Research Briefs

Research Activities, February 2010, No. 354

Baron, S., McPhaul, K., Phillips, S., and others (2009). "Protecting home health care workers: A challenge to pandemic influenza preparedness planning." American Journal of Public Health 99 (Supp 2), pp. S301-S307. Reprints (AHRQ Publication No. 10-R015) are available from the AHRQ Publications Clearinghouse.

The home health care sector is a critical element in the response to a pandemic influenza emergency. The authors summarize findings from a national stakeholder meeting that highlighted the need to integrate home health care employers, workers, community advocates, and labor unions into the preparedness planning process. The meeting discussed the following topics: the home health care workforce; in-home care provider models; home health care workers in a pandemic; ability and willingness to respond; communication and training; prevention and protection; and economic, legal, and ethical issues. The stakeholders recommended that: (1) Federal, State, and local pandemic preparedness planners should consider approaches to help home health care workers maintain their ability to work during a pandemic; (2) all home health workers should receive high priority for vaccinations and treatments related to protecting their health; and (3) communication strategies should emphasize maintaining a constant state of preparedness during the regular influenza season.

Berner, E. S. (2009). "Diagnostic error in medicine: Introduction." (AHRQ grant HS17406). Advances in Health Science Education 14, pp. 1-5.

The author summarizes a journal supplement containing reports from a conference on diagnostic errors in medicine. The May 2008 meeting, cosponsored by the Agency for Healthcare Research and Quality and the American Medical Informatics Association, included experts representing four research areas: medical problem solving, normative decisionmaking, clinical diagnostic decision support research, and patient safety research. It was the first national effort to bring researchers from the disparate research areas together to address this important issue. The following subjects were discussed: the challenge of diagnostic errors, perspectives on clinical decisionmaking, the settings and sources of diagnostic errors, educational approaches to reducing diagnostic errors, and reducing diagnostic errors with clinical decision support. The final part of the supplement discusses what is known from previous research and the questions that remain unanswered.

Clancy, C. M. (2009, June). "Reducing central line-related bloodstream infections." AORN 89(6), pp. 1123-1125. Reprints (AHRQ Publication No. 10-R009) are available from the AHRQ Publications Clearinghouse.

Catheter-related bloodstream infections are common and, too often, deadly. They are also preventable. In the intensive care unit (ICU), providers traditionally viewed the baseline rate of occurrence of such infections-about five per 1,000 catheter days-as the price of inserting central lines. This mindset, thankfully, is changing, according to the author, Director of the Agency for Healthcare Research and Quality (AHRQ). The solution is a simple, low-cost checklist of five basic steps, all of which are backed by scientific evidence. Researchers from Johns Hopkins University introduced five interventions to assist clinicians in complying with the guidelines summarized in the checklist. Over a 5-year period, the Johns Hopkins surgical ICU was able to reduce the infection rate from 11.3 per 1,000 catheter days to zero. Under a new $3 million grant from AHRQ, the Johns Hopkins researchers are taking their comprehensive unit-based safety program to hospitals in 10 States.

Clancy, C. M. (2009). "Healthcare quality and disparities. Attacking problems at their root." Journal of Nursing Care Quality 24(4), pp. 269-272. Reprints (AHRQ Publication No. 10-R012) are available from the AHRQ Publications Clearinghouse.

U.S. health care quality continues to lag. The National Healthcare Quality Report and its companion, The National Healthcare Disparities Report, recently published by the Agency for Healthcare Research and Quality (AHRQ), document that health care quality remains suboptimal and continues to improve at a slow pace, while disparities persist in quality and access. The author, the Director of AHRQ, notes that the reports indicate that the U.S. health care system has many opportunities for improvement. Although there is progress in select areas, the system is not achieving the more substantial strides needed to close the "quality chasm" that exists. The system achieves higher performance on measures related to acute treatment, such as that for heart attacks, as opposed to prevention and anticipatory treatment of chronic illnesses, such as cancer screening and diabetes management. Some of the reasons why quality lags and disparities persist have to do with rising costs, diminished access, and primitive use of information technology. Significant investments in quality improvement are being made through the American Reinvestment and Recovery Act of 2009.

