Tropical Medicine and International Health doi:10.1111/tmi.

12933

volume 00 no 00

Association of Toxoplasma gondii infection with schizophrenia

and its relationship with suicide attempts in these patients
Maryam Ansari-Lari1, Hassan Farashbandi2 and Fahimeh Mohammadi1
1 Department of Food Hygiene and Public Heath, School of Veterinary Medicine, Shiraz University, Shiraz, Iran
2 Department of Psychiatry, School of Medicine, Shiraz University of Medical Sciences, Shiraz, Iran

Abstract objectives To investigate the association between schizophrenia and Toxoplasma gondii, and to
assess the association of infection with suicide attempts and age of onset of schizophrenia in these
patients.
methods Case–control study Fars Province, southern Iran. Cases were individuals with psychiatric
diagnosis of schizophrenia as per Diagnostic and Statistical Manual of Mental Disorders (DSM-IV)
criteria. Controls were healthy blood donors, frequency-matched with patients according to age and
sex. For the detection of IgG antibodies, enzyme-linked immunosorbent assay (ELISA) was used.
Data about demographic information in all subjects and duration of illness and history of suicide
attempts in patients with schizophrenia were collected using a brief questionnaire and hospital
records. Chi-square test and multivariable logistic regression were used for statistical analyses.
results Among 99 cases, 42 individuals (42%) were positive for T. gondii antibody, vs. 41 (27%)
among 152 controls (OR = 2, 95% CI: 1.2–3.4, P = 0.012). We compared the suicide attempts in
patients with schizophrenia based on their T. gondii serologic status. There was a lower rate of
suicide attempts in seropositive male patients than seronegative ones (OR = 0.3, 95% CI: 0.1–0.97,
P = 0.04). Age of onset of schizophrenia did not differ between T. gondii-infected and non-infected
patients.
conclusions These findings may have implications for schizophrenia and suicide prevention
programmes. However, clearly further studies are required to confirm them.

keywords Iran, Schizophrenia, Suicide, Toxoplasma gondii

prevalence of antibodies to T. gondii; the overall odds

Introduction
ratio was 2.71 (95% CI: 1.93–3.80). However, other
Schizophrenia is a neuropsychiatric disorder with world- studies failed to demonstrate a significant association
wide prevalence that affects approximately 1% of the between schizophrenia and toxoplasmosis [5–7]. Several
adult population in the United States and Europe [1]. investigators in various parts of Iran have sought to
Based on the epidemiologic studies, several risk factors, determine the association of schizophrenia with T. gondii
infectious agents among them, have been suggested for seroprevalence. As other parts of the world, the results
schizophrenia. The intracellular parasite Toxoplasma were conflicting: Saraei-Sahnesaraei et al. [8] and Daryani
gondii (T. gondii) is one of these infectious agents due to et al. [9] did not observe any significant association,
its association with brain dysfunction and its neu- whereas Hamidinejat et al. [10] and Alipour et al. [11]
rotropism [2]. found significantly higher seropositive rates among
Toxoplasma gondii is a zoonotic infection whose life patients with schizophrenia than among healthy controls.
cycle can be completed only in cats and other felids as Rodent studies have shown that T. gondii organisms
definitive hosts. Humans and various warm-blooded ani- can impair learning in mice and that it manipulates host
mals are intermediate hosts. Toxoplasma gondii infects behaviour to produce changes that increase the chances
approximately one-third of the world’s population [3]. that a rat will be eaten by a cat, thus enabling the para-
Several studies have investigated the association of site to complete its life cycle [1, 12]. The relationship
T. gondii infection with schizophrenia disorder. A meta- between rates of infection with T. gondii and behaviour
analysis by Torrey et al. [4] of 38 studies showed that impairment in humans is challenging to prove and incon-
individuals with schizophrenia have an increased clusive. There is some evidence that a range of subtle

