National Health Interview Survey NHIS Linked Mortality File

National Health Interview Survey (NHIS) Linked Mortality File Overview Christine S. Cox Special Assistant for Record Linkage Bryan Sayer, Donna Miller, Jim Brittain U. S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Centers for Disease Control and Prevention National Center for Health Statistics

National Health Interview Survey (NHIS) • Nationally representative household based survey of the civilian non-institutionalized U. S. population. • Conducted continously • Surveys approximately 43, 000 households per year for a total person sample of about 100, 000 • Collects information on health conditions and utilization, access to care and health insurance coverage, along with other varying special topics of interest • Survey design materials, questionnaires, and analytic reports available from the NCHS web site. http: //www. cdc. gov/nchs/nhis. htm

National Death Index • A centralized electronic data base containing death information from U. S. death certificates • Available by application for statistical research only • Users submit records containing study participant data to attempt to match to correct death record • Each death record contains: demographic characteristics of the decedent, state of death, underlying/multiple causes of death • Current Coverage: 1979 -2002 • NDI webpage: http: //www. cdc. gov/nchs/r&d/ndi. htm

NHIS-NDI Linkage • Conducting NDI data linkage is both a scientifically valuable and cost effective data collection activity. • Provides the opportunity to conduct a wide variety of mortality outcome studies analyzing NHIS questionnaire data on health conditions, insurance coverage, health behaviors and other risk factors – Example: breast cancer death rates among uninsured vs. insured women

Previous vs. current NHIS Linked Mortality files • Previous NHIS Mortality Data Files – 1986 -1994 NHIS Mortality Data through 1995 – 1986 -1994 NHIS Mortality Data through 1997 • Current NHIS Linked Mortality File – 1986 -2000 NHIS Mortality Data through 2002

NHIS – NDI Mortality Linkage Methodology • Based on a probabilistic match to the NDI • No death certificates were collected for verification purposes • Calibration study was conducted • Best plausible NDI match record provided on user file (Definite non-matches eliminated) • Final match status assigned – Presumed Dead, Presumed Alive, Ineligible

Ineligible Population • All NHIS respondents under age 18 are not eligible for mortality follow-up • All adult NHIS respondents who do not meet the minimum criteria set by NDI are defined as ineligible. – SSN, sex, m/d/y of birth present – Last name, first initial, month of birth, year of birth present – Last name, first initial, SSN present • Analysts should treat ineligibles as lost to followup – New sample weights developed

Percent of ineligible and deceased adult NHIS respondents by survey year : NHIS 1986 -2000

Total NHIS Deaths By Survey Year Total Number of Deaths 19861 8, 492 1994 7, 911 1987 15, 776 1995 5, 762 1988 14, 695 19961 3, 004 1989 12, 956 1997 4, 007 1990 12, 061 1998 3, 009 1991 11, 014 1999 2, 192 1992 10, 596 2000 1, 531 1993 8, 130 Total 121, 136 Survey Year 1 The 1986 and 1996 NHIS were fielded with a reduced sample. Total Number of Deaths

Cumulative survival curves, all males: NHIS 1987 and 1995

NHIS Linked Mortality File Access NCHS Research Data Center (RDC) Purpose – • To provide access to restricted data files which data users would not be able to otherwise access Accessibility – • • On-site access – NCHS, Hyattsville, MD Off-site access - User’s remote location

User Procedures To gain access to NCHS restricted data users must: • Submit a research proposal and receive approval • Sign an affidavit of confidentiality and promise not to use any method to attempt to identify respondents • Submit data files to be linked with RDC data ahead of time – all linking is conducted by RDC staff • Subject all output and/or materials removed from the RDC to a disclosure limitation review • Pay a user’s fee that will be estimated prior to proposal approval

RDC Contact Information For general Questions/Comments Email: rdca@cdc. gov Phone: (301)458 -4732 For On-site Info: Negasi Beyene Email: NBeyene@cdc. gov Phone: (301)458 -4097 For Remote Access Info: Vijay Gambhir Email: vgambhir@cdc. gov Phone: (301)458 -4226

File Structure A standard linked mortality file has been developed to assist RDC users: – All variables on the standard will be linked to approved researcher’s analytic file (no need to specify variables of interest) – Additional variables available by special approval only • Geography • Occupation & Industry

Linked Mortality File Structure • Separate mortality data file for each NHIS survey year containing the total number of NHIS respondents for that survey year • File documentation will be available for download on the NCHS Data Linkage web page (www. cdc. gov/nchs/rd. htm) • Send email queries to (datalinkage@cdc. gov) – Include NHIS Linked Mortality File in Subject line if possible for quicker routing

Standard NHIS Linked Mortality File • • NHIS Public Use ID Eligibility Status NDI match results Date of Birth and Death (to calculate survival period) • Underlying and Multiple Causes of Death • Sample weights

NHIS Survival period Decedents (Presumed Dead) • End date = date of death Survivors (Presumed Alive) • End date = December 31, 2002 • Some respondents appear to have very advanced ages (Dec. 31, 2002 – Date of birth) • Analysts should consider reassigning these cases as lost-to-followup (otherwise immortal) or review the NDI results to see if a “close to match” NDI record exists that could be reassigned as deceased

Cause of death codes Underlying cause of death recode groups ICD-9 (1986 -1998) ICD-10 (1999 -2002) • 282 • 72 • 34 • 358 (all years coded) • 113 (all years coded) • 39