Research Briefs

Research Activities, February 2011, No. 366

Akincigil, A., Munch, S., and Niemczyk, K. S. (2010). "Predictors of maternal depression in the first year postpartum: Marital status and mediating role of relationship quality." (AHRQ grant HS16097). Social Work in Health Care 49(3), pp. 227-244.

This Rutgers University study may help dispel the belief that unmarried women are more likely than married women to suffer from depression in the year after giving birth. In fact, the research team found that the quality of the woman's relationship with the child's father, not her marital status, was a better predictor of whether she would battle depression. The researchers used data collected from 4,348 women just after they gave birth and 1 year later. Just over 12 percent of women became depressed in the year following the birth. Regardless of their marital status, women who reported supportive relationships with the father of their child were less likely to become depressed during the following year than mothers who did not receive emotional support from the child's father. Further, women who were abused by their child's father (26 percent) or had disagreements with the father about the pregnancy (4 percent) were likely to become depressed.

Burris, H. B., and Collins, J. W. (2010, Summer). "Commentary: Race and preterm birth—the case for epigenetic inquiry." (AHRQ grant T32 HS00063). Ethnicity & Disease 20, pp. 296-299.

In the United States, preterm birth and infant mortality disproportionately affect black families. The authors argue that epigenetic (i.e., gene-environment interactions) mechanisms will improve our understanding of such disparities. They propose that now is the time to translate what has been learned from the laboratory and the study of the agouti mouse into the realm of human studies. Studies have shown how impact of dietary differences among agouti mice may affect the health status of their offspring. This effect seems to be mediated by DNA methylation. The "epigenome," or pattern of DNA methylation, is laid down during early fetal life and may determine later health status. Whether intrauterine exposures to both toxins and beneficial dietary supplements can alter phenotypes has not yet been explored in humans.

Clancy, C. (2010). "CER: Tools—not rules, and its effects can be transformational." Reprints (AHRQ Publication No. 11-R010) are available from the AHRQ Publications Clearinghouse.

In this online commentary piece, the director of the Agency for Healthcare Research and Quality (AHRQ) discusses comparative effectiveness research (CER) and its usefulness to the practicing physician. CER is a type of patient-centered research that delivers evidence-based information to those who need to make a decision about care. CER is being funded by the American Recovery and Reinvestment Act of 2009 and is a central element in the new health care reform law known as the Patient Protection and Affordable Care Act. However, AHRQ has already funded and completed dozens of CER projects under its Effective Health Care Program on such subjects as diabetes, breast cancer, and depression. CER is descriptive, not prescriptive, and provides tools that both doctor and patient can use to make the best possible decisions. For more information, visit

Clark, D. E., Hannan, E. L., and Wu, C. (2010). "Predicting risk-adjusted mortality for trauma patients: Logistic versus multilevel logistic models." (AHRQ grant HS15656). Journal of the American College of Surgery 211, pp. 224-231.

Statistical models for predicting hospital performance are increasingly of interest to health planners, regulators, and patients. Multilevel logistic regression models (MLLR), while more complex than standard logistic regression (LR) models, have some theoretical advantages, and have become popular among those in health services research. The major drawback to the use of MLLR modeling has been computational complexity, which is now being overcome. The assumption is that, by employing MLLR models, properly adjusted institutional performance in the recent past can be used to predict institutional performance in the immediate future. Using data from the 2002-2006 National Inpatient Surveys on patients with trauma from 989 hospitals, the researchers compared predictions from LR and MLLR models for 2004, 2005, and 2006. They found that the differences between actual and predicted mortality were smaller with the MLLR models.

Curns, A. T., Steiner, C., Barrett, M., and others (2010, June). "Reduction in acute gastroenteritis hospitalizations among US children after introduction of rotavirus vaccine: Analysis of hospital discharge data from 18 US states." Journal of Infectious Diseases 201(11), pp. 1617-1624. Reprints (AHRQ Publication No. 10-R071) are available from the AHRQ Publications Clearinghouse.

Rotavirus is the most common cause of diarrhea and gastrointestinal upset in infants and young children. Symptoms can be severe enough to require hospitalization. Researchers compared acute gastroenteritis hospitalizations before (2000-2006) and after (2007-2008) the introduction of a new rotavirus vaccine (RotaTeq®), which was recommended in 2006 for the routine vaccination of all infants in the United States. By 2008, hospitalization rates for acute gastroenteritis from rotavirus infection had declined significantly as the result of widespread use of the vaccine. In 2007, there was a 16 percent decrease in hospitalizations and in 2008 a 45 percent decrease compared with the prevaccine period. The data source was the Agency for Healthcare Research and Quality's Healthcare Cost and Utilization Project which contains hospital discharge data representing nearly half of the U.S. population.

Ford, D., Zaa, J., and Gebregziabher, M. (2010, July). "Factors associated with illness perception among critically ill patients and surrogates." (AHRQ grant HS10871). Chest 138(1), pp. 59-67.

