Exposure to Toxoplasma gondii Through Consumption of Raw or Undercooked Meat: A Systematic Review and Meta-Analysis

Abstract

Toxoplasma gondii is a globally distributed protozoan that mainly causes health issues in the fetuses of pregnant women who have never been exposed to this parasite and patients with deficient immune systems. Except in these vulnerable populations, the primary infection generally goes unnoticed in most healthy individuals. Apart from transplant/transfusion, congenital transmission, direct contact with infected cats or their feces, and environmental contamination (i.e., oocysts in food, water, and soil) pathways, humans can acquire the parasite through consumption of animal tissues infected by T. gondii. This meta-analysis estimated the risk of acquiring T. gondii by consuming raw or undercooked meat, regardless of which animal species are eaten. Using a random-effect model, crude and adjusted pooled measures of association (risk and odds ratio) were estimated according to study design (cohort, case–control, and cross-sectional studies). The meta-analysis included measures of heterogeneity as well as quality rating scales for each study design. Our results suggest that individuals who eat raw or undercooked meat have, respectively, 1.2–1.3 times the risk and 1.7–3.0 times the odds of T. gondii infection compared to those who thoroughly cook meat, regardless of the animal species they consume. These results align with the current understanding that adequately cooking meat inactivates the parasite and decreases the risk of transmission. Seroprevalence ranged from 1.3% to 88.6%, while the proportion of individuals eating raw or undercooked meat fluctuated from 0.7% to 98.3% across the studies in the meta-analysis. These numbers reflect various preferences with regard to eating meat (i.e., eating tartar, sausages, or salamis) as well as individual, cultural and religious food habits, and personal awareness.

Introduction

Rationale

Approximately 30% of the world's population has been exposed to Toxoplasma gondii, a globally distributed protozoan (Pappas et al. 2009). When immune-competent individuals acquire the parasite for the first time, the infection goes unnoticed for ∼85% of them, leaving ∼15% with mild and nonspecific signs or symptoms (fever and lymphadenopathy) (Montoya and Lisenfield 2004). Very rarely infected individuals exhibit clinical presentations such as posterior uveitis or acute systemic infections (myocarditis, encephalitis, pneumonitis, hepatitis, myalgia, etc.), which are more frequently observed in immunosuppressed patients (Biedermann et al. 1995, Akturk et al. 2017, Wang et al. 2017). Because the parasite can cross the placenta barrier, an infection during pregnancy can also have significant effects on the health of the fetus (e.g., neurological and ocular impairments) and gestation (miscarriage, stillbirth, pregnancy interruption, burden of an infected child) (McAuley 2014).

Three life cycle stages of the T. gondii are infectious to humans: (1) sporozoites contained in oocysts shed by domestic or wild felids, the only known definitive hosts of the protozoon; (2) tachyzoites in the rapidly multiplying stage encountered during the acute phase of a primary infection in intermediate hosts (humans and animals); and (3) bradyzoites in latent tissue cysts within intermediate hosts (Dubey 2010). Humans can acquire the parasite through four main pathways: medical intervention (e.g., organ transplantation or blood transfusion), mother to fetus (congenital), and by ingestion of oocysts or consumption of animal-infected tissues (cysts) (Montoya and Lisenfield 2004). Ingestion of oocysts happens either directly by contact with felids or their stools or indirectly through environmental contamination (e.g., sandbox, gardens, cat litter, water, or fresh produce such as vegetables or fruits) (Dubey 1998, Jones and Dubey 2010). As for animals, most domestic or wild species are potential carriers of T. gondii (Dubey 2010). Most public health agencies promote strategies to inactivate the parasite in meat (by cooking and freezing) or to decrease the risk of acquiring it through manipulation of meat (e.g., washing hands, wearing gloves, not tasting meat while cooking, and so on) (Dubey et al. 1990, Kijlstra and Jongert 2008, Herrero et al. 2017). Not all individuals and cultures practice these strategies, however, as some prefer their meat undercooked or raw. European culinary culture includes steak tartar, for example, and Arctic Inuit communities prefer raw or aged meat (Blanchet and Rochette 2008).

Objective

The objective of this systematic literature review is to estimate the probability of being exposed to T. gondii by raw or undercooked meat consumption, regardless of animal species.

Methodology

Information sources, search strategy, and eligibility criteria

This study followed the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines (Moher et al. 2015). No publication period was imposed and the literature search was last updated on February 6, 2020. Articles in languages other than English, French, Spanish, and Portuguese were excluded. Supplementary Table S1 describes the search strategies used for each bibliographic database. Exposure was defined as eating raw or undercooked meat, and outcome refers to the presence of antibodies against T. gondii. Meat was defined as muscular tissue from an animal (terrestrial or aquatic mammals or birds) or meat byproducts (e.g., sausages).

Data management

All references were imported directly to Endnote X8 software; duplicates were eliminated, and the final citations were exported into Excel software. Titles and abstracts were screened by two independent reviewers and classified as eligible (1), noneligible (0) or uncertain (?). An agreement value of 89.1% was reached and discrepancies were resolved by consensus after publications were further scrutinized for eligibility.

Selection process

A total of 3023 publications were identified, of which 779 duplicates and 1754 studies that, based on their abstract, did not cover risk factors associated with T. gondii were excluded (Supplementary Fig. S1). Then, combined with 18 additional studies found by examining other systematic reviews, 570 publications were reviewed to assess their eligibility. Data from 197 retained publications were extracted. As 18 of these publications presented 2 or more measures of effect (resulting from separate multivariate analysis for each population or study design), a database with 215 entries was obtained (Supplementary Fig. S1). The randomized trial portion from Desmont et al. (1965) was excluded from the meta-analysis based on ethical concerns and deficient methodology, while the data from the cross-sectional portion were maintained (Vergnes et al. 2010). Four additional studies were excluded from the meta-analysis because of null values that impeded the calculation of a measure of association. The final database contained 210 entries (Supplementary Fig. S1).

