Testing for Survey Attrition Effects on Chronic Disease Care Cost Estimates in a National Longitudinal Medical Expenditure Survey
*Steven B Cohen, AHRQ
Keywords: MEPS, chronic conditions, expenditures, attrition
With health care expenses accounting for ~16% of the nation's GDP, the need for accurate and high quality national survey estimates is clear. Furthermore, medical care expenses are highly concentrated among a relatively small proportion of individuals in the population. The Medical Expenditure Panel Survey is the core health care expenditure survey in the US that serves as the primary source for these essential health care expenditures estimates. In this study, the impact of nonresponse attributable to survey attrition is assessed with respect to the national medical care expenditure estimates derived from the MEPS. Particular attention is given to assessing the impact of attrition on estimates of the medical expenditures for individuals with chronic conditions. The overlapping panel design of the MEPS and its longitudinal features are particularly well suited to inform these analyses.