Skip to main content Accessibility help
×
Home

Information:

  • Access
  • Cited by 6

Actions:

      • Send article to Kindle

        To send this article to your Kindle, first ensure no-reply@cambridge.org is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

        Note you can select to send to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Cancer mortality in patients with schizophrenia: systematic review and meta-analysis
        Available formats
        ×

        Send article to Dropbox

        To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

        Cancer mortality in patients with schizophrenia: systematic review and meta-analysis
        Available formats
        ×

        Send article to Google Drive

        To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

        Cancer mortality in patients with schizophrenia: systematic review and meta-analysis
        Available formats
        ×
Export citation

Abstract

Background

Previous studies have reported conflicting results on the association between schizophrenia and cancer mortality.

Aims

To summarise available evidence and quantify the association between schizophrenia and cancer mortality using meta-analysis.

Method

We systematically searched literature in the PubMed and Embase databases. Risk estimates and 95% confidence intervals reported in individual studies were pooled using the DerSimonian–Laird random-effects model.

Results

We included 19 studies in the meta-analysis. Among them, 15 studies reported standardised mortality ratios (SMRs) comparing patients with schizophrenia with the general population, and the pooled SMR was 1.40 (95% CI 1.29–1.52, P<0.001). The other four studies reported hazard ratios (HRs) comparing individuals with schizophrenia with those without schizophrenia; the pooled HR was 1.51 (95% CI 1.13–2.03, P = 0.006).

Conclusions

Patients with schizophrenia are at a significantly increased risk of cancer mortality compared with the general population or individuals without schizophrenia.

Footnotes

Declaration of interest

None.

Schizophrenia is a serious mental illness that has a profound effect on the patients, their families and society. 1 Despite its low prevalence in populations, schizophrenia is associated with an enormous economic burden worldwide. 2 In the USA, the economic burden of schizophrenia was estimated at 62.7 billion dollars in 2002 3 and 155.7 billion dollars in 2013. 4 Patients with schizophrenia are known to have a significantly higher risk of premature death, 57 with nearly 20% shorter life expectancy than the general population. 8 Although unnatural causes of death such as suicide, homicide and accidents partly contribute to the excess mortality of schizophrenia, more patients actually died from natural causes, such as cardiovascular diseases, respiratory diseases and cancers. 912 In Sweden, natural causes accounted for 90.9% and 82.3% of all deaths among women and men with schizophrenia respectively. 13 In a previous systematic review of mortality in schizophrenia, the median standardised mortality ratios (SMRs) were 2.58, 2.41 and 7.50 for mortality from all causes, natural causes and unnatural causes respectively, when comparing patients with schizophrenia with the general population. 5

During recent decades, there has been an immense interest in estimating the risk of cancer mortality after a diagnosis of schizophrenia. Findings from previous studies have been mixed with positive, null and inverse associations between schizophrenia and cancer mortality. Several studies, in particular early studies, reported a lower or similar risk of cancer mortality among patients with schizophrenia compared with the general population. For instance, in a retrospective cohort study (1957–1986) in Denmark, Mortensen & Juel reported a 15% lower risk of cancer mortality in men but a 17% higher risk of cancer mortality in women. 14 Similar results were found in subsequent studies in Japan 15 and West Australia. 16 However, a positive association between schizophrenia and cancer mortality was observed in other studies in a Danish population 17,18 and studies in other populations. 9,10,13,1927 In a large national cohort in the USA, Olfson et al 9 found that adults with schizophrenia had a 1.8-fold chance of dying from cancers compared with adults in the general population. Similarly, paradoxical findings were also reported in literature pertaining to cancer incidence after a diagnosis of schizophrenia. 2831 Cancers are usually invasive and life-threating; thus, it is important to accurately characterise cancer mortality patterns after a diagnosis of schizophrenia, which may help inform changes in clinical care to reduce cancer-related deaths in patients with schizophrenia. However, because the prevalence of both schizophrenia and cancer mortality are very low, a robust estimate of the association between schizophrenia and cancer mortality may not have been achievable in some previous studies. Therefore, this study was performed to systematically review the currently available evidence regarding cancer mortality in patients with schizophrenia and to quantify the association between schizophrenia and cancer mortality through a comprehensive meta-analysis.

