An increased risk of stroke among young schizophrenia patients
Herng-Ching Lina, Fei-Hsiu Hsiaob, Stefani Pfeifferc,
Yi-Ting Hwangd, Hsin-Chien Leee,f,⁎
aSchool of Health Care Administration, Taipei Medical University, 250 Wu-Hsing St., Taipei 110, Taiwan
bTaipei Medical University, College of Nursing, Taipei, Taiwan
cDepartment of History, Rutgers University, New Jersey, USA
dNational Taipei University, Department of Statistics, Taipei, Taiwan
eTaipei Medical University Hospital, Department of Psychiatry, Taipei, Taiwan
fTaipei Medical University, School of Medicine, Department of Psychiatry, Taipei, Taiwan
Received 23 October 2007; received in revised form 30 November 2007; accepted 23 December 2007
Available online 8 February 2008
Objective: This study sets out to estimate the risk of stroke developing among young schizophrenia patients during a five-year
follow-up period after hospitalization for episodes of acute exacerbation.
Methods: Hospitalized schizophrenia patients under 45 years of age were identified from the Taiwan National Health Insurance
Research Database for the year 1998 (n=5001). Two age-matched cases were randomly selected for each schizophrenia patient
from among patients who underwent appendectomies in the same year (n=10,002). Each individual patient was retrospectively
followed up from 1998 until the end of 2003 to determine whether any had developed strokes. Cox proportional hazard regressions
were carried out to compute the adjusted five-year survival rate.
Results: A total of 219 patients (1.46%) developed strokes during the five-year follow-up period, with the attacks occurring among
2.46% of schizophrenia patients and 0.94% of the comparison cohort. Following adjustment for patients' demographic
characteristics, select comorbid medical disorders and substance abuse, schizophrenia patients were found to be 2.02 times
( pb0.001) more likely to develop strokes during the follow-up period than age-matched appendectomy patients. The adjusted
hazard ratios of developing stroke for male and female schizophrenia patients were, respectively, 1.64 ( pb0.001) and 2.87
( pb0.001) times greater than their counterparts in the comparison group.
Conclusions: As compared with the comparison group, young schizophrenia patients demonstrated a two-fold increased risk of
developing stroke during the five-year period after hospitalization. The risk of developing stroke among schizophrenia patients was
found to be much higher for females than males.
© 2008 Elsevier B.V. All rights reserved.
Keywords: Schizophrenia; Stroke; Young stroke
Higher medical morbidity and mortality rates have
been reported in a number of studies on schizophrenia
patients (Brown, 1997; Osby et al., 2000; Brown et al.,
Available online at www.sciencedirect.com
Schizophrenia Research 101 (2008) 234–241
⁎Corresponding author. Department of Psychiatry, Taipei Medical
University & Hospital, 252 Wu-Hsing St., Taipei, 110, Taiwan. Tel.:
+886 2 2737 2181x3667; fax: +886 2 2378 9788.
E-mail address: email@example.com (H.-C. Lee).
0920-9964/$ - see front matter © 2008 Elsevier B.V. All rights reserved.
2000; Auquier et al., 2006), with the lines of evidence
suggesting that, compared to the general population, all
causes of mortality among this particular group of
patients are two to three times higher. In addition to
unnatural deaths (including both accidents and sui-
cides), the leading causes of excessive mortality in
schizophrenia patients are cardiovascular, cerebrovas-
cular and respiratory diseases.
Although numerous studies have examined the in-
cidence or prevalence of mortality from cardiovascular
and respiratory diseases among schizophrenia patients
over the past decade (Hennekens et al., 2005; McCrea-
die and Scottish Schizophrenia Lifestyle Group, 2003;
Davidson, 2002; Filik et al., 2006), little attention has
been given to any investigation of the incidence or risk
of developing cerebrovascular diseases within this
particular population. Curkendall et al. (2004) reported
that the adjusted risk of developing stroke was sig-
nificantly higher among schizophrenia patients (odds
ratio 2.1) than among comparable age- and sex-matched
individuals. Nevertheless, their study was heavily reliant
upon data from a single province, such that the findings
may not be generalized to the population as a whole.