Data extraction

The following variables were extracted: first author's last name, year of publication, study design, country, seroprevalence, sample size, number of women and men, age, diagnostic tests, antibodies measured, type of statistical analysis, $α - l e v e l$ , number of seronegative and seropositive according to eating raw or undercooked meat, the crude and adjusted measures of association, the type and extent of statistical analysis, and the list of potential confounders included in multivariable analysis (Aiello and Larson 2002). When publications presented multiple associations according to different study design or populations, the data were entered in separate lines. If multiple sources of raw or undercooked meat were studied, the most frequently eaten meat was retained. All categories besides well-cooked meat were aggregated together to harmonize with most studies presenting a binary exposure level (Supplementary Table S2). Markers of acute or recent infection (IgM, IgA, IgE, or low IgG avidity) were preferred over IgG without avidity testing. Each diagnostic tool used by researchers was deemed appropriate and accurate enough to meet scientific publications criteria.

Quality of studies and bias

The quality of studies was evaluated using a Newcastle-Ottawa Scale for each study design, containing specific criteria about selection of individuals, exposure, and outcome. The maximum points granted was nine, which represents the highest quality and lowest bias (Wells et al. 2019). The scale was customized for comparability criteria for a maximum of two points. One point was attributed if (1) multivariable regression models adjusted for a sociodemographic variable (age, gender, education, socioeconomic status, or residential area) as they can reflect unmeasured exposure to the parasite or knowledge of hygiene measures (Cutler and Lleras-Muney 2010); and (2) if authors controlled for other source of exposure to T. gondii other than raw meat. These other sources of exposure were regrouped as (1) having contacts with cats; (2) being exposed to soil; (3) consumption of unwashed fruits and vegetables; (4) traveling or eating outside the home; (5) occupational risk; and (6) medical interventions. In the absence of a validated questionnaire, two points were attributed if a pretesting phase or standardization occurred and one point was added if the questionnaire was available.

Statistical methods

Four pooled measures of associations were calculated: crude odds ratio (OR), prevalence ratio (PR), incidence ratio (IR), and adjusted odds ratio (aOR). All measures of association were transformed on the natural logarithmic scale to conserve symmetry during the analysis (Borenstein et al. 2009). Standard error of log-transformed measures were calculated using 95% confidence intervals (CIs) or with the variance, and measures of effect were converted back to their original scale. Random-effect computational models were used based on the assumption that studies would reflect variability in the force of association; T. gondii prevalence and parasite burden vary between animal species, concentration of cats in the environment and geographical areas (Belluco et al. 2016). The study-weights were based on the inverse of the within-study variance $(V_{i}$ ) added to the between-studies variance ( $τ^{2}$ ). Therefore, the weighted mean is the sum of the product of each association measure multiplied by weight and divided by the sum of the weights. Only the forest plots of pooled adjusted measures of effects for cohorts and case–controls are presented in the article; all others are in the Supplementary Figs. S2–S6.

Heterogeneity was evaluated with the observed variation between studies that reflects a real difference in effect size ( $I^{2})$ (Borenstein et al. 2009). Publication bias was evaluated with Egger's regression test for zero intercept, using a random trim and fill approach (Egger et al. 1997). A sensitivity assessment was performed by excluding one study at a time and reestimating the pooled measures of effects. All data analyses were performed using MIX 2.0 software and only 95% exact CI are presented. For all subgroup comparisons, α level was adjusted downward using the sequential Bonferroni procedure to maintain the overall probability of a type I error at 0.05 (Bland and Altman 1995).

Results

Characteristics of eligible publications

All 210 entries retained were observational studies published from 1965 to 2019; most were cross-sectional (n = 196), followed by case–control (n = 11) and cohort (n = 3) (Supplementary Table S2). All studies included in the meta-analysis are referenced in the Supplementary Material. The T. gondii antibody seroprevalences reported ranged from 1.3% to 98.4%, while the proportion of those eating raw meat varied from 4.4% to 86.5%. The publications spanned 64 countries with the exception of 1 multicenter case–control study design based in 6 large European cities. The number of participants in the studies included in the meta-analysis ranged from 23 to 4436 individuals, while age of the participants varied from 6 months to 60 years.

Measures of effect between eating raw or undercooked meat and past exposure to T. gondii

The three prospective cohorts that measured incident cases included a total of 1218 individuals. The OR of 2.0 (95% CI; 0.7 to 5.8) overestimated the true association in a context of high seroprevalence such as with T. gondii compared to the IR of 1.3 (95% CI; 1.0 to 1.6) (Table 1 and Supplementary Figs. S2 and S3). The proportion of individuals eating raw or undercooked meat ranged from 8.6% to 41.9%, while the reported prevalence of antibodies against T. gondii varied from 6.2% to 67.1%.

Table 1.

Pooled Measures of Effect According to Study Design

Study design	Cohort	Case–control	Cross sectional
Crude pooled measures of effect
Odds ratio (95% CI)	2.0 (0.7 to 5.8) ^a	3.0 (2.1 to 4.3)	1.6 (1.4 to 1.7)
No. of database entries	3	10	203
No. of individuals	1218	4438	109,185
I ² (% with 95% CI)	—	62.9 (26.5 to 81.3)	80.3 (77.7 to 82.7)
Egger's regression coefficient β (95% CI)	—	1.1 (0.4 to 1.7)	0.3 (0.1 to 0.4)
Risk ratio (prevalence or incidence ratio) (95% CI)	1.3 (1.0 to 1.6)	—	1.2 (1.2 to 1.3)
No. of database entries	3	—	156
No. of individuals	1218	—	86,874
I ² (% with 95% CI)	88.4 (68.0 to 95.8)	—	83.4 (81.1 to 85.5)
Egger's regression coefficient β (95% CI)	0.1 (−0.1 to 0.4)	—	0.1 (0 to 0.1)
Adjusted pooled measure of effect
Adjusted odds ratio (95% CI)	—	2.9 (1.6 to 5.2)	1.7 (1.5 to 1.9)
No. of database entries	—	7	58
No. of individuals	—	2426	44,103
I ² (% with 95% CI)	—	69.5 (33.0 to 86.1)	72.8 (64.7 to 79.0)
Egger's regression coefficient β (95% CI)	—	1.2 (−0.2 to 2.5)	0.3 (0.1 to 0.5)

All numbers highlighted in bold correspond to the pooled measure of effect.

