Misclassification of Maternal Deaths -- Washington State
To determine more accurately the number of maternal deaths in
the
State of Washington in the period 1977-1984, death certificates for
all reproductive-age women who died in those years were linked to
birth and fetal-death files. The purpose of this linkage was to
identify women who had died within a year after having a live birth
or
fetal death. For this 8-year period, the record linkage identified
23
maternal deaths (Table 3) in addition to the 34 maternal deaths
reported through the state's vital statistics system. This total
of
57 represents a 68% increase in ascertainment of maternal deaths.
To establish linkage, the names of the 2,073 women, ages 15-44
years, who had died in Washington State in 1977-1984 from causes
other
than cancer and intentional or unintentional injuries, were
compared
with the names of women on birth and fetal-death files. When these
names matched, birth and death certificates were compared to verify
each match. The woman's cause of death, the interval between
delivery
and death, and other medical information about the case were
reviewed
to determine whether the death should be classified as maternal
(Table
3). Maternal mortality was defined as either a direct or indirect
maternal death occurring within 42 days of delivery, based on the
codes defined by the International Classification of Diseases
(1,2).
Of the 34 maternal deaths reported through the state's vital
statistics system, 20 were linked to birth or fetal-death
certificates; 14 women had a maternal condition listed as the cause
of
death, but were not linked to birth or fetal-death certificates.
In
addition, death certificates for 39 other women were matched with
birth or fetal-death certificates, but these deaths had not been
classified by the state's vital statistics system as maternal
deaths.
For 16 of these 39 deaths, a causal association with pregnancy was
unclear or unlikely; the other 23 were classified as maternal.
Seven
of the 23 maternal deaths that were originally misclassified as
non-maternal were concurrent with the reported birth or fetal
death.
Reported by P Starzyk, PhD, F Frost, PhD, JM Kobayashi, MD, State
Epidemiologist, Washington Dept of Social and Health Svcs;
Pregnancy
Epidemiology Br, Research and Statistics Br, Div of Reproductive
Health, Center for Health Promotion and Education, CDC.
Editorial Note
Editorial Note: There was a decrease in misclassified deaths in
Washington State between 1977 and 1984. In the period 1977-1981,
19
(53%) of 36 maternal deaths were originally misclassified as
non-maternal (3). In contrast, in 1982-1984, only four (19%) of 21
maternal deaths were originally misclassified. The decrease in
misclassified maternal deaths in 1982-1984 probably reflects an
increased number of queries by the nosologists during this period.
A
review of maternal death certificates in Washington showed that 60%
of
the maternal deaths were queried from 1982 through 1984, in
contrast
to 15% from 1977 through 1981.
Based on the state's vital statistics, the maternal mortality
rate
in Washington was 6.5 maternal deaths/100,000 live births for this
8-year period. However, the rate calculated from both vital
statistics and record linkage was 10.9/100,000 live births.
Underestimation of maternal mortality is a problem that has long
been
recognized (4,5). The Washington State study--as well as recent
studies in Georgia, New Jersey, and Puerto Rico--suggests that the
problem persists (6,7,8). The percentage of maternal deaths not
obtainable from routine coding of death certificates ranged from
27%
in Georgia to 71% in Puerto Rico. As was shown in Georgia (6),
this
study demonstrates that more complete counts of maternal deaths can
be
obtained from routine linkage of birth and fetal-death certificates
with death certificates of reproductive-age women.
Several categories of maternal deaths may not be found through
record linkage if they are misclassified on death certificates.
The
first category includes maternal deaths that may not generate a
record
of pregnancy outcome. This includes maternal deaths from ectopic
pregnancy, gestational trophoblastic disease, induced abortion, and
first-trimester spontaneous abortion, as well as maternal deaths
that
occur during pregnancy. The second category includes maternal
deaths
that are misclassified but are not identified by record linkage
because of incorrect identifiers (e.g., name, address, and birth
date)
or because of unreported births or fetal deaths. Some of these
maternal deaths can be identified by supplementing information from
vital statistics with annual maternity service reports from
hospitals
and with individual reports from physicians, hospitals, and medical
examiners (7). Furthermore, as was shown in Puerto Rico (8), a
review
of death certificates and selected medical records of
reproductive-age
women who died could probably identify even more of these maternal
deaths.
In 1983, the reported maternal mortality rate in the United
States
was 8.0/100,000 live births (9). The U.S. Public Health Service
1990
Objective for maternal mortality is a maternal death rate not to
exceed 5/100,000 live births for any county or for any ethnic
group.
Accurate and complete data on maternal deaths may facilitate the
development of strategies to reduce the maternal mortality rate.
In
the near future, the Center for Health Promotion and Education,
CDC,
plans to establish a National Maternal Mortality Surveillance
System
to assist in achieving this objective.
References
National Center for Health Statistics. Instructions for
classifying the underlying cause of death, 1976-1978. Vital
statistics, part 2A. Hyattsville, Maryland: Department of
Health
and Human Services, Public Health Service, National Center for
Health Statistics, 1976.
National Center for Health Statistics. Instructions for
classifying the underlying cause of death, 1979. Vital
statistics, part 2A. Hyattsville, Maryland: Department of
Health
and Human Services, Public Health Service, National Center for
Health Statistics, 1978.
Benedetti TJ, Starzyk P, Frost F. Maternal deaths in
Washington
State. Obstet Gynecol 1985;66:99-101.
Barno A, Freeman DW, Bellville TP. Minnesota maternal
mortality
study; five-year general summary, 1950-1954. Obstet Gynecol
1957;9:336-44.
Jewett JF. Changing maternal mortality in Massachusetts. N
Engl
J Med 1957;256:395-400.
Rubin G, McCarthy B, Shelton J, Rochat RW, Terry J. The risk
of
childbearing re-evaluated. Am J Public Health 1981;71:712-6.
Ziskin LZ, Gregory M, Kreitzer M. Improved surveillance of
maternal deaths. Int J Gynaecol Obstet 1979;16:281-6.
Speckhard ME, Comas-Urrutia AC, Rigau-Perez JG, Adamsons K.
Intensive surveillance of pregnancy-related deaths, Puerto
Rico,
1978-1979. Bol Asoc Med PR 1985;77:508-13.
National Center for Health Statistics. Advance report of final
mortality statistics, 1983. Monthly Vital Statistics Report
1985;34:38.
Disclaimer
All MMWR HTML documents published before January 1993 are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.
**Questions or messages regarding errors in formatting should be addressed to mmwrq@cdc.gov.