Clearly, the lack of studies on the association be-
tween schizophrenia and the risk of developing cere-
brovascular diseases prevents mental healthcare and
related healthcare professionals from identifying further
the simultaneous contributions of biochemical, meta-
bolic and immune factors to the excessive morbidity and
mortality rates among schizophrenia patients. This study
therefore sets out to estimate the risk of stroke de-
veloping among young schizophrenia patients during a
five-year follow-up period after hospitalization for
episodes of acute exacerbation. Estimation of the risk
of developing stroke is also undertaken for another
cohort of patients who underwent appendectomies
during the same period. The risks for these two cohorts
are subsequently calculated and compared after adjust-
ing for demographic characteristics, select comorbid
medical conditions, and substance abuse.
This study used data from the National Health
Insurance Research Dataset (NHIRD), published by the
National Health Research Institute in Taiwan, covering
the years 1996 to 2003. The dataset includes all claims
data from Taiwan's National Health Insurance (NHI)
healthcare for all Taiwanese citizens. The NHI program
currently has over 21 million enrollees, representing
around 96% of the island's population, and is character-
ized by a single-payer payment system with unrestricted
access to any mental healthcare provider of the patient's
choice. Thus, the NHRID offers a unique opportunity to
identify the risk of stroke development among young
schizophrenia patients.Since the NHIRDconsistedofde-
identified secondary data released for public access for
research purposes, the study was exempt from fullreview
by the Internal Review Board (IRB).
2.2. Study sample
Our study design features a study cohort and a com-
parison cohort. The study cohort comprised all patients
under 45 years old who were hospitalized for schizo-
phrenia between January and December 1998 (any ICD-
9-CM 295 code other than 295.7-schizoaffective dis-
order). Inorder toavoid the potential confoundingfactors
of institutionalization and chronicity, those who had been
hospitalized for schizophrenia or who received any
psychiatric treatment for schizophrenia during the
previous two-year period were excluded from the study
cohort, as were patients previously diagnosed with stroke
(ICD-9-CM codes 430–438). To ensure the validity of
consensus schizophrenia diagnoses during the follow-up
period after the index hospitalization.
The comparison cohort was chosen from among
patients younger than 45 years old who were hospitalized
for an appendectomy (ICD-OP code 47.0) between
January and December 1998. The reason for the selection
of appendectomy patients as the comparison group was
that patients undergoing appendectomies were relatively
indistinguishable from the general population.
There were no statistically significant differences
population in Taiwan with regard to either gender
(pN0.05) or age (pN0.05). Furthermore, the procedures
involved in an appendectomy have no known long-term
impact on brain functioning, and indeed, there has never
been any study reporting any increased risk of schizo-
phrenia among patients undergoing an appendectomy.
Appendectomy patients were, however, excluded if they
had ever been diagnosed with any major psychiatric
disorder in the two years preceding appendectomy (ICD-
9-CM codes 290, 294, 295, 296 or 297). Again, patients
with a previous diagnosis of stroke were also excluded.
A total of 5001 patients suffered from schizophrenia
and 18,754 patients underwent appendectomies during
the study period. Since there is a significant difference in
the mean age between schizophrenia and appendectomy
235H.-C. Lin et al. / Schizophrenia Research 101 (2008) 234–241
patients (31.1 years for schizophrenia and 24.1 years for
appendectomies), we further refined our criteria for the
comparison group by randomly selecting 10,002
appendectomy patients (two for every schizophrenia
patient) matched with the study group in terms of age
(b18, 18–24, 25–34 and ≥35 years). Tracking of each
patient was subsequently undertaken, from their hospi-
talization in 1998 until the end of 2003, using ad-
ministrative data to identify all patients who developed
strokes. The data was also linked to mortality data in
Taiwan to calculate the stroke-free survival time after
hospitalization for a five-year period, with cases cen-
sored if individuals died from non-stroke causes during
that time (440 patients had already died from non-stroke
causes, comprising of 316 from the study cohort and
124 from the comparison cohort).
The regression modeling also adjusted for socio-
demographic characteristic variables including gender,
income-related insurance payment amount as a proxy for
income (0, NT$1–NT15,840, NT15,841–NT25,000,
≥NT25,001), level of urbanization and the geographical
location of the community in which the patient resided
(Northern, Central, Eastern and Southern Taiwan). The
reason for selecting NT$15,840 as the cutoff point was
that this value is stipulated by the government as the
minimum wage for full-time employees in Taiwan.