For cohorts, the odds ratio are presented only for comparison purposes with the other study designs since they represent a biased estimation of the true effect in the context of Toxoplasma gondii seroprevalence above 10% (odds ratio can approximate the risk ratios only in the presence of the “rare disease assumption”). For this reason, only heterogeneity measures of the risk ratios are presented.

CI, confidence interval.

The 10 case–control studies providing data to estimate a pooled crude measure of effect encompassed 4438 individuals and yielded an OR of 3.0 (95% CI; 2.1 to 4.3), while the 7 case–control studies retained to estimate an adjusted pooled measure yielded an aOR of 2.9 (95% CI; 1.6 to 5.2) and included 2426 individuals (Table 1). All point-estimates of OR observed in these case–controls were above the null value (Fig. 1 and Supplementary Fig. S4). These studies targeted high-risk individuals, either patients with acute toxoplasmosis or pregnant woman with recent seroconversion. The proportion of individuals consuming raw or undercooked meat varied from 6.6% to 57.4% and prevalence of T. gondii antibodies in the study and reported by authors ranged from 6.3% to 74.5%.

FIG. 1.

Forest plot^a of case–control studies included in the estimation of the pooled aOR measuring the association between Toxoplasma gondii seroprevalence and eating raw or undercooked meat. ^aInterpretation of a forest plot: each gray box reflects the measure of association (aOR) in the study with its 95% confidence intervals represented by the black lines on each side whereas the size of the box reflects the weight of each study. The thick black line traversing the graphic from top to bottom represents the pooled aOR with the 95% confidence intervals depicted by the width of the diamond shape at the bottom of the graph. aOR, adjusted odds ratio.

Of all the cross-sectional entries, three pooled measures of effect were estimated: an OR of 1.6 (95% CI; 1.4 to 1.7), a PR of 1.2 (95% CI; 1.2 to 1.3), and an aOR of 1.7 (95% CI; 1.5 to 1.9) (Table 1). The forest plot for the aOR is presented in Fig. 2, and the others are available in Supplementary Figs S5 and S6. The number of individuals involved in the cross-sectional studies varied according to the estimated measures of effect (Table 1). The proportion of participants eating raw or undercooked meat ranged from 0.7% to 98.3%, while seroprevalence of antibodies against T. gondii varied from 1.3% to 88.6%.

FIG. 2.

Forest plot^a of cross-sectional studies included in the estimation of the pooled aOR measuring the association between T. gondii seroprevalence and eating raw or undercooked meat. ^aInterpretation of a forest plot: Each gray box reflects the measure of association (aOR) in the study with its 95% confidence intervals represented by the black lines on each side whereas the size of the box reflects the weight of each study. The thick black line traversing the graphic from top to bottom represents the pooled aOR with the 95% confidence intervals depicted by the width of the diamond shape at the bottom of the graph.

None of the pooled measure of effect varied according to the quality score, climatic zone, human development index, serological test, or the type of immunoglobulins detected (results not shown). None of the exclusion sensitivity assessment yielded pooled measures of effect that changed the direction and strengths of the associations.

Heterogeneity bias

The within-study error ( $I^{2})$ of all study designs is greater than what would be expected if all studies shared a common effect (Table 1). The Egger's regression coefficients were above one for the case–controls and slightly above the null value for cohorts and cross-sectional designs (Table 1). These values demonstrate the presence of a slight dissemination bias, favoring studies with measures of association, when the 95% CI does not include the null value (Table 1).

Quality of studies

On a scale of nine points, cohorts ranked from six to seven points; case–controls ranked from zero to eight points (for crude measure) and from six to eight points (for adjusted measures), while cross-sectional studies ranked from one to seven points (crude measure) and from four to seven (adjusted measure) (Table 2). More detailed explanations of the quality of the studies are available in the Supplementary Data.

Table 2.

Evaluation of Quality of Publications According to Study Design

Criterion items by study design—maximum nine points	No. of points for each item	No. of entries with the attributed points (%)
Criterion items by study design—maximum nine points	No. of points for each item	Crude measures of effect	Adjusted measures of effect
Cohort studies			—
Selection
Representativeness of the exposed cohort	1	3 (100)	—
Selection of the nonexposed cohort	1	3 (100)	—
Ascertainment of exposure to raw/undercooked meat	1	3 (100)	—
Absence of T. gondii antibodies at the beginning	1	3 (100)	—
Comparability
Another source of exposure and a SD^a variables	2	1 (33.3)	—
Another source of exposure or a SD variables	1	—
Outcome
Assessment of T. gondii antibodies	1	3 (100)	—
Long enough follow-up period	1	2 (66.7)	—
Adequate follow-up period	1	1 (33.3)	—
Case–control studies
Selection
Adequate case definition	1	10 (100)	7 (100)
Representativeness of cases	1	10 (100)	7 (100)
Selection of controls from community	1	1 (10.0)	1 (14.3)
Definition of controls with no history of recent seroconversion/acute toxoplasmosis	1	8 (80.0)	7 (100)
Comparability
Another source of exposure and a SD variable	2	7 (70.0)	7 (100)
Another source of exposure or a SD variable	1	—	—
Exposure to raw or undercooked meat
Ascertainment of exposure by blinded interview	1	3 (30.0)	1 (14.3)
Same method used between cases and controls	1	8 (80.0)	7 (100)
Potential biases
Non-respondents information	1	—	1 (14.3)
Cross-sectional studies
Selection
Representativeness of the sample	1	61 (32.3)	21 (36.2)
Nonrespondents information	1	3 (1.6)	1 (1.7)
Ascertainment of exposure (two points maximum)
By pretested questionnaire	2	9 (4.8)	7 (12.1)
Description of variables in the questionnaire	1	173 (91.5)	51 (87.9)
Comparability (adjusted in statistical models)
Another source of exposure and a SD variable	2	76 (40.2)	49 (84.5)
Another source of exposure or a SD variable	1	15 (7.9)	9 (15.5)
Outcome (detection of T. gondii immunoglobulin)
Adequate assessment	1	189 (100)	58 (100)
Appropriateness of statistical tests	1	89 (47.1)	58 (100)

Sociodemographic variable (e.g., age, gender, education, socioeconomic status or residential area).

