Research Briefs

Bell, C.M., Urbach, D.R., Ray, J.G., and others (2006, February). "Bias in published cost effectiveness studies: Systematic review." (AHRQ grant HS10919). British Medical Journal 332, pp. 699-703.

Many insurers now require some form of cost effectiveness analysis before covering health interventions. This study found that most published analyses reported favorable incremental cost effectiveness ratios, with medical interventions costing below $20,000, $50,000, and $100,000 per quality adjusted life year (QALY) gained. Studies by industry were more likely to report ratios below the three thresholds. Studies of higher methodological quality and those conducted in Europe and the United States rather than elsewhere were less likely to report ratios below $20,000/QALY. These findings were based on a review of 494 studies measuring health effects in QALYs published up to December 2001.

Blake, S.C., Kohler, S., Rask, K., and others (2006, September). "Facilitators and barriers to 10 National Quality Forum safe practices." (AHRQ contract 290-00-0011). American Journal of Medical Quality 21, pp. 323-334.

The National Quality Forum (NQF) has listed 30 safe practices for hospitals to adopt to improve health care. Implementing safety practices can be a difficult process, replete with organizational, financial, and professional barriers. Interviews with Georgia hospital administrators identified several key factors that resulted in the adoption and/or nonadoption of 10 NQF safe practices in 2 areas: medication safety and the culture of safety. Medication safety practices ranged from pharmacist consultation with prescribers and review of medication orders to recording of verbal medication orders, use of standardized abbreviations and dosage designations, implementation of a computerized order entry system, and identification of all high-alert drugs. Culture of safety practices ranged from a blame-free, user-friendly, and confidential error-reporting system to safe staffing ratios, positive feedback from hospital supervisors/managers, teamwork within hospital units, and smooth transfer and handoff of patients within and among hospital units.

Brauer, C.A., Rosen, A.B., Greenberg, D., and Neumann, P.J. (2006, July). "Trends in the measurement of health utilities in published cost-utility analyses." (AHRQ grant HS10919). Value In Health 9(4), pp. 213-218.

Many experts and consensus groups have recommended cost-utility analysis (CUA) as the gold standard for conducting economic evaluations of medical interventions. Increasingly, analysts conducting CUAs are using generic, preference-weighted instruments, and relying on community-based preferences. The catalog of utility weights developed by these authors provides a useful reference tool for producers and consumers of CUAs. The catalog also highlights the continued need for improvement in methods and transparency, note the authors. This paper presents an update, through 2001, to their current registry of utility weights, and documents recent changes in methods used for utility weight calculation.

Clancy, C.M. (2006). "Enhancing quality improvement." Healthcare Papers 6(3), pp. 46-50.

In this commentary, the Director of the Agency for Healthcare Research and Quality notes that a substantial gap exists between the best possible care and the care that is routinely delivered in most developed nations. Numerous research studies and reports from authoritative organizations such as the U.S. Institute of Medicine have provided compelling evidence that care is not consistently safe, timely, effective, equitable, efficient, or patient-centered. In fact, a landmark study published in 2003 reported that Americans receive recommended care only 55 percent of the time. A recent survey of sicker adults conducted by the Commonwealth Fund in six developed countries underscores the pervasive challenges of providing high-quality care. There is room for significant improvement in all developing nations, concludes the author. Reprints (AHRQ Publication No. 07-R005) are available from the AHRQ Publications Clearinghouse.

Clancy, C. (2006, September). "The intensive care unit, patient safety, and the Agency for Healthcare Research and Quality." American Journal of Medical Quality 21(5), pp. 348-351.

There are about 6,000 intensive care units (ICUs) in the United States, caring for about 55,000 patients each day. Some estimates indicate that about 85,000 medical errors occur each day in American ICUs, of which 24,650 are potentially life-threatening. In a commentary, the Director of the Agency for Healthcare Research and Quality (AHRQ) highlights some of the major issues in ICU safety and describes ICU patient safety studies and activities that AHRQ supports. Study topics have ranged from the impact of sleep deprivation on staff performance to incident reporting and staffing with intensivists (physicians specialized in intensive care). A group of AHRQ-supported researchers has developed an ICU Safety Reporting System, a Web-based incident reporting system that collects information about adverse events and near misses reported by clinical staff in ICUs. Another project, the Keystone ICU Patient Safety Project, designed and implemented unit-based safety programs and daily goals sheets to help eliminate bloodstream infections and eradicate ventilator-associated pneumonia. Finally, AHRQ's 2001 report, Making Health Care Safer, included a chapter on the importance of using intensivists in hospital ICUs, an approach later studies have shown improves ICU patient health and care costs. Reprints (AHRQ Publication No. 07-R001) are available from the AHRQ Publications Clearinghouse.

