National Health Interview Survey, 2004 (ICPSR 4349)

Published: May 4, 2006

Principal Investigator(s):
United States Department of Health and Human Services. Centers for Disease Control and Prevention. National Center for Health Statistics


Version V2

The purpose of the National Health Interview Survey (NHIS) is to obtain information about the amount and distribution of illness, its effects in terms of disability and chronic impairments, and the kinds of health services people receive. Implementation of a redesigned NHIS, consisting of a basic module, a periodic module, and a topical module, began in 1997 (see NATIONAL HEALTH INTERVIEW SURVEY, 1997 [ICPSR 2954]). The 2004 NHIS contains the Household, Family, Person, Sample Adult and Sample Child files from the basic module. Each record in the Household-level File (Part 1) contains data on type of living quarters, number of families in the household responding and not responding, and the month and year of the interview for each sampling unit. The Family-level File (Part 2) is made up of reconstructed variables from the person-level data of the basic module and includes information on sex, age, race, marital status, Hispanic origin, education, veteran status, family income, family size, major activities, health status, activity limits, and employment status, along with industry and occupation. As part of the basic module, the Person-level File (Part 3) provides information on all family members with respect to health status, limitation of daily activities, cognitive impairment, and health conditions. Also included are data on years at current residence, region variables, height, weight, bed days, doctor visits, hospital stays, and health care access and utilization. A randomly-selected adult in each family was interviewed for the Sample Adult File (Part 4) regarding respiratory conditions, use of nasal spray, renal conditions, AIDS, joint symptoms, health status, limitation of daily activities, and behaviors such as smoking, alcohol consumption, and physical activity. The Sample Child File (Part 5) provides information from an adult in the household on medical conditions of one child in the household, such as respiratory problems, seizures, allergies, and use of special equipment like hearing aids, braces, or wheelchairs. Also included are variables regarding child behavior, the use of mental health services, and Attention Deficit Hyperactivity Disorder (ADHD) as well as responses to the SDQ-EX, which is the extended version of Strengths and Difficulties questionnaire on child mental health. Several changes have occurred in the 2004 NHIS. The Child Immunization Section (CIM) has been dropped. Also new in 2004, questionnaires have been provided in both English and Spanish. The Injury/Poison Episode File (Part 6) is an episode-based file that contains information about the external cause and nature of the injury or poisoning episode and what the person was doing at the time of the injury or poisoning episode, in addition to the date and place of occurrence. The Injury/Poison Episode Verbatim File (Part 7) contains edited narrative text descriptions of the injury or poisoning provided by the respondent. Imputed income files for 2004 are now available through the NCHS Web site at

United States Department of Health and Human Services. Centers for Disease Control and Prevention. National Center for Health Statistics. National Health Interview Survey, 2004. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2006-05-04.

Export Citation:

  • RIS (generic format for RefWorks, EndNote, etc.)
  • EndNote

In preparing the data files for this collection, the National Center for Health Statistics (NCHS) has removed direct identifiers and characteristics that might lead to identification of data subjects. As an additional precaution, NCHS requires, under Section 308(d) of the Public Health Service Act (42 U.S.C. 242m), that data collected by NCHS not be used for any purpose other than statistical analysis and reporting. NCHS further requires that analysts not use the data to learn the identity of any persons or establishments and that the director of NCHS be notified if any identities are inadvertently discovered. ICPSR member institutions and other users ordering data from ICPSR are expected to adhere to these restrictions.



(1) Users should refer to the Survey Description for changes regarding the structure of the 2004 NHIS. In previous years, one household equaled one "case." In 2004, each family is considered a separate case. For multiple-family households, there is a "parent" case designated for the first family identified, and each subsequent family becomes a "spawned" case from the parent, with its own unique ID number. (2) Because variables are not repeated on data files in the 2004 NHIS, users will need to merge files in order to analyze the data. Instructions on merging files are found in the Survey Description. Also included in 2004 are Variable Summary Reports that describe the variable and the question number on which it was based, and Variable Layout Reports that provide the actual question that generated the variable along with the universe and response values. (3) The data from the Household-level File can be merged with any of the other files, and other files can be merged as well. For further information on merging data, consult the Survey Description. (4) The Injury/Poison Episode Verbatim File (Part 7) is available only with SAS setup files. (5) To learn more about the National Health Interview Survey, visit the following Web site of the Centers for Disease Control and Prevention (CDC). At that site you can join the HISUSERS e-mail list by providing your name and e-mail address, selecting the item, "National Health Interview Survey (NHIS) researchers," and clicking on "subscribe."

The NHIS uses a stratified multistage probability design. The sample for the NHIS is redesigned every decade using population data from the most recent decennial census. A redesigned sample was implemented in 1995. This new design includes a greater number of primary sampling units (PSUs) (from 198 in 1994 to 358), and a more complicated nonresponse adjustment based on household screening and oversampling of Black and Hispanic persons, for more reliable estimates of these groups.

Civilian, noninstitutionalized population of the 50 states and the District of Columbia.

survey data, and clinical data



2006-05-04 Parts 6 and 7 were added.

These data contain multiple weight variables for each part. Users should refer to the technical documentation for further information regarding the weight and their derivation. Additionally, users may need to weight the data prior to analysis.


  • The public-use data files in this collection are available for access by the general public. Access does not require affiliation with an ICPSR member institution.

NACDA logo

This study is maintained and distributed by the National Archive of Computerized Data on Aging (NACDA), the aging program within ICPSR. NACDA is sponsored by the National Institute on Aging (NIA) at the National Institutes of Heath (NIH).