SD, sociodemographic variable.

Discussion

To our knowledge, this is the first systematic review and meta-analysis of the association between T. gondii past exposure and eating raw or undercooked meat, regardless of the animal species consumed. Our meta-analysis suggests that individuals who eat raw or undercooked meat have higher risks and odds of having antibodies against T. gondii, compared to those who thoroughly cook meat. These results align with the current understanding that adequately cooking meat leads to a decreased risk of transmission (Dubey 2000, Kijlstra and Jongert 2008, Herrero et al. 2017). Our results also align with the findings of another meta-analysis that used only case–control studies, which estimated the same type of association by animal species. Belluco et al. (2016) obtained an overall pooled OR of 3.4 (95% CI; 1.3 to 9.2) for consumption of raw or undercooked meat compared to well-cooked meat (2.2 for beef and of 3.9 for sheep).

The proportion of individuals who eat raw or undercooked meat fluctuated across studies and reflects various preferences of how to consume meat (e.g., as tartar, sausages or salamis) according to individual, cultural or religious food habits, and personal awareness (Buffolano et al. 1996, Kapperud et al. 1996, Bobic et al. 1998, Baril et al. 1999, Roghmann et al. 1999). These expected variations are reflected in the heterogeneity measures, demonstrating a wide dispersion of associations in the forest plots.

For our purposes, participants were dichotomized as eating raw or undercooked meat or not, which represents a crude assessment of exposure. Different recall periods and varying frequency of the consumption of raw or undercooked meat between studies impeded the use of a finer scale to measure exposure. The few studies that defined the terms “well-cooked,” “raw,” or “undercooked” meat usually referred to the color and/or the presence of blood in meat. Even when individuals are aware of possible foodborne illnesses in undercooked meat, few use a thermometer to assess the temperature of the middle of the piece of meat (Murray et al. 2017). Hence, a superior ratio of misclassification of exposure in our control group is expected compared to the exposed (eating raw or undercooked meat), which suggests that our pooled measures of association underestimate the true effect.

Evaluating past or recent exposure to T. gondii with antibody titers was preferred since serological testing serves as initial screening in most studies. These tests display higher sensitivity and specificity values for past exposure than molecular techniques to diagnose acute toxoplasmosis, its progressions, and the treatment efficacy (Liu et al. 2015). However, serology lacks information about the moment individuals were first exposed to the parasite (temporality). Typically, following a primary T. gondii infection, IgM will first rise, followed by IgG in the first to third weeks postinfection (Fricker-Hidalgo et al. 2013). IgM and IgG are most frequently measured compared to the other markers (i.e., IgA and IgE or IgG with avidity testing). However, atypical seroconversion patterns have been occasionally observed, which suggests that misclassification can occur (Fricker-Hidalgo et al. 2013). Different serological tests were used between studies; mostly enzyme immune-assay (68%) followed by agglutination (16%), fluorescent antibody (13%), Sabin–Feldman dye (3%), and skin antigen (1%) tests. Only rarely did studies provide information on sensitivity and specificity of their serological test. Therefore, it was not possible to validate their level of accuracy or the direction and strength of this potential information bias.

The main strength of this meta-analysis is that it includes all types of study design that evaluate the association between eating raw or undercooked meat and T. gondii infection. Crude or adjusted pooled measures of effect for each study design are consistently above the null value, indicating that eating raw or undercooked meat increased the probabilities of having antibodies against T. gondii. The adjusted measures of effect decrease the potential confounding bias and increase the validity of results compared to crude measures (Mickey and Greenland 1989).

Our meta-analysis also has limitations. First, many studies did not include the “eating raw or undercooked meat” variable in their multivariate analyses, leading to a potential selection bias when estimating the pooled adjusted measures of effect (Meireles et al. 2015). This type of bias was mainly observed with cross-sectional designs. The estimation of a pooled OR also overestimates the strengths of the association in the context of a disease of high seroprevalence such as T. gondii, and relative risks should be preferred when interpreting the results (Mickey and Greenland 1989). A few studies had significant associations with 95% CIs below the null value, indicating a lower seroprevalence in the exposed (eating raw or undercooked meat) compared to the control group. This suggests the presence of uncontrolled confounding or of random variability (Egger et al. 1998). These cross-sectional studies have potentially biased the pooled measures toward the null values, but the extent of this effect remains unknown. Finally, the different quality scales used to evaluate the studies did not capture all the features of internal and external validity—namely, information on validation of regression model postulates such as linearity, independence between subjects, collinearity, and extreme data. Violation of these assumptions can also bias the interpretation of the measure of association (Vittinghoff et al. 2005). Some studies may have ignored collinearity between variables, a reality that often occurs in nutritional epidemiology when collecting information about multiple food items (Elmsthal and Gullberg 1997).

Conclusion

The consistency of the estimates of measures of effect above the null value across all study designs suggests that eating raw or undercooked meat is frequently linked to the presence of antibodies against T. gondii. Our study supports the notion that more individuals who eat raw or undercooked meat have antibodies against T. gondii than those who thoroughly cook meat, regardless of what animal they consume. These pooled measures of association are believed to reflect that eating raw or undercooked meat contributes to exposing humans to T. gondii, despite the presence of limitations associated with this meta-analysis.