Method

This study was conducted in accordance with the guidelines in the Meta-analysis of Observational Studies in Epidemiology statement. 32

Literature searches

We searched the PubMed and Embase databases for literature that was published up to 16 October 2016. The search terms were a combination of key words and standard subheading terms relevant to schizophrenia, cancer and mortality. Specifically, we used the following search terms in PubMed: (“Schizophrenia”[Mesh] or “schizophrenia”[tiab] or “schizophrenic”[tiab]) AND (“Neoplasms”[Mesh] or “cancer”[tiab] or “tumor”[tiab]) AND (“Mortality”[Mesh] or “mortality”[tiab] or “death”[tiab]). Similar search terms were constructed and used in the Embase database. Additionally, the references listed in any relevant articles or reviews were screened. No language restrictions were applied for the searches or study inclusion.

Study selection and data extraction

Any published article that reported the risk of cancer mortality in patients with schizophrenia was eligible for inclusion in this systematic review. During the screening steps, we excluded reviews, editorials or protocols that did not include original data. We also excluded studies on animals or cell lines, studies that did not evaluate schizophrenia as an exposure variable, and studies that did not include cancer mortality as an outcome variable. After detailed evaluation, we excluded studies if risk estimates and/or confidence intervals for the association between schizophrenia and cancer mortality were not reported and were unable to be calculated. We also excluded two studies 33,34 in which the results were updated by later reports 10,35 with longer follow-up in the same population. Another study was excluded because the participants already had cancer at baseline, with a focus on cancer fatality rather than mortality from cancer. 36

The following data were extracted from each included article: title, author, publication year, location, study design, number of participants, methods used for the assessment of schizophrenia and cancer death, statistical methods used for the analysis, risk estimates and 95% confidence intervals after adjustment for covariates, and any covariates that were adjusted or matched for in the multivariate model. When the original studies reported the results separately in men and women we considered them independent populations and extracted the risk estimates separately.

Statistical analysis

Risk estimates and 95% confidence intervals reported in individual studies were pooled in a meta-analysis using the DerSimonian–Laird random-effects model, which incorporates between-study heterogeneity in addition to sampling variation. 37 The majority of studies on cancer mortality in patients with schizophrenia used SMRs compared with the general population as their risk estimates. The relative risk (RR) of cancer mortality in comparing patients with schizophrenia with the general population, as reported in one study, 19 was assumed to approximate SMRs. Several studies reported hazard ratios (HRs) using a Cox proportional hazards model to compare individuals with schizophrenia with those who did not have schizophrenia. 13,23,25,26 Therefore, we separately summarised the results as pooled SMRs and pooled HRs in the meta-analysis given the difference in comparator populations. In one study, 26 the 95% confidence intervals were not directly reported, and therefore we calculated these on the basis of reported risk estimates and P-values following the method of Altman & Bland. 38

Heterogeneity across studies was assessed by both the χ2-based Cochran's Q statistic and the I 2 metric. 39 To explore the potential sources of heterogeneity, we conducted meta-regression analyses with the following factors: geographical location, sample size (⩾3000 v. <3000), follow-up duration (⩾10 v. <10 years) or adjustment for covariates (age and gender only v. age, gender and other factors). Additionally, we conducted stratified analyses by gender because of the observed gender disparity in several earlier studies.

The possibility of publication bias was visually inspected by funnel plot and statistically assessed using the Egger regression asymmetry test. 40 Sensitivity analyses were performed by omitting one study at a time and calculating a pooled estimate for the remainder of the studies to determine whether the results were markedly affected by a single study. We also used the fixed-effect model for all above analyses as another set of sensitivity analyses. All statistical analyses were performed using Stata software (version 14.0). A P < 0.05 was considered statistically significant.

Results

Characteristics of the included studies

Through a systematic search in literature databases and reference lists of relevant articles, we identified 356 potentially relevant articles. After implementing the screening process, we finally included 19 studies 9,10,1327,35,41 that fulfilled our eligibility criteria in the meta-analysis (Fig. 1). Two 10,35 of the included studies updated the data of their earlier reports 33,34 in the same population; thus, the earlier studies 33,34 were not included in our list. Of the 19 included studies 11 were conducted in Europe, 4 in Australia, 3 in North America, and 1 in East Asia (Table 1). The majority of the included studies had a retrospective cohort study design with population-based record linkage data. In those studies, schizophrenia was usually defined according to clinical diagnosis in medical records, register, or administrative data and cancer death was ascertained from national or regional registries of vital statistics. The number of patients with schizophrenia in the included studies ranged from 370 to 1 138 853, with most studies having over 1000 patients. The follow-up period varied from 6 to 37 years, and the majority of the included studies had a follow-up duration of 10 years or longer.

Fig. 1 Flow diagram for literature search and study selection.