Prior studies have reported that the occurrence of
stroke is associated with both geographical regions and
levels of urbanization; these two variables are therefore
also taken into account in the regression modeling. In
National Health Research Institute, urbanization levels in
‘most urbanized’ communities and level 7 referring to the
‘least urbanized’ communities. However, given that there
were only very small numbers of schizophrenia cases in
levels 5, 6 and 7, these three levels were combined into
one singlegroup,whichwas thereafterreferredtoaslevel
5. Details on selectcomorbid medical disorders including
hypertension, diabetes, hyperlipidemia and substance
abuse were also extracted from the claims data at the
time of the index discharge, essentially because these
conditions may exacerbate the risk of stroke.
2.3. Statistical analysis
The SAS statistical package (SAS System for
Windows, Version 8.2) was used to perform the statistical
analyses in this study, with descriptive analyses being
carried out on all of the identified variables, including the
frequency and percentage. Pearson χ2tests were used to
of socio-demographic characteristics, select comorbid
medical disorders, substance abuse and the risk of stroke
development. The five-year stroke-free survival rate was
then estimated under the Kaplan–Meier method, with the
log-rank test being used to examine the differences
Demographic characteristics and comorbid medical disorders of
schizophrenia and appendectomy patient samples in Taiwan, 1998
Total no.%Total no.%
Income-related insured amount
236H.-C. Lin et al. / Schizophrenia Research 101 (2008) 234–241
between the two cohorts. Cox proportional hazard re-
gressions were also carried out as a means of computing
the adjusted five-year survival rate following adjustment
for the abovementioned variables. Finally, hazard ratios
(HR) are presented along with the 95% confidence
intervals (95%CI), with a significance level of 0.05
having been adopted for this study.
Of the total sample of 15,003 patients under the age
of 45 years, 219 patients (1.46%) had developed strokes
during the five-year follow-up period. The details of the
distribution of the demographic characteristics, select
comorbid medical disorders, and substance abuse for
these two cohorts are provided in Table 1.
The mean age of the sampled patients was 31.1 years.
As compared to the appendectomy patients, the schizo-
phrenia patients were more likely to be male, cormobid
with hypertension, diabetes or hyperlipidemia, and
residing in the less urbanized areas or in the southern
part of Taiwan at the time of the index discharge (all
The distribution of stroke for the two cohorts during
the five-year follow-up period, by demographic char-
acteristics, is provided in Table 2. The Pearson χ2tests
reveal that, as compared to the appendectomy patients,
with the exception of those residing in eastern Taiwan,
schizophrenia patients were more likely to develop
stroke during the follow-up period in every segment,
whether in terms of gender, age, income-related insured
amount, urbanization level or geographical region
The crude hazard ratios of stroke development for the
two cohorts are presented in Table 3, which reveals that
2.46% of the schizophrenia patients and 0.94% of the
appendectomy patients suffered from strokes between
1998 and 2003 (pb0.001).
For schizophrenia patients, the risk of stroke devel-
opment during the follow-up period was 2.721 times
(95% CI=2.142–3.517, pb0.001) that for appendect-
omy patients. Rather surprisingly, the risk of stroke
development among female schizophrenia patients was
significantly higher than that for female appendectomy
patients, by a multiple of 4.078 (95% CI=2.857–5.727,
The log-rank test indicating that schizophrenia
patients had significantly lower five-year survival rates
(pb0.001). Details of the adjusted hazard ratios of
Stroke development among schizophrenia and appendectomy patient
samples in Taiwan during the five-year follow-up period
Variable Stroke developmentp value
Yes NoYes No
66 2.3 2860 97.7 56 1.1 5015 98.9 b0.001
57 2.8 2018 97.3 40 0.8 4891 99.2 b0.001
44 2.1 2089 97.9 34 0.8 4247 99.2 b0.001
64 3.6 1720 96.4 75 2.1 3486 97.9 b0.001
917 98.5 13 0.7 1837 99.3
2 0.6308 99.4 0.381
Income-related insured amount
NT–15,84069 2.7 2521 97.3 22 1.7 1248 98.3
≥NT$25,001 1 0.6160 99.4 21 1.1 1879 98.9
21 2.3890 97.7 14 0.4 3617 99.6 b0.001
32 2.4 1307 97.6 35 1.1 3166 98.9 b0.001
27 2.2 1192 97.8 22 0.9 2419 99.1 b0.001
32 2.3 1378 97.7 25 0.9 2796 99.1 b0.001
13 1.6808 98.4 13 0.8 1627 99.2
15 2.0736 98.0 22 1.5 1458 98.5
36 4.5764 95.5 10 0.6 1610 99.4 b0.001
50 2.3 2089 97.7 35 0.7 4986 99.3 b0.001
41 3.5 1123 96.5 25 1.2 2085 98.8 b0.001
27 1.8 1446 98.2 26 1.0 2605 99.0
5 2.2220 97.8 4 1.8
Note: Total sample number=15,003.