Support

The methodology follows the PRISMA-P 2015 (Moher et al. 2015). Details of the protocol have been registered on PROSPERO (No. CRD42015029253) and can be accessed at (www.crd.york.ac.uk/PROSPEROFILES/29253_PROTOCOL_20151017.pdf).

Authors' Contributions

J.D. and A.S. have elaborated the protocol, performed the systematic review, extracted, and analyzed the data. J.D. wrote the article, and all others have revised the text and made comments. The first author conducted this systematic review as a requisite for a doctoral course.

Footnotes

Acknowledgments

We thank Jacques Brisson and Hervé Tchala Vignon Zomahoun from Université Laval (Québec, Canada) for their ongoing support and feedback throughout the progress of this article. Many thanks also to Myrto Mondor and Elhadji Anassour Sidi for their guidance as biostatisticians and to Sara Carazo for her review of Spanish, Portuguese, and Italian articles. In addition, we offer a warm welcome to the following authors who generously wrote back with additional information about their studies: Digsu Negese, Abeer Mahgoub, Dominika Salamon, Olgica Djurkovic-Djakovic, Edwin Ogendi and Jorge Enrique Gomez Marin.

Author Disclosure Statement

The authors declare no conflict of interest.

Funding Information

Financial support was provided by the Fond de recherche du Québec—Santé, ArcticNet and the Nasivvik Research Chair in Ecosystemic Approach to Northern Health.

Supplementary Material

Supplementary Data

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

Supplementary Figure S4

Supplementary Figure S5

Supplementary Figure S6

Supplementary Table S1

Supplementary Table S2

References

Achas

, Tesfa

, Zeleke

, Worku

, et al. Sero-prevalence of Toxoplasma gondii and associated risk factors among psychiatric outpatients attending University of Gondar Hospital, Northwest Ethiopia. BMC Infect Dis, 2019; 19:581.

Aiello

, Larson

. Causal inference: The case of hygiene and health. Am J Infect Control, 2002; 30:503–511.

Akturk

, Sotello

, Ameri

, Abuzaid

, et al. Toxoplasma infection in an immunocompetent host: Possible risk of living with multiple cats. Cureus, 2017; 9:e1103.

Al-Mohammad

, Amin

, Balaha

, Al-Moghannum

. Toxoplasmosis among the pregnant women attending a Saudi maternity hospital: Seroprevalence and possible risk factors. Ann Trop Med Parasitol, 2010; 104:493–504.

Alvarado-Esquivel

, Estrada-Martinez

, Liesenfeld

. Toxoplasma gondii infection in workers occupationally exposed to unwashed raw fruits and vegetables: A case control seroprevalence study. Parasit Vectors, 2011; 4:1–7.

Alvarado-Esquivel

, Estrada-Martinez

, Pizarro-Villalobos

, Arce-Quinones

, et al. Seroepidemiology of Toxoplasma gondii infection in general population in a northern Mexican city. J Parasitol, 2010a; 97:40–43.

Alvarado-Esquivel

, Cruz-Magallanes H

, Esquivel-Cruz

, Estrada-Martínez

, et al. Infection in human adults from three rural communities in Durango state, Mexico. J Parasitol, 2008; 94:811–816.

Alvarado-Esquivel

, Pacheco-Vega

, Hernandez-Tinoco

, Saldana-Simental

, et al. Lack of association between Toxoplasma gondii infection and occupational exposure to animals. Eur J Microbiol Immunol, 2014a; 4:184–192.

Alvarado-Esquivel

, Pacheco-Vega

, Hernandez-Tinoco

, Sanchez-Anguiano

, et al. Seroprevalence of Toxoplasma gondii infection and associated risk factors in Huicholes in Mexico. Parasites Vectors, 2014b; 7:301.

10.

Alvarado-Esquivel

, Rojas-Rivera

, Estrada-Martinez

, Sifuentes-Alvarez

, et al. Seroepidemiology of Toxoplasma gondii infection in a Mennonite community in Durango State, Mexico. J Parasitol, 2010b; 96:941.

11.

Alvarado-Esquivel

, Torres-Castorena

, Liesenfeld

, Estrada-Martinez

, et al. High seroprevalence of Toxoplasma gondii infection in a subset of Mexican patients with work accidents and low socioeconomic status. Parasites Vectors, 2012b; 5:13.

12.

Alvarado-Esquivel

, Liesenfeld

, Burciaga-Lopez

, Ramos-Nevarez

, et al. Seroepidemiology of Toxoplasma gondii infection in elderly people in a northern Mexican city. Vector Borne Zoonotic Dis, 2012a; 12:568–574.

13.

Awoke

, Nibret

, Munshea

. Sero-prevalence and associated risk factors of Toxoplasma gondii infection among pregnant women attending antenatal care at Felege Hiwot Referral Hospital, northwest Ethiopia. Asian Pac J Trop Med, 2015; 8:549–554.

14.

Baril

, Ancelle

, Goulet

, Thulliez

, et al. Risk factors for Toxoplasma infection in pregnancy: A case-control study in France. Scand J Infect Dis, 1999; 31:305–309.

15.

Belluco

, Mancin

, Conficoni

, Simonato

, et al. Investigating the determinants of Toxoplasma gondii prevalence in meat: A systematic review and meta-regression. PLoS One, 2016; 11:1–24.

16.

Biedermann

, Flepp

, Fierz

, Joller-Jemelka

, et al. Pregnancy, immunosuppression and reactivation of latent toxoplasmosis. J Perinat Med, 1995; 23:191–203.

17.

Blanchet

, Rochette

. Nutrition and food consumption among the Inuit of

Nunavik

. Nunavik Inuit Health

Survey

. 2004. Qanuippitaa? How are we? 2008. Available at https://www.inspq.qc.ca/sites/default/files/publications/762_esi_nutrition_report_ma.pdf

18.

