A systematic review of mortality in schizophrenia: is the differential mortality gap worsening over time?

Abstract

Despite improvements in mental health services in recent decades, it is unclear whether the risk of mortality in schizophrenia has changed over time.
To explore the distribution of standardized mortality ratios (SMRs) for people with schizophrenia.
Broad search terms were used in MEDLINE, PsychINFO, Web of Science, and Google Scholar to identify all studies that investigated mortality in schizophrenia, published between January 1, 1980, and January 31, 2006. References were also identified from review articles, reference lists, and communication with authors.
Population-based studies that reported primary data on deaths in people with schizophrenia.
Operationalized criteria were used to extract key study features and mortality data.
We examined the distribution of SMRs and pooled selected estimates using random-effects meta-analysis. We identified 37 articles drawn from 25 different nations. The median SMR for all persons for all-cause mortality was 2.58 (10%-90% quantile, 1.18-5.76), with a corresponding random-effects pooled SMR of 2.50 (95% confidence interval, 2.18-2.43). No sex difference was detected. Suicide was associated with the highest SMR (12.86); however, most of the major causes-of-death categories were found to be elevated in people with schizophrenia. The SMRs for all-cause mortality have increased during recent decades (P = .03).
With respect to mortality, a substantial gap exists between the health of people with schizophrenia and the general community. This differential mortality gap has worsened in recent decades. In light of the potential for second-generation antipsychotic medications to further adversely influence mortality rates in the decades to come, optimizing the general health of people with schizophrenia warrants urgent attention.

Full text

Page 1

META-ANALYSIS
A Systematic Review of Mortality in Schizophrenia
Is the Differential Mortality Gap Worsening Over Time?
Sukanta Saha, MSc, MCN; David Chant, PhD; John McGrath, MD, PhD, FRANZCP
Context: Despite improvements in mental health ser-
vices in recent decades, it is unclear whether the risk of
mortality in schizophrenia has changed over time.
Objective: To explore the distribution of standardized
mortality ratios (SMRs) for people with schizophrenia.
Data Sources: Broad search terms were used in
MEDLINE, PsychINFO, Web of Science, and Google
Scholartoidentifyallstudiesthatinvestigatedmortality
inschizophrenia,publishedbetweenJanuary1,1980,and
January 31, 2006. References were also identified from
reviewarticles,referencelists,andcommunicationwith
authors.
Study Selection: Population-based studies that re-
ported primary data on deaths in people with schizo-
phrenia.
Data Extraction: Operationalized criteria were used to
extract key study features and mortality data.
Data Synthesis:WeexaminedthedistributionofSMRs
and pooled selected estimates using random-effects
meta-analysis. We identified 37 articles drawn from 25
different nations. The median SMR for all persons for
all-cause mortality was 2.58 (10%-90% quantile, 1.18-
5.76),withacorrespondingrandom-effectspooledSMR
of2.50(95%confidenceinterval,2.18-2.43).Nosexdif-
ferencewasdetected.Suicidewasassociatedwiththehigh-
est SMR (12.86); however, most of the major causes-of-
deathcategorieswerefoundtobeelevatedinpeoplewith
schizophrenia.TheSMRsforall-causemortalityhavein-
creased during recent decades (P=.03).
Conclusions:Withrespecttomortality,asubstantialgap
exists between the health of people with schizophrenia
and the general community. This differential mortality
gap has worsened in recent decades. In light of the po-
tentialforsecond-generationantipsychoticmedications
to further adversely influence mortality rates in the de-
cades to come, optimizing the general health of people
with schizophrenia warrants urgent attention.
Arch Gen Psychiatry. 2007;64(10):1123-1131
I
Less widely appreciated is the fact that
people with schizophrenia are at in-
creased risk for premature death associ-
ated with comorbid somatic conditions.4
Apartfromadverseeffectsrelatedtomedi-
cation, schizophrenia can trigger a cas-
cade of socioeconomic and lifestyle fac-
torsthat,inturn,cantranslateintoadverse
physical health outcomes. These comor-
bid physical conditions contribute to in-
creasedmortalityrisksamongpeoplewith
schizophrenia.
