Research Briefs

Bazzoli, G. J., Chen, H-F., Zhao, M., and Lindrooth, R. C. (2008, August). "Hospital financial condition and the quality of patient care." (AHRQ grant HS13094). Health Economics 17(8), pp. 977-995.

The mounting concerns about patient quality of care that were occurring during a period (1995-2000) of sustained hospital financial weakness raised the question of whether patients are harmed by declining hospital financial performance. The researchers examined the relationship between hospital financial condition and patient care. Hospital financial performance was measured by two measures: operating margin and the ratio of cash flow to total revenues. Patient care was measured by patient safety and quality of care measures that were developed from the Agency for Healthcare Research and Quality's Healthcare Cost and Utilization Program State Inpatient Database. The study included more than 1,544 non-federal general acute care hospitals in 11 States. The researchers found that operating margin did not have a significant effect on the incidence of adverse events in the following areas: surgical-related patient safety events, nursing-related patient safety events, and in-house mortality for diagnosis-related groups with typically low mortality rates. However, the poorest and second poorest performing groups (hospitals were divided into quartiles) in relation to cash flow to total revenues did have higher excess incidents in two of these areas.

Berdahl, T. A. (2008, December). "Racial/ethnic and gender differences in individual workplace injury risk trajectories: 1988-1998." American Journal of Public Health 98(12), pp. 2258-2263. Reprints (AHRQ Publication No. 09-R020) are available from the AHRQ Publications Clearinghouse.

Very little is known about how individual workplace injury risk changes across occupations or how racial/ethnic and gender disparities in risk change over time. The researcher estimated individual workplace injury and illness risk over time ("trajectories") for a group of American workers who participated in the National Longitudinal Survey of Youth (1988-1998). The study found that white men had a high risk of injury relative to the other groups (white women, black men, black women, Latino men, Latino women) and experienced the greatest decline over time. Among women, black women had the greatest risk of injury. Workers who moved into jobs requiring more work hours had increased odds of injury. Working in higher-wage jobs did not protect against injury; moreover, unionized jobs and jobs with health insurance were associated with increased injury odds. Finally, environmental hazards were associated with elevated injury risk but the level of physical demand was not associated with a higher risk of physical injury.

Brady, J., Ho, K., and Clancy, C. M. (2008). "State snapshots—A picture of unacceptable variation: Are we destined to live with ‘geography is destiny’?" (2008, November/December). American Journal of Medical Quality 23(6), pp. 492-495.

Variations in the quality of health care are based on region as well as on other factors. The Agency for Healthcare Research and Quality (AHRQ) annually produces a Web tool highlighting state-level data from its quality and disparities reports. The "State Snapshots" generated with this Web tool include information on performance strengths and weaknesses for each state. Analysis of these State-level data provide concrete, actionable information about where to focus quality improvement efforts, according to Carolyn M. Clancy, M.D., AHRQ's director, and colleagues. For example, comparisons of two States (Maryland and Missouri) of similar populations show that Maryland performs weakly in preventive measures and in respiratory disease measures, while Missouri is below average in diabetes and heart disease measures. State-level data in their totality demonstrate two points: even States with comparatively strong overall health care quality have areas that need improvement and attention to variations in quality can help target resources for quality improvement efforts where they are needed most in each State.

To compare how a State fares against regional and national averages, visit AHRQ's State Snapshots Web page at statesnapshots.ahrq.gov.

Chernew, M., and Gibson, T. B. (2008, December). "Cost sharing and HEDIS performance." (AHRQ grant HS10771). Medical Care Research and Review 65(6), pp. 713-728.

