To conduct this systematic review, the protocol was developed and registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the number CRD42024596261. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines [18] were used as a writing guide. In this study, the research question focused on: Is there a difference between ethnic-racial groups in the prevalence of toxoplasmosis in pregnant women?

The inclusion and exclusion criteria for articles obtained during database searches followed the PECO strategy – Population, Exposure, Comparison, and Outcome, in which the Population is pregnant women, including adolescents, without restriction of nationality and at any gestational age; Exposure – infection with Toxoplasma Gondii during pregnancy; Comparison – pregnant women of different races and ethnicities; Outcome – Prevalence of toxoplasmosis during pregnancy in black, white, and other ethnic women—type of studies – Cross-sectional studies. Cohort studies, case-control studies, and literature reviews (integrative, narrative, or systematic with or without meta-analysis), case series, case reports, and qualitative analyses were not included in this research.

Given the scarcity of studies specifically designed to investigate racial inequalities in gestational toxoplasmosis, we chose to include investigations that reported the race/ethnicity variable in the description of infection prevalence, regardless of whether adjusted analyses were performed. This methodological decision was grounded in the objective of the present review, which was to map and synthesize the current state of available evidence regarding the distribution of toxoplasmosis during pregnancy across ethno-racial groups. Consequently, both studies featuring adjusted analyses and those with exclusively descriptive presentations were considered eligible, allowing for the identification of methodological gaps, heterogeneities in variable operationalization, and analytical limitations within the existing literature.

In September 2024, the following electronic databases were searched: PubMed, Web of Science, LILACS, Scopus, CINAHL, and Google Scholar to identify articles in the gray literature, in addition to manual searches. The searches were conducted without restrictions on the language of the articles or date of publication.

A librarian specializing in the field was consulted to develop the search strategies. To capture a wide range of articles related to the topic, the following health sciences descriptors (DeCS/MeSH) were used “Toxoplasmosis,” “Toxoplasma infection,” “Toxoplasma gondii,” “Pregnant women,” “Pregnancy,” and “Prenatal Care”, in addition to related free terms, combined with the Boolean operators “AND” and “OR” to obtain the most appropriate search key in each database, according to the controlled vocabulary. The search strategies used in this study are presented in Table 1.

Initially, exploratory search tests were conducted including terms related to race or skin color. However, the mandatory inclusion of these descriptors substantially reduced the sensitivity of the search strategy. In some tests, the strategy that combined terms for toxoplasmosis in pregnancy with race/ethnicity descriptors resulted in very limited retrieval of records and failed to identify articles that were previously known and considered relevant to the topic.

This result may be explained by the well-recognized terminological heterogeneity and the inconsistent indexing of the race/ethnicity variable in bibliographic databases. In many epidemiological studies, race or ethnicity is not the primary exposure variable and is often reported only in the sample characterization or included as a covariate in multivariable analyses. In such situations, these variables do not always appear in titles, abstracts, or indexed descriptors, which reduces the probability of retrieval when race/ethnicity-related terms are required in the search strategy.

Indeed, it was observed that key studies on toxoplasmosis in pregnancy that report information on race or ethnicity were not retrieved when these descriptors were included in the search strategy, but were identified when the search was conducted using only terms related to gestational toxoplasmosis. This finding suggests that requiring these descriptors could lead to the inadvertent exclusion of potentially relevant studies.

Therefore, a more sensitive search strategy was adopted, centered on terms related to toxoplasmosis during pregnancy. The identification of the race/ethnicity variable and its analytical role in the included studies was subsequently performed during the screening and full-text review stages, according to the criteria previously defined in the registered protocol.

The results obtained from each database were exported to the Mendeley Desktop® reference manager, version 2.85.0, to check and eliminate duplicate articles. Subsequently, they were deposited in the Rayyan QCRI® software for blind reading of the articles, which was carried out in two stages: in the first stage, screening was performed according to the reading of titles and abstracts, and subsequently, in the second stage, the articles were read in their entirety. In both stages, eligibility criteria were considered, and the articles were read by two independent reviewers (SRVC and RSB) trained in study selection, use of analysis software, and data extraction. Conflicts were resolved in consensus meetings by a third reviewer (IRB).

To assess the methodological quality of each included study, the Joanna Briggs Institute’s critical appraisal tool for systematic reviews of cross-sectional studies was used. This tool consists of eight questions that assess the methodological quality of the research and determine the extent to which the study addresses the possibility of bias in its design, conduct, and analysis [29].