© 2017 John Wiley & Sons Ltd 1

Tropical Medicine and International Health volume 00 no 00

M. Ansari-Lari et al. T. gondii, schizophrenia and suicide attempts

behavioural alterations comparable with those observed standard curve for predetermined titres, the antibody titre
in infected rodent intermediate hosts is associated with of all samples was determined.
T. gondii infection [13, 14]. However, others failed to
show similar associations [7, 15].
Data collection
Regarding schizophrenia, the clinical picture of
schizophrenia in toxoplasma-infected and toxoplasma- In the control group, data about age, sex, education,
free patients differs in several respects. Toxoplasma- marital status, and history of psychiatric illness or other
infected individuals show more severe positive symptoms major health problems were collected at the time of
of schizophrenia and are more likely to experience a con- blood donation using a brief questionnaire. In patients
tinuous course of disease than toxoplasma-free patients with schizophrenia, we also noted duration of illness,
[16–18]. Several studies investigating the relationship number of hospitalisations, history of suicide attempts
between T. gondii and suicide in infected persons or in and history of smoking/drug from hospital records. His-
patients with schizophrenia gave inconclusive results [19– tory of suicide attempts were defined as explicit state-
22]. It seems that the truth about the association of ments referring to attempted suicide in the patient’s
T. gondii with schizophrenia and behavioural alterations record. Duration of illness was defined based on the first
in infected individuals remains elusive and that inconsis- occurrence of psychotic symptoms as stated in the
tent results are common. patient’s history. Age of onset of schizophrenia symptoms
Thus, our aims were to determine the association was calculated based on the subtraction of duration of ill-
between schizophrenia and T. gondii in Fars Province, ness from the current age.
southern Iran; to assess the association of T. gondii with
suicide attempts in these patients; and to compare the age
Statistical analysis
at onset of schizophrenia in T. gondii-infected patients to
that of non-infected individuals. Statistical analysis was conducted using SPSS software
(Version 16.0; SPSS, Inc., Chicago, USA). There were
missing cases for duration of illness (and age of onset)
Materials and methods and family history of psychiatric illness, which were
deleted case-wise in the statistical analyses. The associa-
Study population
tion of seroprevalence of T. gondii with schizophrenia
This is a case–control study which was conducted in Shi- was determined using chi-square test also calculation of
raz, the capital of Fars Province, southern Iran. Cases odds ratio (OR). Comparison of serologic titres between
were individuals with psychiatric diagnosis of case and control groups was performed using nonpara-
schizophrenia based on Diagnostic and Statistical Man- metric Mann–Whitney U test due to the non-normal dis-
ual of Mental Disorders, fourth edition (DSM-IV), crite- tribution of data. Also, to evaluate the potential
ria who were recruited from the Ostad Moharrari confounding effect of sex and education, serologic titres
hospital, a centre for psychiatric disorders in Fars Pro- were compared between sex and education groups. Mul-
vince. Control subjects were consenting, apparently tivariable logistic regression analysis was used to investi-
healthy blood donors with no history of psychiatric ill- gate the association of the serologic titres with
ness or any other health problem, frequency-matched schizophrenia. Schizophrenia status was considered as a
with patients according to age and sex. In total, 99 cases dependent variable, and antibody titres and potential
and 152 controls from both sexes were selected for the confounders were considered as covariates. Due to the
study. very small number of patients with schizophrenia in the
higher education group (eight cases), education was cate-
gorised as two groups. Using backward elimination, the
Serological examination
final logistic model was constructed.
Blood samples were collected from the study population, For comparison of age of onset in seropositive and
and sera were stored at 20 °C until use. For serological seronegative patients with schizophrenia, two indepen-
examination and detection of IgG antibodies, commercial dent samples’ t-tests were used. Association of suicide
enzyme-linked immunosorbent assay (ELISA) kits (Gene- attempts with toxoplasma serologic status in patients
sis, UK) were used according to the instructions of the with schizophrenia was examined using multivariable
manufacturer. Samples with an absorbance value above logistic regression analysis to control for probable con-
the cut-off level (greater than or equal to 8 IU/ml OD) founding factors. Here, age was categorised into <40 and
were considered positive. Furthermore, based on the ≥40 years based on the average age of patients and a

