How should we take care of the people who have positive Medicaid spending but not with Medicaid coverage

Hello there,

When working with the MEPS data, I noticed that there are some inconsistency between the MEPS Medicaid/Medicare coverage and expenditures paid by Medicaid/Medicare. I guess it makes sense for us to assume that people who reported on Medicaid/Medicare can have 0 those expenditures. But how about the people who have positive expenditures paid by Medicaid/Medicare but did not report to have Medicaid/Medicare coverage? How do people usually take care of those cases? Any suggestions? Thank you!

The IPUMS MEPS user note on expenditure definition and measurement notes that:

Missing data for events where HC data were not complete and MPC data were not collected or complete were derived through an imputation process. A series of logical edits were applied to both the HC and MPC data to correct for several problems including, but not limited to, outliers, copayments or charges reported as total payments, and reimbursed amounts that were reported as out-of-pocket payments. In addition, edits were implemented to correct for misclassifications between Medicare and Medicaid and between Medicare HMOs and private HMOs as payment sources. Data were not edited to insure complete consistency between health insurance and source of payment variables on the file.

While these non-zero expenditures paid by Medicare and Medicaid for persons who do not report the corresponding insurance coverage do seem suspect, the documentation indicates that this is not wholly unexpected based on editing procedures. I will also note that the vast majority of persons who do not report Medicare or Medicaid coverage report $0 of Medicare or Medicaid expenditures, respectively.

How to handle these cases is up to each individual researcher. I suggest running your analysis both with and without them and comparing your results.

1 Like

Thank you for the information Kari. That makes sense. I will try different ways when working with the data!

1 Like