The intensive care unit (ICU) is a setting where patients are seriously ill and often die. Communication between patients, families, and clinicians is particularly important in an ICU, and decisionmaking is clearly affected by perceptions of patient prognosis and treatment options. The researchers interviewed 100 patients (or their surrogate decisionmakers) who were in a university hospital medical ICU for at least 3 days. Questions focused on their beliefs about the timeline and consequences of illness, emotional reaction to the illness, beliefs about personal control over the illness, beliefs about the efficacy of treatments, and comprehension of their illness. They found that blacks were more optimistic than whites on five of six domains of illness perception even after adjusting for faith/religion. Blacks also tended to perceive the illness as less enduring and reported more confidence in treatment efficacy. In addition, they tended to report the illness as less serious, having less emotional impact, and themselves as having greater personal control. However, they also reported lower illness comprehension.

Friedman, B., and Jiang, H. J. (2010). "Do Medicare Advantage enrollees tend to be admitted to hospitals with better or worse outcomes compared with fee-for-service enrollees?" International Journal of Health Care Finance and Economics 10, pp. 171-185. Reprints (AHRQ Publication No. 10-R042) are available from the AHRQ Publications Clearinghouse.

Persons eligible for Medicare can enroll in the traditional fee-for-service (FFS) plan or the newer Medicare Advantage (MA) plans. These plans are managed by outside insurance companies who receive a monthly capitation fee per enrollee from Medicare. Since there is no need to submit claims to Medicare for reimbursement, the Medicare program does not receive hospital discharge summaries. Researchers from the Agency for Healthcare Research and Quality (AHRQ) sought to discover whether MA patients tend to go to hospitals with better outcomes than FFS patients. Using 2006 hospital data from 13 statewide databases that are part of AHRQ's Healthcare Cost and Utilization Project, they found that Medicare Advantage patients tend to be admitted to hospitals that have lower resource costs and higher mortality rates (adjusted for patient risk factors). However, no significant differences were observed for surgical patients in terms of hospital cost or mortality rates.

Glurich, I., Burmester, J. K., and Caldwell, M. D. (2010). "Understanding the pharmacogenetic approach to warfarin dosing." (AHRQ grant HS16335). Heart Failure Review 15, pp. 239-248.

Despite its association with serious drug-related adverse events, warfarin remains the drug of choice for long-term anticoagulation management in a variety of conditions. Its current underutilization is attributable to its being ranked among the top 10 drugs associated with the greatest number of serious adverse drug events. This review summarizes the continued importance of warfarin to the reduction of the morbidity and mortality associated with thromboembolic disease. It also provides a brief historical overview of the emergence of the pharmacogenetics era of warfarin dosing and explores the existing controversies that prevent the full implementation of personalized medicine approaches to warfarin dosing. Variations in individual responses to warfarin have been traced to differences in age, body size, presence of other illnesses, and use of other medications. However, more recent research has highlighted a more important group of factors—the combined effects of polymorphisms in three genes. Ongoing studies and international collaborative efforts that have been generated by these recent breakthroughs hold the potential of making warfarin a safer drug for patients.

Holman, R. C., Christensen, K. Y., Belay, E. D., and others (2010, August). "Racial/ethnic differences in the incidence of Kawasaki syndrome among children in Hawaii." Hawaii Medical Journal 69, pp. 194-197. Reprints (AHRQ Publication No. 11-R004) are available from the AHRQ Publications Clearinghouse.

Kawasaki syndrome (KS) is a rare childhood disease affecting the blood vessels, which can lead to heart problems. It usually occurs in children under the age of 5 and is most likely to manifest itself in children of Japanese and other Asian ancestry in Hawaii and the continental United States. To better understand the racial/ethnic-specific incidence of KS among children living in Hawaii, researchers analyzed hospital discharge data for patients with KS younger than 18 years from 1996 to 2006. For this period, they found that 528 children accounted for 582 hospitalizations with KS. Children less than 5 years old accounted for 441 (83.5 percent) of cases. Among all racial/ethnic groups less than 5 years of age in Hawaii, Japanese children had the highest incidence per 100,000 people (210.5), followed by Native Hawaiian children (86.9), other Asian children (84.9), and Chinese children (83.2).

Holman, R. C., Belay, E. D., Christensen, K. Y., and others (2010, June). "Hospitalizations for Kawasaki syndrome among children in the United States, 1997-2007." The Pediatric Infectious Disease Journal 29(6), pp. 483-488. Reprints (AHRQ Publication No. 10-R074) are available from the AHRQ Publications Clearinghouse.

To examine trends in hospitalization of children 18 years of age or younger with Kawasaki syndrome (KS), researchers mined two national databases for hospital discharge records between 1997 and 2007: the Kids' Inpatient Database and the Nationwide Inpatient Sample of the Agency for Healthcare Research and Quality. In 2006, the hospitalization rate for children aged 5 and younger with KS was 20.8 per 100,000 children. Annual rates remained constant during the study period, except for a peak in 2005. Most hospitalizations were for children 3 years old and under, but there were very few hospitalizations during the first 2 months of life. Higher hospitalization rates were observed for boys, children from higher income families, and children with private insurance.

Karsh, B.-T., and Brown, R. (2010). "Macroergonomics and patient safety: The impact of levels on theory, measurement, analysis, and intervention in patient safety research." (AHRQ grant HS13610). Applied Ergonomics 41, pp. 674-681.

