Implications of the Accuracy of MEPS Prescription Drug Data for Health Services Research

Article excerpt

This paper assesses the quality of the Medical Expenditure Panel Survey (MEPS) drug data and the impact that misreporting prescription drug data has on descriptive and behavioral analyses. It does this by matching MEPS participants with Medicare Part D coverage during the period 2006-2007 to their Part D claims data. In the validation sample, the number of drug fills and total expenditures are reasonably accurate compared with claims. Household respondents tended to underreport the number of different drugs taken, but tended to overreport the number of fills of each drug. Behavioral analyses of the determinants of medication use and expenditures were largely unaffected because underreporting cut across most sociodemographic groups.

**********

The Medical Expenditure Panel Survey (MEPS) is a unique, nationally representative source of micro data on health care use and expenditures for all payers. Household respondents report drugs and the number of times each drug was obtained, while follow-back surveys of pharmacies are the primary source of price and expenditure data. The validity of the data is critical because the MEPS is widely used for national estimates, behavioral modeling, and policy simulations, including many analyses of prescription drug markets.

Previous studies have found that MEPS respondents reported inpatient hospital stays well, but tended to underreport ambulatory services (office, hospital outpatient department, and emergency department visits) (Zuvekas and Olin 2009a, 2009b). A validation study of the Medicare Current Beneficiary Survey (MCBS) found that Part D expenditures were underreported (Centers for Medicare and Medicaid Services [CMS] 2010). However, studies of the MEPS also suggest that biases introduced into behavioral and distributional analyses are likely small, and that simple adjustment strategies can correct effectively for underreporting.

We extend these previous studies to validate prescription drug use reported by household survey respondents and expenditures reported by pharmacies, using Medicare beneficiaries in the 2006 and 2007 MEPS who were linked to their Medicare administrative records. The analytic sample contains 1,271 observations of MEPS sample individuals who reported Medicare coverage and who had Medicare Part D drug coverage in the linked administrative data set for an entire year.

The Part D program began on January 1, 2006, and had 28 million enrolled Medicare beneficiaries in April 2010. Beneficiaries may obtain Part D coverage through either a prescription drug plan (PDP) or a Medicare Advantage plan. Medicare beneficiaries who also have Medicaid ("dual eligibles") are automatically enrolled in a PDP. Enrollment is voluntary for other Medicare beneficiaries.

Benefits for 2009 were budgeted at $53 billion.

The initial analyses of this sample assess: 1) the concordance between the number of drugs and prescription drug fills reported by the household in the MEPS and the numbers in the administrative data, and 2) the concordance between MEPS total drug expenditures and expenditures found in the Part D claims. We further investigate the personal and interview characteristics (such as self versus proxy response and interview language) associated with the level of concordance.

Next, we investigate whether reporting errors in the MEPS lead to systematic biases in behavioral analyses by estimating pairs of drug use regressions employing the claims and MEPS household-reported use measures, respectively, as the dependent variable and comparing the results. We formally test whether the marginal effects of each covariate are the same in the pairs of regressions. For example, does poor health increase drug spending by the same magnitude whether using the household-reported or claims-based measure?

Validation Literature

While the accuracy of other forms of service use in the MEPS has been the subject of validation studies, the accuracy of MEPS prescription drug data has been assessed by comparing aggregate estimates from the MEPS with other sources. …