A Half Century of Health Data for the U.S. Population: The Integrated Health Interview Series

Abstract

The U.S. National Health Interview Survey (NHIS) is the world's longest survey time series of health data and a rich source of information on health conditions, behaviors, and care from the 1960s to the present. NHIS public-use files are difficult to use for long-term analysis, due to complex file structure, changes in questionnaire content, and evolving variable names and coding schemes. Researchers at the Minnesota Population Center have created the Integrated Health Interview Series (IHIS) to overcome these problems. IHIS provides access to thousands of consistently coded and well-documented NHIS variables on the Internet and makes it easy to analyze health trends and differentials. IHIS multiplies the value of NHIS data by allowing researchers to make consistent comparisons over half a century and thus to study U.S. health status as a dynamic process. This article describes the main features of IHIS and suggests fruitful avenues for historical research using these invaluable health data.

Keywords:

• demography
• health data
• microdata
• Minnesota Population Center
• NHIS
• United States

Notes

The Integrated Health Interview Series project is supported by a grant to the Minnesota Population Center, University of Minnesota, from the National Institutes of Health, 5RO1HD046697.

1. For an informative summary of the origin and evolution of the NHIS, including major milestones in NHIS history, see Jane F. Gentleman 2010.

2. The NCHS uses the slogan “Monitoring the Nation's Health” as the title of a Web-based newsletter reporting upcoming statistical releases and findings from recent NCHS surveys, such as the NHIS. See http://www.cdc.gov/nchs/pressroom/MNH.htm.

3. To match the complex stratified sample design of the NHIS, the version of SDA set up on the IHIS Web site specifies strata and cluster variables and calculates standard errors using the Taylor series method. A short user note providing an overview of the NHIS sample design is available on the IHIS Web site: http://www.ihis.us/ihis/

4. The data conversion program that most MPC data harmonization projects use handles two levels of hierarchy: households and persons within households. MPC information technology staff developed for IHIS a data conversion program that can handle three levels of hierarchy: households, persons within households, and varying numbers and types of episode records (e.g., condition records) for persons. The IHIS program is thus structured to handle the episode-level condition records and use their information to create new pre-1997 person-level variables covering the presence or absence of specific conditions.

5. The variables based on condition records will encompass only those pre-1997 condition records that were generated in response to the specific condition questions appearing on the survey form (for 1963–77) or on condition checklists administered to households (for 1978–96).

6. For examples of the types of health topics that can be studied using NHIS and IHIS data, see Catherine Cubbin, Felicia B. LeClere, and Gordon S. Smith 2000; Jay S. Kaufman et al. 1998; Ellen Johnson Silver and Ruth E. K. Stein 2001; Frances Chevarley and Emily White 1997; Paul W. Newacheck et al. 2003; Laura E. Montgomery, John L. Kiely, and Gregory Pappas 1996; LeClere and Mah-jabeen Soobader 2000; Michael J. Kallan 1997; Gopal K. Singh and Mohammed Siahpush 2002; Arlinda Rolett et al. 2001; John Gary Collins and LeClere 1996; and Richard G. Rogers, Charles B. Nam, and Robert A. Hummer 2002.

7. For examples of research on disability using NHIS data, see Scott Bilder and David Mechanic 2003; Eileen M. Crimmins, Sandra L. Reynolds, and Yasuhiko Saito 1999; Timothy A. Waidmann, John Bound, and Michael Schoenbaum 1995; Mark D. Hayward and Melonie Heron 1999; and Richard V. Burkhauser 2002.

8. George Kaplan, Nalini Ranjit, and Sara Burgard's 2008 work on “Did Civil Rights Improve the Health of African-American Women in the 1960s and 1970s?” is one such provocative effort to link broad social changes to health outcomes.