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

Research Methodology

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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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

AHRQ researchers explore ways to improve the accuracy and design of health care surveys

The accuracy and quality of a health care survey are invariably linked to the quality of the survey design, its ability to achieve targeted response rates and precision, and control over sources of survey error. The Medical Expenditure Panel Survey (MEPS), sponsored by the Agency for Healthcare Research and Quality, was designed to produce national and regional estimates of health care use, expenditures, sources of payment, and insurance coverage of the U.S. civilian noninstitutionalized population.

The scope and depth of data collected in this survey reflect the needs of government agencies, legislative bodies, and health professionals for comprehensive national estimates that can be used in the formulation and analysis of national health care policies. Researchers in AHRQ's Center for Cost and Financing Studies recently published four articles in a special issue of the Journal of Economic and Social Measurement. The articles examine ways to improve the design of health care surveys, as well as enhancements and innovations that characterize the current MEPS design. The studies are summarized here.

Cohen, S.B., and Machlin, S.R. (2000). "Survey attrition considerations in the Medical Expenditure Panel Survey." Journal of Economic and Social Measurement 26, pp. 83-98.

In this study, the authors identify the characteristics that distinguish MEPS participants across waves of the survey from those that only participated in initial rounds before discontinuing survey participation (nonrespondents). They also examine the impact of survey attrition (nonrespondents) on resultant survey estimates of health insurance coverage. The findings described in this article provide insights into the efficacy of the MEPS nonresponse adjustment strategies by comparing the survey estimates from the second year of the longitudinal panel with those from a new panel for the same time period. The researchers conclude that the data collection and estimation strategies that were implemented to mitigate the impact of nonresponse bias associated with survey attrition in the MEPS should serve as an effective model for other national surveys.

Reprints (AHRQ Publication No. 01-R057) are available from the AHRQ Publications Clearinghouse.

Cohen, S.B., and Yu, W.W. (2000). "The impact of alternative sample allocation schemes on the precision of survey estimates derived from the National Medical Expenditure Panel Survey." Journal of Economic and Social Measurement 26, pp. 111-128.

The 1996 MEPS sample consisted of 195 primary sampling units (PSUs), which contained 10,597 responding households. These researchers compared the precision and cost of survey estimates derived from a 195 PSU design with precision results for alternative sample allocation schemes that preserved the number of sample respondents and the over-sampling of minorities, while varying the number of PSUs and segments. The results provide insights on the impact of alternative sample allocation schemes on the precision of national health care estimates. The authors conclude with a discussion of trade-offs between cost and precision in a national survey with a geographically dispersed multistage sample design.

Reprints (AHRQ Publication No. 01-R056) are available from the AHRQ Publications Clearinghouse.

Winglee, M., Valliant, R., Brick, J.M., and Machlin, S. (2000). "Probability matching of medical events." Journal of Economic and Social Measurement 26, pp. 129-140.

This paper addresses sources of measurement error in household reports of medical care use and expenses and identifies methods to reduce error in match rates between household and medical provider event level data covering the same set of individuals. The medical events reported by household members and providers are subject to reporting differences. Thus, probability linkage methods are used to determine if pairs of medical events represent the same entities. The researchers used three approaches to provide estimates of linkage errors: manual reviews, cumulative weight curves, and simulation approaches. The linked events have been used to help handle missing data and to adjust for household response errors when estimating U.S. medical expenditures.

Reprints (AHRQ Publication No. 01-R063) are available from the AHRQ Publications Clearinghouse.

Machlin, S.R., Cohen, J.W., and Thorpe, J.M. (2000). "Measuring inpatient care use in the United States: A comparison across five Federal data sources." Journal of Economic and Social Measurement 26, pp. 141-151.

The U.S. Department of Health and Human Services currently sponsors five data collection efforts that can be used to estimate the use of inpatient hospital care in the United States. These authors used these five Federal data sources to compare estimates of the use of inpatient care in 1996. They found that surveys with similar target populations and methodologies produced similar estimates. Hospital discharge surveys produced substantially higher estimates of total discharges than household surveys. This could be attributed in part to differences in target populations or underestimates from household surveys. The authors stress the need to ensure standardization in target populations and underlying units of measurement when comparing data sources.

Reprints (AHRQ Publication No. 01-R059) are available from the AHRQ Publications Clearinghouse.

Return to Contents
Proceed to Next Article

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


AHRQ Advancing Excellence in Health Care