Patient cost sharing is among the most commonly used levers to control health care costs. Therefore, efforts should be made to assess the extent to which cost sharing contributes to the failure of patients to receive high quality care. This study is the first to investigate the impact of cost sharing on the most widely used measures of performance contained in the Health Plan Effectiveness Data and Information Set (HEDIS). Specifically, the researchers focus on all of the HEDIS measures related to prescription medications. Their analysis uses the 2000-2003 MarketScan Commercial Claims and Encounters database, which includes 6 million employees with employer-sponsored health insurance each year. Three types of patients were included: those with persistent asthma, those with a diagnosis of major depression, and those discharged from the hospital after a heart attack. Copayments for office visits or prescription drug payments had no appreciable effect on heart disease performance measures. By contrast, office visit copayments but not prescription drug copayments affected the asthma performance measure. The strongest copay effects were found for depression, where a $10 increase in copayments yielded a reduction of 3.6 percent in the performance measure.

Clancy, C. M. (2008, December). "Medicare policy marks new link between hospital payment, patient safety." Journal of Patient Safety 4(4), pp. 215-216.

Under a policy rule that took effect in October 2008, the Centers for Medicare & Medicaid Services will stop paying hospitals the extra costs they incur for eight hospital-acquired conditions whose serious consequences, experts believe, could be prevented. This rule is a much-anticipated change in the Federal government's response to patient safety events. Aligning hospital payments with improved patient safety has also gained steam among private health care purchasers, according to Carolyn M. Clancy, M.D., director of the Agency for Healthcare Research and Quality (AHRQ). A recent study by AHRQ researchers has estimated the costs to insurers of preventable patient safety events among surgical patients during and after a hospital stay. These potentially preventable medical errors could cost employers $1.5 billion a year. Insurers paid an additional $28,218 for surgical patients who experienced acute respiratory failure compared with patients who did not. Medicare's new payment policy demands that we focus our energies on conditions that evidence shows can be drastically reduced.

Conwell, L. J., and Boult, C. (2008). "The effects of complications and comorbidities on the quality of preventive diabetes care: A literature review." (AHRQ grant HS16219). Population Health Management 11(4), pp. 217-228, 2008.

The association between complications and comorbidities and the quality of preventive diabetes care is unclear but it may affect either the treatment of diabetes or other comorbid conditions. This literature review categorizes measures of complications and comorbidities in the studies of the quality of diabetes care, to document whether these studies adjusted for complications and comorbidities and to determine the extent to which these measures are valid for assessing the quality of care delivered to people with diabetes. The review identified and categorized 34 studies in which the quality of diabetes preventive care was assessed with process measures and complications or comorbidities were reported. Because of cross-study variation among measures of complications and comorbidities and because very few studies address the independent effects of complications and comorbidities, the effects of complications and comorbidities on processes of care are unclear. The results of the review suggest that the effects of complications and comorbidities on the delivery of preventive services are complex, only partly understood, and not yet reliably quantified.

Gorelick, M., Scribano, P. V., Stevens, M. W., and others. (2008, November). "Predicting need for hospitalization in acute pediatric asthma." (AHRQ grant HS09825). Pediatric Emergency Care 24(11), pp. 735-744.

Traditionally, children treated for acute asthma on an emergency basis faced two possible dispositions: hospital admission or discharge to home in the care of a parent. Either of these dispositions could be in error, according to earlier studies. Recently, many emergency departments have incorporated short-stay units (SSUs) where patients may receive more intensive care for periods less than 24 hours. The researchers sought to develop a clinical prediction model, using explicit criteria for appropriateness, to assign an accurate disposition (discharge, SSU, or inpatient care) for children with acute asthma. Using only two variables—final clinical score at the time of disposition and number of albuterol treatments given in the emergency department (ED)—the researchers were able to develop a prediction score that predicted accurately those patients who could be discharged successfully to home from the ED without relapse versus those requiring further care in the hospital. The prospective study was of children aged 2 years and older treated at two pediatric EDs for acute asthma.

Haggerty, C. L., Totten, P. A., Astete, S. G., and others. (2008, October). "Failure of cefoxitin and docycycline to eradicate endometrial Mycoplasma genitalium and the consequence for clinical cure of pelvic inflammatory disease." (AHRQ grant HS08358). Sexually Transmitted Infections 84(5), pp. 338-342.