Clancy, C.M., McNeill, D., Moy, E., and Dayton, E. (2006, June). "Agency for Healthcare Research and Quality's National Quality and Disparities Reports emphasize patient safety." Journal of Patient Safety 2(2), pp. 70-71.

This commentary reviews the implications for patient safety of the Agency for Healthcare Research and Quality's two Congressionally mandated reports, the National Healthcare Quality Report (NHQR) and the National Healthcare Disparities Report (NHDR). Even in the relatively brief span of 2 years, these annual reports showed improvements in patient safety. For example, the NHQR's five core measures of patient safety increased at an annual median rate of improvement of 10.2 percent. Similarly, this year's third edition of the NHDR indicates that ethnic/racial disparities in patient safety are less substantial than disparities in quality. Overall, minority hospital inpatients tended to have more nosocomial infections, potentially avoidable deaths, and complications of care compared with whites. On the other hand, minority patients tended to have fewer hospital injuries, adverse events related to technical errors, and birth-related traumas. Reprints (AHRQ Publication No. 07-R013) are available from the AHRQ Publications Clearinghouse.

Coben, J.H., Steiner, C.A., Barrett, M., and others (2006). "Completeness of cause of injury coding in healthcare administrative databases in the United States." Injury Prevention 12, pp. 199-201.

External cause of injury codes (E-codes) describe the mechanism and intent of an injury. E-code reporting in administrative databases is relatively complete, but there is significant variation in completeness in State databases. This study found that States with mandates for the collection of E-codes and with a mechanism to enforce those mandates had the highest rates of E-code reporting. Nine Statewide emergency department (ED) data systems showed consistently high E-coding completeness. The findings were based on analysis of the 2001 Healthcare Cost and Utilization Project State Inpatient Databases, a Nationwide Inpatient Sample, and nine State ED databases. Reprints (AHRQ Publication No. 07-R009) are available from the AHRQ Publications Clearinghouse.

Fongwa, M.N., Cunningham, W., Weech-Maldonado, R., and others (2006, October). "Comparison of data quality for reports and ratings of ambulatory care by African American and white Medicare managed care enrollees." (AHRQ grant HS09204). Journal of Aging and Health 18(5), pp. 707-721.

More data is missing for elderly black than white Medicare managed care enrollees on all items of the Consumer Assessment of Healthcare Providers and Systems (CAHPS®) survey of health plans. Higher missing data rates and lower plan-level reliability estimates for black Medicare managed care enrollees suggest caution in making race/ethnicity comparisons across health plan evaluations. Future efforts are needed to enhance the quality of data collected from older blacks, conclude the researchers. They analyzed CAHPS health plan survey data collected from 109,980 Medicare managed care enrollees in 321 health plans to compare missing data and reliability of health care evaluations.

Gorelick, M.H. (2006). "Bias arising from missing data in predictive models."(AHRQ grant HS11395). Journal of Clinical Epidemiology 59, pp. 1115-1123.

This author examined the effect of three common approaches to handling missing data on the results of a predictive model of hospital admission. The study used logistic regression analysis of complete data to predict hospital admission based on white blood cell count (WBC), presence of fever, or procedures performed (PROC). The author performed a series of simulations in which WBC data were deleted for varying proportions (15 to 85 percent) of patients under various patterns of missingness. Three approaches to handling missing data on the results of the predictive model were used: analysis restricted to cases with complete data, missing data assumed to be normal, and use of imputed values. Results showed that all three methods of handling large amounts of missing data can lead to biased estimates of the odds ratio for WBC, fever, and PROC, and overall predictive model performance.

Gupta, D., Potthoff, S., Blowers, D., and Corlett, J. (2006, July). "Performance metrics for advanced access." (AHRQ grant HS13023). Journal of Healthcare Management 51(4), pp. 246-258.