The questions were classified independently by the authors as “Yes,” “No,” “Unclear,” or “Not/Applicable.” The classification of the risk of bias results will be evaluated as: (1) low risk, if the studies achieved more than 70% “yes” scores; (2) moderate risk, if the “yes” score is between 50% and 69%; and (3) high risk of bias, if the bias score is less than 49% [30]. Review Manager software (RevMan; version 5.4.1) was used to prepare the figure.

The authors extracted the data using Microsoft Excel. The data extracted from the included articles were authors, year of publication, study location, sample size, age group of the population, race/ethnicity of the population, diagnostic test method, Sero reaction, prevalence of toxoplasmosis, measures of association (odds ratio), and adjustment variables.

Given the lack of international standardization in the definition and operationalization of race and ethnicity, as well as the conceptual heterogeneity across different sociocultural contexts, these variables were analyzed in this review as reported in the included primary studies. For analytical purposes, race and ethnicity were considered collectively as sociodemographic markers of population stratification, maintaining the original terminology adopted by the authors. No reclassification or terminological harmonization was performed between studies to preserve methodological fidelity and avoid artificial interpretations resulting from recategorization.

Searches of databases and gray literature resulted in 4,652 studies. After removing duplicate articles, the studies were screened by title and abstract, resulting in 677 articles potentially eligible for full-text reading. Of these, 15 were excluded because the full text was unavailable. At the end of the selection process, 10 studies met the eligibility criteria and were included in this systematic review (Figure 1).

Table 2 summarizes the characteristics of the included studies. Most of the articles included (09) originate from the American continent, including Brazil and Mexico, as well as the Asian continent. There was only one study from Europe, conducted in the United Kingdom [25]. Brazil stands out with several regions represented, with studies conducted in the cities of Ilhéus-Bahia, Foz do Iguaçu-Paraná, in the state of Goiás, and in the city of Porto Alegre, in Rio Grande do Sul. The age ranges of the participants varied widely, ranging from adolescents under 18 to women over 50, as in studies conducted in Brazil and the United Kingdom. Other studies, such as that by Chandrasena et al. [19] in Sri Lanka, included women aged 16 to 40 [19], while in Aguascalientes, Mexico [20], the age range was 13–42. This age range reflects the search for a broad analysis, covering different reproductive profiles and their associated risks for toxoplasmosis.

The analysis of the studies included the detection of IgG, IgM, or both antibodies, depending on the objectives of each study. Most studies used IgG Sero reaction, which indicates past infections and is a valuable marker for understanding the cumulative prevalence of toxoplasmosis over time. In some cases, such as in the study conducted in Porto Alegre [23], both IgG and IgM were evaluated, allowing the identification of recent infections, which are especially relevant in contexts of gestational risk. Studies that analyzed IgM often complemented the results with avidity tests, ensuring the differentiation between recent and old infections.

The studies included in the analysis used different serological methods to detect the presence of antibodies against T. gondii, with emphasis on the ELISA (Enzyme-Linked Immunosorbent Assay) test, used in most studies due to its high sensitivity and specificity for IgG and IgM. Other methods used were MEIA (Microelisa Automated Immunoassay), LFA (Lateral Flow Assay), and EIA (Enzyme Immunoassay). These tests allowed the identification of previous, recent, or acute infections, depending on the detection of IgG, IgM, or both antibodies. In some cases, avidity tests were performed to distinguish acute infections from past infections. The choice of method varied according to the resources available, the context of the study, and the research objectives.

High heterogeneity was observed in the operationalization of the race and ethnicity variables across the ten included studies. Chandrasena et al. and Alvarado-Esquivel et al. [19, 20] adopted dichotomous classifications, such as “White vs. Non-white,” or nationality-based categories obtained through questionnaires and self-reporting [19, 20]. In contrast, Costa et al, Sartori et al. and Varella et al. [21–23] utilized polytomous (multicategorical) categories following the Brazilian Institute of Geography and Statistics (IBGE) criteria, also based on questionnaires and self-declaration [21–23].

Ramsewak et al., Flatt and Shetty, Chemoh et al., Mohamed et al. and Enter et al. [24–28] employed multicategorical categories grounded in nationality, geographic region of origin, or ethnic-based groups [24–28]. These variables were used as proxy markers for ethnicity, lacking a clear conceptual definition regarding the distinction between race, ethnicity, and nationality.

