Research Activities, September 2009
Bender, H. A., Cole, J. R., Aponte-Samalot, M., and others (2009). "Construct validity of the Neuropsychological Screening Battery for Hispanics (NeSBHIS) in a neurological sample." (AHRQ grant HS16819). Journal of the International Neuropsychological Society 15, pp. 217-224.
The authors describe testing of a cross-cultural assessment instrument, the Neuropsychological Screening Battery for Hispanics (NeSBHIS), in a sample of native Spanish-speaking immigrants with epilepsy. This Spanish-language assessment battery—which evaluates the domains of language, memory functioning, visuospatial functioning, mental control, psychomotor functioning, and reasoning—was originally normed using a standardization sample of 300 community-referred, Spanish-speaking Hispanics. The authors' primary goal was to determine whether this screening battery remains useful in a clinical sample of 127 Hispanic participants with epilepsy, recruited over a 6-year period from Hispanic patients at a large urban Comprehensive Epilepsy Center. Although the authors used a different psychomotor test than had the original developers of NeSBHIS, they found their results on the clinical sample consistent with those found by the original developers. The authors suggest that further studies be undertaken using NeSBHIS in patients with other neurological problems.
Bradford, D. W., Kim, M. M., Braxton, L. E., and others (2008, August). "Access to medical care among persons with psychotic and major affective disorders." (AHRQ grant HS13353). Psychiatric Services 59(8), pp. 847-852.
This study found that people with psychotic disorders had a 45 percent lower likelihood and those with bipolar disorder had 26 percent lower likelihood of having a primary care doctor compared with people not suffering from mental disorders. Researchers compared access and barriers to medical care among 156,475 adults reporting psychotic and mood disorders or without mental disorders, who completed the National Health Interview Survey (NHIS) and NHIS-Disability Component for 1994 and 1995. People with psychotic disorders, bipolar disorder, or major depressive disorder had 2.5 to 7 times greater odds of any barriers to care, ranging from delaying medical care because of costs to being unable to get needed medical care or a needed prescription medication. However, those with major depression were as likely to report having a regular source of care as those who did not report psychiatric disorders.
Bundy, D. G., Rinke, M. L., Shore, A. D., and others (2008, September). "Medication errors in the ambulatory treatment of pediatric attention deficit hyperactivity disorder." (AHRQ grant HS16774). Joint Commission Journal on Quality and Patient Safety 34(9), pp. 552-560.
The incidence of medication error related to the treatment of attention deficit hyperactivity disorder (ADHD) is significant, concludes this study. Researchers searched the U.S. Pharmacopeia MEDMARX database for reports involving medications used in the outpatient treatment of ADHD in children between 2003 and 2005. Of 361 error reports, 329 involved medications used only in the treatment of ADHD and 32 involved medications used for ADHD as well as other conditions. Among first-listed generic medications, methylphenidate (MPH) and its derivatives (43 percent) and dextroamphetamine, alone and combined with amphetamine salts (41 percent), accounted for more than four out of five error reports. Improper dose, wrong dosage form, and prescribing errors were the 3 most common errors listed in the 361 reports. Improper dose was a significantly more common error with MPH. Wrong dosage form was the second most common error type. This is particularly likely when multiple formulations of the same medication have names that sound or look similar.
Clancy, C. M. (2009, May/June). "More work is needed to protect medical residents from fatigue and potential errors, IOM report finds." American Journal of Medical Quality 24(3), pp. 259-261. Reprints (AHRQ Publication No. 09-R065) are available from the AHRQ Publications Clearinghouse.