Pelvic inflammatory disease (PID) is associated with the pathogen Mycoplasma genitalium; however, the efficacy of commonly used PID antimicrobials in treating M. genitalium upper genital tract infection is unknown. In the PID Evaluation and Clinical Health Study (PEACH), 682 women treated with cefotoxin and doxycycline for clinically suspected PID had stored cervical and endometrial specimens available for analysis. This study, a substudy of PEACH, is the first to investigate treatment failure among PID patients with M. genitalium identified in the endometrium. This pathogen persisted among 44 percent of women after 30 days despite a standard Centers for Disease Control and Prevention-recommended treatment of cefotoxin and doxycycline. The researchers conclude that M. genitalium is associated with endometritis and short-term PID treatment failure, as evidenced by persistent endometritis and continued pelvic pain.

Halm, E. A., Press, M. J., Tuhrim, S., and others. (2008, November/December). "Does managed care affect quality? Appropriateness, referral patterns, and outcomes of carotid endarterectomy." (AHRQ grant HS09754). American Journal of Medical Quality 23(6), pp. 448-456.

The New York Carotid Artery Surgery (NYCAS) study sought to determine whether Medicare patients enrolled in Medicare managed care plans (Medicare Choice MC plans) had lower rates of carotid endarterectomies (CEAs), operations more frequently performed in high-volume hospitals, or better perioperative outcomes compared with patients enrolled in fee-for-service (FFS) Medicare. The study group consisted of 8,691 Medicare beneficiaries with FFS coverage and 897 with MC coverage. The study found that there were no differences in rates of inappropriate surgery between FFS and MC cases (8.6 percent vs. 8.4 percent). FFS patients were more likely to be operated on in higher volume hospitals, mostly because fewer MC patients were operated on at the highest quintile facilities. Rates of death, nonfatal stroke, and myocardial infarction were similar for the two groups. The researchers concluded that MC plans did not have a positive impact on inappropriateness, referral to high-volume providers, or clinical outcomes.

Le Cook, B., McGuire, T. G., and Zuvekas, S. H. (2009, February). "Measuring trends in racial/ethnic health care disparities." Medical Care Research and Review 66(1), pp. 23-48. Reprints (AHRQ Publication No. 09-R019) are available from the AHRQ Publications Clearinghouse.

This study compares trends in disparities by three definitions of racial/ethnic disparities and assesses the influence of changes in socioeconomic status (SES) among racial/ethnic minorities on disparity trends. The researchers use data (1996-2005) from the Agency for Healthcare Research and Quality's (AHRQ's) Medical Expenditure Panel Survey (MEPS). The three definitions used in the comparison are AHRQ's definition, the Residual Direct Effect (RDE) definition, and the Institute of Medicine's (IOM) definition. The researchers found that black-white disparities in having an office-based or outpatient visit were roughly constant between 1997 and 2005 and Hispanic-white disparities increased for office-based or outpatient visits and for medical expenditure during this period. All three definitions tell basically the same story; however, for most analyses measuring disparities at a point in time, the AHRQ unadjusted approach estimated the largest disparities and the RDE measured the smallest disparities. After discussing empirical differences among the definitions and the debate about the different definitions, the researchers state a preference for the IOM's definition (health care disparities are all differences not due to health status or need) because the definition captures what most researchers and policymakers are concerned about.

Liu, L., Ma, J. Z., and Johnson, B. A. (2008). "A multi-level two-part random effects model, with application to an alcohol-dependence study." (AHRQ grant HS16543). Statistics in Medicine 27, pp. 3528-3539.

The researchers extend the two-part random effects model for clustered semi-continuous data to the multilevel setting. They apply a novel multilevel two-part model to the efficacy trial of topiramate for alcohol-dependent subjects. The estimation and inference are carried out through Gaussian quadrature technique, which is available in free software. The model takes into account the preponderance of zeros as well as the multilevel structure. The efficacy trial was conducted to compare the safety and efficacy of oral topiramate and placebo in a group of 150 alcohol-dependent subjects. When compared with the results of two simple models with respect to the positive number of drinks outcome, the novel model performed better in terms of Akaike information criteria. An important advantage of the new model is that it can be applied to other substance addiction studies because of its efficiency and simplicity.

