Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner


This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Glauber, J.H. (2001, June). "Does the HEDIS asthma measure go far enough?" (AHRQ National Research Service Award training grant T32 HS00063). American Journal of Managed Care 7(6), pp. 575-579.

The asthma measure of the Health Plan Employer Data and Information Set (HEDIS) 2000 may lead to overlooking important dimensions of quality, risking unintended negative consequences on the overall quality of asthma care, according to this author. HEDIS 2000 measures the percentage of individuals who meet a claims-based definition of persistent asthma and receive at least one controller medication (for example, inhaled corticosteroids or cromolyn sodium) in the measurement year. It also identifies those with persistent asthma by such measures as at least one emergency department visit or hospitalization for asthma during the prior year, dispensing of asthma medication on at least four occasions, or at least four outpatient asthma visits and at least two asthma medication dispensing events. This new HEDIS asthma measure may be setting the bar too low and encouraging more casual prescribing of controller medications, rather than encouraging the more painstaking disease severity assessment that would define the need for, and level of, controller use, concludes the author.

Jarvik, J.G. (2001, April). "Fundamentals of clinical research for radiologists." (AHRQ grant HS09499). American Journal of Radiology 176, pp. 873-878.

This author asserts that randomized trials focusing on patient outcomes are the only way to investigate the efficacy of diagnostic technologies such as x-rays with absolute assurance that bias is being avoided. Such trials should be conducted when the stakes are high enough. However, three other types of studies can be quite powerful in their own right and, because they are simpler and less expensive, they should be used in certain situations, depending on their relative advantages and disadvantages. Case-control studies are particularly useful for examining rare outcomes because subjects are selected on the basis of having the outcome of interest. Conversely, cohort studies are useful for rare risk factors because subjects are chosen on the basis of having a particular exposure. Measuring all variables at a single time is the distinguishing characteristic of cross-sectional studies, for example, magnetic resonance imaging studies in patients with low back pain. Although cross-sectional studies are relatively easy to perform, it is often impossible to determine if the exposure preceded the disease or vice-versa.

Localio, A.R., Berlin, J.A., Ten Have, T.R., and Kimmel, S.E. (2001, July). "Adjustments for center in multicenter studies: An overview." (AHRQ grant HS10399). Annals of Internal Medicine 135(2), pp. 112-123.

These researchers suggest adjustments for centers in multicenter studies to account for the possible confounding effects of the centers themselves when treatments are administered across several centers. They point out that although convenient and expedient, the simple pooling of data in multicenter studies as if they arose from a single population can produce incorrect results. The researchers suggest, for example, that the correlation or clustering resulting from the similarity of outcomes among patients within a center requires an adjustment to confidence intervals and P values, especially in observational studies and in randomized multicenter studies in which treatment is allocated by center rather than by individual patient. Multicenter designs also warrant testing and adjustment for the potential bias of confounding by center and for the presence of effect modification or interaction by center. Patient populations at different centers might not react the same way, perhaps because of unmeasured population or environmental factors or variation in adherence to protocols. This possibility increases in observational studies and meta-analyses, in which exposures or protocols are likely to vary across centers.

Zhou, X-H., Li, C., Gao, S., and Tierney, W.M. (2001). "Methods for testing equality of means of health care costs in a paired design study." (AHRQ grants HS90217 and HS09083). Statistics in Medicine 20, pp. 1703-1720.

The authors of this paper propose five new tests for the equality of paired means of health care costs. The first two tests are the parametric tests—a Z-score test and likelihood ratio test—both derived under the bivariate normality assumption for the log-transformed costs. The third test (Z-score with jackknife) is a semi-parametric Z-score method, which only requires marginal log-normal assumptions. The fourth and fifth tests are the nonparametric bootstrap tests, one based on a t-test statistic and the other based on Johnson's modified t-test statistic. The authors describe the results of a simulation study they conducted to compare the performance of these tests, along with some commonly used tests when the sample size is small to moderate.

Return to Contents

Current as of September 2001
AHRQ Publication No. 01-0046

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care