The association between severe men-
talillnessandincreasedmortalityrateshas
longbeenrecognized.5Withrespecttothe
T IS NOW WIDELY ACKNOWL-
edged that schizophrenia con-
tributessubstantiallytotheglobal
burdenofdisease.1,2Itisalsowell
known that schizophrenia is
associated with elevated suicide rates.3
group of disorders now labeled schizo-
phrenia, increased mortality rates have
beentheobjectofscrutinysincetheearly
20thcentury.6-8Thequalityofresearchon
this topic has improved greatly in recent
decades, with access to larger, better-
characterized samples and the availabil-
ity of high-quality mortality data for the
general population. Access to these data
allows the calculation of the standard-
ized mortality ratio (SMR), which com-
paresmortalityinpeoplewithschizophre-
nia vs the general population. The SMRs
are calculated by dividing the observed
mortality rates in a given population (eg,
the number of deaths in a group of indi-
viduals with schizophrenia) by the ex-
pected mortality rates in that same group
aspredictedbyage-andsex-specificmor-
talityratesforastandardpopulation.Thus,
an SMR of 2.0 would indicate that people
Author Affiliations:
Queensland Centre for Mental
Health Research, The Park
Centre for Mental Health,
Wacol (Mr Saha and Drs Chant
and McGrath), and Department
of Psychiatry, The University of
Queensland, St Lucia
(Drs Chant and McGrath),
Australia.
(REPRINTED) ARCH GEN PSYCHIATRY/VOL 64 (NO. 10), OCT 2007WWW.ARCHGENPSYCHIATRY.COM
1123
©2007 American Medical Association. All rights reserved.
by GregLaub, on October 8, 2007 www.archgenpsychiatry.com Downloaded from

Page 2

with schizophrenia are twice as likely to die compared
with the general population. The SMRs can be calcu-
lated for overall mortality (all-cause) or for more spe-
cific, widely used categories (eg, cancer, cardiovascular
disease, endocrine disorders, or suicide).
Inrecentyears,severalscholarlyreviews4,9-11havenoted
highermortalityinschizophreniacomparedwiththegen-
eralpopulation.Ameta-analysis,4basedon18studiespub-
lishedbetween1969and1996,reportedanall-causeSMR
for people with schizophrenia of 1.51. Another meta-
analysis,11based on 20 studies published between 1973
and1995,reportedasimilarSMRforpeoplewithschizo-
phrenia (1.57). Although these 2 systematic reviews
agreed on the size of the pooled SMR associated with
schizophrenia, there were discrepancies in the sex dif-
ference of overall mortality ratios. Brown4found a small
but significant male excess in the overall mortality ra-
tio, whereas other studies12,13reported either no sex dif-
ference11or higher mortality ratios in females compared
with males.
In collating data from different sites, systematic re-
viewers need to appreciate the structure of the underly-
ing data. In light of the differing population age struc-
tureanddiseaseprofileamongsites,1,14wewouldexpect
substantialvariationinmortalityratiosamongsites.For
example, one would predict that SMRs for people with
schizophrenia would differ between developed and de-
veloping nations, where the profiles of disease and the
access to services vary markedly.