Table 1 Characteristics of the studies included in the meta-analysis

Authors, year Location Study design Participants with

schizophrenia, n
Assessment

of schizophrenia
Assessment

of cancer death
Follow-up,

years
Risk estimate

(95% CI)
Adjusted covariates
Mortensen &

Juel (1990) 14
Denmark Retrospective cohort

study, 1957–1986
6178 Hospital records based on the

Kraepelinian concept
Danish Register of Causes of

Death
29 SMR: 0.85 (0.76–0.94) in men

and 1.17 (1.06–1.28) in women
Age (reported separately by gender)
Mortensen &

Juel (1993) 41
Denmark Retrospective cohort

study, 1970–1987
9156 Danish Psychiatric Case Register Danish Register of Causes of

Death
17 SMR: 0.81 (0.54–1.19) in men

and 1.01 (0.75–1.33) in women
Age (reported separately by gender)
Saku et al

(1995) 15
Japan Retrospective cohort

study, 1948–1985
4980 Medical records, according to

DSM-III-R 1987
Japanese family registration

system (Koseki) and death

certificate
37 SMR: 0.84 (0.54–1.25) in men

and 1.37 (0.80–2.19) in women
Age (reported separately by gender)
Lawrence

et al (2000) 16
Australia Retrospective cohort

study, 1982–1995
N/A Western Australian Health Services

Research Linked Database
Western Australian Cancer

Register and Death Register
13 SMR: 0.90 (0.71–1.80) in men

and 1.19 (1.05–1.40) in women
Age (reported separately by gender)
Heila et al

(2005) 19
Finland Retrospective cohort

study, 1980–1996
58 761 Finnish National Hospital Discharge

Register
Finnish National Causes of

Death Register
16 RR: 1.50 (1.41–1.58) in men and

1.48 (1.40–1.57) in women a
Age and calendar year (reported

separately by gender)
Laursen

et al (2007) 17
Denmark Retrospective cohort

study, 1973–2000
17 660 Danish Psychiatric Central Register Danish Register of Causes

of Death
27 SMR: 1.24 (1.08–1.43) in men

and 1.32 (1.18–1.48) in women
Age and calendar period (reported

separately by gender)
Tran et al

(2009) 20
France Prospective cohort

study, 1993–2003
3470 Questionnaire and/or hospital

records
National death certificate 11 SMR: 1.5 (1.2–1.9) Age and gender
Brown et al

(2010) 10
UK Prospective cohort

study, 1981–2006
370 Hospital records UK Office of National

Statistics database
25 SMR: 1.93 (1.18–2.98) in men

and 1.02 (0.49–1.88) in women
Age (reported separately by gender)
Daumit et al

(2010) 21
USA Retrospective cohort

study, 1992–2001
N/A Medicaid database National Death Index 9 SMR: 1.4 (1.3–1.5) Age, gender and ethnicity
Talaslahti

et al (2012) 22
Finland Retrospective cohort

study, 1999–2008
9461 Finnish Hospital Discharge Register National Causes of Death

Register of Statistics Finland
9 SMR: 1.9 (1.7–2.1) in men

and 2.0 (1.8–2.1) in women
Age (reported separately by gender)
Castagnini

et al (2013) 18
Denmark Retrospective cohort

study, 1995–2008
4576 Danish Psychiatric Register Danish Register of Causes of

Death
13 SMR: 1.5 (1.0–2.3) Age and gender
Crump et al

(2013) 13
Sweden Retrospective cohort

study, 2001–2009
8277 Swedish Outpatient Registry

and Swedish Hospital Registry
Swedish Death Registry 9 HR: 1.39 (1.11–1.74) in men