This article proposes several performance measures that can help clinic directors monitor and evaluate their advanced access implementation. Advanced access is an outpatient scheduling technique that aims to provide same-day appointment access. It is designed to reduce the time patients must wait for a scheduled appointment and to improve continuity of care by matching daily appointment supply and demand. Factors that make it difficult to sustain initial success in achieving supply-demand balance include different practice styles of doctors, differences in panel compositions and patient preferences, and time-varying demand patterns.

Henriksen, K., Keyes, M.A., Stevens, D.M., and Clancy, C.M. (2006, March). "Initiating transformational change to enhance patient safety." Journal of Patient Safety 2(1), pp. 20-24.

Transformational change is a daunting undertaking, but one that is needed in health care. This paper explores what transformational change means with respect to patient safety and quality initiatives. It draws on the transformational change literature to help identify the distinguishing features of successful transformations, and provides emerging findings and examples in health care that illustrate transformational concepts in practice. For example, introducing new information technology prompts changes in work procedures, work flow, communication networks, performance standards, needed subject matter expertise, personnel decisions, and other activities. Reprints (AHRQ Publication No. 07-R003) are available from the AHRQ Publications Clearinghouse.

Hubbard, H.B. (2006, September). "Interdisciplinary research: The role of nursing education." Journal of Professional Nursing 22(5), pp. 266-269.

Because of health care's increased complexity and subsequent reliance on interdisciplinary collaboration, nursing faculty have the opportunity to encourage emerging nurse scientists to be part of multidisciplinary teams that address a variety of healthcare issues, notes the author. She details the educational requirements necessary for an emerging nurse scientist, as well as the requirements that nurse investigators will encounter from funding agencies along the research trajectory. Through this dual lens, she defines health services research, which by its nature is interdisciplinary and multidisciplinary, examines its importance, and integrates its application with nursing education. Reprints (AHRQ Publication No. 07-R006) are available from the AHRQ Publications Clearinghouse.

Indik, J.H., Pearson, E.C., Fried, K., and Woosley, R.L. (2006, September). "Bazett and Fridericia QT correction formulas interfere with measurement of drug-induced changes in QT interval." (AHRQ grant HS10385). Heart Rhythm 3(9), pp. 1003-1007.

The QT interval on the electrocardiogram is prolonged by more than 50 marketed drugs, an effect that has been associated with sudden cardiac death due to arrhythmia. Because changes in heart rate also change the QT interval, it has become standard practice to use a correction formula, such as the Bazett formula, to normalize the QT interval to a heart rate of 60 beats per minute. This study shows how the Bazett formula and three other formulas influenced assessment of the QT-prolonging effect of the potassium channel-blocking drug ibutilide. The researchers used a standard physical activity protocol to assess the QT interval over a broad range of heart rates before and after an infusion of ibutilide that produced a stable 15- to 20-ms QT prolongation in consenting normal subjects. At heart rates from 60 to 120 beats per minute, the Bazett and Fridericia correction formulas overestimated the change in QT in both men and women. However, the Framingham and Hodges formulas did not alter the accuracy of the assessment of QT interval change.

Localio, A.R., Berlin, J.A., and Have, T.R. (2006). "Longitudinal and repeated cross-sectional cluster-randomization designs using mixed effects regression for binary outcomes: Bias and coverage of frequentist and Bayesian methods." (AHRQ grant HS11481). Statistics in Medicine 25, p. 2720-2736.

This study examined two distinct cluster randomization designs, in which a treated group and a control group are followed over time. These designs can be a powerful tool to measure the effect of medical interventions that must be applied to all individuals in the cluster. In a repeated cross-sectional design, patients who are naturally grouped within clusters or centers are recruited at baseline, but they cannot be followed individually. Rather, different patients within the same center are measured at each subsequent time period. In a second longitudinal design, individual patients are followed longitudinally within clusters, while the treatment is allocated to all patients in the centers. The authors concluded that the performance of common statistical tools for the analysis of cluster randomization designs depends heavily on the precise design, the number of clusters, and the variability of baseline outcomes and treatment effects across centers.

Max, M.B., Wu, T., Atlas, S.J., and others (2006, April). "A clinical genetic method to identify mechanisms by which pain causes depression and anxiety." (AHRQ grants HS06344, HS08194, and HS09804). Molecular Pain 2(4), pp. 14-25.

