Informing Healthcare Policy on Insurance Coverage: Issues of Data Availability, Analytical Capacity and Statistical Quality

National estimates of the population's health insurance status are essential inputs to policymakers to inform assessments of the population's access to medical care and analyses of associated health care expenditures. There are several national surveys used by policy makers to obtain national estimates of health insurance coverage, each with unique capacities and analytic strengths. While all these surveys focus on similar dimensions of health insurance status, there is considerable variation in coverage estimates. These differences have contributed to a greater awareness of the need to better understand the design and content features of these surveys. Several factors contribute to differences in estimates of health insurance status measures across surveys. These factors include survey content and questionnaire design, definitions of coverage, survey methods, sample design, post survey data processing, and estimation techniques. Survey design features such as mode of administration, length of recall period, sample design and response rates also affect the accuracy and precision of survey estimates of coverage. In addition, estimates within and across surveys differ depending on the duration of the time period that the survey estimates cover. Gaining a better understanding of each of these characteristics of the respective surveys is critical to helping policymakers make the best use of statistics that ostensibly differ. This presentation focuses on the issues of data availability, analytical capacity and statistical quality to inform health policy. Particular attention will be given to the capacity of the MEPS and other surveys to support these efforts to advance analytical capability and statistical integrity.