Because of the increased focus on mental health care
seen in many countries during the last few decades, one
might predict that SMRs associated with disorders such
asschizophreniashouldbedecreasingovertime.15,16How-
ever,severalauthorshavesuggestedthatSMRsinschizo-
phreniahavebeenincreasingduringrecentdecades.For
example, Osby et al17found a linear increasing trend of
mortalityduring5-yearperiodsfrom1976to1995among
peoplewithschizophrenia.Themeta-analysisbyBrown4
also reported significantly higher mortality in the 1980s
comparedwiththe1970s.Deinstitutionalizationmayhave
influenced recent secular changes in mortality rates in
schizophrenia.Althoughdeinstitutionalizationstartedin
the 1950s, findings on its relationship to mortality have
been inconsistent.10,11,18
The aims of this study were to undertake a system-
atic review of mortality in schizophrenia and to exam-
ine a limited number of planned sensitivity analyses. In
keeping with our previous systematic review of the in-
cidence19and prevalence20of schizophrenia and consid-
ering that variability is to be expected in systematic re-
views of SMRs,4,21we sought to preserve the expected
variation in the data rather than to focus only on pooled
values derived from meta-analysis. Thus, for the main
analyses, we present distributions of mortality esti-
mates with measures of central tendency (eg, median or
means) and quantiles (10% and 90% quantiles). On the
basis of all-cause SMR, we predicted that the SMRs of
malesandfemaleswouldnotdiffersignificantly.Wealso
predictedthatSMRsfromthedevelopedworldwoulddif-
fer from those from the developing world (nondirec-
tional hypothesis). We wished to explore the impact of
studyqualityonSMRs.Withtheassumptionthathigher-
quality studies would be more likely to identify deaths
in schizophrenia, we predicted that SMRs derived from
suchstudieswouldbehighercomparedwiththosefrom
lower-quality studies. On the basis of previous system-
atic reviews and commentaries, we predicted that SMRs
would increase over time.
METHODS
DATA SOURCES
Mostmortalitystudiesarebasedonrecordlinkage.Peoplewith
schizophrenia are identified via psychiatric case registers and
then subsequently linked to registers of cause of death. Some
studies13,22report mortality ratios based on hospital inpatient
cohorts. Other studies23,24have used community-based fol-
low-up data for people with schizophrenia who are first iden-
tifiedthroughcommunitysurveysandthenfollowedupforex-
tended periods.
IDENTIFICATION OF STUDIES
GuidelinesoutlinedbytheMeta-analysisofObservationalStud-
iesinEpidemiology25werefollowedtoidentifyandcollatemor-
tality studies. The broad search string of (schizo* or psych*)
and(mortalityoroutcomeorfollow-up)wasusedinMEDLINE,
PsychINFO, Web of Science, and Google Scholar to identify
all research studies that investigated mortality in schizophre-
nia. Potentially relevant articles (in all languages) were ac-
cessed to review the full text. Citations from significant ar-
ticles and review articles were scrutinized to locate additional
relevantarticles,bookchapters,andconferencepapers.TheWeb
of Science Cited Reference Search system was also used to lo-
caterelevantarticles.Finally,lettersore-mailsweresenttothe
senior authors of articles that met the inclusion criteria. These
authors were provided with an interim list of included studies
and asked to nominate missing studies.
INCLUSION AND EXCLUSION RULES
Studies were included if they met all the following criteria: (1)
publishedand/oravailablebetweenJanuary1,1980,andJanu-
ary 31, 2006, (2) reported deaths in people with schizophre-
nia as diagnosed by any criteria, (3) studied a population 15
years and older, (4) reported primary data on all-cause mor-
tality and/or cause-specific mortality, and (5) reported SMRs
and/or data on observed and expected deaths sufficient to cal-
culateSMRs.Studieswereexcludedifthey(1)involvedpeople
with a diagnosis other than schizophrenia (ie, studies that re-
ported on broader categories of psychosis were excluded), (2)
reported duplicate data, (3) reported SMRs solely attributable
tosuicide(thiswasthefocusofarecentsystematicreviewand
meta-analysis3),and(4)reportedmortalityinsubgroupsofthe
population (eg, homeless people,26twins,27and those in-
volved in clinical trials).
DATA ABSTRACTION
Once a study was included, data were extracted and entered
into a 3-level, normalized database that included study-level
variables (eg, authors, year of publication, and site), middle-
level variables (eg, age group, recruitment duration, case-
findingmethod,anddiagnosticcriteria),andestimate-levelvari-
ables(eg,generalandspecific-causeSMRsforallpersons,males,
or females). Two or more of the authors checked all data used
in the analysis. When disagreements arose, these were re-
(REPRINTED) ARCH GEN PSYCHIATRY/VOL 64 (NO. 10), OCT 2007 WWW.ARCHGENPSYCHIATRY.COM
1124
©2007 American Medical Association. All rights reserved.
by GregLaub, on October 8, 2007 www.archgenpsychiatry.com Downloaded from

Page 3

solvedbyconsensus.Ifrequired,wecontactedtheoriginalau-
thorsforclarificationofissues.Thefulldatasetisavailablefrom
the authors (www.qcmhr.uq.edu.au/epi).