2 © 2017 John Wiley & Sons Ltd

Tropical Medicine and International Health volume 00 no 00

M. Ansari-Lari et al. T. gondii, schizophrenia and suicide attempts

logistic model constructed in each sex group separately. titres was performed using multivariable logistic regres-
To account for the probable confounding effects of age sion. Serologic titres and education were introduced in
and history of smoking/drug abuse, they were introduced the model using backward elimination, leading to results
in the logistic regression model as covariates. Again, showing that the odds of being schizophrenic increased
using backward elimination, the final model was con- by 2% with each unit of antibody titre (OR = 1.02,
structed. In all analyses, a P-value <0.05 was considered P = 0.041; Table 2).
statistically significant. Age of onset of disease, duration of illness, number of
hospitalisation and suicide attempts were compared in
patients with schizophrenia based on their T. gondii sero-
Results
logic status. Results are listed in Table 3. There was no
We selected 99 patients with schizophrenia and 152 significant difference between seronegative and seroposi-
healthy blood donors for the study. Demographic charac- tive groups for education and age of onset of disease
teristics of the cases and controls are presented in (P > 0.05). However, age and duration of illness in
Table 1. Age ranged between 18 and 59 years in patients seropositive group were significantly higher than seroneg-
with schizophrenia and between 19 and 59 years ative ones (Table 3).
(P = 0.73) in the control group. 42 patients with Twenty-nine patients had a history of suicide attempts:
schizophrenia (42%) were positive for toxoplasma anti- 20 of 72 male (28%) and nine of 27 female patients
body vs. 41 controls (27%) (v2 = 6.46, df = 1, (33%). Suicide attempts were less common in seroposi-
P = 0.011). The odds ratio calculation revealed that odds tive (19%) than seronegative patients (37%); the result
of schizophrenia in seropositive persons were double tends to be significant (P = 0.054). When the comparison
those in seronegative persons (OR = 1.99, 95% CI: 1.2– was replicated in each sex separately, a significant differ-
3.4). ence was observed in male patients (P = 0.039) but not
The mean (median) and standard deviation (SD) of in female ones (P = 1). Mean (median) and SD of anti-
antibody titres in cases and controls were 11.1 (5.9) body titre were 7.7 (4.5) 11.7 and 12.5 (7.5) 13.7 in
13.2 and 6.5 (1.0) 11.5, respectively (P < 0.001). patients with and without suicide attempts, respectively
Serologic titres showed significant differences between (P = 0.055). Comparison of antibody titres in each sex
education levels (P = 0.001) but not between sexes individually showed results relatively similar to those on
(P = 0.35). Therefore, comparison of patients with seropositivity (P = 0.057 and P = 0.63 in male and
schizophrenia to blood donor controls for toxoplasmosis female, respectively).
No association between suicide attempts with cigarette
smoking/drug abuse was observed in male or female
patients with schizophrenia (P > 0.05). However, suicidal
Table 1 Comparison between schizophrenia and control group
for demographic characteristics and Toxoplasma gondii status, attempters were significantly younger than non-attemp-
Fars Province, southern Iran ters (mean difference = 8, 95% CI: 3.7–12.2 years). Due
to probable confounding effect of these variables, they
Schizophrenia Healthy controls were introduced in the multivariable logistic model. Using
(n = 99) (n = 152) P-Value
backward elimination, only toxoplasma serologic status
Age (Mean  SD*) 40.3  10.2 40.6  10.7 0.73 remained in the final logistic model in male patients. The
Toxoplasma gondii odds of suicide attempts in seropositive patients are about
Positive 42 41 0.011
Negative 57 111
Sex Table 2 Results of multivariable logistic regression analysis for
Male 72 110 0.95 comparison of toxoplasma antibody titre in schizophrenia
Female 27 42 patients with control group, Fars Province, southern Iran
Education
Primary School 40 14 <0.001 Risk factors Wald SE P-value OR (95% CI)
High school 51 82
Higher than 8 56 Toxoplasma 4.16 0.01 0.041 1.02 (1.001–1.045)
high school gondii antibody
Marital status titre
Married 28 131 <0.001 Education* 26.83 0.35 <0.001 6.15 (3.09–12.24)
Single 71 21 Constant 31.89 0.18 <0.001 –

*Standard deviation. *Primary school was compared to higher education.

© 2017 John Wiley & Sons Ltd 3

Tropical Medicine and International Health volume 00 no 00

M. Ansari-Lari et al. T. gondii, schizophrenia and suicide attempts

Table 3 Comparison of some characteristics in patients with schizophrenia (n = 99) according to their serologic status for Toxoplasma
gondii, Fars Province, southern Iran

Toxoplasma positive (n = 42) Toxoplasma negative (n = 57) P-value

Age (Mean  SD*) 43.5  8.1 38.0  11.1 0.008

Age at onset (Mean  SD)† 25.0  7.6 25.1  9.8 0.95
Duration of illness (Mean  SD)† 19.0  8.3 12.9  8.5 0.001
Number of hospitalisation (Mean  SD) 5.3  2.9 5.1  3.1 0.76
Education
Primary School 21 19 0.23
High school 19 32
Higher than high school 2 6
Family history of psychiatric disease†
Yes 14 22 0.83
No 23 33
Marital status
Married 14 14 0.34
Single 28 43
Sex
Male 32 40 0.51
Female 10 17
Smoking/drug abuse
Yes 28 36 0.84
No 14 21
Suicide attempts
Yes 8 21 0.055
No 34 36