In this commentary, Carolyn Clancy, M.D., director of the Agency for Healthcare Research and Quality (AHRQ), outlines the issue of medical errors among medical residents due to sleep deprivation as well as regulations put into place to address this problem. She summarizes the findings of a recent Institute of Medicine (IOM) report, funded by AHRQ, on acute and chronic fatigue among medical residents. The report found that the work limits were not being followed, thereby continuing to put patients at risk. The commentary reviews the reactions of clinical educators and lists the new recommendations from the IOM panel. The author notes the usefulness of evidence-based tools from AHRQ that hospitals can employ to assess and improve their patient safety culture and teamwork. Balancing the issues of sleep deprivation and the demanding training requirement of certain specialties will require shifts in organizational and individual thinking, she concludes.
Daniels, T. L., Deppen, S., Arbogast, P. G., and others (2008, November). "Mortality rates associated with multidrug-resistant Acinetobacter baumannii infection in surgical intensive care units." (AHRQ grant HS13833). Infection Control and Hospital Epidemiology 29(11), pp. 1080-1083.
Researchers found that mortality rates are similar for intensive care unit patients with drug-resistant and nonresistant Acinetobacter baumannii. They collected data from July 2003 to June 2006 to determine the 28-day mortality rate of patients who contracted multidrug resistant (MDR) and nonresistant A. baumannii infections in burn, surgical, and trauma intensive care units at a large university-affiliated hospital. Patients with MDR A. baumannii more frequently received two antibiotics (aminoglycosides and quinolones) and were ventilated 9.5 days compared with 6.8 days for patients whose infections were not drug-resistant. However, hospital stay lengths did not differ between the two groups. A total of 10 patients died: 6 infections were drug-resistant and 4 infections were not. The researchers did not find an increased mortality rate between the two groups; however, the small sample size may have limited the study's ability to draw conclusions.
Dossett, L. A., Heffernan, D., Lightfoot, M., and others (2008, November). "Obesity and pulmonary complications in critically injured adults." (AHRQ grant HS13833). CHEST 134(5), pp. 974-980.
Studies have linked obesity with a high risk of pulmonary complications after severe injury. However, a new study found no greater risk of pulmonary complications among critically injured obese adults than their normal-weight counterparts. Nevertheless, severely obese patients stayed in the intensive care unit (ICU) nearly 5 days longer than normal-weight injury victims. Researchers studied the outcomes of 1,291 critically injured adults admitted at 2 trauma ICUs. Overall, 30 percent of the patients were classified as either obese or severely obese. They examined rates of adult respiratory distress syndrome (ARDS), pneumonia, placement of a tracheostomy tube, and in-hospital deaths. Despite obese patients' underlying risk factors for pulmonary complications, severely obese patients had a 64 percent lower rate of ARDS than normal-weight patients. Rates of pneumonia (37 percent), tracheostomy (10 percent), and in-hospital death (11 percent) were similar for obese and normal-weight patients. However, the severely obese group had a nearly 5-day-longer ICU stay.
Farley, D. O., Haviland, A., Champagne, S., and others (2008, December). "Adverse-event-reporting practices by US hospitals: Results of a national survey." Quality and Safety in Health Care 17(6), pp. 416-423. Reprints (AHRQ Publication No. 09-R021) are available from the AHRQ Publications Clearinghouse.
Nearly 95 percent of critical access hospitals (CAHs), typically smaller, rural hospitals that provide 24-hour emergency services but a limited range of other services, had centralized adverse event reporting systems; 98.2 percent of other hospitals had similar systems, according to this study. James B. Battles, Ph.D., and William B. Munier, M.D., of the Agency for Healthcare Research and Quality, and coinvestigators surveyed risk managers from a national sample of 2,050 hospitals, resulting in 1,652 completed surveys. Less than one-third of hospitals (32 percent) had established environments that supported adverse event reporting, and only 13 percent had broad staff involvement (beyond physicians and nurses) in reporting these events. Only one-fifth of hospitals fully distributed summary reports on identified adverse events to departments and committees that could bring about practice changes. All reporting systems collected the type and place of adverse events. Almost as many reporting systems collected time of occurrence (99 percent), patient demographics (95 percent), if any action was taken (94 percent), and needed followup treatment (94 percent).