To assess the impact of overall quality of the distribution
ofSMRs,wedevisedaqualityscore.Onthebasisofoperation-
alized criteria, this score rewarded studies that (1) used supe-
rior research design features (eg, more thorough case ascer-
tainment, published diagnostic criteria, methods to confirm
diagnosis, and longer periods of follow-up) and (2) provided
comprehensive reporting of the study results (eg, provision of
numerator,denominator,SMRs,detailsofsubjectattrition,and
description of the completeness of the data source). Full de-
tails of the quality score used in this review are available from
the authors (www.qcmhr.uq.edu.au/epi).
Insystematicreviews,itisimportanttoavoiddoublecount-
ingoftheindexvariable(deaths)bythesameordifferentstud-
ies. Thus, a key feature of this review is the application of se-
quentialfilterstoidentifydiscretemortalityestimates.Weapplied
a similar sorting algorithm to that used in our previous re-
viewsofschizophrenia.19,20Briefly,themortalityestimateswere
sorted into different causes of death. Study-level and middle-
level filters were applied to isolate data from multiple studies
thatoverlappedinbothtimeandplace.Thethirdfilterwasused
toselect1representativemortalityestimateforinclusioninthe
cumulativedistributionusingthe“mostinformative”rule.For
example, if 1 study presented multiple overlapping ratios, the
ratios based on the largest sample were preferred (ie, the wid-
est age range was preferred over narrower age strata).
Thehighest-order(andmostreliable)categoryofdeath,all-
cause mortality, can be further subdivided according to rules
suchasthosecodifiedbytheInternationalClassificationofDis-
eases, Ninth Revision (ICD-9).28Almost all included studies in
thisreviewwerecodedwiththeICD-9.Althoughdeathcanre-
sult from the combination of many different health problems,
in circumstances in which several codes may be suitable, em-
phasis is given to the underlying cause of death. More specific
causes of death can be allocated to categories according to or-
gan systems (eg, cardiovascular or gastrointestinal) or nature
of disease (eg, cancers are coded together). Apart from codes
for these specific domains, studies occasionally report SMRs
for middle-level categories such as all-unnatural (ICD-9 codes
E800-E999) (which includes codes for suicide, accident, and
homicide) and all-natural (ICD-9 codes 001-799; the remain-
der from all-cause when all-unnatural cause is excluded).
The SMRs were extracted from the publications or calcu-
lated by dividing the sum of observed deaths by the sum of ex-
pected deaths (when sufficient data were available to calculate
these). The distributions of SMRs were assessed in cumulative
plots,witheverySMRcontributingtothedistribution.Thedis-
tributionofthedatawasassessedinrankorderforSMRs(low-
est to highest ranks) with the cumulative percentage of SMRs
shown on the vertical axis. Key features of these distributions
arepresented(eg,median,mean,geometricmean,standardde-
viation, and quantiles at 10%, 25%, 50%, 75%, and 90%).
For all-cause death, we were often able to extract data on
case fatality rate (CFR). The CFR is calculated by dividing the
number of deaths in people with schizophrenia during a cer-
tain period by the number of people with that disorder at the
beginningoftheperiod.AnannualizedCFRwasderivedtoal-
low comparisons among studies of different durations.14
In keeping with definitions from our previous systematic
reviews of schizophrenia,20,29we divided studies according to
the per capita gross national product of the study site (based
on 2004 data)30and used a standard World Bank definition of
country status31: (1) least developed countries,mean income
of less than US $2995; (2) emerging economy countries,mean
incomebetweenUS$2995and$9266;and(3)developedcoun-
tries,mean income of greater than US $9266.