*Standard deviation; †There were one missing case for duration of illness and age of onset each in toxoplasma positive group; missing
cases for family history were 5 and 2 in positive and negative groups, respectively.

one-third compared with seronegative ones in males which revealed an overall OR of 2.7 for the association
(OR = 0.31, 95% CI: 0.10–0.97, Table 4). No significant of T. gondii with schizophrenia [4, 23]. Previous studies
difference in suicide attempts was observed in female from Iran arrived at inconsistent results regarding the
patients, and consequently no logistic model was fitted association between schizophrenia and T. gondii seropos-
for this group. itivity [8–11]. Recent research from other countries failed
to show a significant association between schizophrenia
and T. gondii infection [5–7].
Discussion
From the epidemiological point of view and for the
A significant positive association between T. gondii implementation of good control programme, it is recom-
(seropositivity and serointensity) and schizophrenia was mended to find the suspected risk factors in each area,
observed in the present study: seropositive persons had independently. This is due to the complexity of the causal
twice the odds to have schizophrenia as seronegative web for multifactorial disease; one factor which is not so
ones. This is consistent with two recent meta-analyses important in an area may be an important component
cause in another area. While scientific evidence supports
the idea that genetic factors play an important role in the
Table 4 Results of multivariable logistic regression analysis for
causation of schizophrenia, many studies indicate that
association of Toxoplasma gondii with suicide attempts in
patients with schizophrenia, Fars Province, southern Iran there must be other environmental factors that interact
with genetic ones [23]. Thus, one reason for the variabil-
Risk factors Wald SE P-value OR (95% CI) ity of results for association between T. gondii and
schizophrenia may be the multifactorial nature of the dis-
Toxoplasma gondii* 4.02 0.59 0.045 0.31 (0.10–0.97)
Constant 2.45 0.33 0.118 –
order. Based on the meta-analyses results, it seems that
exposure to T. gondii is a relatively important environ-
*Seropositive patients compared with negative ones. mental risk factor for some cases of schizophrenia in

4 © 2017 John Wiley & Sons Ltd

Tropical Medicine and International Health volume 00 no 00

M. Ansari-Lari et al. T. gondii, schizophrenia and suicide attempts

some regions in conjunction with other environmental our study, some type of selection bias (late look bias)
and genetic susceptibilities. Yet, it could be a trivial fac- may have occurred. Risk of suicide is high among all
tor in other regions, assuming that all pertinent studies patients with schizophrenia [29]. It may be speculated
were well-designed and conducted without any clear bias. that the incidence of successful suicide attempts is higher
Association of T. gondii with suicide was reported by in male toxoplasma-infected schizophrenia patients.
Arling et al. [24] for the first time who noted that in Therefore, in a cross-sectional study, a lower rate of
patients with recurrent mood disorders, suicide attemp- seropositive male patients with a history of suicide
ters had significantly higher T. gondii antibody titres than attempts would be found. The process of change in
non-attempters. However, no significant relationship was infected men is probably different from that in infected
found between T. gondii seropositivity and suicide women in whom a positive relationship between infection
attempts status in their study [24]. Based on our results, with T. gondii and suicide was detected at post-
in male patients with schizophrenia, suicide attempts dif- menopausal age [19]. It should be noted that we observed
fered significantly between seropositive and seronegative the association after adjusting for age and history of
subjects. No such relationship was observed for women. smoking/drug abuse in male patients to account for prob-
Regarding patients with schizophrenia, few studies inves- able confounding effect of these factors. However, deter-
tigated the relationship between T. gondii infection and mining suicide attempts based on hospital records may be
suicide [20, 21, 25, 26]. Okusaga et al. [21] indicated a drawback of our study. Future prospective studies are
that the association of T. gondii seropositivity or seroin- required to verify and validate this relationship; convinc-
tensity with history of suicidal behaviour was found only ing evidence for an association of toxoplasma with sui-
in those younger than the median age of the sample cide attempt risk will require prospective studies in which
(38 years); in the older patients, the relationship was not suicide attempt could be ascertained with more confi-
significant. Dickerson et al. [26] studied patients receiving dence [30].
treatment for schizophrenia, bipolar disorder or major Some studies indicated the association of T. gondii
depression and reported particularly strong odds of a sui- with age of onset of schizophrenia in a sex-dependent
cide attempts history in individuals who had elevated manner [18, 28]. We did not find such a relationship in
levels of IgM but not IgG antibodies to T. gondii. No our study group.
gender difference was reported in their study, and no
data for patients with schizophrenia and suicide were
Conclusion
given separately. In contrast, one study on a group of
outpatients with completely mixed diagnoses of various In conclusion, we observed a positive relationship
psychiatric illnesses showed that suicide attempters had between infection with T. gondii and schizophrenia in
significantly higher IgG levels than controls while no such the present study. Also, we observed a significantly lower
relationship for IgM antibodies was observed [25]. It rate of suicide attempts in T. gondii-infected male
may be worthwhile to mention that in another study on patients than in non-infected men. Age of onset of disease
Turkish subjects with and without the history of suicide did not show significant differences in our study. The evi-
attempts and with no reference to psychiatric illness, dence is equivocal, and clearly more studies, particularly
seroprevalence for T. gondii in suicide attempters was prospective ones, are needed.
significantly higher than in controls based on IgG anti-
bodies; no significant difference based on IgM antibodies
Acknowledgements
was found [22].
Nevertheless, we observed a lower rate of suicide We appreciate the cooperation of the staff of Ostad
attempts in toxoplasma-infected male subjects than in Moharrari Hospital and the Fars Blood Transfusion
non-infected men. Therefore, in addition to gender differ- office. This study was supported by Shiraz University
ence, we observed the relationship between T. gondii and (Grant No. 91GCU1M2003).
suicide attempts in patients with schizophrenia in the
opposite direction found in previous research. Gender dif-
References
ferences in behavioural changes induced by latent toxo-
plasmosis were reported by Lindov
a et al. [27]. Also, 1. Torrey EF, Yolken RH. Toxoplasma gondii and schizophre-
recent studies have indicated a greater severity of clinical nia. Emerg Infect Dis 2003: 9: 1375–1380.
symptoms in T. gondii-infected male patients with 2. Torrey EF, Bartko JJ, Lun ZR, Yolken RH. Antibodies to
Toxoplasma gondii in patients with schizophrenia: a meta-
schizophrenia [28]. Regarding the inverse direction of
analysis. Schizophr Bull 2007: 33: 729–736.
observed association, due to the cross-sectional nature of