Flanders, S. A., Centor, B., Weber, V., and others (2009). "Challenges and opportunities in academic hospital medicine: Report from the Academic Hospital Medicine Summit." (AHRQ grants HS11416 and HS15701). Journal of General Internal Medicine 24(5), pp. 636-641.
The authors summarize the discussions at and results of the Academic Hospital Medicine Summit, a consensus conference on the status of academic hospital medicine and the clinical and financial issues that are limiting its growth. The conference brought together the major stakeholders in academic hospital medicine, including the Society of Hospital Medicine, the Society of General Internal Medicine, and the Association of Chiefs of General Internal Medicine. In particular, the limits placed on resident work hours have led to the use of hospitalists to provide general internal medicine care in teaching hospitals. They are often called on to implement quality or safety improvement initiatives, but little value is typically placed on teaching or research—even at teaching hospitals. Another problem is that the relative youth of the hospitalist field means that there are few senior academic hospitalists to provide leadership to the field and influence staff negotiations with hospitals. Conference participants concluded that there is a need to define a sustainable job description for academic hospitalists, as well as develop a quality improvement portfolio that would be comparable to an educator's portfolio.
Fonkych, K., O'Leary, J. F., Melnick, G. A., and Keeler, E. B. (2008, August). "Medicare HMO impact on utilization at the end of life." (AHRQ grant HS10256). American Journal of Managed Care 14(8), pp. 505-512.
Physicians practicing in HMOs in California have developed approaches that allow the use of fewer inpatient resources at the end of their patients' lives, according to this study. Because of fewer and shorter hospitalizations, Medicare patients in HMOs used 34 to 51 percent less inpatient care in the last 2 years of life than similar patients in Medicare fee-for-service (FFS) plans. Neither the timing of the last hospitalization nor the probability of having at least one hospitalization during the last 2 years of life was substantially different between HMO and FFS patients. These findings suggests that the California Kaiser and Independent Practice Association (IPA) HMOs primarily reduced the number of hospitalizations that preceded the last hospitalization, note the researchers. The findings were based on analysis of data from the Centers for Medicare & Medicaid Services, which were linked to California inpatient discharge data. The researchers determined the HMO-FFS differences in patient use during the last 2 years of life among a sample of aged Medicare beneficiaries who died between January 1998 and June 2001. These patients were enrolled during the 2 years before death in an FFS Medicare plan, an IPA, or Kaiser.
Glance, L. G., Osler, T. M., Mukamel, D. B., and others (2009, April). "Expert consensus vs empirical estimation of injury severity." (AHRQ grant HS16737). Archives of Surgery 144(4), pp. 326-332.
The authors of this paper compare two trauma injury scores that can be used as outcome measures for quality comparisons of hospital trauma care. The smaller the hospital's ratio of observed-to-estimated mortality for injuries, the higher the quality, they note. One score, the Injury Severity Score, depends on expert consensus of the severity of each injury on a scale ranging from 1 (minor) to 6 (currently untreatable). An alternate score, the Trauma Mortality Prediction Model, is based on empirical estimates of injury severity, and can be derived using hospital billing data. Both models were augmented by use of demographic information, mechanism of injury, and the motor function component of the Glasgow Coma Scale. Using data from the National Trauma Data Bank, which collects trauma registry data from hospitals, the authors applied both augmented scoring models to data from 66,214 patients in 68 hospitals. The two score models gave an interclass correlation coefficient of 0.93, indicating almost perfect agreement. The authors conclude that either injury scoring model can be used to give accurate and valid results for calculating estimated mortality, and hence quality of trauma care.
Gooden, K. M., Howard, D. L., Carpenter, W. R., and others (2008, November). "The effect of hospital and surgeon volume on racial differences in recurrence-free survival after radical prostatectomy." (AHRQ grant HS13353). Medical Care 46(11), pp. 1170-1176.