Toassessseculartrends,weusedmeta-regressiontoexam-
inetherelationshipbetweenthemidpointofthefollow-uppe-
riod and all-cause SMR for persons. Study quality scores were
divided into tertiles, and the distribution of all-cause SMR for
persons were compared according to these 3 levels.
We performed statistical analyses for the test of signifi-
cance between distributions of different SMRs. These analyses
takeintoaccount(1)theneedtocontrolforwithin-studyvaria-
tion(estimatesdrawnfromthesamestudytendtobemorealike
than SMRs drawn from different studies) and (2) the use of a
logtransformationtoanalyzedistributionsthatareoftenposi-
tivelyskewed.AnalyseswereperformedwithSASstatisticalsoft-
ware, version 9.2 (SAS Institute Inc, Cary, North Carolina).
We also undertook a secondary analysis based on conven-
tional meta-analytic techniques. Because SMRs are known to
varywidelyamongsitesbecauseofpopulationanddiseasefre-
quency differences, we adopted a random-effects model to es-
timateapooledSMRforall-causemortalityforpersons.21When
necessary, 95% confidence intervals (CIs) were generated ac-
cording to the formula detailed by Rothman and Greenland.21
Heterogeneity among the studies was tested using the Coch-
ran heterogeneity statistic.32Apart from the specific analyses
related to sex differences, we restricted the analyses to per-
sons to limit the number of planned comparisons. The fund-
ing source played no part in the design, analyses, writing, or
submission of this study.
RESULTS
The electronic search identified 1726 articles, whereas
manual reference checking identified an additional 26
references.Wereceivedresponsesfrom16authors,who
provided an additional 11 references. Four articles from
languages other than English were included after trans-
lation.Elevenstudies33-43wereexcludedbecausetheycom-
pletely overlapped with other included studies. Further
details of the results of the search strategy and key fea-
tures of the included studies are available from the au-
thors (www.qcmhr.uq.edu.au/epi).
Thesystematicreviewidentified37studies9,12,13,18,22-24,44-73
that provided data on 561 SMRs for different causes
of deaths drawn from 25 different countries: Australia
(n=2),59,68Brazil (n=1),61Bulgaria (n=1),53Canada
(n=3),50,51,65China(n=1),53Columbia(n=1),53CzechRe-
public (n=1),53Denmark (n=2),63,64Finland (n=3),18,22,23
France(n=2),46,48Germany(n=1),57HongKong(n=1),53
India (n=2),12,53Indonesia (n=1),58Ireland (n=2)53,62
Israel(n=1),73Italy(n=2),60,67Japan(n=3),53,69,71Norway
(n=1),52Russia (n=1),53Sweden (n=2),9,66Taiwan
(n=1),49the Netherlands (n=1),13the United Kingdom
(n=5),44,47,53,54,56and the United States (n=6).24,45,53,55,70,72
The SMRs were based on an estimated total of 22296
discrete deaths. Thirty-seven studies9,12,13,18,22-24,44-73pro-
vided SMRs for all-cause mortality for either all per-
sons, males, or females.
Figure 1showsthedistributionforall-causeSMRsfor
allpersons,males,andfemales.Themedianall-causeSMR
for all persons (based on 38 SMRs) was 2.58, with 10%
and 90% quantiles ranging from 1.18 to 5.76 (Table 1).
In other words, people with schizophrenia had 2.5 times
the risk of dying compared with the general population,
andthecentral80%ofallSMRsvariedovera4-foldrange.
Themedianannualizedall-causeCFRforallpersonswas
(REPRINTED) ARCH GEN PSYCHIATRY/VOL 64 (NO. 10), OCT 2007WWW.ARCHGENPSYCHIATRY.COM
1125
©2007 American Medical Association. All rights reserved.
by GregLaub, on October 8, 2007 www.archgenpsychiatry.com Downloaded from

Page 4

95.4 per 10000 population, with 10% and 90% quantiles
ranging from 57.2 to 301.7 (5-fold range).