© 2017 John Wiley & Sons Ltd 5

Tropical Medicine and International Health volume 00 no 00

M. Ansari-Lari et al. T. gondii, schizophrenia and suicide attempts

3. Halonen SK, Weiss LM. Toxoplasmosis. Handb Clin Neu- characteristic for Toxoplasma gondii-infected patients. Folia
rol. 2013. 114:125–145. Parasitol (Praha) 2015: 62: pii: 2015.015.
4. Torrey EF, Bartko JJ, Yolken RH. Toxoplasma gondii and 17. Flegr J. Schizophrenia and Toxoplasma gondii: an underval-
other risk factors for schizophrenia: an update. Schizophr ued association? Expert Rev Anti Infect Ther 2015: 13:
Bull 2012: 38: 642–647. 817–820.
5. de Witte LD, van Mierlo HC, Litjens M, Klein HC, Bahn S, 18. Holub D, Flegr J, Dragomireck
a E et al. Differences in onset
Osterhaus AD. The association between antibodies to neu- of disease and severity of psychopathology between toxo-
rotropic pathogens and schizophrenia: a case-control study. plasmosis-related and toxoplasmosis-unrelated schizophre-
NPJ Schizophr 2015: 4: 15041. nia. Acta Psychiatr Scand 2013: 127: 227–238.
6. Karabulut N, Bilgicß S, G€ urok MG, Karaboga F. Is there any 19. Ling VJ, Lester D, Mortensen PB, Langenberg PW, Posto-
role of latent toxoplasmosis in schizophrenia disease? J Chin lache TT. Toxoplasma gondii seropositivity and suicide rates
Med Assoc 2015: 78: 533–537. in women. J Nerv Ment Dis 2011: 199: 440–444.
7. Sugden K, Moffitt TE, Pinto L, Poulton R, Williams BS, 20. Okusaga O, Duncan E, Langenberg P et al. Combined Tox-
Caspi A. Is Toxoplasma gondii infection related to brain oplasma gondii seropositivity and high blood kynurenine–
and behavior impairments in humans? Evidence from a pop- linked with nonfatal suicidal self-directed violence in
ulation-representative birth cohort. PLoS ONE 2016: 11: patients with schizophrenia. J Psychiatr Res 2016: 72: 74–
e0148435. 81.
8. Saraei-Sahnesaraei M, Shamloo F, Jahani Hashemi H, Khab- 21. Okusaga O, Langenberg P, Sleemi A et al. Toxoplasma gon-
baz F, Alizadeh SA. Relation between Toxoplasma gondii dii antibody titers and history of suicide attempts in patients
infections and schizophrenia. Iran J Psychiatry Clin Psychol with schizophrenia. Schizophr Res 2011: 133: 150–155.
2009: 15: 3–9. 22. Yagmur F, Yazar S, Temel HO, Cavusoglu M. May Toxo-
9. Daryani A, Sharif SM, Hosseini SH, Karimi SH, Gholami S. plasma gondii increase suicide attempt-preliminary results in
Serological survey of Toxoplasma gondii in schizophrenia Turkish subjects? Forensic Sci Int 2010: 199: 15–17.
patients referred to Psychiatric Hospital, Sari City, Iran. 23. Arias I, Sorlozano A, Villegas E et al. Infectious agents asso-
Trop Biomed 2010: 27: 476–482. ciated with schizophrenia: a meta-analysis. Schizophr Res
10. Hamidinejat H, Ghorbanpoor M, Hosseini H et al. Toxo- 2012: 136: 128–136, Review.
