The Current Population Survey (CPS) is the only source for historical data for national and state estimates of the uninsured in the United States and is relied upon by national and state health policy analysts. Many changes have occurred over the 20 years it has been in use. Among the adjustments made over time for known consistency problems:
- Introduction of a question to verify uninsurance in 2000 that resulted in 8 percent uninsured becoming insured.
- Changes in how individuals with Indian Health Services were classified, moving them from insured to uninsured in 1996 and 1997.
- Improper coding in 2001 of State Children Health Insurance Program-covered children as privately insured.
- How some missing values were coded.
In order to harmonize the years of estimates and understand coverage trends, these researchers enhanced individual-level microdata and compared rates of coverage to the non-enhanced CPS. Overall, the enhancements resulted in a 1 percentage point increase in the number of insured. Notably, with the enhancements, children had lower rates of private coverage than adults, somewhat offset by increases in public coverage.
As changes are made as part of continuous improvement of CPS data measuring health insurance coverage, it is important for researchers to continue to harmonize new and historical data.