Associations of mental disorders and neurotropic parasitic diseases: a meta-analysis in developing and emerging countries

Although parasitic diseases are declining, they are still common in developing and emerging countries. Among the public health issues typically faced by developing and emerging countries (all non high income countries according to World Bank rankings) [1], neurotropic parasitic diseases are very common [2]. Neurotropic parasitic diseases such as malaria, cysticercosis, toxoplasmosis, human African trypanosomiasis (HAT), Chagas disease, and human toxocariasis have a predilection for infesting the central nervous system, which can lead to neurological disorders. Similarly, mental disorders are frequent with pooled 12-month period prevalence estimates of 17.6% (15.5–20.0%) in low and middle-income countries [3]. However, there are few studies regarding the association of mental disorders with neurological parasitoses in these countries.

To date, epidemiological studies have identified several risk factors for the different disease groups studied [4‐9]. The table below (Table 1) provides a summary of the main risk factors of interest diseases in this meta-analysis. We have focused our attention on parasitic diseases due to the ever-present burden of these diseases in developing and emerging countries, and we selected the diseases according to their overall burden [5, 11‐13]. It is estimated that mental disorders rank third among the most frequent diseases encountered in the world, just after cancer and cardiovascular diseases [14].

In developing and emerging countries, surveys have suggested that more than 25% of individuals in their lifetime develop one or more mental or behavioural disorders [15]. In the general population, it has been estimated that 3% of people are affected by severe depression, 2% by generalized anxiety disorders and 1% by schizophrenia [16]. Parasitic diseases remain a major burden to developing and emerging countries, although this type of disease has declined globally. In 2015, out of 95 countries and territories in the world where malaria transmission remains high, it is estimated that the number of malaria cases was 214 million (95% CI, 149–303) and that malaria is responsible for 438,000 deaths per year (95% CI, 236000–635,000) mainly in Africa (88%) [17]. In 2012, eight million individuals were already infected with Trypanosoma cruzi, the parasite that causes Chagas disease, in endemic areas of 21 Latin American countries. In addition, the chronic infection caused by this parasite is incurable, can be disabling, and causes more than 10,000 deaths per year [18]. The seroprevalence of toxocariasis varies from 2.4 to 30.0% in Europe [19], but in tropical countries, higher prevalences have been reported: 7.5 to 92.8% in Africa [20, 21], 6.4 to 52.0% in South America [22, 23], and 5.0 to 84.6% in Asia [24, 25]. Human African trypanosomiasis (HAT) affects 60 million inhabitants mainly living in rural areas of 36 endemic sub-Saharan countries in East, West and Central Africa [26]. Toxoplasmosis remains frequent in these countries and even in developed countries [10]. Finally, the agent responsible for cysticercosis, Taenia solium, is found mainly in Latin America, Asia, sub-Saharan Africa, and the Indian Ocean region. According to the World Health Organization (WHO), cysticercosis is responsible for 50,000 deaths per year with 2.5 to 5 million adult worm carriers and 50 million cysticercal larvae carriers [27, 28]. Neurocysticercosis, a type of brain damage resulting from cysticercosis, is thought to be a factor responsible for more than 50% of late onset epileptic seizures in developing countries [27].

In recent years, there has been an increasing number of studies on co-morbidities between mental disorders and parasitic diseases, particularly those with neurological tropism such as toxoplasmosis [9], human toxocariasis [29, 30], and cysticercosis [31‐34]. However, these have produced heterogenous data and to date, no meta-analysis has been published on the association of mental disorders and parasitic diseases in developing and emerging countries; hence the reason why we decided to perform this study.

The purpose of this study was to estimate the pooled prevalence and pooled association measures of comorbidities between mental disorders and neurotropic parasitic diseases in developing and emerging countries.

To perform this meta-analysis, the diseases of interest are neutropic parasitic diseases: malaria, cysticercosis, toxoplasmosis, HAT, Chagas disease, and human toxocariasis [9, 13]. These are forms of parasitosis that have a predilection for infesting the central nervous system and which can result in neurological disorders. Mental disorders of interest were the same as those investigated in the previous meta-analysis [35]. These disorders interfere with thinking, feeling, mood, communication and daily functioning, which typically lead to a reduction in the ability to perform common daily activities, such as caring for family or working [4]. These were: anxiety, depression, bipolar disorder and schizophrenia [5, 11].

Since this meta-analysis is the continuation of a previously published study on the association of mental disorders and chronic physical diseases in developing and emerging countries, its shares the same research strategy, inclusion criteria and selection of articles, data extraction, article quality assessment and statistical analysis as our previous meta-analysis [35].

The protocol for this meta-analysis was recorded in PROSPERO (N° CRD42017056521) and accessible via the following link: http://​www.​crd.​york.​ac.​uk/​PROSPERO. This meta-analysis follows the recommended methodology for the meta-analysis of observational studies [36] and was performed in accordance with Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines [37].