DeVoe, J. E. (2009, March). "Educaid: What if the US systems of education and health care were more alike?" (AHRQ grant HS16181). Family Medicine 41(9), pp. 652-655.

In order to highlight the failings of the U.S. health care system, the author develops a scenario around a future system of education transformed to include many of the features of today's health care system. A central premise of the scenario is a State insurance program for low-income children called "Educaid." The scenario includes events such as: a child becoming eligible for Educaid (and thus admission to a school) only when his father loses his job; delays of many weeks until the Educaid card arrives in the mail; and many teachers refusing to accept students with Educaid coverage. These events are interlaced with actual quotes from parents. These quotes were originally about the experiences of families in the current health care system and have been slightly altered to apply to the hypothetical education examples (e.g., "clinic" has been changed to "school" and "doctor" to "teacher"). In closing, the author suggests consideration of a primary health care system analogous to the public education "ideal" with safe clinics for every neighborhood, community involvement, regional planning, and adequate funding.

Griffin, M. R., Braun, M., and Bart, K. J. (2009). "What should an ideal vaccine postlicensure safety system be?" American Journal of Public Health 99 suppl. 2, pp. S345-S350.

Preventing infectious diseases through immunization programs depends on ensuring the safety of vaccines and effectively communicating their benefits and risks. In 2007, the National Vaccine Program, along with the Centers for Disease Control and Prevention, the Food and Drug Administration, the National Institutes of Health, and the Health Resources and Services Administration, sponsored a public conference on "Vaccine Safety Evaluation: Post Marketing Surveillance." The conference's objective was to discuss enhanced approaches to postlicensure evaluation of vaccine safety, including active and passive surveillance systems and special studies. The participants reviewed the evolution of vaccine safety assessments, detailed current national approaches to postmarketing safety, and offered new approaches to evaluating vaccine safety. The authors summarize the major meeting presentations and discussions.

He, Y., and Zaslavsky, A. M. (2009, September). "Combining information from cancer registry and medical records data to improve analyses of adjuvant cancer therapies." (AHRQ grant HS09869). Biometrics 65, pp. 646-952.

Cancer registry records contain valuable data on provision of adjuvant therapies for cancer patients. Previous studies, however, have shown that these therapies are underreported in registry systems. Therefore, direct use of the registry data may lead to invalid analysis results. The authors first impute correct treatment status, borrowing information from an additional source such as medical records data collected in a validation sample, and then analyze the imputed data. They extend their models to multiple therapies using multivariate probit models with random effects. Their model takes into account the associations among different therapies in both administration and probability of reporting, as well as the multilevel structure (patients clustered within hospitals) of registry data. They use Gibbs sampling to estimate model parameters and impute treatment status. They apply their proposed methodology to the data from the Quality of Cancer Care Project, in which stage II or III colorectal cancer patients were eligible to receive adjuvant chemotherapy and radiation therapy.

Kelz, R. R., Tran, T. T., Hosokawa, P., and others (2009, October). "Time-of-day effects on surgical outcomes in the private sector: A retrospective cohort study." (AHRQ grant HS11913). Journal of the American College of Surgeons 209(4), pp. 434-445.

It is known that aspects of timing can affect health care outcomes. To inform decisionmaking and resource allocation for surgical procedures, the researchers examined surgical patients' outcomes in the American College of Surgeons Private Sector Study (PSS), according to start, duration, and end times of their surgical procedures during the course of the day. The PSS data set contained records for 56,920 patients who underwent general or vascular surgical procedures at one of 14 academic medical centers. The researchers found that nonemergency cases performed overnight (after 5:30 PM) faced an elevated risk of mortality, and emergency cases (7,370 of the 56,920 cases) performed overnight faced an elevated risk of morbidity. Risk adjustment was made for patient and procedure characteristics. The following factors may contribute to these differences in outcomes: human performance factors such as fatigue, system performance factors such as resource availability, and unidentified biological patient effects.

Lau, D. T. and Kirby, J. B. (2009, October). "Living arrangement and colorectal cancer screening: Updated USPSTF Guidelines." American Journal of Public Health 99(10), pp. 1733-1734. Reprints (AHRQ Publication No. 10-R014) are available from the AHRQ Publications Clearinghouse.