Pain patients are often depressed and anxious, and benefit less from psychotropic drugs than pain-free patients. These authors propose a method to prioritize molecular targets by studying polymorphic genes in patients undergoing lumbar spine surgical procedures associated with a variable pain relief response. Their goal was to identify molecules through which pain alters mood during the first postoperative year. In patients whose pain was reduced by over 25 percent by surgery, symptoms of depression and anxiety improved briskly at the first postoperative measurement. In patients with little or no surgical pain reduction, mood scores stayed about the same on average. Polymorphisms in three pre-specified pain-mood candidate genes were not associated with postoperative mood or with a pain-gene interaction on mood. However, an exploratory survey of 25 other genes illustrates pain-gene interactions on postoperative mood—the mu opioid receptor for short-term effects of acute sciatica on mood, and the galanin-2 receptor for effects of unrelieved post-discectomy pain on mood 1 year after surgery.

Parzen, M., Lipsitz, S.R., Fitzmaurice, G.M., and others (2006). "Pseudo-likelihood methods for longitudinal binary data with non-ignorable missing responses and covariates." (AHRQ grant HS10871). Statistics in Medicine 25, pp. 2784-2796.

In longitudinal studies, it is common for outcomes and any time-varying covariates to be missing due to missed study visits, resulting in nonmonotone patterns of missingness. Furthermore, the reasons for missed visits may be related to the specific values of the response and/or covariates that should have been obtained. In other words, missingness is nonignorable. With nonmonotone, nonignorable missing response and covariate data, a full likelihood approach is quite complicated. Also, maximum likelihood estimation can be computationally prohibitive when there are many occasions of followup. Furthermore, the full likelihood must be correctly specified to obtain consistent parameter estimates. These authors propose a pseudo-likelihood method for jointly estimating the covariate effects on the marginal probabilities of the outcomes and the parameters of the missing data mechanism.

Serwint, J.R., Thomas, K.A., Dabrow, S.M., and others (2006, September). "Comparing patients seen in pediatric resident continuity clinics and national ambulatory medical care survey practices: A study from the continuity research network." (AHRQ grant HS13582). Pediatrics 118(3), pp. e849-e857.

This study found that residents in Continuity Research Network (CRN) practices provide care to more underserved patients than pediatric practices that participate in the National Ambulatory Medical Care Survey (NAMCS). However, the Network practices evaluate problems that are similar to those observed in NAMCS office practices. Thus, CRN practices provide important training experiences for residents, who will serve both minority and nonminority children, concludes this study. The researchers compared data for CRN practice visits during a 1-week period in 2002 with data from the 2000 NAMCS.

Talcott, J.A., Clark, J.A., Manola, J., and Mitchell, S.P. (2006, October). "Bringing prostate cancer quality of life research back to the bedside: Translating numbers into a format that patients can understand." (AHRQ grant HS08208). The Journal of Urology 176, pp. 1558-1564.

Most treatments of clinically localized prostate cancer cause some degree of permanent erectile dysfunction and, less often, urinary or bowel symptoms. Understanding the likely effects of their treatment options is often a crucial step for men toward making a difficult decision with lasting consequences. Sophisticated quality-of-life measures produce purely numerical results that patients find difficult to understand. The authors of this paper present an approach that preserves the methodological strengths of validated multi-item measures, but provides more accessible information for clinical use. They use symptom indexes to define levels of function to produce a quality-of-life metric that is valid, defines quantitative intervals, is transparent, and may be more useful to patients attempting to choose a prostate cancer treatment.

Welch, W.P., Rudolph, B.A. Blewett, L.A., and others (2006). "Management tools for Medicaid and State Children's Health Insurance Program (SCHIP)." Journal of Ambulatory Care Management, 29(4), pp. 272-282.

State Medicaid health insurance programs for low-income families and the State Children's Health Insurance Program (SCHIP) need analytic tools to manage their programs. Drawing upon extensive discussions with experts in various States, this article describes the state of the art in use of analytic tools to manage health insurance programs, and makes several observations. First, several States have linked Medicaid/SCHIP administrative data to other data, such as birth and death records, to measure access to care. Second, several States use managed care encounter data to set payments. Third, the analysis of pharmacy claims data appears widespread. The authors also describe "lessons learned" regarding building capacity and improving data to support the implementation of management tools.

Current as of December 2006
AHRQ Publication No. 07-0011