy for benign gynecologic disease and participated in one of several hysterectomy studies. They measured women's quality of life outcomes at 4 weeks, 6 months, and 2 years after hysterectomy.

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