The search for articles was conducted through four databases: Medline, Embase, Lilacs, and IENT (database of the Institute of Epidemiology and Tropical Neurology of the University of Limoges in France: http://​www-ient.​unilim.​fr/​) by LOD, the principal investigator, from February to May 2017 without linguistic or date restrictions.

The same research equation built on Medline below and used for the previous meta-analysis [35] was used in the other databases to search for articles in each of the 139 countries studied:

“(“Depressive Disorder”[Mesh] OR “Depression” [Mesh] OR “Anxiety Disorders”[Mesh] OR “Anxiety” [Mesh] OR “Bipolar Disorder”[Mesh] OR “Schizophrenia”[Mesh]) AND (“Diabetes metillus”[Mesh] OR “Obesity”[Mesh] OR “Neoplasms”[Mesh] OR “Cardiovascular Diseases”[Mesh] OR “Pulmonary Disease, Chronic Obstructive”[Mesh] OR “Malaria”[Mesh] OR “Cysticercosis”[Mesh] OR “Toxoplasmosis”[Mesh] OR “Toxocariasis”[Mesh] OR “Trypanosomiasis”[Mesh] OR “Chagas Disease”[Mesh]) AND (“Name of a country”[Mesh]).” [35]

When searching for articles in the IENT database, which is specifically dedicated to work on neurotrophic parasitosis in the countries of interest of this study, only the free text terms “comorbidity” or “mental health comorbidity” were used. Finally, the registration and selection of articles was done through the Zotéro software.

Every article included in this meta-analysis had to meet the same criteria as the previous meta-analysis, which were: “ be an original article whose full text was available; be a cross-sectional or analytical study; have been conducted on adult patients, both males and females, and on all age groups (age≥15 years); be a study involving either only hospitalised subjects or only non-hospitalised subjects, but not both hospitalised and non-hospitalised subjects at the same time; specify the method of disease diagnosis. For cross-sectional studies, it had to give the prevalence, or the data from which it could be calculated, and for analytical studies, it had to give the association measures or the data from which they could be calculated. “ [35].

As in the previous meta-analysis [35], the revised Downs and Black evaluation grid was used independently [38, 39] by two researchers (LOD and PEB) to assess the quality of the studies included in this meta-analysis and the data were extracted for each article by LOD, the principal investigator. The statistical tests were performed with a significance threshold of 5% using Comprehensive Meta-Analysis (CMA) Version 3.0 [40]. Q-test and I² [41, 42] were performed to assess the heterogeneity of the studies included in our pooled estimates. The DerSimonian-Laird random effects technique [43] was then used to calculate the pooled estimates and the results obtained were presented in forest plot. In order to investigate publication bias, we constructed a funnel plot and performed a Duval and Tweedie adjustment and filling test [43], and, Egger regression [44]. The robustness of our results was tested using the sensitivity test, which consisted of subtracting the study with the highest weight among the studies included in a pooled estimate and subtracting the lowest quality studies among the studies from a pooled estimate. Finally, we conducted subgroup analyses for the variables: original disease, associated disease, type of subject and continent.

We focused on developing and emerging countries studied due to their high burden in the selected diseases. For mental disorders, these were anxiety, depression, bipolar disorder, and schizophrenia [5, 11] and for parasitic diseases with neurological tropism, these were malaria, cysticercosis, toxoplasmosis, HAT, Chagas disease, and human toxocariasis [9, 13].

In general, the diagnosis of diseases in the studies selected in this meta-analysis was performed using acceptable diagnostic techniques. Indeed, the quality of diagnostic techniques used to screen for mental disorders has been developed in accordance with DSM-5 and ICD-10 [63]. Nevertheless, it has been shown, as in the study of Kirkil et al., that the questionnaire used can also influence the results in terms of diagnosis [64]. The variety of questionnaires used in the studies included in this meta-analysis can be explained by the need to adapt the questionnaire to the population studied [65‐73], as is usually recommended. Regarding laboratory diagnostic techniques, it is recommended to use two complementary diagnostic tests (one very sensitive and therefore very specific) and to have the samples handed by two different technicians who do not know the clinical condition of the subject [74]. However, most of the studies included in this meta-analysis generally use only one fairly sensitive and specific diagnostic method at a time, which is not the best practice. This can be partly explained by financial constraints related to the cost of diagnostic tests, which are often quite expensive in low-income countries, and by the nature of the sample size, which increases the burden of work. This may be acceptable for this type of disease, since parasitic diseases are very commonly found in these countries.