Bland

, Altman

. Multiple significance tests: The Bonferroni method. Br Med J, 1995; 310:170.

19.

Bobic

, Jevremovic

, Marinkovic

, Sibalic

, et al. Risk factors for Toxoplasma infection in a reproductive age female population in the area of Belgrade, Yugoslavia. Eur J Epidemiol, 1998; 14:605–610.

20.

Bobic

, Nikolic

, Klun

, Vujanic

, et al. Undercooked meat consumption remains the major risk factor for Toxoplasma infection in Serbia. Parassitologia, 2007; 49:227–230.

21.

Borenstein

, Hedges

, Higgins

, Rothstrein

. Introduction to Meta-Analysis. United Kingdom: John Wiley and Sons, Ltd., 2009.

22.

Buffolano

, Gilbert

, Holland

, Fratta

, et al. Risk factors for recent toxoplasma infection in pregnant women in Naples. Epidemiol Infect, 1996; 116:347–351.

23.

Chiang

, Kuo

, Chen

, Yang

, et al. Risk factors for acute Toxoplasma gondii diseases in Taiwan: A population-based case-control study. PLoS One, 2014; 9:e90880.

24.

Cong

, Dong

, Bai

, Wang

, et al. Seroprevalence and associated risk factors of Toxoplasma gondii infection in psychiatric patients: A case-control study in eastern China. Epidemiol Infect, 2015a; 143:3103–3109.

25.

Cong

, Liu

, Meng

, Dong

, et al. Toxoplasma gondii infection in cancer patients: Prevalence, risk factors, genotypes and association with clinical diagnosis. Cancer Lett, 2015b; 359:307–313.

26.

Cong

, Dong

, Meng

, Zhou

, et al. Toxoplasma gondii infection in pregnant women: A seroprevalence and case-control study in Eastern China. Epidemiol Infect, 2015c; 143:3103–3109.

27.

Cook

AJG

, Gilbert

, Buffolano

, Zufferey

, et al. Sources of toxoplasma infection in pregnant women: European multicentre case-control study. European Research Network on congenital toxoplasmosis. Br Med J, 2000; 321:14–27.

28.

Cutler

, Lleras-Muney

. Understanding differences in health behaviors by education. J Health Econ, 2010; 29:1–28.

29.

Desmonts

, Couvreur

, Alison

, Baudelot

, et al. Epidemiological study on toxoplasmosis: The influence of cooking slaughter-animal meat on the incidence of human infection. Rev Fr Etud Clin Biol, 1965; 10:952.

30.

Dubey

. Toxoplasma gondii oocyst survival under defined temperatures. J Parasitol, 1998; 84:862–865.

31.

Dubey

. The scientific basis for prevention of Toxoplasma gondii infection: Studies on tissue cyst survival, risk factors and hygiene measures. In: Petersen

, Ambroise-Thomas

, eds. Congenital Toxoplasmosis: Scientific Background, Clinical Management and Control. France: Springer-Velag, 2000:271–275.

32.

Dubey

. Toxoplasma gondii of animals and humans. Boca Ratton, CRC Press, 2010.

33.

Dubey

, Kotula

, Sharar

, Andrews

, et al. Effect of high temperature on infectivity of Toxoplasma gondii tissue cysts in pork. J Parasitol, 1990; 76:201–204.

34.

Egger

, Davey Smith

, Schneider

, Minder

. Bias in meta-analysis detected by a simple, graphical test. Br Med J, 1997; 315:629–634.

35.

Egger

, Schneider

, Smith

. Meta-analysis: Spurious precision? Meta-analysis of observational studies. Br Med J, 1998; 316:140–144.

36.

Egorov

, Converse

, Griffin

, Styles

, et al. Environmental risk factors for Toxoplasma gondii infections and the impact of latent infections on allostatic load in residents of Central North Carolina. BMC Infect Dis, 2018; 18:421.

37.

Elmsthal

, Gullberg

. Bias in diet assessment methods—Consequences of collinearity and measurement errors on power and observed relative risks. Int J Epidemiol, 1997; 26:1071–1079.

38.

Fan

, Liao

, Kao

, Lu

, et al. Toxoplasma gondii infection: Relationship between seroprevalence and risk factors among inhabitants in two offshore islands from Taiwan. Acta Med Okayama, 2001; 55:301–308.

39.

Fan

, Su

, Wu

, Chiou

. Seroepidemiology of Toxoplasma gondii infection among two mountain aboriginal populations and Southeast Asian laborers in Taiwan. Journal of Parasitology, 2002; 88:411–414.

40.

Flatt

, Shetty

. Seroprevalence and risk factors for toxoplasmosis among antenatal women in London: A re-examination of risk in an ethnically diverse population. Eur J Public Health, 2013; 23:648–652.

41.

Flegr

, Hrda

, Tachezy

. The role of psychological factors in questionnaire-based studies on routes of human toxoplasmosis transmission. Central Eur J Public Health, 1998; 6:45–50.

42.

Fricker-Hidalgo

, Cimon

, Chemla

, Darde

, et al. Toxoplasma seroconversion with negative or transient immunoglobulin M in pregnant women: Myth or reality? A French multicenter retrospective study. J Clin Microbiol, 2013; 51:2103–2111.

43.

, Chuang

, Lin

, Wu

, et al. Toxoplasma gondii infection: Seroprevalence and associated risk factors among primary school children in the capital area of the Republic of the Marshall Islands. Jpn J Infect Dis, 2014; 67:405–410.

44.

Gebremedhin

, Agonafir

, Tessema

, Tilahun

, et al. Seroepidemiological study of ovine toxoplasmosis in East and West Shewa Zones of Oromia Regional State, Central Ethiopia. BMC Vet Res, 2013; 9:117.

45.

