Description

For persons in families with private insurance, HI1WORK indicates whether the first insurance plan reported was originally obtained through work (i.e., through a present or former employer or union). Beginning in 1995, HI1WORK also included persons in families whose first-mentioned plan was obtained through self-employment or professional associations.

Insurance "through" an employer refers to employer-sponsored insurance (ESI) plans that are offered as a benefit of employment. Employers generally contribute to monthly plan premiums, which lowers the cost for the individual. Employer-sponsored coverage is often much more affordable than plans bought directly by an individual.

Definition of Private Coverage and Data Collection Process

Prior to 1993 to 1996, single service plans were not considered private insurance coverage and thus are excluded from HI1WORK. For 1993 to 1996, persons in families with single service plans are included in the universe for this variable. Users can select those who had comprehensive coverage by using the variable HI1VARSS, which indicates whether the plan covered a variety of services or only a single service.

The definition, order and wording of private health insurance changed over time to reflect changes in the availability of different types of coverage and issues of interest. Users are strongly encouraged to review the user notes Insurance Data Collection and Private Insurance Definitions.

Prior to 1997, details about the characteristics of insurance plans (such as HI1WORK) reflect plans for any family member in the household. In order to determine if the person (rather than someone else in the family) was covered by this plan, analysts should use the variable HI1PCOV (Health Insurance Plan 1: Person's Coverage Status), which indicates whether the person was covered by the plan.

Information was collected on the details of up to four private insurance plans per family for 1992-1996, on up to five plans per family for 1980, 1982, 1984, 1986 and 1989, and on up to six plans in 1983. Users may also want to see the related variables HI2WORK, HI3WORK, HI4WORK, and HI5WORK. Likewise, the complementary variables HI2PCOV, HI3PCOV, HI4PCOV, and HI5PCOV indicate coverage status for the individual under each plan.

Instead of considering each private insurance plan in turn, users may find it easier to employ the summary variable HIPWORKR, created by IPUMS NHIS staff, which indicates whether the person was covered by any plan obtained through work.

Comparability

Along with changes in the universe, the meaning of HI1WORK varied somewhat over time. Data for 1993-1996 differ from other years in including single service plans and using a reference period of the previous month (rather that at the time of the survey).
Moreover, for years prior to 1997, a "yes" response in HI1WORK indicated that the person was part of a family that obtained insurance coverage through the workplace in plan 1; beginning in 1997, a "yes" response meant the individual had obtained such coverage.

Data Editing and Recoding

For 1997 forward, the NCHS corrected errors in the responses to questions about insurance coverage through back-editing. Such errors might be apparent, for example, in a mismatch between the verbatim name of an insurance plan and the type of insurance coverage the respondent selected from categories on a flashcard.

In the case of HI1WORK, someone might have mistakenly reported that the family members had private insurance coverage, when in fact they did not. Beginning in 1997, only persons verified to have private insurance are included in the data for HI1WORK. A full description of the back editing process is provided in the variable description for HIPRIVATEE (Covered by private health insurance: Recode).

The NCHS strongly advises analysts to use the back-edited insurance variables as a more reliable source of information about the types of insurance coverage than is provided by respondents' unedited answers. Researchers should be cautious about comparing results from before and after 1997, because only the data for 1997 forward are back-edited for correctness.

Universe

Availability

• 1980, 1982-1984, 1986, 1989, 1992-2003

Weights

• 1980, 1982, 1984, 1986, 1989, 1992, 1994-2003 : PERWEIGHT
• 1983, 1993 : SAMPWEIGHT