Mutter, R. L., Wong, H. S., and Goldfarb, M. G. (2008, Fall). "The effects of hospital competition on inpatient quality of care." Inquiry 45, pp. 263-279. Reprints (AHRQ Publication No. 09-R042) are available from the AHRQ Publications Clearinghouse.

Studies thus far have had inconclusive findings about the effects of hospital competition on inpatient quality of care. Applying Quality Indicator software from the Agency for Healthcare Research and Quality to the 1997 Healthcare Cost and Utilization Project State Inpatient Databases, the researchers created 3 versions of 38 measures of inpatient quality. To assess competitiveness, they used 12 different hospital competition measures. The study used data for all patients in 22 States and up to 2,595 hospitals. The findings were not unidirectional. Hospital competition was associated with an improvement in inpatient quality for six Quality Indicators (QIs), including complications of anesthesia; however, it was also associated with a reduction in inpatient quality for six QIs, including decubitis ulcer. In general, the findings at the high and low end of HMO penetration were consistent with each other and with the findings at the mean level of HMO penetration. The exception was iatrogenic pneumothorax for which there were more incidences associated with high managed care penetration; yet, in markets with low managed care penetration, the opposite was true.

Novotny, N. L., and Anderson, M. A. (2008, November/ December). "Prediction of early readmission in medical inpatients using the probability of repeated admission instrument." (AHRQ grant HS15084). Nursing Research 57(6), pp. 406-415.

Up to 25 percent of hospitalized adults experience early readmission, and those within the first few months after discharge are more likely to have been avoidable than later readmissions. Well-developed and validated instruments to predict an individual's risk of readmission are rare. The probability of repeated admission (Pra) instrument incorporates select diagnostic, demographic, and self-rated psychosocial factors and has been shown to be a valid predictor for readmission within 4 years for outpatients older than 70 years. Since there is a need to predict this risk for patients of any age, this study assessed how well the Pra correctly identified and predicted adult medical patients' risk of early readmission. At a Pra score value of .47 or greater, the instrument demonstrates good specificity. In this population, however, the Pra score value needs to be .38 or less to yield satisfactory sensitivity. At this time, the Pra is better than any other known instrument for the purpose of predicting early readmission but it is necessary to develop an instrument that will be more highly predictive of early readmission within a heterogeneous population.

Nyman, J. A., Barleen, N. A., and Kirdruang, P. (2008, November/December). "Quality-adjusted life years lost from nonfatal motor vehicle accident injuries." AHRQ grant HS14097). Medical Decision Making 28, pp. 819-828.

In describing the importance and severity of the health consequences of motor vehicle accident injuries, researchers are turning to quality-adjusted life years (QALYs) as the preferred measure because of their wide acceptance, standardized methodology, and rigorous theoretical origins. QALYs associated with nonfatal injuries are more problematic than those associated with fatalities because they require the researcher to determine the severity of the various injuries and then assign a quality-of-life decrement and a duration for each type of injury. The researchers used 1997-2004 data from the household component of the Medical Expenditures Panel Survey. They found that the QALY decrements associated with a motor vehicle accident injury are 0.0612 QALYs or 0.0360 QALYs, if discounted. The discounted QALY decrement is between 3 and 10 times smaller than the discounted estimates in the literature (0.127 and 0.356). This is probably because the baseline quality of life for calculating the decrement due to injury is not assumed to be 1.00 but rather the actual quality-of-life level before an injury (which averaged 0.865).

O'Malley, A. J., and Zaslavsky, A. M. (2008). "Domain-level covariance analysis for multilevel survey data with structured nonresponse." (AHRQ grant HS09205). Journal of the American Statistical Society 103(484), pp. 1405-1418.