Themedianall-causeSMRformales(3.02)wasslightly
higher than females (2.37); however, these 2 distribu-
tions were not statistically significantly different
(F1,18=0.0003; P=.99). For all persons, the median SMR
for natural causes of death was 2.41, and the 10% and
90% quantiles ranged from 0.99 to 4.10 (Table 1). El-
evated median SMRs were found in all of the specific
causes of death apart from cerebrovascular diseases.
Sevenstudies18,47,49,51,56,65,66publisheddataforthesum-
mary category of unnatural causes of death for all per-
sons, males, or females. Table 1 gives the distributions
ofSMRsforunnaturalcausesofdeath.Peoplewithschizo-
phrenia had 12 times the risk of dying of suicide com-
paredwiththegeneralpopulation(medianSMR,12.86).
Twenty-two studies* were identified that contrib-
uted28SMRsfordevelopedcountries,3studies53,58,61con-
tributed 6 SMRs for emerging economy countries, and 1
study53contributed 4 SMRs for least developed coun-
tries. When divided according to this criterion, the all-
cause SMR distributions were not significantly different
(F2,34=0.30; P=.74); the median all-cause SMRs for least
developed, emerging economy, and developed coun-
tries were 2.02, 2.19, and 2.79, respectively (Table 2).
When the all-cause SMRs for all persons were di-
vided into study quality score tertiles, no significant dif-
ferences were found between SMR distributions
(F2,24=0.61; P=.55). On the basis of follow-up periods,
weidentified8studies24,45,51,54,55,63,71,72withSMRsfromthe
1970s,10studies47,53,57-60,65-67,73withSMRsfromthe1980s,
and7studies22,46,48,53,61,62,68withSMRsfromthe1990s.Con-
cerningsecularchange,meta-regressionconfirmedasig-
nificant positive association between the follow-up pe-
riodmidpointyearandall-causeSMR(slopecoefficient,
0.06; 95% CI, 0.01-0.11; z=2.21; P=.03). The median
SMRs for the 1970s, 1980s, and 1990s were 1.84, 2.98,
and3.20,respectively.ConcerningCFRs,themedianrates
per 10000 population (all-cause mortality) were 162.2,
95.4,and108.3forthe1970s,1980s,and1990s,respec-
tively. The CFRs for the 3 decades were not statistically
significantly different (F2,23=0.38; P=.38).
The 38 studies that report all-cause SMRs for all per-
sons are shown in a traditional forest plot with a pooled
estimate based on a random-effects model in Figure 2.
Usingtraditionalmeta-analytictechniques,wefoundthat
thepooledrandom-effectsall-causeSMR(basedon37SMRs
with finite standard errors) for all persons was 2.50 (95%
CI,2.18-2.83).TheCochranQtestfoundamarginallyac-
ceptablelevelofheterogeneity(Q36=50.72;P=.06).Weun-
dertook several post hoc analyses to explore potential
sources of variation (eg, published vs unpublished diag-
nostic criteria, cohorts based on first-episode patients vs
all patients, cohorts based on inpatient and/or outpatient
samples, sites clustered according to World Health Orga-
nization regions, and SMRs attributable to suicide sorted
by decade). However, none of the post hoc comparisons
resulted in significantly different SMR distributions (data
not shown).
COMMENT
People with schizophrenia have a substantially in-
creased risk of death compared with the general popu-
lation.Overall,peoplewithschizophreniahave2.5times
theriskofdying.Thisreviewwasabletoextractdatafrom
37 studies that were conducted in 25 countries. As pre-
dicted,thedistributionofall-causeSMRsshowedpromi-
nent variability.