and 1.68 (1.36–2.07) in women
Age, marital status, education,

employment status, income and

substance use disorder (reported

separately by gender)
Guan et al

(2013) 23
The

Netherlands
Retrospective cohort

study, 1999–2009
4590 Psychiatric Case Register Middle

Netherlands
Death Register of Statistics

Netherlands
11 HR: 1.61 (1.26–2.06) Age, gender, ethnicity and mean income

of last-registered neighbourhood
Kisely et al

(2013) 24
Australia Retrospective cohort

study, 1988–2007
N/A Hospital Morbidity Data System, and

Mental Health Information System
Registrar General's Death

Registration Data
19 SMR: 2.00 (1.51–2.64) in men

and 1.68 (1.29–2.18) in women
Age and gender
Almeida

et al (2014) 25
Australia Prospective cohort

study, 1996–2010
444 Western Australian Data Linkage

System
Western Australian Data

Linkage System
14 HR: 2.0 (1.8–2.2) Age (all participants were men)
Kredentser

et al (2014) 26
Canada Retrospective cohort

study, 1999–2008
9038 Population Health Research Data

Repository
Population Health Research

Data Repository
10 HR: 1.05 (0.93–1.18) b Age and gender
Perini et al

(2014) 35
Italy Retrospective cohort

study, 1982–2006
695 South Verona Psychiatric Case

Register
Mortality Registry of the Local

Health District of Verona, and

other Registries of Deaths
25 SMR: 0.83 (0.50–1.30) Age and gender
Olfson et al

(2015) 9
USA Retrospective cohort

study, 2001–2007
1 138 853 National Medicaid Analytic eXtract

(MAX) database
National Death Index 7 SMR: 1.7 (1.7–1.8) in men

and 1.8 (1.8–1.8) in women
Age, ethnicity and geographic region

(reported separately by gender)
Kisely et al

(2016) 27
Australia Retrospective cohort

study, 2002–2007
N/A Queensland Hospital Admitted

Patients' Data Collection or

Queensland Client Event Services

Application
Queensland Registrar

General's Death Registration

Data
6 SMR: 2.02 (1.61–2.53) Age, gender, residence, socioeconomic

status and length of mental health

service in-patient stay

SMR, standardised mortality ratio; N/A, not applicable; RR, relative risk; HR, hazard ratio.

a. The risk estimates were pooled from the original values that were separately reported for cancer mortality after 0–5, 5–10 and >10 years after the first admission to hospital with schizophrenia.

b. The 95% confidence intervals were not directly reported in the original article, and therefore they were calculated from the P-value along with the risk estimate following the method by Altman & Bland. 38

Association between schizophrenia and cancer mortality

Among the 19 included studies, 15 studies 9,10,1422,24,27,35,41 reported SMRs comparing patients with schizophrenia with the general population. An inverse, null or positive association between schizophrenia and cancer mortality was observed in those studies, with the reported SMRs ranging from 0.81 to 2.02 (Fig. 2). In the random-effects meta-analysis, the pooled SMR of cancer mortality in patients with schizophrenia compared with the general population was 1.40 (95% CI 1.29–1.52; P < 0.001). There was evidence of heterogeneity across studies (I 2 = 95%, P < 0.001). In meta-regression analyses, we observed no evidence that the heterogeneity was caused by a difference in the geographical location, sample size, follow-up duration or adjustment for covariates. In several 10,1416,41 although not all 9,17,19,22,24 previous studies, a gender difference was noted; however, gender did not appear to be a significant source of heterogeneity in this meta-analysis. Stratified analyses by gender showed that the pooled SMR of cancer mortality was 1.32 (95% CI 1.11–1.57) in men, 1.42 (95% CI 1.24–1.63) in women, and 1.47 (95% CI 1.20–1.79) in studies with both men and women (online Fig. DS1). Further exploration using a cumulative meta-analysis showed evidence of cohort effects; the pooled estimates shifted from an inverse association to a positive association with overall time (online Fig. DS2). There was also evidence for publication bias, as indicated by the funnel plot (online Fig. DS3) and the Egger regression asymmetry test (P < 0.01). Sensitivity analyses by omitting one study at a time did not substantially alter the pooled results, which ranged from 1.37 (95% CI 1.26–1.49) to 1.46 (95% CI 1.35–1.57). Additionally, similar results but with a stronger positive association were obtained when a fixed-effect model was used instead of a random-effects model.

Fig. 2 Forest plot of the risk of cancer mortality in patients with schizophrenia compared with the general population (above) or people without schizophrenia (below).

ES, effect size; HR, hazard ratio; SMR, standard mortality ratio.

The other four studies 13,23,25,26 reported HRs comparing patients with schizophrenia with those without schizophrenia. All of those studies with the exception of the one by Kredentser et al 26 reported a positive association between schizophrenia and cancer mortality (Fig. 2). The pooled HR of cancer mortality in individuals with schizophrenia compared with those without schizophrenia was 1.51 (95% CI 1.13–2.03, P = 0.006). Similarly, there was evidence for heterogeneity across studies (I 2 = 94%, P < 0.001). Only the study by Crump et al 13 reported the HRs separately by gender, in which the HRs were 1.39 (95% CI 1.11–1.74) in men and 1.68 (95% CI 1.36–2.07) in women. We did not perform a meta-regression analysis for those studies because the limited number of included studies did not allow sufficient statistical robustness in meta regression. There was no evidence for significant publication bias (P = 0.84 in the Egger regression asymmetry test). Sensitivity analyses by omitting one study at a time did not substantially alter the pooled results, which ranged from 1.39 (95% CI 1.08–1.80) to 1.69 (95% CI 1.42–2.01). We also observed similar results when a fixed-effect model was used instead of a random-effects model.