In a recent article, the researchers examined the relationship between living arrangement and preventive care use among community-dwelling persons aged 65 years and older in the United States by analyzing the 2002-2005 Medical Expenditure Panel Survey. Of the six preventive services examined, they defined adherence to recommended colorectal cancer screening (either fecal occult blood test within the past year or sigmoidoscopy within the past 5 years) according to the 2002 United States Preventive Services Task Force (USPSTF) guidelines. After the USPSTF revised its guidelines in 2008 to offer separate recommendations for those older than 85 and those between 76 and 85 years, the researchers performed additional analyses on the subgroup of those between 65 and 75 years. They found that elderly persons who lived with a spouse were just as likely to get colorectal cancer screening as those living alone. However, elderly persons who lived with an adult offspring, regardless of the presence of a spouse, were significantly less likely to get screening than either those who lived with a spouse or those who lived alone.

Lautenbach, E. (2009). "Antimicrobial resistance in gram-negative pathogens: Crafting the tools necessary to navigate the long ascent out of the abyss." (AHRQ grant HS10399). The Journal of Infectious Diseases 200, pp. 838-840.

The unrelenting increase in the prevalence of antimicrobial resistance is of great concern. The importance of gram-negative organisms as causes of health care-acquired infections may be resurgent. These resistant organisms include multidrug-resistant (MDR) Pseudomonas aeruginosa, among others. The author reviews a new study in the same issue of this journal that characterizes the importance of person-to-person spread in the emergence of imipenem-resistant P. aeruginosa. The author places this study, the largest of its kind, in the context of other studies in this area. Since this study and others employ different definitions of person-to-person transmission, he argues the need for a common definition. The author also finds that a clear message of this study is how complex the emergence of drug resistance is and how little we really understand it at this time.

Meltzer, D. O., Basu, A., and Meltzer, H. Y. (2009, July). "Comparative effectiveness research for antipsychotic medications: How much is enough?" (AHRQ grant HS16967). July 21, 2009 Health Affairs Web Exclusive, pp. w794-w808.

Second-generation antipsychotic drugs differ from first-generation antipsychotic drugs in having a diminished risk of symptoms such as extreme restlessness and involuntary movements. However, they have other side effects and are much more costly. A study by the National Institute of Mental Health (NIMH) comparing the effectiveness of perphenazine, a first-generation antipsychotic, and all second-generation antipsychotics found that perphenazine could not be rejected as an inferior treatment. However, the design of the NIMH study has been criticized for its use of time to drug discontinuation as the primary outcome. Using data from the literature on schizophrenia prevalence and mortality and data from the NIMH study on the effects of treatments on quality of life, the authors of this study estimated that the NIMH finding that initial assignment to typical antipsychotics is cost-effective had a 55 percent chance of being wrong. The authors conclude that future research on this class of medications is likely to be of immense value to people with schizophrenia today and those who will develop it over the next 20 years.

Ngo, V. K., Asarnow, J. R., Lange, J., and others (2009). "Outcomes for youths from racial-ethnic minority groups in a quality improvement intervention for depression treatment." Psychiatric Services 60(10), pp. 1357-1364.

Evidence-based mental health interventions have been found effective for youths from ethnic minorities. However, they are less likely to receive high-quality mental health services. The researchers studied the impact of a quality improvement intervention designed to improve access to evidence-based depression care for minority youths. The quality improvement program featured cognitive behavioral therapy and care management services. Of the 418 initially enrolled youths, 344 completed the 6-month followup assessment. To diagnose depression, the researchers used the Composite International Diagnostic Interview; for the 6-month followup, they used the Center for Epidemiological Studies Depression Scale. Results showed a significant reduction in depression symptoms among blacks in the intervention group. Among Latinos in the intervention group, the only significant improvement was in care satisfaction. No intervention effects were found for whites.

Plantinga, L. C., Fink, N. E., Coresh, J., and others (2009). "Peripheral vascular disease-related procedures in dialysis patients: Predictors and prognosis." (AHRQ grant HS 08365). Clinical Journal of the American Society of Nephrology 4, pp. 1637-1645.