Confirmingthehypothesisthattherelativemortality
riskassociatedwithschizophreniaisincreasing,wefound
that SMRs have increased in a linear fashion during the
3 decades examined in this study. This finding is con-
sistentwithearlierstudies.4,17Consideringthat(1)CFRs
for schizophrenia did not significantly differ among the
decadesand(2)age-standardizedmortalityratesaregen-
erallydecreasinginmostnations,74-76thesefindingssug-
gest that people with schizophrenia have not fully ben-
efitedfromtheimprovementsinhealthoutcomesavailable
to the general population. The SMRs are ratio measures
and thus reflect differential mortality. If mortality rates
in the general population decrease over time at a faster
ratethanthoseforpeoplewithschizophrenia,thenSMRs
forpeoplewithschizophreniawillincreaseovertime.The
evidence from the current study suggests that this dif-
ferential mortality gap has widened over time.
Mental health services have advanced in many parts
of the world during the past few decades. Apart from a
different mix of community-based care, the introduc-
tion of the second-generation antipsychotic medica-
tions in the early 1990s was initially found to be associ-
ated with better quality of life and reduced risk of
relapse.77-79More recent trials have questioned the clini-
calsuperiorityofsecond-generationantipsychoticmedi-
cation,80,81andconcernisnowwidespreadaboutthead-
verse effects associated with these medications.82In
particular, compared with typical antipsychotics, sev-
eral of the second-generation antipsychotics are more
likely to cause weight gain and metabolic syndrome.83
Because the metabolic syndrome is associated with a 2-
*References 22, 24, 45-48, 51, 53-55, 57, 59, 60, 62, 63, 65-68,
71-73.
100
90
80
70
60
50
40
30
20
10
0
123
456789
SMR
SMR, Cumulative %
All persons
Males
Females
Figure 1. Cumulative plots of standardized mortality ratios (SMRs) for
all-cause mortality associated with schizophrenia by sex.
(REPRINTED) ARCH GEN PSYCHIATRY/VOL 64 (NO. 10), OCT 2007WWW.ARCHGENPSYCHIATRY.COM
1126
©2007 American Medical Association. All rights reserved.
by GregLaub, on October 8, 2007 www.archgenpsychiatry.comDownloaded from

Page 5

to3-foldincreaseincardiovascularmortalityanda2-fold
increase in all-cause mortality,84these adverse effects
wouldbeexpectedtocontributetoevenhigherSMRsin
the next few decades.85,86Unfortunately, we are unable
to explore the role of atypical medications as a contrib-
uting factor for the increasing SMRs associated with
schizophrenia (eg, deaths related to clozapine-induced
agranulocytosisordeathsrelatedtoatypicalantipsychotic-
induced weight gain). Adverse health outcomes associ-
ated with weight gain and/or metabolic syndrome (eg,
myocardialinfarction,cerebrovascularaccidents,orcan-
cer)maytakedecadestofullyemerge.Thus,itseemslikely
that studies undertaken in the 1990s (ie, the most re-
centstudiesincludedinthisreview)wouldcaptureonly
a small fraction of the eventual burden of mortality as-
sociated with the adverse effect profile of the second-
generationantipsychoticmedications.Inlightoftheris-
ing secular trends in SMRs already identified by this
review,theprospectoffurtherincreasesinmortalityrisks
for schizophrenia is alarming.
In keeping with the findings of Harris and Barra-
clough11andSimpson,10wefoundnosignificantsexdif-
ference in all-cause SMRs. Thus, although many well-
documentedsexdifferencesexistintheepidemiological
featuresofschizophrenia,19,87,88theincreasedriskofmor-
tality associated with schizophrenia affects men and
women equally.
Ofthespecific-causeSMRs,suicidewasassociatedwith
thehighestestimate:12timesgreaterthanexpectedfrom
the general population. In keeping with previous re-
views, the SMRs associated with many different types of
natural causes of death were elevated in people with
schizophrenia.Curiously,thecategoryneoplasticdisor-
der had one of the lowest median SMRs (1.37). Al-
though the median was still greater than 1, several rec-
ordlinkagestudies89havesuggestedthatcancersmaybe
significantly less prevalent in people with schizophre-
nia.Thecurrentreviewexaminesonlymortality,andstud-
ies that examine morbidity would be better able to ex-
plore this issue.90
We found no significant difference in SMRs among
siteswhensortedbyeconomicstatus.However,thismeta-
analysis identified just 3 studies53,58,61that provided dis-
crete SMRs from the least developed and emerging
economycountries;thus,cautionshouldbeexercisedin
the interpretation of this finding. Furthermore, a single
derivedvariablewasusedtodefineeconomicstatus,which
was applied at the ecological level.
