Editorial
How Do We Know If We Are Making Progress in Reducing
the Public Health Burden of Stroke?
Virginia J. Howard, PhD; Brett M. Kissela, MD
colleagues have made the clever observation that the any
mention cause of death, in which there can be multiple entries
of diseases and conditions contributing to death, may not be
as subject to these challenges. They note that for both any
mention cause of death and UCOD, the age-adjusted stroke
mortality rates declined by a similar 33%, making the
important suggestion that changes in the approaches for
coding the single UCOD are less likely to contribute to these
observed declines in stroke mortality rates.
Although revisions of International Classification of Diseases can result in discontinuities in cause of death trends, the
10th Revision (International Classification of Diseases,
Tenth Revision) was in use for the entire period examined by
Burke and colleagues (2000 –2008). Although a second edition of International Classification of Diseases, Tenth Revision was adopted in 2004, the changes did not impact coding
of stroke (but could have affected some of the other diseases
in the report). However, the report of Burke and colleagues
must also be interpreted within the context of a much
longer-term decline in stroke mortality that has been nearly
monotonic since 1900 and covers all 10 editions of International Classification of Diseases coding.10 Importantly, although the coding of the UCOD is unlikely to contribute to
recent declines in stroke mortality, the report does not offer
insights to the longer-term pattern.
This work by Burke and colleagues also provides an
excellent summary of the processes by which disease-specific
mortality data are compiled in the United States. Doctors are
rarely, if ever, trained in how to fill out death certificates and
thus death certificate data should always be viewed with
caution because the input data may be suboptimal. Furthermore, concern exists as to whether there are biases by
race– ethnic group or region. Prior studies have suggested this
is not the case,.11,12 but validation studies have not been
reported that looked for recent changes. Regardless, the
process for coding disease-specific causes of mortality takes
the death certificate data and applies a continuously reviewed
and updated algorithm to assign the most likely cause. The
authors note that for some diseases (including stroke and
diabetes), it can be difficult to determine the UCOD. In
persons with stroke, this may be due to substantial long-term
survival, for example. Burke and colleagues conclude that the
combination of these declines in national mortality, lesser
declines in regional stroke incidence, and a slight increase in
regional case fatality as documented in the Greater Cincinnati/Northern Kentucky Stroke Study during a similar time
period13 is “challenging to explain.”9 There are other issues
that should be considered in reviewing the results of Burke
and colleagues and others examining stroke’s changing rank
as a cause of death. In the technical notes section of the
See related article, p, XXX
uch of the description of the differential burden of
specific diseases and temporal changes in these burdens is based on mortality data. Stroke mortality rates have
declined dramatically in the 20th century and into the new
millennium. This has been acclaimed as one of the top 10
public health achievements of these eras.1,2 In 1999, the
age-adjusted stroke death rate was 61.6 per 100 000, but it
decreased by 37% in only 11 years to 38.9 per 100 000 in
2009 (most recent year available).3 Stroke has been stable as
the third leading cause of death throughout most of the 20th
century but in 2008, it declined to the fourth leading cause of
death in the United States.4 – 6 Preliminary mortality data for
2009 and 2010 show that although stroke remains the fourth
leading cause of death, the age-adjusted death rates for stroke
continue to decrease, specifically by 4.3% between 2008 and
2009 and by 1.5% between 2009 and 2010.7,8
This dramatic and sustained decline in stroke mortality is
indeed remarkable and raises questions regarding the
underlying causes of the decline, including the possibility
that it could be an artifact of the classification system for
causes of death.
What define disease-specific causes of death are codes on
death certificates related to “underlying cause of death”
(UCOD). In this issue of Stroke, Burke and colleagues9 report
on their examination of 2000 to 2008 death certificate data of
both UCOD and “any mention cause of death” to investigate
whether systematic changes in the processes of mortality
assignment could be driving this decline in stroke mortality
rates. Using the Centers for Disease Control and Prevention
multicause mortality files, they examined the assignment of
stroke and the other 5 most common organ- and diseasespecific causes of death from 2000 to 2008. Because the
outcome after stroke can include sequelae that subsequently
lead to death, the coding of the single UCOD is challenging,
requires judgment, and is potentially fraught with error and
changes over time (although as noted by the authors, the
algorithm is complex and annually re-evaluated) Burke and
M
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
The opinions in this editorial are not necessarily those of the editors or
of the American Heart Association.
From the Department of Epidemiology, School of Public Health,
University of Alabama at Birmingham, Birmingham, AL (V.J.H.); and
the Department of Neurology, University of Cincinnati College of
Medicine, Cincinnati, OH (B.M.K.).
Correspondence to Virginia J. Howard, PhD, Professor of Epidemiology, School of Public Health, University of Alabama at Birmingham,
210F Ryals Building, 1720 2nd Avenue S, Birmingham, AL 35294-0022.
E-mail [email protected].
(Stroke. 2012;43:00-00.)
© 2012 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org
DOI: 10.1161/STROKEAHA.112.663237
1
2
Stroke
August 2012
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
release of death statistics for 2008 and 2009, caution is
expressed related to changes in coding rules implemented in
2008 that impact comparability for selected causes of death
between 2007 and 2008.4 One of those impacted is cerebrovascular diseases (International Classification of Diseases,
Tenth Revision codes I60 –I69.) Some of the deaths that
would have been coded as subarachnoid hemorrhage (I60) in
2007 were coded as vascular dementia (F01) in 2008. At the
same time, coding rules changed for the subcategory of
chronic obstructive pulmonary disease with acute lower
respiratory infection, resulting in an increase in deaths assigned to the larger category of chronic lower respiratory
diseases. The combination of these 2 changes likely then has
contributed to chronic lower respiratory disease being identified as the third leading cause of death in 2008 and stroke
moving to fourth. This does not diminish the main point of
this work, which is to show that coding is not a primary
contributor to the decline in stroke mortality.
