Survey
description
The National Mortality Followback Survey (NMFS) includes a number of
surveys, the two most recent relating to 1986 and 1993, designed to
supplement information from death certificates in the vital statistics
file with information on important characteristics of the decedent. These
include the use of health services prior to death, socioeconomic status,
aspects of life style, health care utilization prior to death, and other
factors that may affect when and how death occurs. Supplemental
information is obtained from a variety of sources. Information is
collected from proxy next-of-kin informants identified on the death
certificate filed in State and independent vital registration areas in
each survey.
In the 1986 NMFS,
additional information is collected from hospital and other health care
facilities used in the last year of life. The 1993 NMFS collects
additional information from medical examiner/coroner offices if death is
due to homicide, suicide, or accidental injury, is certified by a medical
examiner or coroner, and permission to release the records is given by the
proxy respondent. The foci of the each survey reflects current research
interests and needs of the Public Health Service, other Federal agencies,
and private health researchers in addressing salient issues related to
health and mortality.
The foci of the 1986
NMFS are: (1) socioeconomic status and mortality; (2) associations between
risk factors and mortality; and (3) health care sought and provided in the
last year of life. Foci of the 1993 NMFS include: (1) socioeconomic
differentials in mortality; (2) prevention of premature death by inquiring
into the association between risk factors and the cause of death; (3)
health care services provided in the last year of life; and (4) disability
in the last year of life. A major objective of both the 1986 and 1993 NMFS
is to examine the reliability of items reported on the death certificate
by comparing these items with the same items reported by the survey proxy
respondent. This includes age, race, gender, veteran status, education,
occupation, and industry.
Data
file description
The 1986 NMFS file consists of four types of linked records for 18,733
decedents: (1) data from death certificates; (2) the 24-page proxy
respondent survey questionnaire; (3) the 16-page nursing home, hospital,
and hospice questionnaire; and (4) the facility abstract record. The 1986
NMFS is a nationally representative sample of adults 25 years of age and
over who died in 1986.
The 1993 NMFS file
consists of three types of linked records for 22,957 decedents: (1) data
from death certificates; (2) the 75-page proxy respondent survey
questionnaire; and (3) the medical examiner/coroner abstract record for
deaths due to homicide, suicide, and accidental injury. The 1993 NMFS is a
nationally representative sample of persons 15 years of age and over who
died in 1993.
List
of data file variables at a glance
(1993 and 1986
files)
Death certificate
file (1993)
Race
Age (decedent and respondent in years)
Hispanic origin
Year of birth
Place of birth
Educational attainment
Usual occupation
Kind of business/industry
Veteran status
Marital status
Month of death
Manner of death
Underlying cause of death
Multiple cause of death
Proxy respondent
file (1993)
Race
Age (in years)
Hispanic origin
Place of birth
Educational attainment (decedent and respondent)
Usual occupation (decedent and spouse)
Kind of business/industry (decedent and spouse)
Veteran status
Marital status
Place of death
Manner of death
Firearm related injury
Circumstances of injury
Use of alcoholic beverages
Drug use
Tobacco use
Co-morbid condition
Utilization of services (in last year of life)
Payment sources
Access to care
Cognitive functioning
Activities of daily living (ADL) and instrumental ADL's
Assistive devices used
Medical devices used/implanted
Health indicators
Motor vehicle/driving behavior
Firearm ownership/access
Organ donorship
Participation in social activities (in last year of life)
Life events
Problem behaviors
Income and assets
Authorizations
Medical
examiner/coroner abstract record file (1993)
Race
Age (estimated, in years)
Month of death
Place of death
Manner of death
Firearm related injury
Circumstances of injury (transport/nontransport
related
injury, homicide, suicide, undetermined
manner
of death)
Examination/autopsy of body
Toxicological findings
Co-morbid conditions
Activities and impairments
Medical devices implanted
Organ donorship
Death Certificate
file (1986)
Race
Sex
Age (in years)
Birth date
Education level
Date of death
Place of death
Birthplace
Usual occupation
Type of business/ industry
Veteran status
Marital status
Informant
Informant
questionnaire file (1986)
Race
Sex
Age (in years)
Education of decedent and spouse
Total family income
Inventory of all facilities used in last year of life
Time in facility
Medical history (for example, cancer, heart,
cerebrovascular,
lung disease,
gynecological
history)
Medical care (in last year)
Problems getting care
Activities in daily living
Costs of care
Source of payment for care
Out-of-pocket cost
Main source of payment
Life style and health (exercise, medicines, health
history,
alcohol and smoking history)
Social support (provider, type)
Kind of work done longest
Number of years doing work
Activity on job
Spouse's occupational history
Marital status at time of death
Marriage length
Number of spouses (ever had)
Family medical history (for example, heart disease)
Relationship to decedent
Time lived with decedent
Facility abstract
record file1 (1986)
Birth date
Date of death
Discharge/admission dates
Diagnoses (ICD-9-CM)
Principal/other
Surgical/diagnostic procedures in each admission
(ICD-9-CM)
Type of facility
Other facility used (identifier not on tape)
Names/addresses of facilities (used for mailing facility
abstract
records/hospital/nursing home
questionnaires)
_____________________________________________________ 1Similar
to hospital discharge summary. ICD-9-CM
is International Classification of Diseases, Ninth Revision, Clinical
Modification.