Discussion

Main findings

In this study, we found that patients with schizophrenia had a higher risk of cancer mortality. Specifically, the risk of cancer mortality in patients with schizophrenia was 40% higher than the general population and 51% higher than individuals without schizophrenia.

The risk of cancer mortality in patients with schizophrenia has not been previously quantified using a meta-analysis. The present study, based on a systematic review of epidemiological studies, was conducted to provide an updated estimate for the risk of cancer mortality in patients with schizophrenia by pooling original data from individual studies.

Interpretation of our findings and comparison with other studies

In the current study, we observed heterogeneity in previous individual studies looking at the association between schizophrenia and cancer mortality, which was not convincingly explained by differences in gender, geographical location, sample size, follow-up duration or adjusting for covariates. We noted possible cohort effects showing a relatively consistent and positive association in recent publications; this was in contrast to inconsistent associations in earlier publications. However, this may be a result of publication bias, and such findings should be interpreted with caution.

Interestingly, there appeared to be a paradox in the associations between schizophrenia and cancer incidence versus cancer mortality. 2831 Whereas our study showed a significantly increased risk of cancer mortality in patients with schizophrenia, a previous meta-analysis on the association of schizophrenia and cancer incidence showed no significant association in general, although there were variations in the risk of specific cancer sites. 30 Several factors have been suggested to explain the divergent associations between schizophrenia and cancer incidence and mortality. Cancer mortality is influenced by not only cancer incidence, but also the survival of those who develop cancer. 15 Several studies on cancer fatality in schizophrenia 36,42,43 have consistently indicated that the presence of schizophrenia increases cancer fatality in patients who had cancer (online Table DS1). Compromised accessibility to treatment facilities and lower quality of care may be the primary reasons for the increased cancer mortality observed in patients with schizophrenia, indicating an imperative need to increase access and popularise cancer screening and detection in this patient population. 42,44 Reduced cancer screening and delayed cancer diagnosis in those with schizophrenia, which results in a late staging of cancer and a higher prevalence of metastasis at the time of cancer diagnosis, may also contribute to worse cancer survival. 24,45,46 Patients with schizophrenia are also more likely to have physical health multimorbidity, 47 engage in more smoking, and are less likely to receive smoking cessation advice, 48 which can increase the risk of cancer mortality. Additionally, different types of antipsychotic drugs may also complicate the risk of cancer mortality in patients with schizophrenia. 49 In female patients with schizophrenia, prolactin and antipsychotic-induced hyperprolactinaemia have been hypothesised to play a role in the development and progression of breast cancer, but available evidence remains controversial and inconclusive. 5052

Strengths and limitations

The major strength of this study is the use of a systematic approach to identify and analyse available evidence. Additionally, the inclusion of data from a large number of identified studies ensures a robust pooled estimate with a high statistical power. There appears to be a high validity of death status and causes of death in the included studies. Information on death status and causes in all the included studies was ascertained from well-established general death registries (such as the National Death Index in the USA and Swedish Death Registry), specific registries of cause of death (for example Danish Register of Causes of Death and Finnish National Causes of Death Register) or directly from national or local death certificates. Death certificates and other official documents were referred to in order to establish the causes of death in all those death registries. Furthermore, there was evidence indicating that the causes of death on death certificates in patients with schizophrenia were probably more accurate than in the general population, because rates of post-mortem examination (54% v. 22%) and coroner's inquest (15% v. 6%) were higher than the national average. 10

There are also several limitations. First, the use of SMRs in most previous studies may underestimate the true risk of cancer mortality in patients with schizophrenia, because in calculating SMRs, the comparator group is usually the general population, which is comprised of individuals with and without schizophrenia. As shown in previous methodological papers, 53,54 this bias was obvious when the risk was assessed in cohort studies of people with common diseases or exposures and/or when large SMRs were observed. However, because the prevalence of schizophrenia is low (~1%) in the general population and the observed SMRs for cancer mortality after a diagnosis of schizophrenia in most studies were modest, the underestimation of true risk by SMRs in the current scenario would be minor. 53,54 In this meta-analysis, such concerns were further reduced because we presented not only the pooled SMRs, but also the pooled HRs, which appeared to be stronger than the pooled SMRs. Second, the risk of cancer mortality by specific types/sites (i.e. lung cancer, breast cancer, etc.) was not summarised in this study because a possible selective reporting bias was observed for the risk of mortality from certain types of cancer in previous studies. In addition to the risk of overall cancer mortality, many studies simply chose to report significant findings for certain types of cancer. Some other studies may be unable to derive a risk estimate for certain types of cancer because the sample size was too small. Apparently, pooling the results from those studies will lead to a biased estimation. In the large study by Olfson et al 9 increased mortality was consistently observed in all cancer subtypes (i.e. lung, colon, breast, liver, pancreas, haematological), with nearly identical SMRs in men and women. This indicates that variations in the risk of mortality from different types of cancer may not be substantial. Further investigation on this is warranted in future studies. Third, information on antipsychotic medication and smoking status was not available in most studies. Whether these factors moderate or mediate the association between schizophrenia and cancer mortality needs further investigation.