Gelaye

, Kebede

, Hailu

. High prevalence of anti-toxoplasma antibodies and absence of Toxoplasma gondii infection risk factors among pregnant women attending routine antenatal care in two Hospitals of Addis Ababa, Ethiopia. Int J Infect Dis, 2015; 34:41–45.

46.

Gyang

VPA

, Akinwale

, Lee

, Chuang

, et al. Toxoplasma gondii infection: Seroprevalence and associated risk factors among primary schoolchildren in Lagos City, Southern Nigeria. Rev Soc Bras Med Trop, 2015; 48:56–63.

47.

Hamza

, El-Taweel

, Abou-Holw

, Khalil

, et al. Toxoplasma gondii seropositivity in renal patients: Rate, pattern, predictors and related morbidity. J Egypt Soc Parasitol, 2015; 45:7–15.

48.

Herrero

, Gracia

, Perez-Arquillue

, Lazaro

, et al. Toxoplasma gondii in raw and dry-cured ham: The influence of the curing process. Food Microbiol, 2017; 65:213–220.

49.

Hofhuis

, Van Pelt

, Van Duynhoven

, Nijhuis

CDM

, et al. Decreased prevalence and age-specific risk factors for Toxoplasma gondii IgG antibodies in The Netherlands between 1995/1996 and 2006/2007. Epidemiol Infect, 2011; 139:530–538.

50.

Jiang

, Zhao

, Jiang

, Chen

, et al. Toxoplasma gondii infection in dustmen in Northeastern China: A case-control seroprevalence study. BioMed Res Int, 2018; 2018:1–15.

51.

Jones

, Dargelas

, Roberts

, Press

, et al. Risk factors for Toxoplasma gondii infection in the United States. Clin Infect Dis, 2009; 49:878–884.

52.

Jones

, Dubey

. Waterborne toxoplasmosis: Recent developments. Exp Parasitol, 2010; 124:10–25.

53.

Jones

, Muccioli

, Belfort

Jr. , Holland

, et al. Recently acquired Toxoplasma gondii infection, Brazil. Emerg Infect Dis, 2006; 12:582–587.

54.

Kapperud

, Jenum

, Stray-Pedersen

, Melby

, et al. Risk factors for Toxoplasma gondii infection in pregnancy: Results of a prospective case-control study in Norway. Am J Epidemiol, 1996; 144:405–412.

55.

Kijlstra

, Jongert

. Control of the risk of human toxoplasmosis transmitted by meat. Int J Parasitol, 2008; 38:1359–1370.

56.

Kolbekova

, Kourbatova

, Novotna

, Kodym

, et al. New and old risk-factors for Toxoplasma gondii infection: Prospective cross-sectional study among military personnel in the Czech Republic. Clin Microbiol Infect, 2007; 13:1012–1017.

57.

, Dong

, Li

, Zhang

, et al. Seroepidemiology of Toxoplasma gondii infection in Bai and Han ethnic groups in southwestern China. Epidemiol Infect, 2015; 143:881–886.

58.

Liu

, Wang

Z-D

, Huang

S-Y

, Zhu

. Diagnosis of toxoplasmosis and typing og Toxoplasma gondii . Parasit Vectors, 2015; 28:292.

59.

Liu

, Wei

, Gao

, Jiang

, et al. Toxoplasma gondii infection in pregnant women in China. Trans R Soc Trop Med Hyg, 2009; 103:162–166.

60.

Lopes

, Dubey

, Moutinho

, Gargate

, et al. Seroepidemiology of Toxoplasma gondii infection in women from the North of Portugal in their childbearing years. Epidemiol Infect, 2012; 140:872–877.

61.

Mahmoudvand

, Saedi Dezaki

, Soleimani

, Baneshi

, et al. Seroprevalence and risk factors of Toxoplasma gondii infection among healthy blood donors in south-east of Iran. Parasite Immunol, 2015; 37:362–367.

62.

McAuley

. Congenital toxoplasmosis. J Pediatr Infect Dis Soc, 2014; 3(Suppl 1):S30–S35.

63.

Meireles

, Ekman

, Andrade

, Luna

. Human toxoplasmosis outbreaks and the agent infecting form: Findings from a systematic review. Rev Inst Med Trop Sao Paulo, 2015; 57:369–376.

64.

Messerer

, Bouzbid

, Gourbdji

, Mansouri

, et al. [Seroprevalence of toxoplasmosis in pregnant women in Annaba, Algeria]. Rev Épidemiol Sante Publique, 2014; 62:160–165.

65.

Mickey

, Greenland

. The impact of confounder selection criteria on effect estimation. Am J Epidemiol, 1989; 129:125–137.

66.

Moher

, Shamseer

, Clarke

, Ghersi

, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev, 2015; 4:1–9.

67.

Montoya

, Lisenfield

. Toxoplasmosis. Lancet, 2004; 363:1965–1976.

68.

Murebwayire

, Njanaake

, Ngabonziza

, Njunwa

, et al. Seroprevalence and risk factors of Toxoplasma gondii infection among pregnant women attending antenatal care in Kigali, Rwanda. Tanzania J Health Res, 2017; 19(1). https://doi.org/10.4314/thrb.v19i1.2

69.

Murray

, Glass-Kaastra

, Gardhouse

, Marshall

, et al. Canadian consumer food safety practices and knowledge: Foodbook Study. J Food Protect, 2017; 80:1711–1718.

70.

Negero

, Yohannes

, Woldemichael

, Tegegne

. Seroprevalence and potential risk factors of T. gondii infection in pregnant women attending antenatal care at Bonga Hospital, Southwestern Ethiopia. Int J Infect Dis, 2017; 57:44–49.

71.

Nissapatorn

, Lee

, Cho

, Rohela

, et al. Toxoplasmosis in HIV/AIDS patients in Malaysia. Southeast Asian J Trop Med Public Health, 2003; 34 Suppl 2:80–85.

72.