Crude hazard ratios of stroke development during the five-year follow-
up period for schizophrenia and appendectomy patient samples in
Note: Total sample number=15,003.
237H.-C. Lin et al. / Schizophrenia Research 101 (2008) 234–241
stroke development, by cohort, are provided in Table 4.
Asthe table shows,following adjustmentfor the patients'
gender, income, level of urbanization, geographical
location and comorbid medical disorders, schizophrenia
patients were found to be 2.02 times more likely to
develop strokes during the follow-up period than their
counterparts in the appendectomy comparison group
(95% CI=1.53–2.54, pb0.001).
The adjusted hazard ratios of stroke development
during the five-year follow-up period were 1.64 (95%
CI=1.24–2.43, pb0.001) times more for male schizo-
phrenia patients than those for male appendectomy
patients and 2.87 (95% CI=1.82–4.36, pb0.001) times
more for female schizophrenia patients than those for
patients complicated with hypertension (hazard ratios
Adjusted hazard ratio of stroke development during the five-year follow-up period for schizophrenia and appendectomy patient samples in Taiwan
Total Male Female
Hazard ratio95% CI Hazard ratio95% CI Hazard ratio 95% CI
Appendectomy (reference group)
Female (reference group)
No (reference group)
No (reference group)
No (reference group)
No (reference group)
Income-related insured amount
0 (reference group)
1 (reference group)
Northern (reference group)
Note: Total sample number=15,003.
238H.-C. Lin et al. / Schizophrenia Research 101 (2008) 234–241
ranging between 10.66 and 13.98) and diabetes (hazard
ratios ranging between 3.97 and 4.05) had a greater
likelihood of the development of a stroke.
After adjusting for demographic characteristics, select
comorbid medical disorders and substance abuse, the
results of this study indicate that, as compared with the
comparison group of appendectomy patients, young
schizophrenia patients have a two-fold higher risk of
developing stroke within 5 years of their hospitalization
for acute exacerbation of schizophrenia. For schizophre-
nia patients, the risk of developing stroke was, rather
surprisingly, much greater among females than males.
Curkendall et al. (2004) reported a 2.1 adjusted hazard
ratio for stroke between schizophrenia patients and a
comparison group. Given that their study selected patients
from all age groups, the similarity with the figures
presented in this study reveal an urgent need for medical
care aimed at improved prevention, evaluation and
treatment options for young schizophrenia patients.
Furthermore, a recent study by Osborn et al. showed that
there was a significant increase in the risk of death from
mental illnesses, but no increased risk for mentally ill
mortality rates may reflect both the occurrence of disease
and its outcome, the increased risk of stroke mortality for
older patients with severe mental illness could possibly
indicate a higher morbidity rate for stroke during their
both the victims and their families. For schizophrenia
patients who have strokes, the emotional, social, or
their psychiatric condition and further compromise the
quality of life for both the patients and their caregivers.
Unhealthy lifestyles including poor diet, a lack of
exercise, smoking and obesity are considered to be major
causes of the excess morbidity and mortality rates for
schizophrenia patients (Brown et al., 2000). Indeed,
schizophrenia patients have a reported tendency to make
poor dietary choices, largely attributable to their social
disadvantages and schizophrenic symptoms (McCreadie,
2003). Inappropriate dietary habits, namely a low intake
of fruit and vegetables, together with a lack of exercise,
predispose them to stroke in the long run.
Despite the lack of information about diet within our
dataset, given that older patients and those who received
psychiatric treatment over the previous 2 years were
excluded from this study in order to avoid the
confounding factors of institutionalization and chroni-
city, the increased risk of developing stroke is unlikely
to be explained solely by the impact of unhealthy life-
styles. In fact, as compared with the lifestyle study of
community-dwelling schizophrenia patients undertaken
by McCreadie (2003), our study cohort had a much
lower unemployment rate (18.2% versus 98.0%),
making their lifestyles more similar to those of other
individuals in their communities.
Cigarette smoking is frequently seen as an important
risk factor for cardiovascular and cerebrovascular events
among schizophrenia patients (Brown et al., 2000;
Hennekens et al., 2005; Goff et al., 2005). However, a
prior study in Taiwan demonstrated only a slight in-
crease in the prevalence of tobacco use among schizo-
phrenia inpatients, compared to the high smoking rate
among the general population (Liao et al., 2002).
