Explaining Poorer Stroke Outcomes in Women: Women
Surviving 3 Months Have More Severe Strokes Than Men
Despite a Lower 3-Month Case Fatality
Tom Skyhøj Olsen, MD, PhD1; Zorana Jovanovic Andersen, MS, PhD2; and
Klaus Kaae Andersen, MS, PhD3
1The Stroke Unit, Frederiksberg University Hospital, Frederiksberg, Denmark;
Screening, Department of Public Health, University of Copenhagen, Copenhagen, Denmark; and
of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark
2Center for Epidemiology and
Background: Women who survive stroke are more disabled and more often institutionalized than men.
Objective: We explore this phenomenon by studying case fatality and stroke severity in stroke survivors
separately for men and women.
Methods: A Danish stroke registry (2000?2007) contains information about 26,818 patients with first-
ever ischemic stroke, including stroke severity (Scandinavian Stroke Scale, 0 worst to 58 best), computed
tomography scan, cardiovascular risk factors, and death 3 months after stroke. We modeled stroke severity
by generalized additive linear model and 3-month case fatality with logistic model adjusting for age and
cardiovascular risk factors.
Results: Male to female ratio was 51.5% to 48.5%. Mean age was 68.8 (SD 12.6) years in men; 73.7 (13.8)
years in women. Stroke was more severe in women (mean [SD] Scandinavian Stroke Scale, 42.2 [16.0]) than
in men (mean [SD] Scandinavian Stroke Scale, 45.6 [14.2]) also after adjustment for age and cardiovascular
risk factors; significant in patients older than 75 years. In survivors at 3 months, stroke was more severe in
women than men, given same age and cardiovascular risk factor profile; significant in patients older than
75 years. More women (11.9%) had died within 3 months than men (8.6%). However, adjusting for age,
stroke severity, and risk factor profile, 3-month case fatality was lower in women than men; significant in
patients older than 78 years.
Conclusions: Although 3-month case fatality was lower in women than men, strokes were more severe
among survivors at 3 months in women than in men. In addition, strokes were more severe in women. Our
data help elucidate why women survive stroke better but have poorer functional outcomes that require
more care than men. (Gend Med. 2012;9:147–153) © 2012 Elsevier HS Journals, Inc. All rights reserved.
Key words: acute stroke, brain infarction, case fatality, outcome, sex.
Accepted for publication March 5, 2012.
© 2012 Elsevier HS Journals, Inc. All rights reserved.
1550-8579/$ - see front matter
GENDER MEDICINE/VOL. 9, NO. 3, 2012
In the Western world, life expectancy is 5 years
longer in women than in men.1Female superiority
in survival of disease is to be expected, unless fe-
male superiority in preventing disease2entirely ex-
plains women’s longer life expectancy. However,
studies reporting better survival among women
after stroke are few.3–6In the vast majority of stud-
ies, case fatality was either higher in women after
stroke or no sex difference was seen.7In a few
studies, functional outcomes after stroke were the
same in men and women, but most reports con-
clude that women who survive stroke are more
disabled and more often institutionalized than
Worse functional outcomes after stroke in
women than in men might be explained by more-
severe strokes in women than men. However, most
studies do not report differences in stroke severity
If women, with all things being equal (ie, age,
stroke severity, cardiovascular risk factor profile),
are more likely to die after stroke than men, it
would mean that men survive strokes from which
women die. In that case, male stroke survivors,
with all things being equal, have to live with more-
severe strokes than women, and functional out-
comes in men, therefore, would be expected to be
worse than in women. This is not consistent with
findings of both lower case fatality and better func-
tional outcomes in men,7unless men are able to
compensate stroke deficits better than women.
Keeping in mind the considerations mentioned,
we find reports of higher case fatality and worse
functional outcomes in women contradictory to
the fact that women live considerably longer than
men. This study was undertaken to investigate, at
the same time, initial stroke severity, case fatality,
and stroke severity in stroke survivors, taking into
account differences between men and women
with regard to initial stroke severity, age, and pres-
ence of cardiovascular risk factors.