Implications

This study has important clinical implications. Because of the high social and economic burden associated with schizophrenia, 2 it is important to clearly understand schizophrenia-related clinical outcomes such as morbidity and mortality risk. Findings from our study emphasise the need for clinicians to be aware of the increased risk of cancer mortality in people with schizophrenia. It appears to be imperative to address health disparities and improve cancer survival among patients with schizophrenia through an integrated approach, which may involve an improvement in access and quality of care, early cancer screening and diagnosis, as well as smoking cessation services.

In conclusion, this systematic review and meta-analysis found a significantly increased risk of cancer mortality in patients with schizophrenia. Future cohort studies with a large sample size and long follow-up are warranted to confirm our findings and to elucidate the risk of mortality from specific types/sites of cancers.

Funding

This work was supported by Found of Tianjin Health Bureau ( to C.Z.), Chinese Postdoctoral Science Foundation ( to C.Z.), Jiangsu Haosen pharmaceutical Limited by Share Ltd (2016-Young scholar support project to C.Z.), Hainan Liou pharmaceutical Limited by Share Ltd (2016-Young scholar support project to C.Z.), Xuzhou Enhua pharmaceutical Limited by Share Ltd (2016-Young scholar support project to C.Z.), Shanghai Zhongxi pharmaceutical Limited by Share Ltd (2016-Young scholar support project to C.Z.).