Researchers examined data from the 1991 to 2002 Surveillance, Epidemiology, and End-Results—Medicare databases for 962 black and 7,387 white men who underwent prostatectomy (surgical removal of the prostate) for prostate cancer within 6 months of diagnosis. They wanted to determine if less experience with prostatectomy among hospitals and surgeons treating black men contributed to racial disparities in postoperative survival. Prostate cancer recurrence-free survival rates improved with hospital and surgical volume of prostatectomies performed. Black men were 34 percent more likely to suffer a cancer recurrence than white men, even those treated at high- and medium-volume hospitals by high- and medium-volume surgeons.
Guise, J.-M. D. and Segel, S. (2008, October). "Teamwork in obstetric critical care." (AHRQ grants HS15800 and HS16673). Best Practice & Research Clinical Obstetrics & Gynaecology 22(5), pp. 937-951.
The authors provide a history of teamwork approaches in industry, reviewing the key components and delivery methods of the teamwork programs being used in medicine today to reduce safety risk. The aviation industry developed Crew Resource Management (CRM) after determining that human error caused most aviation accidents. Focusing on obstetric critical care, the authors provide an overview of 11 currently available medical team training programs that use many CRM principles. At the core of CRM are communication techniques, situational awareness, and leadership. Communication techniques in medical teamwork training include SBAR, in which a team member provides the situation, background, assessment, and response to ensure all information is conveyed to a receiving team member. Situational awareness ensures team members know what resources are available and where they are, so that they can be obtained quickly. Good leadership ensures every team member feels comfortable speaking up.
Heslin, K. C., Gore, J. L., King, W. D., and Fox, S. A. (2008, December). "Sexual orientation and testing for prostate and colorectal cancers among men in California." (AHRQ grant HS14022). Medical Care 46(12), pp. 1240-1248.
Gay/bisexual men differ from heterosexual men in their use of colorectal cancer (CRC) screening and prostate-specific antigen (PSA) screening for prostate cancer, according to this study. The researchers analyzed responses of 19,410 men to the California Health Interview Survey to correlate the sexual orientation of men age 50 and over with their receipt of PSA and CRC screening tests. More men received CRC than PSA tests, with nearly 80 percent of gay/bisexual men ever receiving a CRC test. After adjusting for race/ethnicity, education, and language proficiency, gay/bisexual men were nearly twice as likely as heterosexual men to have ever received a CRC test. In contrast, gay/bisexual men were 39 percent less likely than heterosexual men to have up-to-date PSA testing. However, interactions between sexual orientation and living situation showed that gay/bisexual men who lived alone had nearly twice the likelihood of receiving PSA tests than did other men.
Horowitz, C. R., Robinson, M., and Seifer, S. (2009, May). "Community-based participatory research from the margin to the mainstream. Are researchers prepared?" (AHRQ grant HS16471). Circulation 119, pp. 2633-2642.
The authors review the usefulness and growth of community-based participatory research (CBPR), an approach to clinical research that involves community members beyond being research subjects. Because representatives of community organizations are involved in the selection of the research topic, design of the study, recruitment of participants, writing up the results, and communicating the findings, the community has a stake in the project and implementing the results. The article compares the traditional approach with CBPR, explains why an approach partnering with the community is important—especially in addressing health care disparities—and highlights the potential benefits of CBPR. Using examples from cardiovascular medicine, the authors discuss key issues, such as team building, study selection and design, funding and ethics review, research conduct and analysis, dissemination of findings and their translation into policy and practice, and sustaining research partnerships with the community.
Howard, D. L., Taylor, Y. J., and Ross, L. E. (2008, October). "Differences in lower urinary tract symptoms, treatment and mortality among African-American and white elderly men." (AHRQ grant HS13353). Journal of the National Medical Association 100(10), pp. 1146-1152.