Peripheral vascular disease (PVD) is common among dialysis patients, many of whom receive PVD-related procedures such as nontraumatic amputation and revascularization. The authors of this study examined the risk factors for and prognosis after such procedures were performed in dialysis patients with and without diabetes. For those patients without diabetes, only a history of PVD and increased fibrinogen were associated with PVD-related procedures. For those patients with diabetes, increased serum phosphate along with decreased albumin, increased C-reactive protein and fibrinogen, and lower systolic blood pressure were associated with the risk of PVD-related procedures. Patients who had PVD-related procedures after the start of dialysis had more cardiovascular events, hospitalizations related to infection, PVD-related mortality, and all-cause mortality than those who did not have these procedures. For example, 68 percent of those who had a PVD-related procedure experienced a cardiovascular event versus 30 percent for those who did not.

Rassen, J. A., Brookhart, M. A., Glynn, R. J., and others (2009). "Instrumental variables I: Instrumental variables exploit natural variations in nonexperimental data to estimate causal relationships." (AHRQ grant HS10881). Journal of Clinical Epidemiology 62, pp. 1226-1232.

In many cases where randomized controlled trials are impractical or unethical, instrumental variable (IV) analysis offers a nonexperimental alternative based on many of the same principles. The authors introduce the use of IV analysis as a supplement to standard epidemiologic methods. They outline the analytical method and the assumptions required for IV analysis and offer several examples to illustrate the strengths and potential pitfalls of the IV approach. Topics discussed include: interventional and IV approaches; the three basic assumptions of IV analysis; compliance and the effect on the marginal subject; treatment-effect heterogeneity; IV variables in epidemiology; and making use of natural variation in treatment choice. The authors believe that with proper design and due caution, IV analysis is a sensible addition to the toolbox of clinical epidemiology.

Rassen, J. A., Brookhart, M. A., Glynn, R. J., and others (2009). "Instrumental Variables II: Instrumental variable application-in 25 variations, the physician prescribing preference generally was strong and reduced covariate imbalance." (AHRQ grant HS10881). Journal of Clinical Epidemiology 62, pp. 1233-1241.

Instrumental variable (IV) analysis joins other techniques that attempt to mitigate the bias introduced by measured and unmeasured confounding present in nonexperimental data. To provide reliably consistent estimates of effects, IVs should be both valid and reasonably strong. Physician prescribing preference (PPP) is an IV that uses variation in doctors' prescribing to predict drug treatment. As reduction in covariate imbalance may suggest increased IV validity, the researchers examined the covariate balance and instrument strength of 25 formulations of the PPP IV in two cohort studies. The PPP IV was applied to assess antipsychotic medication use and subsequent death among two cohorts of elderly patients. The researchers varied the measurement of PPP, performed cohort restriction and stratification, and modeled risk differences with two-stage least square regression. They concluded that most of the 25 formulations of the PPP IV were strong IVs and resulted in a strong reduction of imbalance in many variations.

Seow, H., Snyder, C. F., Mularski, R. A., and others (2009, December). "A framework for assessing quality indicators for cancer care at the end of life." (AHRQ Contract No. 290-05-0034). Journal of Pain and Symptom Management 38(6), pp. 903-912.

The lack of readily available data on quality of care for patients with advanced cancer has been a major barrier to improving palliative and end-of-life care. Measuring and improving the quality of cancer end-of-life care requires indicators that are reflective of the domains of quality cancer care, feasible to implement, and supported by experts and research evidence. A framework conceptualizing quality end-of-life cancer care and informing the development and evaluation of quality indicators could advance the field of cancer quality measurement and improvement, note these authors. To develop this framework, they built on previous initiatives, updated reviews of existing indicators and data sources, and obtained input from experts through a national symposium. The framework presents the population of focus, 10 broad quality domains, 4 steps in the care process, and evaluation criteria for quality indicators.

Seow, H., Snyder, C. F., Shugarman, L. R., and others (2009, September). "Developing quality indicators for cancer end-of-life care." (AHRQ Contract No. 290-05-0034). Cancer 115(17), pp. 3820-3829.