References

1 Owen, MJ, Sawa, A, Mortensen, PB. Schizophrenia. Lancet 2016; 388: 8697.
2 Chong, HY, Teoh, SL, Wu, DB, Kotirum, S, Chiou, CF, Chaiyakunapruk, N. Global economic burden of schizophrenia: a systematic review. Neuropsychiatr Dis Treat 2016; 12: 357–73.
3 Wu, EQ, Birnbaum, HG, Shi, L, Ball, DE, Kessler, RC, Moulis, M, et al. The economic burden of schizophrenia in the United States in 2002. J Clin Psychiatry 2005; 66: 1122–9.
4 Cloutier, M, Aigbogun, MS, Guerin, A, Nitulescu, R, Ramanakumar, AV, Kamat, SA, et al. The economic burden of schizophrenia in the United States in 2013. J Clin Psychiatry 2016; 77: 764–71.
5 Saha, S, Chant, D, McGrath, J. A systematic review of mortality in schizophrenia: is the differential mortality gap worsening over time? Arch Gen Psychiatry 2007; 64: 1123–31.
6 Bushe, CJ, Taylor, M, Haukka, J. Mortality in schizophrenia: a measurable clinical endpoint. J Psychopharmacol 2010; 24: 1725.
7 Laursen, TM, Nordentoft, M, Mortensen, PB. Excess early mortality in schizophrenia. Annu Rev Clin Psychol 2014; 10: 425–48.
8 Chou, FH, Tsai, KY, Wu, HC, Shen, SP. Cancer in patients with schizophrenia: what is the next step? Psychiatry Clin Neurosci 2016; 70: 473–88.
9 Olfson, M, Gerhard, T, Huang, C, Crystal, S, Stroup, TS. Premature mortality among adults with schizophrenia in the United States. JAMA Psychiatry 2015; 72: 1172–81.
10 Brown, S, Kim, M, Mitchell, C, Inskip, H. Twenty-five year mortality of a community cohort with schizophrenia. Br J Psychiatry 2010; 196: 116–21.
11 Reininghaus, U, Dutta, R, Dazzan, P, Doody, GA, Fearon, P, Lappin, J, et al. Mortality in schizophrenia and other psychoses: a 10-year follow-up of the SOP first-episode cohort. Schizophr Bull 2015; 41: 664–73.
12 Walker, ER, McGee, RE, Druss, BG. Mortality in mental disorders and global disease burden implications: a systematic review and meta-analysis. JAMA Psychiatry 2015; 72: 334–41.
13 Crump, C, Winkleby, MA, Sundquist, K, Sundquist, J. Comorbidities and mortality in persons with schizophrenia: a Swedish national cohort study. Am J Psychiatry 2013; 170: 324–33.
14 Mortensen, PB, Juel, K. Mortality and causes of death in schizophrenic patients in Denmark. Acta Psychiatr Scand 1990; 81: 372–7.
15 Saku, M, Tokudome, S, Ikeda, M, Kono, S, Makimoto, K, Uchimura, H, et al. Mortality in psychiatric patients, with a specific focus on cancer mortality associated with schizophrenia. Int J Epidemiol 1995; 24: 366–72.
16 Lawrence, D, Holman, CD, Jablensky, AV, Threlfall, TJ, Fuller, SA. Excess cancer mortality in Western Australian psychiatric patients due to higher case fatality rates. Acta Psychiatr Scand 2000; 101: 382–8.
17 Laursen, TM, Munk-Oisen, T, Nordentoft, M, Mortensen, PB. Increased mortality among patients admitted with major psychiatric disorders: a register-based study comparing mortality in unipolar depressive disorder, bipolar affective disorder, schizoaffective disorder, and schizophrenia. J Clin Psychiatry 2007; 68: 899907.
18 Castagnini, A, Foldager, L, Bertelsen, A. Excess mortality of acute and transient psychotic disorders: Comparison with bipolar affective disorder and schizophrenia. Acta Psychiatr Scand 2013; 128: 370–5.
19 Heila, H, Haukka, J, Suvisaari, J, Lonnqvist, J. Mortality among patients with schizophrenia and reduced psychiatric hospital care. Psychol Med 2005; 35: 725–32.
20 Tran, E, Rouillon, F, Loze, JY, Casadebaig, F, Philippe, A, Vitry, F, et al. Cancer mortality in patients with schizophrenia: an 11-year prospective cohort study. Cancer 2009; 115: 3555–62.
21 Daumit, GL, Anthony, CB, Ford, DE, Fahey, M, Skinner, EA, Lehman, AF, et al. Pattern of mortality in a sample of Maryland residents with severe mental illness. Psychiatry Res 2010; 176: 242–5.
22 Talaslahti, T, Alanen, HM, Hakko, H, Isohanni, M, Häkkinen, U, Leinonen, E. Mortality and causes of death in older patients with schizophrenia, Int J Geriatr Psychiatry 2012; 27: 1131–7.
23 Guan, NC, Termorshuizen, F, Laan, W, Smeets, HM, Zainal, NZ, Kahn, RS, et al. Cancer mortality in patients with psychiatric diagnoses: a higher hazard of cancer death does not lead to a higher cumulative risk of dying from cancer. Soc Psychiatry Psychiatr Epidemiol 2013; 48: 1289–95.
24 Kisely, S, Crowe, E, Lawrence, D. Cancer-related mortality in people with mental illness. JAMA Psychiatry 2013; 70: 209–17.
25 Almeida, OP, Hankey, GJ, Yeap, BB, Golledge, J, Norman, PE, Flicker, L. Mortality among people with severe mental disorders who reach old age: a longitudinal study of a community-representative sample of 37,892 men. PloS One 2014; 9: e111882.
26 Kredentser, MS, Martens, PJ, Chochinov, HM, Prior, HJ. Cause and rate of death in people with schizophrenia across the lifespan: a population-based study in Manitoba, Canada. J Clin Psychiatry 2014; 75: 154–61.
27 Kisely, S, Forsyth, S, Lawrence, D. Why do psychiatric patients have higher cancer mortality rates when cancer incidence is the same or lower? Aust NZ J Psychiatry 2016; 50: 254–63.