There is a difference between the racial groups studied, with the prevalence of toxoplasmosis varying widely among populations of different ethnic and racial origins. In studies conducted in Brazil, such as in Ilhéus, in the state of Bahia [21], the prevalence was higher among brown-skinned individuals (74.54%) and indigenous individuals (83.33%). The prevalence of toxoplasmosis among black pregnant women in a study conducted in Goiás-GO was 72.1% [22], while in Porto Alegre-RS it was 63.0% [23].

In Trinidad and Tobago, the prevalence of toxoplasmosis was 31.1% for African pregnant women [24], and in the United Kingdom, the prevalence was 31.48% among Afro-Caribbean women [25], highlighting differences between continents. On the other hand, in Sri Lanka, the prevalence was higher in Sinhalese women (13.56%) compared to non-Sinhalese women (6.3%) [19].

Beyond the prevalence distribution across ethno-racial groups, several studies conducted multivariable analyses to investigate factors associated with toxoplasmosis seropositivity. Statistically significant associations between race/ethnicity and infection were observed in different contexts. In Ilheus [21], non-white women showed a higher likelihood of seropositivity compared to white women (OR = 2.26; 95% CI: 1.25–4.06). In Goiás [22], Pardo (OR = 1.4; 95% CI: 1.1–1.6) and Black women (OR = 1.6; 95% CI: 1.1–1.7) had a greater probability of infection relative to white women. In the United Kingdom [25], Afro-Caribbean pregnant women exhibited higher odds of seropositivity (OR = 2.67; 95% CI: 1.83–3.88), even after adjusting for dietary habits. Furthermore, factors such as age, education level, income, occupation, and dietary behaviors were also associated with the infection across various adjusted models.

These data highlight significant variation between regions, suggesting that factors such as socioeconomic conditions, environmental exposure, and cultural differences may influence the observed rates. In addition, due to methodological and population heterogeneity, it was not possible to perform the meta-analysis planned in the protocol.

Regarding the risk of bias, most studies were classified as low risk [20–23, 25–28] and only two as moderate risk [19, 24]. Among the individual items, all studies measured exposure validly and reliably, used objective standards and criteria to measure the condition, measured outcomes with valid and reliable methods, and performed appropriate statistical analyses. However, more than half of the studies did not identify confounding factors [19, 22, 25–28]. Among these studies, only four report strategies for dealing with confounding factors [21–23, 25]. These aspects are shown in Table 3; Figure 2.

Despite the relevance of the findings, this review presents limitations that must be considered when interpreting the results. High methodological heterogeneity was observed among the included studies regarding study designs, population characteristics, and analytical strategies, which precluded the performance of a meta-analysis and limits direct comparability between findings.

The geographic representation was concentrated in the Americas and Asia, with an underrepresentation of European and African countries, thus restricting the generalizability of the results to distinct global contexts.

Furthermore, inconsistencies were identified in the ethno-racial classifications used, involving different categorization criteria and measurement methods, which may introduce classification biases. Variability in the diagnostic methods employed may also have influenced the reported prevalence estimates.

It is also noteworthy that some studies did not perform adequate adjustment for potential confounding variables—such as socioeconomic conditions, access to healthcare services, and other social determinants—which may impact the magnitude of the observed associations between race/ethnicity and gestational toxoplasmosis.

Despite these limitations, this review was conducted with methodological rigor and contributes to highlighting gaps in the literature, reinforcing the need for future investigations with greater methodological standardization and an intersectional approach to social determinants.

The studies included in this review suggest variations in the prevalence of gestational toxoplasmosis across different ethno-racial groups and geographic contexts. However, the interpretation of these differences must be conducted with caution, considering that a significant portion of the studies did not perform consistent adjustment for potential socioeconomic, environmental, and structural confounding factors.

The observed methodological heterogeneity—including variations in the operationalization of race/ethnicity, diagnostic methods, and analytical strategies—limits causal inferences regarding the independent role of race/ethnicity in the occurrence of toxoplasmosis during pregnancy. In this sense, the findings of this review should be understood as indicative of possible inequalities in the distribution of the infection, rather than as conclusive evidence of an independent association.

Above all, this review highlights significant gaps in the literature concerning the conceptual and methodological standardization of the race/ethnicity variable and the adequate control of social determinants of health. Such limitations hinder the identification of explanatory mechanisms and the adequate interpretation of the observed differences across populations.

Future investigations should prioritize more robust analytical designs, featuring systematic adjustment for social determinants, standardization in race/ethnicity classification, and transparent reporting of the operationalization of these variables. Longitudinal studies and stratified analyses with control for confounders may contribute more consistently to understanding the relative role of race/ethnicity in the epidemiology of gestational toxoplasmosis.