Health care quality surveys such as the Consumer Assessments of Healthcare Providers and Systems (CAHPS®) are administered to individual respondents to evaluate the performance of hospitals and health plans. For a better understanding of dimensions of quality, the researchers analyze relationships among quality measures at the domain (i.e., hospital or health plan) level. They first fit generalized variance-covariance functions that take into account nonresponse patterns in the survey responses, then specify a likelihood function for the domain mean responses using these generalized variance-covariance functions. This allows them to model directly the relationships among domain means for different items. After calculating maximum likelihood estimates using the EM algorithm and sample under Bayesian models using Markov chain Monte Carlo, they perform factor analysis on the estimated or sampled between-domain covariance matrixes. Since this approach accommodates missing data at both the domain and individual levels, it is particularly useful when measures may have extensive data, very little data, or no data in various domains.

Paddock, S. M., and Ebener, P. (2008). "Subjective prior distributions for modeling longitudinal continuous outcomes with non-ignorable dropout." (AHRQ grant HS14805). Statistics in Medicine 28, pp. 659-678.

Substance abuse treatment research is complicated by the pervasive problem of important missing data related to unobserved outcomes. This occurs because some clients leave prior to completion of treatment. Missing data of this type are problematic when the goal is to measure the treatment process and its effects on post-treatment outcomes. To account for such factors, researchers have frequently used pattern-mixture models (PMMs) to jointly model the outcome and the missing data mechanism. Despite the widespread use of PMMs for longitudinal data analysis, no attention has yet been devoted to eliciting prior distributions from subject-matter experts about the identification of the rate of change (slope) parameter for persons who drop out of a study after completing just one assessment. Interviews with five substance abuse clinical experts revealed that their opinions differed dramatically from assumptions widely used to identify parameters in the PMM. The researchers concluded that those who plan to conduct sensitivity analyses using PMMs or selection models should make a serious effort to incorporate expert opinion into the model in order to address concerns about nontestable assumptions.

Powers, B.J., Olsen, M.K., Oddone, E.Z., and others. (2008). "Literacy and blood pressure—do healthcare systems influence this relationship? A cross-sectional study," (AHRQ grant T32 HS00079). BioMed Central Health Services Research 8(219), pp. 1–9.

Limited literacy (reading below the 9th grade level) is associated with poorer health care outcomes. In this study, the researchers found evidence that differences in health care systems may influence the relationship between literacy and blood pressure in primary care patients with hypertension. They analyzed baseline data for 588 patients enrolled in a hypertension control trial within the Veterans Affairs health care system (VAHS) and 636 patients enrolled a hypertension control trial at the Duke University Health System (UHS). Overall, 38.4 percent of the VAHS patients and 27.5 percent of the UHS patients had limited literacy. There was a significant difference only in systolic blood pressure between the two health care settings. Systolic blood pressure for VAHS patients with limited literacy was 1.2 mm Hg lower than for patients with adequate literacy, but patients in UHS with limited literacy had systolic pressure 6.1 mm Hg higher than for patients with adequate literacy. This finding of sensitivity to the patient's health system was not true of diastolic blood pressure or blood pressure control.

Raab, S. S., Grzybicki, D. M., Condel, J. L., and others. (2008). "Effect of Lean method implementation in the histopathology section of an anatomical pathology laboratory." (AHRQ grant HS13321). Journal of Clinical Pathology 61, pp.1193-1199.

The lack of standardization in American anatomical pathology laboratories results in less than optimal quality, inefficiencies, and increased health care costs. The researchers sought to measure the effects of a Lean quality improvement process on the efficiency and quality of a histopathology section of an anatomical pathology laboratory. They selected a Lean process known as Perfecting Patient Care (PPC). The setting for the study was a large urban hospital in Pittsburgh. One efficiency metric used was specimen turnaround time (TAT), defined as the time from when the gross examination was complete to the time when the slides from a case were verified and sent to a pathologist. The other efficiency metric was productivity, defined as the total number of work units (i.e., tissue blocks and slides) divided by the number of personnel full-time equivalents. The study found that the implementation of Lean processes decreased specimen TAT and increased productivity.