28 Hodgson, R, Wildgust, HJ, Bushe, CJ. Cancer and schizophrenia: is there a paradox? J Psychopharmacol 2010; 24: 5160.
29 Bushe, CJ, Hodgson, R. Schizophrenia and cancer: in 2010 do we understand the connection? Can J Psychiatry 2010; 55: 761–7.
30 Catala-Lopez, F, Suarez-Pinilla, M, Suarez-Pinilla, P, Valderas, JM, Gomez-Beneyto, M, Martinez, S, et al. Inverse and direct cancer comorbidity in people with central nervous system disorders: a meta-analysis of cancer incidence in 577,013 participants of 50 observational studies. Psychother Psychosom 2014; 83: 89105.
31 Dalton, SO, Laursen, TM, Mellemkjaer, L, Johansen, C, Mortensen, PB. Risk for cancer in parents of patients with schizophrenia. Am J Psychiatry 2004; 161: 903–8.
32 Stroup, DF, Berlin, JA, Morton, SC, Olkin, I, Williamson, GD, Rennie, D, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000; 283: 2008–12.
33 Brown, S, Inskip, H, Barraclough, B. Causes of the excess mortality of schizophrenia. Br J Psychiatry 2000; 177: 212–7.
34 Grigoletti, L, Perini, G, Rossi, A, Biggeri, A, Barbui, C, Tansella, M, et al. Mortality and cause of death among psychiatric patients: a 20-year case-register study in an area with a community-based system of care. Psychol Med 2009; 39: 1875–84.
35 Perini, G, Grigoletti, L, Hanife, B, Biggeri, A, Tansella, M, Amaddeo, F. Cancer mortality among psychiatric patients treated in a community-based system of care: a 25-year case register study. Soc Psychiatry Psychiatr Epidemiol 2014; 49: 693701.
36 Batty, GD, Whitley, E, Gale, CR, Osborn, D, Tynelius, P, Rasmussen, F. Impact of mental health problems on case fatality in male cancer patients. Br J Cancer 2012; 106: 1842–5.
37 DerSimonian, R, Laird, N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 177–88.
38 Altman, DG, Bland, JM. How to obtain the confidence interval from a P value. BMJ 2011; 343: d2090.
39 Higgins, JP, Thompson, SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21: 1539–58.
40 Egger, M, Davey Smith, G, Schneider, M, Minder, C. Bias in meta-analysis detected by a simple, graphical test, BMJ 1997; 315: 629–34.
41 Mortensen, PB, Juel, K. Mortality and causes of death in first admitted schizophrenic patients. Br J Psychiatry 1993; 163: 183–9.
42 Chou, FH, Tsai, KY, Su, CY, Lee, CC. The incidence and relative risk factors for developing cancer among patients with schizophrenia: a nine-year follow-up study. Schizophr Res 2011; 129: 97103.
43 Bergamo, C, Sigel, K, Mhango, G, Kale, M, Wisnivesky, JP. Inequalities in lung cancer care of elderly patients with schizophrenia: an observational cohort study. Psychosom Med 2014; 76: 215–20.
44 Beary, M, Hodgson, R, Wildgust, HJ. A critical review of major mortality risk factors for all-cause mortality in first-episode schizophrenia: clinical and research implications. J Psychopharmacol 2012; 26: 5261.
45 Cunningham, R, Sarfati, D, Stanley, J, Peterson, D, Collings, S. Cancer survival in the context of mental illness: a national cohort study. Gen Hosp Psychiatry 2015; 37: 501–6.
46 Mitchell, AJ, Pereira, IE, Yadegarfar, M, Pepereke, S, Mugadza, V, Stubbs, B. Breast cancer screening in women with mental illness: comparative meta-analysis of mammography uptake. Br J Psychiatry 2014; 205: 428–35.
47 Stubbs, B, Koyanagi, A, Veronese, N, Vancampfort, D, Solmi, M, Gaughran, F, et al. Physical multimorbidity and psychosis: comprehensive cross sectional analysis including 242,952 people across 48 low- and middle-income countries. BMC Med 2016; 14: 189.
48 Mitchell, AJ, Vancampfort, D, De Hert, M, Stubbs, B. Do people with mental illness receive adequate smoking cessation advice? A systematic review and meta-analysis. Gen Hosp Psychiatry 2015; 37: 1423.
49 Tiihonen, J, Lonnqvist, J, Wahlbeck, K, Klaukka, T, Niskanen, L, Tanskanen, A, et al. 11-year follow-up of mortality in patients with schizophrenia: a population-based cohort study (FIN11 study). Lancet 2009; 374: 620–7.
50 De Hert, M, Peuskens, J, Sabbe, T, Mitchell, AJ, Stubbs, B, Neven, P, et al. Relationship between prolactin, breast cancer risk, and antipsychotics in patients with schizophrenia: a critical review. Acta Psychiatr Scand 2016; 133: 522.
51 Froes Brandao, D, Strasser-Weippl, K, Goss, PE. Prolactin and breast cancer: the need to avoid undertreatment of serious psychiatric illnesses in breast cancer patients: a review. Cancer 2016; 122: 184–8.
52 De Hert, M, Vancampfort, D, Stubbs, B, Sabbe, T, Wildiers, H, Detraux, J. Antipsychotic treatment, prolactin, and breast tumorigenesis. Psychiatr Danub 2016; 28: 243–54.
53 Jones, ME, Swerdlow, AJ. Bias in the standardized mortality ratio when using general population rates to estimate expected number of deaths. Am J Epidemiol 1998; 148: 1012–7.
54 Card, TR, Solaymani-Dodaran, M, Hubbard, R, Logan, RF, West, J. Is an internal comparison better than using national data when estimating mortality in longitudinal studies? J Epidemiol Community Health 2006; 60: 819–21.