Therefore, smoking may not be a major reason con-
tributing to increased stroke rates found by our study
among young, community-dwelling schizophrenia
patients. Obesity—in particular the weight gain induced
by the use of antipsychotics—has been seen as con-
tributing significantly to the risk for numerous diseases,
including insulin resistance, lipid abnormalities and
hypertension (Newcomer et al., 2007), all of which can
result in vascular events such as stroke.
The lack of general healthcare among schizophrenia
patients, including recognition and treatment of rapidly
developing health conditions, may place this patient
population at greater risk for stroke; indeed, our prior
study found that only a small proportion of schizophrenia
patients with diabetes in Taiwan received appropriate
medical care (Lin et al., 2007). Another recent study also
indicated that inequalities remained in primary care for
coronary heart disease for patients with serious mental
illness, particularly in terms of identifying and treating
higher cholesterol levels among schizophrenia patients
comorbidity rates for hypertension, diabetes and hyperli-
medical care for such conditions far exceeds the number
of patients currently being treated.
The significantly higher risk of developing stroke
among young female schizophrenia patients represents a
significant departure from widely-held beliefs about the
relationship between gender and stroke occurrence.
Stroke incidence is reportedly higher in men than in
women, and women are, on average, several years older
than men at the time of their first stroke (Wyller, 1999).
However, it has been found that in schizophrenia
patients, obesity is more common among females than
males (Homel et al., 2002). Young female schizophrenia
239 H.-C. Lin et al. / Schizophrenia Research 101 (2008) 234–241
patients may exhibit disproportionate weight gain from
the use of psychotropic agents, especially atypical
antipsychotics. It is therefore suggested that a further
large-scale clinical follow-up study should be initiated
in order to clarify the mechanisms involved in the high
risk of developing stroke among young female schizo-
study had not taken any psychotropic agents in the two-
drug effects on the cerebrovascular system may develop
development of cerebrovascular disease after starting
psychotropic medications could be a cause of worry for
both patients and clinicians alike.
By using a five-year baseline, starting from the first
diagnosed episode of schizophrenia, we have probably
excluded patients who have used such medications long
term, Because the chronic and socially disruptive nature
of the disease generally brings schizophrenia patients
into frequent contact with medical professionals (espe-
cially true in Taiwan, where younger people are likely to
have parents involved in such situations), we feel that by
excluding those who saw a doctor in the preceding two-
year period and those over age 45, it is very unlikely that
the patients in our study had previously received
treatment (i.e., medication), even if they had previous
(probably untreated) episodes of the disease.
the context ofthreelimitations.Firstofall,schizophrenia,
stroke or any secondary diagnoses that are totally reliant
those undertaken in person through a standardized
procedure. However, the NHI Bureau randomly samples
a fixed percentage of claims from every hospital each
year, so as to verify the diagnosis validity and the quality
of care through chart review by an independent peer
group. Any hospital with outlier charges or outlier
patterns for any diagnosis group faces the risk of audit
and subsequent heavy penalties by the NHI Bureau when
discrepancies or overcharging are discovered. In order to
ensure the validity of the schizophrenia diagnoses in this
study, we ensured that all of the study cohort patients had
at least two consensus schizophrenia diagnoses after the
index hospitalization. Furthermore, virtually all hospitals
in Taiwan capable of admitting stroke patients are
equipped with CT or MRI scanners which can consider-
ably increase the validity of stroke diagnosis.
Our second limitation is that some variables such as
dietary habits, cigarette smoking and body mass index,
which may contribute to stroke development, were not
available in the dataset, which could compromise our
findings. Thirdly, because the database starts in 1995,
we cannot definitively identify and exclude the schizo-
phrenia patients who were chronically disabled or even
institutionalized before 1995. The NHIRD contains no
information regarding the patient's duration of illness
and number of episodes. Those variables may make
certain contributions to the development of stroke.
However, we specifically selected younger schizo-
phrenia patients who are less likely to have been
institutionalized or have had the disease for an extended
time for the study. This approach may result in problems
factors. Though sampled patients were free from
psychotropic agents for at least 2 years before their
index admission, no information is available on their
medication use before the two-year period.