The study is based on data from the Danish Na-
tional Indicator Project, described in detail else-
where.6,11All Danish hospitals are committed to
reporting a predefined set of data into the National
Indicator Project database on all patients admitted
to hospital with acute stroke, including age, sex,
admission stroke severity measured by the Scandi-
navian Stroke Scale (SSS),12stroke subtype, and a
predefined cardiovascular profile.
The SSS is a validated neurological stroke scale
that evaluates stroke severity on a score from 0 to
58, with lower scores indicating more-severe
strokes.12Distinction between ischemic and hem-
orrhagic stroke was determined after computed
tomography or magnetic resonance scan. The
cardiovascular profile included information on al-
cohol consumption (?14 and 21 and ?14 and 21
drinks per week for women and men, respectively,
representing under and over limit values set by the
Danish National Board of Health), current daily
smoking, diabetes mellitus, atrial fibrillation (chr-
onic or paroxysmal), arterial hypertension, previous
myocardial infarction, previous stroke, and intermit-
tent arterial claudication. Diagnosis of diabetes mel-
litus, atrial fibrillation, arterial hypertension, previ-
ous myocardial infarction, previous stroke, and
intermittent arterial claudication was based on cur-
rent Danish standards11and was either known be-
fore onset of stroke or diagnosed during hospitaliza-
tion. Stroke was defined according to World Health
This study included patients with first-ever isch-
emic stroke admissions only by excluding those
who have been hospitalized for stroke before May
8, 2000. For patients with ?1 stroke hospitaliza-
tion, after May 8, 2000, only the first event was
included in the analysis. Patients with hemor-
rhagic stroke, transient ischemic attacks, or pa-
tients younger than 18 years were excluded from
the study, as were patients in whom computed
tomography and magnetic resonance scans were
not performed (0.4%) or were unavailable (0.7%).
Time origin for the analysis was the date of hospi-
The National Indicator Project registry coverage
was estimated by professional consensus to be
about 80% of all stroke admissions in Denmark.14
The very high proportion of stroke patients admit-
ted to hospital (90%) is due to free hospital care in
Survival of the patients included in the Danish
National Indicator Project database was followed
through the Danish Central Person Registry, with
information on the vital status, including date of
death or emigration. We studied all-cause mortal-
ity only. Inclusion of patients started on May 8,
2000 and end of the study follow-up was on Jan-
uary 16, 2007. Less than 0.2% of the patients were
lost to follow-up. The study was approved by the
board of Danish National Indicator Project and the
Danish Data Protection Agency.
We calculated overall and sex-specific preva-
lence of cardiovascular risk factors at stroke onset
for 26,818 study subjects and 3-month (90 days)
case fatality. We modeled stroke severity by gen-
eralized additive linear model and adjusted for age
using restricted cubic splines, and for cardiovascu-
lar risk factors as factor variables. We presented
graphically the functional form of the relationship
between stroke severity and age separately for
males and females. We modeled 3-month case fa-
tality with generalized additive logistic model, ad-
justed for age and stroke severity using restricted
cubic splines, and for cardiovascular risk factors as
factor variables, and presented the results of this
model graphically. Although emphasis was put on
describing the sex-specific differences for all ages
and levels of SSS, statistical significance was
proven whenever the 95% confidence intervals did
not include the null hypothesis of no effect, that
is, odds ratio of 1. The statistical software R16was
used for the statistical analyses.
Of the 26,818 patients included in the study,
13,005 (48.5%) were women and 13,813 (51.5%)
were men. Within 3 months (90 days), 1547
women (11.9%) and 1186 men (8.6%) died (P ?