Quality indicators applicable to cancer end-of-life care exist, but have not been widely implemented. In-depth discussions based on a conceptual framework and specific domains may help to identify cross-cutting issues and future priorities for the field, note these authors. They worked with the Agency for Healthcare Research and Quality, the National Cancer Institute, experts in the field, and other stakeholders to organize a national symposium on this topic. They report that the symposium discussed eight domains from which cross-cutting themes emerged (pain; dyspnea; communication, care planning and decisionmaking; psychosocial issues; communication about chemotherapy; depression; continuity, coordination, and care transitions; and spirituality and closure) along with each domain's priority issues and potential solutions. Symposium participants concluded that only by developing better quality indicators and improving their use can it be determined where providers most need to improve.

Spetz, J., and Keane, D. (2009, September). "Information technology implementation in a rural hospital: A cautionary tale." (AHRQ grant HS10960). Journal of Healthcare Management 54, pp. 337-348.

A growing number of hospitals are implementing electronic medical records and other information technology (IT) systems, and national policy is focused on fostering expansion of these systems. The researchers evaluated what happened when a 100-bed acute care hospital implemented an integrated hospital IT system with electronic medical records and computerized physician order entry. The goals were to improve overall patient safety, decrease medication errors, and offer physicians remote access to data. However, the introduction of the Patient Care Documentation System was followed by increased rates of medication errors, patient care incidents, and procedure errors. Problems occurring during the implementation period included changes in nurse leadership, inadequate staff preparation, computer malfunctions, an overly aggressive schedule, and a vendor whose products were not ready in time. The researchers concluded that the hospital suffered a number of setbacks during the implementation that could provide lessons to other hospitals and may explain the substantial boost in adverse patient events at the hospital.

Tsai, C., Clark, S, Sullivan, A. F., and Camargo, C. A. (2009). "Development and validation of a risk-adjustment tool in acute asthma." (AHRQ grant HS13099). HSR: Health Services Research 44(5), pp. 1701-1717.

Risk adjustment is an important method in health services research, particularly when profiling provider performance and adjusting capitation-based payment. However, risk-adjustment tools for acute respiratory disorders, such as acute asthma, are very limited. The researchers developed and prospectively validated risk-adjustment tools for acute asthma using data from two large multicenter studies. They chose hospital admission as a potentially important outcome measure and profiled admission practices across more than 60 emergency departments (EDs). A model was constructed from the derivation cohort (3,515 patients with acute asthma) comprised of 9 variables that included demographics, chronic asthma-related factors, acuity at ED presentation, and initial ED treatments. When the derivation model was applied to the validation cohort (3,986 patients), in all deciles of admission probability the predicted probabilities of admission were fairly consistent with actual risks of admission. The researchers concluded that this tool could be used for profiling admission practices across hospitals.

Whelan, C. T., Kaboli, P., Zhang Q., and others (2009). "Upper gastrointestinal hemorrhage: Have new therapies made a difference? (AHRQ grant HS10597). Journal of Hospital Medicine 4(7), pp. E6-E10.

Upper gastrointestinal hemorrhage (UGH) is a common cause of acute admission to the hospital. In contrast to previous studies, this study examined the distribution of etiologies and risk factors of UGH in the era of widespread use of effective preventive therapy for erosive disease (ED) and peptic ulcer disease (PUD). Of the 227 patients with UGH, 44 percent had ED, 33 percent had PUD, and 17 percent had variceal bleeds. The most common risk factors for UGH among the patients were the use of aspirin (25 percent), nonsteroidal anti-inflammatory drugs, and COX-2 inhibitors (5 percent). Also, prior history of UGH (43 percent) was a major risk factor. ED was more common among patients from academic medical center 1 (59 percent) than academic medical center 2 (19 percent), while variceal bleeding was more common among patients from center 2 (34 percent) than center 1 (6.5 percent).

Page last reviewed February 2010
Internet Citation: Research Briefs: Research Activities, February 2010, No. 354. February 2010. Agency for Healthcare Research and Quality, Rockville, MD.