January 28, 2009
From: Bradford D. Gessner, MD, MPH
Director, Maternal-Child Health Epidemiology Unit
Section of WOmens, Childrens, and Family Health
Alaska Division of Public Health
Suite 424
3601 C Street
Anchorage, Alaska 99524
Tel. 907-269-3446
Email. Brad.Gessner@alaska.gov
To: Editors
Pediatrics
Dear Editors,
Shapiro-Mendoza et al. report an increase in rates of infant
mortality from accidental suffocation and strangulation (1) based on
analysis of data from the National Center for Health Statistics. As the
authors likely know, data from NCHS are provided by State Bureaus of Vital
Statistics, which in turn must interpret causes of death recorded on death
certificates. While NCHS data may be adequate for reporting overall infant
mortality and for broad categories of causal mechanisms (congenital
anomalies, infections, sudden unexpected infant death) they usually are
inappropriate for looking at specific mechanisms such as accidental
suffocation and strangulation. Data recorded on death certificates are
affected by, among other things, the training of the recorder, diagnostic
standards, funding for State Medical Examiners’ offices, existence of and
adherence to requirements for autopsies, standardization of and use of
death scene investigations by first responders, existence and composition
of mortality review committees, and other issues. For example, we reported
years ago an outbreak of infant viral myocarditis/pneumonitis mortality
that resulted from misinterpretation of pathology specimens (2); at the
time, the database used by the authors (CDC WONDER) indicated spuriously
that Alaska had the highest infant myocarditis mortality rates in the
nation by several-fold. Others, including the authors themselves (3), have
demonstrated changes in classification of causes of infant deaths. The
trends reported in the current study (1) most likely also derive from
shifts in cause of death classification, the last few years
notwithstanding.
While it is less work to analyze a pre-existing database, it may also
provide uninterpretable results. In this case, if the US Centers for
Disease Control and Prevention wants to evaluate the contribution to
infant mortality of accidental suffocation and strangulation, they should
consider actually reviewing medical records and autopsy reports.
Furthermore, the Editors of Pediatrics should scrutinize carefully results
and claims based on cause-specific analysis of vital records data. Readers
unfamiliar with the nuances of such data may interpret the results Shapiro
-Mendoza’s study as actually documenting an increase in suffocation and
strangulation deaths. At best these results form the basis for generating
a hypothesis that requires testing with a more robust study.
Sincerely,
Bradford D. Gessner, MD
Alaska Division of Public Health
1. Shapiro-Mendoza CK, Kimball M, Tomashek KM, Anderson RN, Blanding
S. US Infant mortality trends attributable to accidental suffocation and
strangulation in bed from 1984 through 2004: are rates increasing?
Pediatrics 2009;123:533-539.
2. Centers for Disease Control and Prevention. Misclassification of infant
deaths—Alaska, 1990-1991. Morb Mort Wkly Rep 1992; 41:584-5,591.
3. Shapiro-Mendoza CK, Tomashek KM, Anderson RN, Wingo J. Recent national
trends in sudden, unexpected infant deaths: more evidence supporting a
change in classification or reporting. Am J Epidemiol. 2006;163:762–769.
Conflict of Interest:
None declared
eLetters: ![[Read eLetters]](/icons/misc/sectionDOWN.gif){kind=icon}