0.001). Mean age in men was 68.9 (SD 12.6) years
and in women was 73.9 (13.8) (P ? 0.001). Strokes
were more severe in women (mean [SD] SSS score,
42.2 [16.0]) than in men (mean [SD] SSS score, 45.6
[SD 4.2]) (P ? 0.001). Mean time from symptoms
to hospitalization did not differ by sex (men, 0.90
days; women, 0.88 days).
Prevalence of cardiovascular risk factors and
3-month case fatality in men and women can be
seen in the Table. Men were more often smok-
ers and alcohol consumers and more often had
diabetes, previous myocardial infarction, and in-
termittent arterial claudication. Women were
more often diagnosed with hypertension and
Figure 1 shows admission stroke severity re-
lated to age in men and women, adjusted for all
cardiovascular risk factors, for all 26,818 patients
(Figure 1A) and for 24,085 who survived 3
months after stroke (Figure 1B). Almost identical
results for 2 groups show more-severe strokes in
women. Admission stroke severity increased sig-
nificantly with age in men and in women. From
the age of about 50 years (all) and 60 years (3-
month survivors), strokes were more severe in
women than men; the difference was significant
when age was older than 75 years.
A multiple logistic regression model adjusting
for age, stroke severity, and all cardiovascular risk
factors revealed nonsignificantly lower case fatal-
ity in women compared with men (odds ratio ?
0.88; 95% confidence interval, 0.77?1.02). This
calculation expresses the average odds ratio within
the entire age spectrum. We then calculated the
effect of age from the multiple logistic regression
model (Figure 2). Mortality curves are shown for
fixed risk factors and a stroke severity score of 50
(corresponding to median SSS of the total study
population). Survival was better for women; signif-
icant within the age range of 78 to 90 years.
Given the same age and risk-factor profile, this
study shows that strokes are more severe in women
than in men; significantly when older than 75
years. Nevertheless, 3-month case fatality is lower
in women than in men of the same age, initial
stroke severity, and risk factor profile; significantly
when older than 78 years. As a result, women who
have survived 3 months after stroke had more-
severe strokes on admission than male survivors
with the same age and risk factor profile, the dif-
ference between men and women was significant
from the age of 75 years. All things being equal,
T.S. Olsen et al.
women appear to survive stroke better than men.
Because of this and because strokes are more severe
in women than men, women who survive stroke
are expected to be living with the consequences of
more-severe strokes than men of the same age and
with the same cardiovascular risk factor profile.
Our study helps explain why functional out-
comes are worse in women and why more women
than men are in nursing homes after stroke. Not
only is stroke more severe in women, women also
survived stroke better than men, resulting in more-
severe strokes among female survivors. The result
of our study is in line with several recent reports
on worse functional outcomes of stroke in women
compared with men.3,7,8We did not measure func-
tional disability at 3 months after stroke in this
study. However, initial stroke severity measured by
SSS is directly related to functional disability in
stroke survivors as measured by the Barthel Index
at discharge from hospital after end of stroke reha-
bilitation.17The absolute difference in mean ad-
mission SSS score between men and women was
3.3 points. The clinical meaning of the SSS score
was studied in the Copenhagen Stroke Study. A
mean 5-point increase in admission SSS score cor-
related to an estimated overall reduction of in-
Table. Distribution of risk factors (overall and by sex) and 3-month case-fatality among 26,818 patients with first-ever
(n ? 26,818), n (%)
Died Within 3 Months
(n ? 2733), n (%)
(n ? 13,813), n (%)
(n ? 13,005), n (%)
Current daily smoking*
*P ? 0.001, ?2test for difference between sexes.