Moderate to severe lower urinary tract symptoms (LUTS), usually associated with benign prostatic hyperplasia (BPH) or enlarged prostate, affect an estimated third of older men. This study found a 40 percent higher probability of use of prostate surgery for elderly black than white men, after adjusting for other factors. However, it found no racial differences in treatment by catheter insertion to relieve blockages or drug therapy for BPH/LUTS. All three treatment types increased from 1994 to 1998, with drug therapy showing the largest increase. Catheter insertion increased over time and increased for men who had ever resided in a nursing home and men with no named physician. Men with coexisting conditions were less likely to receive drug therapy. Prostate surgery increased for black men, older aged men, those in poor self-reported health, men with cancer, and men having regular rectal exams.
Jaar, B. G., Plantinga, L. C., Crews, D. C., and others (2009, February). "Timing, causes, predictors and prognosis of switching from peritoneal dialysis to hemodialysis: A prospective study." (AHRQ grant HS08365). BioMedCentral Nephrology 10(3), pp. 1-12.
The authors investigated factors associated with dialysis patients switching from peritoneal dialysis (PD) to hemodialysis (HD), and whether this switch influences mortality. Although PD is capable of being done at home, its use has fallen, while that of clinic- or hospital-based HD is on the increase. To understand why this has changed, the authors undertook a prospective study of 262 PD patients enrolled at 28 PD clinics in 13 States. Among the patients in the study, 25 percent switched to HD during the study period. The majority of switchers (70 percent) did so within the first 2 years. The leading cause of switching (37 percent) was infection, either peritonitis or catheter-related. Cardiovascular problems (fluid overload) accounted for nearly 19 percent of the switchers. Black patients were three times more likely than white patients to switch, and this association became stronger after adjusting for demographic factors, diabetes status, body-mass index, and baseline serum creatinine. No statistically significant difference in mortality was observed between the switchers and nonswitchers. Further studies are needed to understand why black PD patients are more likely to switch to HD.
Keyser, D. J., Dembosky, J. W., Kmetik, K., and others (2009, May). "Performance improvement: Using health information technology-related performance measures and tools to improve chronic care." (AHRQ grant HS13690). The Joint Commission Journal on Quality and Patient Safety 35(5), pp. 248-255.
A collaborative initiative described by the authors sought to explore the usefulness of health information systems in collecting clinical performance measures to improve the quality of care for patients with selected chronic diseases. The initiative, led by the American Medical Association and assessed by the RAND Corporation, consisted of four pilot projects. Pilot 1, which focused on physician practices with electronic health information systems, found that paper-based data collection systems and an attempt to create a community-based central data warehouse were not effective. Pilot 2, conducted at an integrated network of outpatient clinics with electronic disease registries, was able to enhance the use of a limited set of clinically based quality measures and tools, such as measurement of glycosolated hemoglobin in patients with diabetes, resulting in improvements in outcomes of care. Pilot 3, conducted at a multispecialty physician's organization affiliated with a major medical school, used the practice's office-based electronic medical record to collect information to analyze the quality of care. Physicians in the practice received regular feedback on selected performance measures, and their compliance improved. Pilot 4, conducted at a large specialty group, documented the challenges of exporting quality performance data from physician practices to external organizations (health plans and Quality Improvement Organizations).
Lyons, M. S., Lindsell, C. J., Haukoos, J. S., and others (2009, February). "Nomenclature and definitions for emergency department human immunodeficiency virus (HIV) testing: Report from the 2007 Conference of the National Emergency Department HIV Testing Consortium." (AHRQ HS17526). Academic Emergency Medicine 16(2), pp. 168-177.
