|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
National Health Interview Survey Multiple Cause of Death Public-Use Data Files The National Center for Health Statistics (NCHS) announces the availability of the updated National Health Interview Survey (NHIS) Multiple Cause-of-Death Public-Use Data Files. Beginning with survey year 1986, linkage information was collected of NHIS respondents 18 years of age and over to allow for matching to other data systems, such as the National Death Index (NDI). Linkage of the NHIS respondents with the NDI provides a longitudinal component to the NHIS that allows for ascertainment of vital status. The addition of vital status permits the use of NHIS data for survival analyses, mortality rates, and life expectancy while using the richness of the NHIS questionnaires, core and supplements, as covariates. The NHIS Multiple Cause-of-Death CD-ROM for survey years 1986-94 contains vital status through the end of 1997, the ICD-9 cause of death code, the NCHS 72 cause recode, the NCHS 282 cause recode for the underlying cause of death, and ICD-9 codes for the entity-axis and record-axis multiple causes of death. Additional information included on the CD-ROM includes a complete description of the matching methodology employed and several SAS examples demonstrating the use of the multiple cause of death files and linkage to NHIS public-use data files. The files in this release completely replace previous releases of linked files. The CD-ROM cannot be used in a stand alone mode and must be linked directly to the corresponding NHIS public-use data file. NHIS survey data for 1986-94 are available on CD-ROM. WARNING Information About the Hispanic Oversample 1991 Supplemental Sample to the 1992 NHIS At the request of the National Cancer Institute, one of the cosponsors of the 1992 NHIS, the Hispanic population was oversampled for the 1992 data collection year. The supplemental sample consists of persons reinterviewed in 1992 at addresses where, in the 1991 NHIS, one or more Hispanic persons 17 years or older were found. It is likely that many of the persons in the supplemental sample are also included in the 1991 NHIS public-use files. However, due to persons and households moving in/out of the supplemental sample addresses between 1991 and 1992, some of the persons in the supplemental sample are not included in the 1991 NHIS. It is not possible to identify which persons were interviewed in both 1991 and 1992. Therefore, any analysis that requires the removal of the overlapping records requires the removal of the entire supplemental sample. The recontacted Hispanic households were assigned for the interview in 1992 in the same week they were assigned for interview in the 1991 NHIS. They can be identified in the 1992 sample by the code for the processing year variable (the code is "91" for the supplemental Hispanic respondents and "92" for all other respondents). Thus, for 1992, processing year must be used in combination with other variables to identify a unique household or individual. Analysts need to keep in mind this oversample feature when using the 1992 NHIS core data sets. The sampling weights provided on the CD-ROMs take into account this oversampling. Those in the oversample may have also participated in the 1991 NHIS and analysts are cautioned to consider the impact of this if they combine the 1991 and 1992 data sets. The oversample respondents can be identified in the 1992 data sets and potentially eliminated for analysis; however, the weights for the remainder of the 1992 respondents on the files would no longer be correct and the files would need to be reweighted. Files containing the adjusted weights are available on the NHIS web site under the heading "1992 NHIS." To order the NHIS Multiple Cause-of-Death CD-ROM free of charge contact: Information
Dissemination Staff
This page last reviewed
January 11, 2007
|