Staff person
questionnaire file (nursing home, hospice) (1986)
Race
Sex
Age
Education of decedent
Inventory of all facilities used in last year of life
Time in facility
Other facility used (identifier not on tape)
Names/addresses of facilities (used for mailing facility
abstract
records/hospital/nursing home
questionnaires)
Medical history (for example, cancer, heart,
cerebrovascular,
lung disease,
gynecological
history)
Medical care (in last year)
Activities in daily living
Costs of care
Source of payment for care
Main source of payment
Health (medicines, health history)
Kind of work done longest
Number of years doing
Activity on job
Veteran status
Marital status at time of death
Marriage length
Geographic
coverage
Although all 50 State vital registration areas and the independent areas
of the District of Columbia and New York City granted approval for a
sample of their death certificates to be used in the 1986 NMFS, deaths
from Oregon are not included in the 1986 NMFS sample because of the
State's respondent consent requirements. For the 1993 NMFS, 49 of the 50
State vital registration areas and the independent areas of the District
of Columbia and New York City granted approval for a sample of their death
certificates to be used. South Dakota declined to participate in the 1993
NMFS due to State law restricting the use of death certificate information
for purpose of followback.
For each survey, death
certificates are sampled from the Current Mortality Sample (CMS) for the
respective calendar years. The CMS is a systematic 10-percent sample of
the States' death certificates, which NCHS, CDC, receives about 3 to 4
months after the deaths occur. Approximately 2,500 of the 18,733 sample
death certificates from the 1986 CMS were selected with certainty to meet
certain specific research needs. A little over 45 percent of the 22,957
sample death certificates from the 1993 CMS were selected with certainty.
Data
file technical characteristics
The code scheme of the data tape is numeric/alphabetic/blank. Labels are
Standard label. The record format is blocked, fixed.
Data
year
Number
of
records
1986
35,757
Data
availability
Data from the NMFS are available on CD-ROM. Survey data are
currently available for 1986. Public-use data files containing underlying
and multiple cause-of-death data
can be purchased as well.
Ordering
information
The NMFS can only be purchased through the National Technical
Information Service (NTIS).
Contact NTIS for current pricing and availability.
Documentation
contents
The 1986 NMFS documentation includes a list of data elements; ICD-9 codes;
occupation codes; and a technical appendix including sources of data,
history, and classification of data schema.
Documentation for the
1993 NMFS provisional file can be downloaded from the CDC/FTP
server. The documentation include sections on general
NCHS, CDC, guidelines for file use, overview of file structure,
introduction to the survey, data preparation and processing, sample
weights, methodological notes for users, sample SAS and SUDAAN procedures,
variable lists, and file format and response frequencies. The latter
section includes tape location, field size, unweighted and weighted
response frequencies, and the item and response categories.
Related
reports
Publications using the 1986 NMFS include: Advance Data Nos. 172,173, and 180. Vital and Health Statistics Series 1,
No. 29, Series 2,
No. 118, and Series 20,
No. 19.
Contact
Mortality Statistics Branch
Division of Vital Statistics
National Center for Health Statistics
Centers for Disease Control and Prevention
3311 Toledo Road, Room 7318
Hyattsville, Maryland 20782
(301) 458-4666