Little, if any, patient safety research has purposefully sought to understand how level issues might impact patient outcomes, such as quality, errors, adverse drug events, or patient harm. Levels are typically discussed in two ways: (1) individuals, groups, or organizations and (2) levels of hierarchy (e.g., the vice president of nursing, nurse supervisors, nurse managers, charge nurses, and staff nurses). The authors explore the implications of their ideas for medical errors theory and research. They believe that medication errors may be the result of nurse behavior, leadership decisions, group dynamics, poor workflow, safety culture, the lack of use of health information technology or some combination of all of these factors. Definite knowledge is not yet available because of a lack or research to disentangle such causes. The authors advise that mixed-level theories should be used to help find answers to this and related questions.

Kirby, J. B., Hudson, J., and Miller, G. E. (2010). "Explaining racial and ethnic differences in antidepressant use among adolescents." Medical Care Research and Review 67(3), pp. 342-363. Reprints (AHRQ Publication No. 10-R072) are available from the AHRQ Publications Clearinghouse.

This study examined the extent to which antidepressant use among adolescents varies across racial and ethnic groups. The researchers found that white adolescents were more than twice as likely as Hispanic adolescents and almost five times as likely as black adolescents to use these medications. These ethnic/racial differences in medication use follow a similar pattern of use of all medications by children. Much of the Hispanic/white gap may be explained by two-parent families, higher education levels, family income, health insurance, and having a usual source of care. However, the black/white gap may be the result of the way minorities perceive mental health difficulties and use of antidepressants in adolescents. The researchers used national data on sociodemographics, insurance coverage, and health care use from the 2000-2004 Medical Expenditure Panel Survey sponsored by the Agency for Healthcare Research and Quality.

Levtzion-Korach, O., Frankel, A., Alcalai, H., and others (2010, September). "Integrating incident data from five reporting systems to assess patient safety: Making sense of the elephant." (AHRQ grant HS10002). The Joint Commission Journal on Quality and Patient Safety 36(9), pp. 402-409.

Hospitals gather safety-related information through an array of approaches, including incident reporting, patient complaints, risk management, medical malpractice claims, and executive walk rounds. The researchers examined these five systems to assess safety at one large academic hospital with a history of patient safety awareness. Their objectives were to evaluate what type of information is received by each system; develop a common framework for representing the identified safety issues; assess the correlation between types of information collected by the different systems; and evaluate the overall safety picture. They found that communication problems were common among patient complaints and malpractice claims and that clinical judgment was the leading category for malpractice claims. Walk rounds identified safety issues with equipment and supplies. Adverse event reporting systems highlighted event identification issues. Hospitals need to synthesize the messages from all these individual approaches into a collated and cohesive whole, suggest the researchers.

Meltzer, D., Chung, J., Khalili, P., and others (2010). "Exploring the use of social network methods in designing healthcare quality improvement teams." (AHRQ grant HS16967). Social Science and Medicine 71, pp. 1119-1130.

Teams, defined as groups of individuals working interdependently to achieve a shared goal, are an important part of quality improvement efforts in healthcare organizations. To improve the design and construction of such teams, the authors draw upon a large body of research on social networks, which has demonstrated how a person's location within a social network can affect the volume, quality, and timeliness of information to which she has access. The authors next review the essential concepts needed to convey the value of social network analysis (SNA) for guiding team composition. These concepts include clusters, degree, betweenness, and density. Using sociometric, sociodemographic, and professional data from social network surveys, they then describe the structure of interaction among the 71 physicians attending on the general medical services at the University of Chicago Medical Center in 2006-2007. Finally, they conclude that their use of SNA provides actionable insight into design of quality improvement teams.

Rivard, P. E., Elixhauser, A., Christiansen, C. L., and others (2010, June). "Testing the association between patient safety indicators and hospital structural characteristics in VA and nonfederal hospitals." Medical Care Research and Review 67(3), pp. 321-341. Reprints (AHRQ Publication No. 10-R027) are available from the AHRQ Publications Clearinghouse.

Researchers sought to discover associations between hospital structural characteristics (number of staffed patient beds, teaching status, location, and nurse staffing) and the likelihood of potentially preventable patient safety events, such as accidental puncture or postoperative respiratory failure. They looked for associations between the characteristics of Veterans Affairs hospitals and non-Federal hospitals and scores on 12 of the Agency for Healthcare Research and Quality's (AHRQ's) Patient Safety Indicators (PSIs). The study found that, for non-Federal hospitals, being a major teaching hospital was associated with worse safety problems (higher values for 7 of 12 PSIs )—a stronger association than any other structural characteristic tested by the researchers. However, the lack of any consistent relationship between hospital structural characteristics and PSIs may indicate that key structural characteristics were not measured in the study, or it may reflect that PSIs have a stronger relationship to patient characteristics than hospital structural characteristics.

Current as of February 2011
Internet Citation: Research Briefs: Research Activities, February 2011, No. 366. February 2011. Agency for Healthcare Research and Quality, Rockville, MD.