Schiff, G.D., and Galanter, W.L. (2009). "Promoting more conservative prescribing." (AHRQ grant HS16973). Journal of the American Medical Association 301(8), pp. 865–867.

In this commentary, the authors set out the reasons for and a set of principles to guide more conservative prescribing by clinicians. Such guiding principles, they say, can help reduce the prevalence of medication-related harm to patients and inappropriate prescribing. The commentary offers 25 principles to enlighten the thinking of clinicians-in-training about pharmacotherapy to emphasize care, caution, and reliance on evidence-based information. The authors organize their principles into six categories: (1) think beyond drugs; (2) practice more strategic prescribing; (3) heighten vigilance regarding adverse effects; (4) act with caution and skepticism regarding new drugs; (5) share your agenda with patients; and (6) weigh long-term, broader aspects of care. The authors then discuss barriers to conservative prescribing, such as time pressure, patient expectations, and the biases built into industry-funded research studies and educational programs.

Selim, A. J., Rogers, W., Fleishman, J. A., and others. (2008). "Updated U.S. population standard for the Veterans RAND 12-item Health Survey (VR-12)." Quality of Life Research 18, pp. 43-52. Reprints (AHRQ Publication No. 09-R039) are available from the AHRQ Publications Clearinghouse.

The Veterans RAND 12-item Health Survey (VR-12) is a Health-Related Quality of Life (HRQoL) survey instrument used in assessments of quality improvement activities and health care system accountability. The researchers' objective was to update the nonproprietary 1990 scoring algorithms for scoring the VR-12. To do this, they used data collected between 2000 and 2002 by the Agency for Healthcare Research and Quality's Medical Expenditure Panel Survey (MEPS). They found that changes in the U.S. population between 1990 and today make the old standards obsolete for the VR-12. The researchers present a robust method for deriving physical and mental summary scores for the VR-12 based upon an updated standard from the U.S. population. This standard has a contemporary mean of 50 in the general U.S. population, as represented in the MEPS 2000-2002. The updated standard made available here is widely available to serve as a contemporary standard for future applications for HRQoL assessments.

Short, V. L., Totten, P. A., Ness, R. B., and others. (2009, January 1). "Clinical presentation of Mycoplasma genitalium infection versus Neisseria gonorrhoeae infection among women with pelvic inflammatory disease." (AHRQ grant HS08358). Clinical Infectious Diseases 48(1), pp. 41-47.

The pathogens Neisseria gonorrhoeae and Chlamydia trachomatis cause 30 to 50 percent of cases of pelvic inflammatory disease (PID) in women. Mycoplasma genitalium may be at the root of PID cases that are neither gonococcal nor nonchlamydial. Left untreated, PID caused by M. genitalium can lead to infertility, ectopic pregnancies, and chronic pain. Researchers evaluated 722 women enrolled in a PID study to compare markers of PID caused by M. genitalium with those of N. gonorrhoeae. They found that the former's markers were closer to chlamydial infection, which tends to have no symptoms. In contrast, women whose PID was caused by N. gonorrhoeae had symptoms such as high pelvic pain scores and inflammation markers including elevated oral temperatures and elevated white blood counts.

Wang, H. E., Marroquin, O. C., and Smith, K. J. (2009, February). "Direct paramedic transport of acute myocardial infarction patients to percutaneous coronary intervention centers: A decision analysis." (AHRQ grant HS13628). Annals of Emergency Medicine 53(2), pp. 233-240.

For patients with acute ST-segment elevation myocardial infarction (STEMI), consensus guidelines recommend rapid primary percutaneous coronary intervention. In following this guideline, emergency medical services (EMS) may bypass nearby community hospitals that offer fibrinolytic therapy in favor of transporting the patient to a more distant specialty center able to perform primary percutaneous coronary intervention. There are many factors to be considered, including travel time to the nearest percutaneous coronary intervention center, the expected survival benefit of each STEMI treatment option, the anticipated benefit decay with elapsed time, and the uncertainties in total treatment time. Using decision analysis based on parameter values from meta-analyses and North American clinical studies of STEMI and chest pain care published after 2001, the researchers found that 30-day survival rates were slightly higher for standard percutaneous coronary intervention when compared with standard community hospital fibrinolytic therapy. However, the survival rates were slightly lower when compared with best-case community hospital fibrinolytic therapy.

