1976, 1980, 1982-1984, 1989, 1993-1996

• Many people do not carry health insurance for various reasons. Hand Card Which of these statements describes why -- is not covered by any health insurance (or Medicare)? Any other reason? (Anything else?)Circle all responses given.

• Which of these are reasons {you/subject's name} stopped being covered or do not have health insurance? Show card

• 2000-2018 *Enter up to 5 reasons.

• Are you currently uninsured because coverage is not affordable?

The related variables from the original NHIS public use files include:

• HINOREXPEN (Reasons for no insurance: Too expensive)

• HISTOP10A (Why coverage stopped: Insurance plan raised premium cost)

• HISTOP11 (Why coverage stopped: Cost is too high)

HINOCOSTR excludes persons with health care coverage through public programs, such as Medicaid or Military health care, whereas the variables listed above include such individuals for some years. For example, for 1993 to 1996, individuals could respond that a reason for "no insurance" was coverage through a public program. Unlike the original public use variables, HINCOSTR also excludes all those covered by AFDC or SSI, which improves comparability between earlier years and 1997 forward.

Note that those with coverage only through single service plans are not included in the universe for 1993 to 1996 (i.e., they were considered to have insurance and hence not asked reasons for no insurance). For 1997 only, the universe includes only those without insurance for less than 3 years.

As noted in the variable description, HINOCOSTR includes people who selected the category "too expensive, can't afford health insurance" for 1976-1996, the category "insurance plan raised cost of premiums" in 1997, the category "cost is too high" for 1998-2018, and the category "coverage is not affordable" for 2019 forward. Because these options all imply a cost barrier, these slight differences in question wording should not greatly affect comparability across years.

For 1997 forward, whether people were classified as uninsured (and thus included in the universe for HINOCOSTR) was affected by back-editing carried out by the NCHS staff. (See the variable description for HINOTCOVE for a description of this back-editing process.) Along with finding errors in the type of insurance coverage reported, the NCHS staff determined that some persons who were reported as lacking insurance were actually covered, while others were reported as insured but actually lacked coverage (e.g., had single service coverage only).

For 1997 forward, data for HINOCOSTR were back-edited to correctly classify peoples' insurance coverage status, while the pre-1997 data were not back-edited. Users may thus want to exercise caution when comparing results from before and after 1997 for HINOCOSTR.