Pappas

, Roussos

, Falagas

. Toxoplasmosis snapshots: Global status of Toxoplasma gondii seroprevalence and implications for pregnancy and congenital toxoplasmosis. Int J Parasitol, 2009; 39:1385–1394.

73.

Roghmann

, Faulkner

, Lefkowitz

, Patton

, et al. Decreased seroprevalence for Toxoplasma gondii in Seventh Day Adventists in Maryland. Am J Trop Med Hygiene, 1999; 60:790–792.

74.

Rostami

, Seyyedtabaei

, Aghamolaie

, Behniafar

, et al. Seroprevalence and risk factors associated with Toxoplasma gondii infection among rural commuities in Northern Iran Rev Inst Med Trop São Paulo. 2016; 58:70.

75.

Said

, Halsby

, O'Connor

, Francis

, et al. Risk factors for acute toxoplasmosis in England and Wales. Epidemiol Infect, 2017; 145:23–29.

76.

Sang-Eun

, Hong

, Jeong

, Lee

, et al. Cross-sectional analysis of the seropositivity and risk factors of Toxoplasma gondii infection among veterinarians, in relation to their public professional activities. Vet Parasitol, 2014; 203:29–34.

77.

Sharbatkhori

, Dadi Moghaddam

, Pagheh

, Mohammadi

, et al. Seroprevalence of Toxoplasma gondii infections in pregnant women in Gorgan City, Golestan Province, northern Iran-2012. Iran J Parasitol, 2014; 9:181–187.

78.

Sharif

, Daryani

, Barzegar

, Nasrolahei

. A seroepidemiological survey for toxoplasmosis among schoolchildren of Sari, Northern Iran. Trop Biomed, 2010; 27:220–225.

79.

Silva

, Camara

, Vinaud

, Castro

. Epidemiological factors associated with seropositivity for toxoplasmosis in pregnant women from Gurupi, State of Tocantins, Brazil. Rev Soc Bras Med Trop, 2014; 47:469–475.

80.

Tegene

, Abdurahaman

, Mosissa

, Yohannes

. Anti-Toxoplasma antibodies prevalence and associated risk factors among HIV patients. Asian Pac J Trop Med, 2016; 9:460–464.

81.

Teweldemedhin

, Gebremichael

, Geberkirstos

, Hadush

, et al. Seroprevalence and risk factors of Toxoplasma gondii among pregnant women in Adwa district, northern Ethiopia [Article]. BMC Infect Dis, 2019; 19:327.

82.

Tian

, Gu

, Zhou

, Cong

, et al. Seroprevalence of Toxoplasma gondii infection in arthritis patients in eastern China. Infect Dis Poverty, 2017a; 6:153.

83.

Tian

, Li

, Elsheikha

, Gardner

, et al. Seroepidemiology of Toxoplasma gondii infection in patients with liver disease in eastern China. Epidemiol Infect, 2017b; 145:2296–2302.

84.

Vergnes

J-N

, Marchal-Sixou

, Nabet

, Maret

. Ethics in systematic reviews. BMC Med Ethics, 2010; 36:771–774.

85.

Vittinghoff

, Glidden

, Shiboski

, McCulloch

. Regression Methods in Biostatistics Linear, Logistic, Survival, and Repeated Measures Models. New York: Springer, 2005.

86.

Walle

, Kebede

, Tsegaye

, Kassa

. Seroprevalence and risk factors for toxoplasmosis in HIV infected and non-infected individuals in Bahir Dar, Northwest Ethiopia. Parasit Vectors, 2013; 6:15.

87.

Wam

, Sama

, Ali

, Ebile

, et al. Seroprevalence of Toxoplasma gondii IgG and IgM antibodies and associated risk factors in women of child-bearing age in Njinikom, NW Cameroon. BMC Res Notes, 2016; 9:406.

88.

Wang

Z-D

, Liu

H-H

, Ma

Z-X

, Ma

H-Y

, et al. Toxoplasma gondii infection in immunocompromised patients: A systematic review and meta-analysis. Front Microbiol, 2017; 8:389.

89.

Wells

, Shea

, O'Connell

, Peterson

, et al. (2019). The Newcastle-Ottawa Scale ( NOS) for assessing the quality of nonrandomised studies in meta-analyses. Available at http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp (accessed March 17, 2020 ).

90.

Yang

, Wang

, Xing

, Li

, et al. Seroepidemiology and risk factors of Toxoplasma gondii infection among the newly enrolled undergraduates and postgraduate students in China. Front Microbiol, 2017; 8:2092.

91.

Yohanes

, Debalke

, Zemene

. Latent Toxoplasma gondii infection and associated risk factors among HIV-infected individuals at Arba Minch Hospital, South Ethiopia. AIDS Res ment, 2014; 2014:652941.

92.

Zeleke

, Melsew

. Seroprevalence of Toxoplasma gondii and associated risk factors among HIV-infected women within reproductive age group at Mizan Aman General Hospital, Southwest Ethiopia: A cross sectional study. BMC Res Notes, 2017; 10:70.

93.

Zemene

, Yewhalaw

, Abera

, Belay

, et al. Seroprevalence of Toxoplasma gondii and associated risk factors among pregnant women in Jimma town, Southwestern Ethiopia. BMC Infect Dis, 2012; 12:337.

94.

Zhang

, Zhao

, Shi

, Yang

, et al. Seroprevalence and associated risk factors of Toxoplasma gondii infection in the Korean, Manchu, Mongol and Han ethnic groups in eastern and northeastern China. Epidemiol Infect, 2016; 144:2018–2024.

95.

Zhao

, Zhao

, Zhang

, Sun

, et al. First report of Toxoplasma gondii infection in tuberculosis patients in China. Vector Borne Zoonotic Dis, 2017; 17:799–803.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.05 MB

0.02 MB

0.06 MB

0.31 MB

0.36 MB

0.31 MB

0.32 MB

0.00 MB