The major strength of using mortality data to evaluate the
burden of stroke is the mandatory reporting of deaths. This
allows calculation of stroke mortality rates at regional (eg,
county or state) and national levels, by demographic factors
(eg, age, race/ethnicity, sex), and allows for monitoring
trends over time. Although we celebrate that stroke mortality
has declined overall, it has not declined equally across all
race– ethnic groups and regions of the country.14,15 Unfortunately, coding stroke as arising from nonspecific causes (I64)
implies that we know little about stroke mortality across
stroke subtypes, even at the level of distinguishing rates of
death from ischemic stroke versus hemorrhage.15 In the
United States, there is no system similar to the National Vital
Statistics System (for deaths) for the reporting and collection
of official stroke records from local communities to allow for
national statistics on incident strokes. In fact, the Million
Hearts Initiative, targeted to prevent one million cardiovascular events (including strokes) over the next 5 years (http://
millionhearts.hhs.gov/), has its own challenge to be able to
define today how it will determine, at the end of the 5-year
period, whether or not it has achieved its goal. The recommendation of a recent Institute of Medicine report to develop
a nationwide surveillance system that can track progress in
preventing stroke and other cardiovascular events is gaining
support in the cardiovascular community.16 Burke and colleagues have provided us confirmation of the usefulness of
national mortality data; it is past time, and the next logical
step, to move toward development of a well-designed, effective system to capture national incidence data.
Disclosures
Dr Howard was the founding chair of the American Heart Association/American Stroke Association Stroke Statistics Committee of the
CVD Epidemiology and Prevention Council and is a current member
of the Committee. Dr Kissela is current chair of the Stroke Statistics
Committee.
References
1. Centers for Disease Control and Prevention. Ten great public health
achievements—United States, 1900 –1999. MMWR Morb Mortal Wkly
Rep. 1999;48:241–243.
2. Koppaka R. Ten great public health achievements—United States,
2001–2010. MMWR Morb Mortal Wkly Rep. 2011;60:619 – 623.
3. Deaths: final data for 2009. Available at: www.cdc.gov/nchs/data/dvs/
deaths_2009_release.pdf. Accessed May 29, 2012.
4. Miniño AM, Murphy SL, Xu JQ, Kochanek KD. Deaths: Final Data for
2008. National Vital Statistics Reports. Vol 59. Hyattsville, MD: National
Center for Health Statistics; 2011.
5. Leading causes of death, 1900 –1998. Available at: www.cdc.gov/nchs/
data/dvs/lead1900_98.pdf. Accessed May 29, 2012.
6. Towfighi A, Saver JL. Stroke declines from the third to fourth leading
cause of death in the United States: historical perspective and challenges
ahead. Stroke. 2011;42:2351–2355.
7. Kochanek KD, Xu JQ, Murphy SL, Miniño AM, Kung H. Deaths:
Preliminary Data for 2009. National Vital Statistics Reports. Vol 59.
Hyattsville, MD: National Center for Health Statistics; 2011.
8. Murphy SL, Xu J, Kochanek KD. Deaths: Preliminary Data for 2010.
National Vital Statistics Reports. Vol 60. Hyattsville, MD: National
Center for Health Statistics; 2012.
9. Burke JF, Lisabeth LD, Brown DL, Reeves MJ, Morgenstern LB. Determining stroke’s rank as a cause of death using multi-cause mortality.
Stroke. 2012;43:xxx–xxx.
10. Centers for Disease Control and Prevention. Decline in deaths from heart
disease and stroke—United States, 1900 –1999. MMWR Morb Mortal
Wkly Rep. 1999;48:1– 8.
11. Nefzger MD, Acheson RM, Heyman A. Mortality from stroke among US
veterans in Georgia and 5 western states. I. Study plan and death rates.
J Chron Dis. 1973;16:393– 404.
12. Lanska DJ, Peterson PM. Geographic variation in reporting of stroke
deaths to underlying or contributing causes in the United States. Stroke.
1995;26:1999 –2003.
13. Kleindorfer DO, Khoury J, Moomaw CJ, Alwell K, Woo D, Flaherty Ml,
et al. Stroke incidence is decreasing in whites but not in blacks: a
population-based estimate of temporal trends in stroke incidence from the
Greater Cincinnati/Northern Kentucky Stroke Study. Stroke. 2010;41:
1326 –1331.
14. Gillum RF, Kwagyan J, Obisesan TO. Ethnic and geographic variation in
stroke mortality trends. Stroke. 2011;42:3294 –3296.
15. Howard VJ, Howard G. Distribution of stroke: heterogeneity of stroke by
age, race, and sex. In: Mohr JP, Choi D, Grotta JC, Weir B, Wolf PA, eds.
Epidemiology and Prevention in Stroke: Pathophysiology, Diagnosis, and
Management. V ed. Philadelphia, PA: Elsevier Saunders; 2011.
16. Institute of Medicine. A Nationwide Framework for Surveillance of Cardiovascular and Chronic Lung Diseases. Washington, DC: The National
Academic Press; 2011.
KEY WORDS: cause of death
䡲
epidemiology
䡲
stroke
䡲
stroke mortality
How Do We Know If We Are Making Progress in Reducing the Public Health Burden of
Stroke?
Virginia J. Howard and Brett M. Kissela
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
Stroke. published online July 3, 2012;
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2012 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/early/2012/06/28/STROKEAHA.112.663237.citation
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/