hospital mortality from 20.3% to 15.5%, a reduc-
tion in the percentage of patients discharged to
nursing home from 14.3% to 13.1%, and an in-
crease in discharge to own home from 65.4% to
71.4%. The difference of 3.3 points between men
and women in this study can be considered of
Most studies on survival after stroke do not
show significant differences between sexes.7Re-
cently, however, studies including large numbers
of cases (?20.000) do report better survival of fe-
males, in agreement with this study.4,6,18,19As
women survived stroke better than men, when
adjusting for age and cardiovascular risk factor
profile, our study points to female superiority in
There is no agreement in the literature on
stroke severity in men and women. Some studies
do not observe differences in stroke severity be-
tween sexes,7–10and others, as in this study,
report strokes to be more severe in women than
in men.20In another large nationwide Scandina-
vian registry, level of consciousness on hospital
admission was lower in women than in age-
matched men, an indication of more-severe
The reason why stroke was more severe in
women than men remains obscure. Differences be-
tween sexes in response to cerebral infarction (ex-
pressed by the SSS), as well as sex differences in
infarct volumes as the result of arterial occlusion,
are possible explanations. Information on infarct
volumes, which was not available in this study, is
needed to further elucidate the issue.
We consider our study population to be repre-
sentative of the Danish hospital stroke population:
the study is nationwide, covering about 80% of all
hospitalized stroke patients in Denmark,14and
data completeness is high, exceeding 85% for all
individual variables. Most patients with acute
stroke are hospitalized in Denmark15; however, a
minority are not. It should be stressed that our
4050 60 7080 90100
Figure 2. Age-specific 3-month case fatality in patients
with Scandinavian Stroke Scale score 50 by sex,
adjusted for cardiovascular risk factors. Median
Scandinavian Stroke Scale score of all 26,818
patients was 50; 95% confidence intervals are
Scandinavian Stroke Scale
20 4060 80100
20 4060 80 100
Scandinavian Stroke Scale
Figure 1. Age and sex-specific stroke severity (Scandina-
vian Stroke Scale, 0 worst to 58 best) in all
26,818 patients with first-ever ischemic stroke
(A) and in 24,085 patients surviving at 3 months
(B), adjusted for cardiovascular risk factors;
95% confidence intervals are indicated.
T.S. Olsen et al.
observations apply to hospitalized stroke patients
only. It could be argued that outcomes in women
are poorer because of less-intensive treatment of
women with stroke. Studies from Denmark,4Can-
ada,9and Australia10did not find evidence of sex-
related differences in stroke management and care.
In addition, in our study, time from symptoms to
hospitalization was not influenced by sex and sur-
vival was better in women. Admission stroke se-
verity increased significantly with age in men and
in women, and more women than men were ad-
mitted with severe strokes, thereby arguing against
the idea that the prospect of poor outcomes (eg,
more-severe stroke) made elderly female stroke pa-
tients abstain from hospitalization.
At the time of stroke, women were older and had
more-severe strokes. Our study demonstrates the
importance of adjusting for age and stroke severity
when studying the effect of sex on survival from
stroke, otherwise case fatality in women will be
overestimated. Female superiority in stroke sur-
vival, as demonstrated here, should be considered
when studying functional outcomes. If not, wom-
en’s ability to recover functionally after stroke will
Prestroke disability and institutional living are
more common in women than in men3,8and fur-
ther contribute to understanding why functional
outcomes are worse in women after stroke.
Our study examines the paradox that women, de-
spite the fact they are better able to survive stroke,
end up being more disabled. Higher degree of dis-
ability in women after stroke cannot be taken as an
indication of women being more fragile than men.
It is rather an expression of the opposite. When
women survive strokes men die of, it implies that
they subsequently have to live with more-severe
strokes and have greater disability. This is en-
hanced by stroke being more severe in women
than in men.
Drs. Olsen, Z.J. Andersen, and K.K. Andersen con-
tributed to the conception and design of the study,
acquisition and interpretation of data, critical re-
vision for important intellectual content, and final
approval of the submitted version. Dr. K.K. Ander-
sen performed the statistical analysis. Dr. Olsen
drafted the article.
CONFLICTS OF INTEREST
The authors have indicated that they have no con-
flicts of interest regarding the content of this
1. United Nations. Demographic Yearbook. United
Nations: Geneva; 2000.