Emergency departments (EDs) are increasingly being asked to make testing for human immunodeficiency virus (HIV) a routine part of emergency care. As HIV testing of patients grows beyond a small number of pioneering centers and demonstration projects, the need has increased for a common set of definitions for reporting HIV and other information. The authors of this paper summarize the consensus process and results of the first conference of the National Emergency Department HIV Testing Consortium, held in Baltimore in November 2007. They report the consensus definitions related to setting (geographic and epidemiologic, facility, and the HIV testing program), patient recruitment and consent, testing methods, preresult and postresult communication with the patient, outcome measures, and key summary measures. The authors discuss some of the important distinctions included in the definitions, such as the difference between "testing" and "screening," and defining "patient" for outcome measures as a distinct individual rather than a distinct visit to the ED.
Needham, D. M., Sinopoli, D. J., Dinglas, V. D., and others (2009, February). "Improving data quality control in quality improvement projects." (AHRQ grant HS14246). International Journal for Quality in Health Care 21(2), pp. 145-150.
The authors emphasize the importance of data quality control in projects aimed at improving quality of care. They use as their example the successful statewide project in Michigan to reduce the rate of central line-associated infections among patients in intensive care units. The authors note that data quality issues for quality improvement (QI) projects should be addressed during the project design phase. Standardized forms with clear definitions of data items and written instructions for collecting each data item are very important to the ultimate quality of the data, as are quality assessment reviews and use of an electronic database. The authors recommend the use of statistical methods to review the data for errors. They also urge researchers to account for missing data during data analysis, manage outlier data, and convey the precision of the QI results. Researchers also should test for the influence of confounding variables (such as teaching status of the hospital and bed size). High quality QI data are vital to appropriate decisionmaking based on the research, the authors conclude.
Newsome, B. B., McClellan, W. M., Allison, J. J., and others (2008, August). "Racial differences in the competing risks of mortality and ESRD after acute myocardial infarction." (AHRQ grant HS13852). American Journal of Kidney Disease 52(2), pp. 251-261.
Blacks have a higher risk than whites of developing end-stage renal disease (ESRD), almost complete kidney failure, but they have lower rates of kidney disease. Researchers used baseline information from 1,261 Medicare patients who had heart attacks between February 1994 and July 1995 and developed ESRD after their hospital discharge to explore the reasons for this "renal paradox." Black patients tended to have lower prevalence of kidney disease but greater incidence and risk of developing ESRD in the 10 years after their heart attacks than white patients. The incidence rate for developing ESRD for black patients was 9.25 per 1,000 person years and 3.39 per 1,000 person years for white patients. Both blacks (77.4 percent) and whites (76.3 percent) died at similar rates before their kidney disease progressed to ESRD. The study does not explain why blacks have less kidney disease at baseline but have a higher risk of developing ESRD than whites. The presence of other diseases, such as diabetes and high blood pressure, which are the most important risk factors for ESRD and are prevalent in blacks, did not account for the differences. The authors suggest that ESRD may be associated with poverty and health care access.
Volpp, K. G., Loewenstein, G., Troxel, A. B., and others (2008). "A test of financial incentives to improve warfarin adherence." (AHRQ grant HS11530). BioMedCentral Health Services Research 8(272), pp. 1-6.
Financial incentives, in the form of a lottery, appear effective in getting patients on anticlotting therapy to take their medicine as prescribed. The authors describe a pilot study to test whether a lottery with a daily expected value of either $3 or $5 improves adherence to warfarin, an anticlotting medication. One set of 10 patients was entered into a daily lottery with a 1 in 5 chance of winning $10 and a 1 in 100 chance of winning $100; a second group of 10 patients had a 2 in 5 chance of winning $10 and a 1 in 100 chance of winning $100. If the lottery winner had not taken his or her medicine that day, they forfeited the prize. In both cases, the percentage of incorrect pills taken fell from the range of 26-28 percent before the study to a range of 3-6 percent during the study. Improvement was also observed for out-of-range international normalized ratios on the prothrombin time test, a measure of anticlotting therapy effectiveness. These studies provide initial evidence that lottery-based financial incentives may be useful in improving medication adherence in patients taking warfarin, as well as for other medications used to treat chronic conditions, the authors say.