Wilt, T. J., Brawer, M. K., Barry, M. J., and others. (2009, January). "The prostate cancer intervention versus observation trial: VA/NCI/AHRQ cooperative studies program #407 (PIVOT): Design and baseline results of a randomized controlled trial comparing radical prostatectomy to watchful waiting for men with clinically localized prostate cancer." Contemporary Clinical Trials 30(1), pp. 81-87.

Undergoing surgery or radiation therapy to treat prostate cancer can leave men with bladder or erectile problems. As a result, some physicians and patients instead choose to monitor the cancer's progression, an option called watchful waiting. Researchers from the U.S. Department of Veterans Affairs, the National Cancer Institute, and the Agency for Healthcare Research and Quality are currently conducting a study comparing the options of watchful waiting and surgery to remove the prostate gland (radical prostatectomy). Called the Prostate Cancer Intervention Versus Observation Trial (PIVOT), the study enrolled 731 men from 1994 to 2002 whose prostate cancer was detected with the prostate specific antigen blood test. Almost a third of the men are black, which is significant because these men are at high risk of getting and dying from prostate cancer. Researchers hope to determine whether early surgical intervention improves the length and quality of life compared with watchful waiting combined with noncurative therapies. This article summarizes the study's rationale, design, recruitment, and enrollee characteristics. Study results are expected in mid 2010.

Wong, C., Mouanoutoua, V., and Chen, M.-J. (2008). "Engaging community in the quality of hypertension care project with Hmong Americans." (AHRQ grant HS110276). Journal of Cultural Diversity 15(1), pp. 30–36.

The culture of a minority community can pose tremendous challenges to researchers striving to improve the quality of health care in these communities. Because of its history of refugee status, low proportion of English speakers, and cultural beliefs, the Asian Hmong community in central California has been cautious about its involvement with health care institutions. In this study, the researchers used collaboration with leaders of the Hmong community to develop a community-sensitive survey on hypertension care. They developed questions in English and translated them into the two main Hmong dialects. After checking by independent back-translation into English, the survey questions were discussed with a group of Hmong community leaders. The survey was then presented to Hmong focus group participants to remove ambiguity or to point out areas of cultural sensitivity. For example, a question whether "your doctor [told] you about what side effects the blood pressure medicine might have?" was dropped because of the lack of a concept equivalent to side effects in Hmong. Based on input from these community members, the researchers were able to develop culturally and linguistically appropriate survey instruments while allowing the Hmong community to effectively voice its own health care needs.

Unruh, L., Russo, L., Jiang, H. J., and Stocks, C. (2009, February). "Can state databases be used to develop a national, standardized hospital nursing staffing database?" Western Journal of Nursing Research 31(1):66-88. Reprints (AHRQ Publication No. 09-R040) are available from the AHRQ Publications Clearinghouse.

Reliable data on hospital nurse staffing are difficult to find and researchers are asking for more valid and reliable national data. To respond to this need, the researchers conducted a State-by-State review of data reporting systems. They found that at least 25 States collect nurse staffing data; however, detailed information is not available from all of them. Only 12 States meet the availability, completeness, and usability criteria for a Level 1 database. Level 2 databases must meet further data quality and specificity criteria, which are to have separately delineated registered nurse (RN) and licensed practical nurse staffing categories measured in full-time employee-calculable numbers, with clearly delineated units of nursing work, levels of measurement, time frames, and nursing roles included in the RN measure. Only five States (Arizona, Pennsylvania, Tennessee, Virginia, and West Virginia) meet these more rigorous criteria. It is from these five States that a Level 1, State-by-State database could begin to be built.

Return to Contents
Proceed to Next Article