The quality scores of the studies included in this meta-analysis are in line with those found by other authors [75]. There was publication bias and great heterogeneity in our estimates. In both the prevalence studies and the analytical studies, the quality scores were all largely above the average score (i.e. above half the maximum score: 11 for prevalence studies and 12.5 for case-control studies). But since our estimates did not change after we performed the sensitivity analysis. In our pooled estimates after withdrawing the study with the highest weight on the one hand, and the study with the lowest quality score on the other, using the two recommended methods, the sensitivity test had almost no impact on our results and the results remained robust. However, the different levels of heterogeneity found in the pooled estimates after analysis of the selected subgroups (type of original disease, type of associated disease, type of subjects included, and continent) implied that there were other covariates in play and that the latter could be the source of these heterogeneities.

This meta-analysis revealed that, despite the small number of studies included, the prevalence of anxiety and/or depression was almost 50% in people with Chagas disease and/or neurocysticercosis. “In addition, toxocariasis and/or toxoplasmosis was associated with an increased risk of schizophrenia and/or bipolar disorders (odds ratio = 2.3). More specifically, through subgroup analysis, we were able to show that toxocariasis (odds ratio = 2.7) and toxocariasis and/or toxoplasmosis (odds ratio = 2.4) were associated with increased risk of schizophrenia. In hospitalised subjects, the results of the subgroup analysis showed that in the presence toxocariasis and/or toxoplasmosis, the increased risk of mental disorders (schizophrenia and/or bipolar disorders) was 130%, and, in non-hospitalised subjects the increased risk of mental disorders (schizophrenia and/or bipolar disorders) in presence of toxoplasmosis was 150%. This similarity of results in the two types of subjects might reflect the difficulties of access to mental and neurological health care in developing and emerging countries, with mental and neurological diseases being grossly under-diagnosed, and suffering both from a low ranking in terms of public health priority.

Due to the well-known neurotropic characteristics of Toxoplasma gondii, it is commonly accepted that many psychiatric symptoms, like mental retardation, may be due to Toxoplasma gondii infections [76]. Much interest exists in determining a causative relationship between this parasite and some mental disorders, in particular schizophrenia [77] and bipolar disorders [78]. Our investigation into the presence of IgG antibodies in people with mental disorders produced results that are similar to those obtained by other authors. People living with schizophrenia were found to have a risk ratio of 1.43 to 2.73 for toxoplasmosis and/or toxocariasis [77, 79], while people with bipolar disorders were found to have a risk ratio of 1.26 to 1.52 [78, 79].

The analytical studies on the associations of mental disorders with neurotropic parasitic diseases didn’t allow some statistical analyses, such as subgroup analysis. Only Asia, which accounted for most of the studies, made it possible to estimate that, in presence of toxoplasmosis and/or toxocariasis, there was a 130% increased risk of the onset of schizophrenia and/or bipolar disorders. The lack of data on mental disorders in developing and emerging countries for other continents could potentially be explained by the lack of medical consultations for people with mental disorders, and the lack of knowledge and skills regarding mental illness among primary healthcare professionals; it could also be due to the insufficient number of appropriate health centres, which, when they exist, do not seem to be accessible or have staff trained to manage mental disorders and neurotropic parasitic diseases. Religion and traditional beliefs among communities can also be additional barriers to the diagnosis of mental disorders in these countries, with these diseases often getting attributed to spiritual causes. It has been reported that up to 80% of people with mental disorders and their families might prefer to seek care from religious leaders, traditional healers or exorcist-priests [80]. Finally, poverty and the lack of national and international funding for mental health often aggravate the situation. As a result, a high proportion of people with severe mental disorders (76 to 85%) do not receive treatment in developing and emerging countries [81].

Our results suggest that further studies on the co-morbidities of mental disorders and neurotropic parasitic diseases in developing and emerging countries may be needed to highlight the true estimations of these often-neglected diseases, especially in these countries.

Despite our efforts to limit bias, our study has some limitations that were already described in our previous meta-analysis [35] which shared the same method.

The limitations of this work are related to the small sizes of the different study samples included in our pooled estimates. The fact that the identification of mental disorders was not always confirmed by a specialist in each of the studies is also a limitation. Asia, which accounted for many of the studies, could also well be an obstacle in the generalization of our results to other continents. Finally, the lack of studies meeting all inclusion criteria could be considered as a limitation in the calculation of some estimates for the co-morbidities studied.

Nevertheless, these criteria have enabled us to provide reliable and generalizable pooled estimates for these co-morbidities. This meta-analysis was actually carried out according to the PRISMA 2015 guidelines and the recommendation for conducting sensitivity analyses by subtracting the study with the highest weight and the studies with the lowest quality among the studies included in each pooled estimate.

In this meta-analysis, the results show that the pooled estimates of co-morbidities between mental disorders and neurotropic parasitic diseases in developing and emerging countries are relatively high. Most of the included studies were conducted in Asia. In conclusion, our findings support the hypothesis that toxoplasmosis could be the cause of schizophrenia and/or bipolar disorders. We hope that this meta-analysis, the first of its kind to focus on developing and emerging countries, can provide some guidance to researchers to further explore and understand the associations between mental disorders and neurotropic parasitic diesases in developing and emerging countries.