Whatfactorshavecontributedtothedifferentialmor-
tality risk associated with schizophrenia? Many demo-
Table 1. SMRs for Schizophrenia by Cause of Death for All Persons
Causes of Death
No. of
SMRs
Quantile
Mean (SD)
Geometric
Mean 10%25% Median 75%90%
All-Cause and Middle-Level Categories
38 1.18
6 0.99
3 5.56
All-cause (ICD-9 codes 001-799/E800-E999)
All-natural cause (ICD-9 codes 001-799)
All-unnatural cause (ICD-9 codes E800-E999)
1.87
1.04
5.56
2.58
2.41
7.50
3.64
2.90
12.73
5.76
4.10
12.73
2.98 (1.75)
2.31 (1.18)
8.60 (3.71)
2.68
2.03
8.10
Natural Causes, Cause Specific
1.11
0.61
1.79
2.20
1.60
1.54
0.71
1.60
2.20
1.45
Cardiovascular diseases (ICD-9 codes 390-429)
Cerebrovascular diseases (ICD-9 codes 430-438)
Digestive diseases (ICD-9 codes 520-579)
Endocrine diseases (ICD-9 codes 250-259)
Infectious diseases (ICD-9 codes 001-139)
Genitourinary diseases (ICD-9 codes 580-629)
Neoplastic diseases (ICD-9 codes 140-239)
Nervous diseases (ICD-9 codes 345-349)
Respiratory diseases (ICD-9 codes 460-519)
Other diseases (ICD-9 codes 1-389/630-799)
7
3
5
3
3
3
7
4
6
3
1.40
0.61
2.24
2.20
1.60
1.54
1.00
1.95
2.39
1.45
1.79
0.69
2.38
2.63
4.29
3.70
1.37
4.22
3.19
2.00
2.49
1.30
2.50
11.66
7.80
4.29
2.01
6.57
3.80
3.40
3.60
1.30
17.50
11.66
7.80
4.29
2.40
7.00
9.30
3.40
2.01 (0.83)
0.87 (0.38)
5.28 (6.84)
5.50 (5.34)
4.56 (3.11)
3.18 (1.45)
1.44 (0.60)
4.26 (2.70)
4.01 (2.66)
2.28 (1.01)
1.88
0.82
3.34
4.07
3.77
2.90
1.33
3.55
3.51
2.14
Unnatural Causes, Cause Specific
1.20
0.66
Accident (ICD-9 codes E800-E949)
Suicide (ICD-9 codes E950-E959)
6 1.63
5.90
1.73
12.86
5.10
21.43
8.40 3.30 (2.88)
43.47 (95.11)
2.51
16.13 10 174.25
Abbreviations: ICD-9, International Classification of Diseases, Ninth Revision (ICD-9); SMRs, standardized mortality ratios.
Table 2. SMRs for Schizophrenia of All-Cause Mortality by Economic Development Status for All Personsa
Economic Development Status
Least developed countries
Emerging economy countries
Developed countries
No. of
SMRs
4
6
28
Quantile
Mean (SD)
2.32 (0.70)
3.52 (3.03)
2.96 (1.52)
Geometric
Mean
2.25
2.57
2.77
10%
1.88
1.04
1.18
25%
1.89
1.31
1.97
Median
2.02
2.19
2.79
75%
2.75
5.98
3.74
90%
3.36
8.43
5.69
Abbreviation: SMRs, standardized mortality ratios.
aF2,34= 0.30; P = .74.
(REPRINTED) ARCH GEN PSYCHIATRY/VOL 64 (NO. 10), OCT 2007 WWW.ARCHGENPSYCHIATRY.COM
1127
©2007 American Medical Association. All rights reserved.
by GregLaub, on October 8, 2007 www.archgenpsychiatry.com Downloaded from