Suchergebnisse

The authors validate reported Part D coverage in the Medical Expenditure Panel Survey (MEPS) and assess the impact of misreporting on descriptive and behavioral analyses. MEPS participants with Medicare coverage during 2006 to 2007 were matched to Medicare administrative data. A summary measure of Part D coverage based on several questions has substantial validity (κ = .70) and an agreement rate of 85.1%. Some beneficiaries confused Part D and private drug coverage, leading to both under- and overreported Part D coverage. Accuracy varies little by sociodemographic group. Standard regression models of the determinants of Part D enrollment were estimated with both MEPS-based and administrative data–based measures of Part D enrollment. In this analysis, the signs of the marginal effects were the same, the magnitudes were similar, and mostly the same variables had statistically significant effects in both regressions. Thus, behavioral analyses are largely unaffected by misreporting.

We explore the determinants of usage of six different types of health care services, using the Medical Expenditure Panel Survey data, years 1996-2000. We apply a number of models for univariate count data, including semiparametric, semi-nonparametric and finite mixture models.
We find that the complexity of the model that is required to fit the data well depends upon the way in which the data is pooled across sexes and over time, and upon the characteristics of the usage measure. Pooling across time and sexes is almost always favored, but when more heterogeneous data is pooled it is often the case that a more complex statistical model is required.

Abstract
It is well established that survey respondents imperfectly recall health care use in surveys. However, careful attention to both survey design and fielding procedures can enhance recall. We examine the effects of a comprehensive, multi-pronged approach to changing field procedures in the Medical Expenditure Panel Survey (MEPS) to improve quality of health care use reporting. Conducted annually since 1996, the MEPS is the leading large-scale nationally representative health survey with detailed individual and household information on health care use and expenditures. These survey enhancements were undertaken in 2013–2014 because of concerns over a drop in the quality of reporting in 2010 that persisted into 2011–2012. The approach combined focused retraining of field supervisors and interviewers, developing quality metrics and reports for ongoing monitoring of interviewers, and revising advanced letters and materials sent to respondents.
We seek to determine the extent to which changes in field procedures and trainings improved interviewer and respondent behaviors associated with better reporting, and more importantly, improved reporting accuracy. We use longitudinal MEPS data from 2008 through 2015, combining household reported use with sociodemographic and health status characteristics, and paradata on the characteristics of the interviews and interviewers. We exploit the longitudinal data and timings of major trainings and changes in field procedures in regression models, separating out the effects of the trainings and other fielding changes to the extent possible.
We find that the 2013–2014 data quality improvement activities substantially improved reporting quality. Positive interviewer behaviors increased substantially to above pre-2010 levels, and utilization reporting has recovered to above pre-2010 levels, returning MEPS to trend. Importantly, these substantial gains occurred in 2013, prior to extensive in-person training for most of the field force.
We examine the lessons learned from this data quality initiative both for the MEPS program and for other large household surveys.

Survey nonsampling errors refer to the components of total survey error (TSE) that result from failures in data collection and processing procedures. Evaluating nonsampling errors can lead to a better understanding of their sources, which in turn, can inform survey inference and assist in the design of future surveys. Data collected via supplemental questionnaires can provide a means for evaluating nonsampling errors because it may provide additional information on survey nonrespondents and/or measurements of the same concept over repeated trials on the same sampling unit. We used a supplemental questionnaire administered to cancer survivors to explore potential nonsampling errors, focusing primarily on nonresponse and measurement/specification errors. We discuss the implications of our findings in the context of the TSE paradigm and identify areas for future research.

Sampling and reporting biases in the Medical Expenditure Panel Survey (MEPS) survey could render data on inpatient utilization that are not representative for individuals with severe psychiatric conditions. The authors assessed the representativeness of MEPS data on psychiatric inpatient utilization, by comparing MEPS estimates of total annual psychiatric and nonpsychiatric inpatient admissions and bed days, and mean length of stay, for nonelderly U.S. adults in calendar years 2005 to 2010 ( N = 9,288) to estimates from the Nationwide Inpatient Sample (NIS), a nationally representative inpatient care database derived from hospitals' administrative records ( N = 21,934,378). Compared with the NIS, the MEPS indicated 34% as many psychiatric admissions and 86% as many nonpsychiatric admissions, while mean psychiatric length of stay was greater in MEPS than in NIS. In MEPS data, underrepresentation of psychiatric inpatient utilization at community hospitals may result in measurement distortions for commonly used statistics on psychiatric inpatient utilization and costs.