The five-year baseline used in our study has one
particular advantage; the association we have identified
between schizophrenia and stroke suggests physicians
should attend to this possibility starting from a patient's
first known episode of the disease. Further studies
therefore need to be initiated to verify whether schizo-
phrenia patients are at an higher risk for stroke. The
increased burden of stroke among young schizophrenia
patients may elevate the importance of developing early
intervention and prevention programs to reduce the
morbidity and mortality from stroke. Our findings stress
the urgent need for more active monitoring and treatment
for schizophrenia patients of the well-recognized risk
factors associated with stroke, including hypertension,
diabetes and hyperlipidemia; furthermore, this may need
to be undertaken at an earlier age in order to get the
maximum possible benefit.
Role of funding source
Author Herng-Ching Lin designed the study and wrote the draft.
Authors Fei-Hsiu Hsiao and Hsin-Chien Lee managed the literature
searches and analyses. Authors Senyeong Kao and Yi-Ting Hwang
undertook the statistical analysis. All authors contributed to and have
approved the final manuscript.
Conflict of interest
This study is based in part on data from the National Health
Insurance Research Database provided by the Bureau of National
Health Insurance, Department of Health and managed by the National
Health Research Institutes, Taiwan. The interpretations and conclu-
sions contained herein do not represent those of the Bureau of National
Health Insurance, Department of Health, or the National Health
240 H.-C. Lin et al. / Schizophrenia Research 101 (2008) 234–241
Auquier, P., Lancon, C., Rouillon, F., Lader, M., Holmes, C., 2006.
Mortality in schizophrenia. Pharmacoepidemiol. Drug Saf. 15,
Brown, S., 1997. Excess mortality of schizophrenia. A metaanalysis.
Br. J. Psychiatry 171, 502–508.
Brown, S., Inskip, H., Barraclough, B., 2000. Causes of the excess
mortality of schizophrenia. Br. J. Psychiatry 177, 212–217.
Curkendall, S.M., Mo, J., Glasser, D.B., et al., 2004. Cardiovascular
disease in patients with schizophrenia in Saskatchewan, Canada.
J. Clin. Psychiatry 65, 715–720.
Davidson, M., 2002. Risk of cardiovascular disease and sudden death
in schizophrenia. J. Clin. Psychiatry 63, 5–11.
Filik, R., Sipos, A., Kehoe, P.G., et al., 2006. The cardiovascular and
respiratory health of people with schizophrenia. Acta Psychiatr.
Scand. 113, 298–305.
Goff, D.C., Cather, C., Evins, E., et al., 2005. Medical morbidity and
mortality in schizophrenia: guidelines for psychiatrists. J. Clin.
Psychiatry 66, 183–194.
Hennekens, C.H., Hennekens, A.R., Hollar, D., Casey, D.E., 2005.
Schizophrenia and increased risks of cardiovascular disease. Am.
Heart J. 150, 1115–1121.
Hippisley-Cox, J., Parker, C.,Coupland,C.A., et al., 2007. Inequalities
in the primary care of patients with coronary heart disease and
serious mental health problems: a cross sectional study. Heart 93,
Homel, P., Casey, D., Allison, D.B., 2002. Changes in body mass
index for individuals with and without schizophrenia, 1987–1996.
Schizophr. Res. 55, 277–284.
Liao, D.L., Yang, J.Y., Lee, S.M., et al., 2002. Smoking in chronic
Lin, H.C., Chen, C.S., Liu, T.C., et al., 2007. Lack of care for diabetes
among schizophrenia patients. Schizophr. Res. 89, 353–354.
McCreadie, R.G., 2003. Diet, smoking and cardiovascular risk in
people with schizophrenia. Br. J. Psychiatry 183, 534–539.
McCreadie, R.G., Scottish Schizophrenia Lifestyle Group, 2003. Diet,
smoking and cardiovascular risk in people with schizophrenia:
descriptive study. Br. J. Psychiatry 183, 534–539.
Newcomer, J.W., Haupt, D.W., Fucetola, R., 2007. Abnormalities in
Arch. Gen. Psychiatry 59, 337–545.
Osborn, D.P., Levy, G., Nazareth, I., et al., 2007. Relative risk of
cardiovascular and cancer mortality in people with severe mental
illness from the United Kingdom's General Practice Rsearch
Database. Arch. Gen. Psychiatry 64, 242–249.
Osby, U., Correia, N., Brandt, L., et al., 2000. Mortality and causes of
death in schizophrenia in Stockholm county, Sweden. Schizophr.
Res. 45, 21–28.
Wyller, T.B., 1999. Stroke and gender. J. Gend. Specif. Med. 2, 41–45.
241H.-C. Lin et al. / Schizophrenia Research 101 (2008) 234–241