2. Kruger DJ, Nesse RM. Sexual selection and the
male: female mortality ratio. Evol Psychol. 2004;2:
3. Glader E-L, Stegmayer B, Norrving B, et al. Sex
differences in management and outcome after
stroke. A Swedish perspective. Stroke. 2003;34:
4. Palnum KD, Andersen G, Ingeman I, et al. Sex-
related differences in quality and short-term mor-
tality among patients with acute stroke in Den-
mark. A nation-wide follow-up study. Stroke. 2009;
5. Vaartjes I, Reitsma JB, Berger-van Sijl M, Bots ML.
Gender differences in mortality after hospital ad-
mission for stroke. Cerebrovasc Dis. 2009;28:564–
6. Olsen TS, Dehlendorff C, Andersen KK. Sexrelated
time-dependent variations in poststroke survi-
val—evidence of a female stroke survival advan-
tage. Neuroepidemiology. 2007;29:218–225.
7. Reeves MJ, Bushnell CD, Howard G, et al. Sex
differences in stroke: epidemiology, clinical pre-
sentation, medical care, and outcomes. Lancet
8. Petrea RE, Beiser AS, Seshadri S, et al. Gender
differences in stroke incidence and poststroke
disability in the Framingham Heart Study. Stroke.
9. Kapral MK, Degani N, Hall R, et al. Gender differ-
ences in stroke care and outcomes in Ontario.
Women’s Health Issues. 2011;21–2:171–176.
10. Gall SL, Donnan G, Dewey HM, et al. Sex differ-
ences in presentation, severity, and management
of stroke in a population-based study. Neurology. Download full-text
11. Mainz J, Krog BR, Bjørnshave B, Bartels P. Nation-
wide continuous quality improvement using clin-
ical indicators: the Danish National Indicator Proj-
ect. Int J Qual Health Care. 2004;16(Suppl I):
12. Lindenstrøm E, Boysen G, Christiansen LW, et al.
Reliability of Scandinavian Stroke Scale. Cerebro-
vasc Dis. 1991;1:103–107.
13. Report of the WHO Task Force on Stroke and
Other Cerebrovascular Disorders: Stroke?1989:
recommendations on stroke prevention, diagno-
sis, and therapy. Stroke. 1989;20:1407–1431.
14. Nip-Apopleksi. National audit report 2009. Final ver-
sion 1.108.04. 2010. Available at: www.sundhed.
national_sfa2009.pdf. Accessed August 10, 2011.
15. Jørgensen HS, Plesner AM, Hûbbe P, Larsen K.
Marked increase of stroke incidence in men be-
tween 1972 and 1990 in Frederiksberg, Denmark.
16. Ihaka I, Gentleman R. R. A language for data analysis
and graphics. J Computat Graph Stat. 1996;5:299–314.
17. Jørgensen HS, Nakayama H, Raaschou HO, Olsen
TS. Acute stroke: prognosis and a prediction of the
effect of medical treatment on outcome and
health care utilization. The Copenhagen Stroke
Study. Neurology. 1997;49:335–1342.
18. Ogbu UZ, Westert GP, Slobbe LCJ, et al. A multi-
faceted look at time of admission and its impact
on case-fatality among a cohort of ischaemic
stroke patients. J Neurol Neurosurg Psychiatry.
19. Vaartjes I, Reitsma JB, Berger-van Sijl M, Bots ML.
Gender differences in mortality after hospital admis-
sion for stroke. Cerebrovasc Dis. 2009;28:564–571.
20. Appelros P, Stegmayr B, Terént A. Sex differences
in stroke epidemiology. A systematic review. Stroke.
21. Eriksson M, Glader E-L, Norrving B, et al. Sex dif-
ferences in stroke care and outcome in the Swed-
ish National Quality Register for Stroke Care.
Address correspondence to: Tom Skyhøj Olsen, MD, PhD, The Stroke Unit, Frederiksberg University Hospital,
DK-2000 Frederiksberg, Denmark. E-mail: firstname.lastname@example.org
T.S. Olsen et al.