plasma gondii infection in first-episode and inpatient individ- 24. Arling TA, Yolken RH, Lapidus M et al. Toxoplasma gon-
uals with schizophrenia. Int J Infect Dis 2010: 14: 978–981. dii antibody titers and history of suicide attempts in patients
11. Alipour A, Shojaee S, Mohebali M, Tehranidoost M, Abdi with recurrent mood disorders. J Nerv Ment Dis 2009: 197:
Masoleh F, Keshavarz H. Toxoplasma infection in 905–908.
schizophrenia patients: a comparative study with control 25. Alvarado-Esquivel C, S
anchez-Anguiano LF, Arnaud-Gil CA
group. Iran J Parasitol 2011: 6: 31–37. et al. Toxoplasma gondii infection and suicide attempts: a
12. Webster JP. The effect of Toxoplasma gondii on animal case-control study in psychiatric outpatients. J Nerv Ment
behavior: playing cat and mouse. Schizophr Bull 2007: 33: Dis 2013: 201: 948–952.
752–756. 26. Dickerson F, Wilcox HC, Adamos M et al. Suicide attempts
13. Flegr J. Influence of latent Toxoplasma infection on human and markers of immune response in individuals with serious
personality, physiology and morphology: pros and cons of mental illness. J Psychiatr Res 2017: 87: 37–43.
the Toxoplasma-human model in studying the manipulation 27. Lindov
a J, Novotn
a M, Havl
ıcek J et al. Gender differences
hypothesis. J Exp Biol 2013: 216: 127–133. in behavioural changes induced by latent toxoplasmosis. Int
14. Webster JP, Kaushik M, Bristow GC, McConkey GA. Toxo- J Parasitol 2006: 36: 1485–1492.
plasma gondii infection, from predation to schizophrenia: 28. Esshili A, Thabet S, Jemli A et al. Toxoplasma gondii infec-
Can animal behaviour help us understand human behaviour? tion in schizophrenia and associated clinical features. Psychi-
J Exp Biol 2013: 216: 99–112. atry Res 2016: 245: 327–332.
15. Lanchava L, Carlson K, Seb
ankov
a B, Flegr J, Nave G. No 29. Laursen TM, Munk-Olsen T, Vestergaard M. Life expec-
evidence of association between Toxoplasma gondii infec- tancy and cardiovascular mortality in persons with
tion and financial risk taking in females. PLoS ONE 2015: schizophrenia. Curr Opin Psychiatry 2012: 25: 83–88.
10: e0136716. 30. Coryell W, Yolken R, Butcher B et al. Toxoplasmosis
16. Celik T, Kartalci S, Aytas O, Akarsu GA, Gozukara H, Unal titers and past suicide attempts among older adolescents
S. Association between latent toxoplasmosis and clinical initiating SSRI treatment. Arch Suicide Res 2016: 20:
course of schizophrenia – continuous course of the disease is 605–613.

Corresponding Author Maryam Ansari-Lari, Department of Food Hygiene and Public Heath, School of Veterinary Medicine,
Shiraz University, P.O. Box 71441, Shiraz 69155, Iran. Tel.: +98-713-2286950; Fax +98-713-2286940, E-mails: ansari@shirazu.
ac.ir, maryamansarilari@yahoo.com

6 © 2017 John Wiley & Sons Ltd