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

Research Briefs

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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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.

Gardiner, J.C., Bradley, C.J., and Huebner, M. (2000). "The cost-effectiveness ratio in the analysis of health care programs." (AHRQ grant HS09514). In P.K. Sen, and C.R. Rao, Eds., Handbook of Statistics 18, pp. 841-869.

This book chapter presents an overview of the statistical estimation of cost-effectiveness ratios (CERs) in the economic evaluation of health care interventions. The CER is the ratio of the net difference in the costs of two interventions to the net difference in their effectiveness. The CER is a useful aid to decisionmaking for policymakers faced with the allocation of health care dollars across several competing interventions. Because the CER is assessed from inputs on costs and effects that are subject to variation, sensitivity analyses are used to assess the extent of the uncertainty in the CER. These authors describe four methods for constructing confidence intervals for CERs and comparing their properties.

Gardiner, J.C., Huebner, M., Jetton, J., and Bradley, C.J. (2000). "Power and sample size assessments for tests of hypotheses on cost-effectiveness ratios." (AHRQ grant HS09514). Health Economics 9, pp. 227-234.

The cost-effectiveness ratio (CER) is an important summary statistic for comparing the costs and effectiveness of competing interventions. These authors constructed tests of hypotheses on the CER from the net cost and incremental effectiveness measures. They also constructed a test of the joint hypothesis of cost-effectiveness and effectiveness and derived an expression connecting power and sample size. Their methods accounted for the correlation between cost and effectiveness and led to smaller sample size requirements than comparative methods that ignored the correlation. Compared with trials designed to demonstrate effectiveness alone, their results indicate that a trial appropriately powered to demonstrate cost-effectiveness might require sample sizes many times greater.

Samsa, G.P., and Matchar, D.B. (2001, March). "Have randomized controlled trials of neuroprotective drugs been underpowered?" (AHRQ PORT contract 290-91-0028). Stroke 32, pp. 669-674.

The use of neuroprotective drugs to treat acute ischemic stroke has been supported by animal studies and phase I and phase II trials. However, the results of phase III studies have been consistently disappointing. This may be because these phase III trials were too small to have the statistical power to detect clinically meaningful effects of these drugs, according to this study. The researchers used computer simulations to calculate the relationship among true outcome rates, assumed outcome rates, and statistical power. They used as examples the results of trials of neuroprotective agents that were published in the journal Stroke from 1996 to 2000. They calculated that even a 2 percent overestimate of the efficacy of an intervention could lead to a serious reduction in statistical power, and that the use of data from phase II studies tends to lead to such overestimation. The researchers recommend placing more emphasis on minimum clinically important differences when planning stroke trials. Even small benefits, when averaged over a sufficiently large number of cases, will accrue to a large positive impact on public health.

Wong, H.S., and Hellinger, F.J. (2001, April). "Conducting research on the Medicare market: The need for better data and methods." Health Services Research 36(1), pp. 291-308.

The Medicare insurance program is experimenting with different payment policies—including implementation of the Health Care Financing Administration's new risk-adjusted principal inpatient diagnostic cost group payment system—to pay insurance plans that enter into a risk-based contract. If Medicare does not appropriately adjust for the health status of extremely ill beneficiaries, insurance plans that enroll a large share of these patients will be underpaid. On the other hand, if Medicare payment policy overpays insurance plans because their Medicare enrollees are healthier, total Medicare costs would be higher than they would be without the new payment system. Nearly all existing studies on these issues have been hampered by a lack of complete data and difficult methodological issues. These researchers highlight existing data limitations, the need for improved and complete data and better analytical methods, and the need to use alternative data sources to conduct Medicare-related research to better evaluate Medicare policies. They introduce a new approach that combines hospital discharge data, State inpatient data, and managed care market penetration data to create an analytic database to assess competition, risk selection, and costs in Medicare HMOs.

Reprints (AHRQ Publication No. 01-R060) are available from the AHRQ Publications Clearinghouse.

Return to Contents

Current as of June 2001
AHRQ Publication No. 01-0033

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


AHRQ Advancing Excellence in Health Care