Background

Human ascariasis is one of the most important neglected tropical diseases (NTDs) worldwide [1]. It is caused by the intestinal nematode of the genus Ascaris—a soil-transmitted helminth (STH) [2]. Transmission occurs as a consequence of the accidental ingestion of embryonated eggs from contaminated soil, food and/or water. Ascaris eggs are resistant and have the potential to survive for long periods of time, particularly under warm and moist conditions [3]. Ascariasis is prevalent and has been estimated to affect ~ 819 million people [4]. The intensity of Ascaris infection is highest in children of 5 to 15 years of age, and has an over-dispersed or aggregated distribution, with most individuals harbouring light infections, and a relatively small proportion of the population harbouring heavy infection [5]. Furthermore, there is consistency in the pattern of re-infection or predisposition in humans [6].

Ascariasis can have acute and chronic manifestations, the latter being associated with significant nutritional and growth deficits [7]. Moreover, there is an association between ascariasis and impaired cognitive development, but the mechanism/s is/are not yet well understood [8]. Ascaris undergoes a larval migration via the liver and the lungs before establishing in the small intestine as a dioecious adult stage. This hepatopulmonary migration has significant, but underestimated pathological and public health impact, such as hepatic and pulmonary disorders [9]. Acute complications caused by adult worms might include intestinal impaction and/or obstruction of biliary and/or pancreatic ducts [7]. Despite the apparent rarity of acute ascariasis, such complications might have a high fatality rate [10].

The commonest method used for the diagnosis for Ascaris infection is the microscopic detection of eggs in faecal samples [11]. Counting adult worms upon expulsion following treatment is regarded as a “gold standard” for estimating the intensity of infection [11], but this approach is very rarely performed and only usually in the context of research projects, rather than routine parasite surveys or monitoring during large-scale deworming programmes.

Key epidemiological measures used to determine the extent of Ascaris infection are the prevalence of infection (% of persons infected in a particular population) and the intensity of infection. Intensity can be expressed as the arithmetic mean number of worms (“worm burden”), of eggs per gram of faeces (EPG), geometric mean EPG, or median EPG. The central importance of knowledge of prevalence and intensity of infection to the understanding of the epidemiology and impact of macroparasites in humans and other animals cannot be underestimated [5]. However, the number of routine surveys and monitoring programmes that utilise a measure of intensity remains low. In 2000, World Health Organization (WHO) recommended EPG thresholds (“cut-offs”) that can be used to classify infection intensity as low, moderate and high EPG [12].

Over the last few decades, large-scale deworming programmes have been implemented to “control” STHs in endemic regions. Despite variations in treatment efficacy among STHs, albendazole and mebendazole appear to remain efficacious anthelmintics against human ascariasis [13]. However, anthelmintic treatment alone will not lead to a marked reduction or elimination of ascariasis without accompanying improvements in socioeconomic conditions and the provision of clean water, improved sanitation and hygiene [14]. Thus, WHO set new targets that included the elimination of morbidity due to STHs (defined as the prevalence of moderate and heavy infection intensities of < 2%) in preschool- and school-age children by 2030 and universal access to at least basic sanitation and hygiene by 2030 in STH-endemic areas [15].

Pigs infected with Ascaris can represent a reservoir for human infection [16, 17], despite controversies about the species status of Ascaris of humans and pigs [16, 18, 19]. Part of the motivation to undertake the present review was the need to critically re-evaluate the extent of ascariasis in humans in endemic regions, since the last estimate in 2010, published in 2014 [4]. Here, we report a comprehensive systematic review and meta-analysis to estimate the prevalence and intensity of Ascaris infection in the general human population in endemic regions from 1 January 2010 to 1 January 2022. We also evaluated the impact of geographical, climatic and socio-demographic factors on the prevalence of Ascaris infection in different countries and regions.

Methods

Search strategy and selection criteria

This meta-analysis study was designed and conducted in accordance with the Cochrane Handbook of Systematic Reviews [20], and is reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [21] (Additional file 5: Checklist S1). Five investigators (C.H., H.B., A.A., G.D. and A.R.) began the search of the relevant studies on April 2020, through MEDLINE/PubMed, and Scopus, without language restriction. An updated search was performed on February 2022. Searches were limited to studies published between 1 January 2010 and 1 January 2022. The search terms used included: “ascariasis”, “Ascaris”, “Ascaris lumbricoides”, “intestinal parasites”, intestinal helminths”, “soil transmitted helminths”, “prevalence”, “incidence”, “epidemiology” and “occurrence”. These keywords were combined using Boolean operators ‘OR’ and/or ‘AND’ in database searches (Additional file 4: Fig. S1). To identify additional studies, the same investigators independently searched the Google Scholar engine, and scrunitised the reference lists of eligible studies. All citations retrieved were imported to Endnote software X8 (Thompson and Reuters, Philadelphia, USA), and 10,185 duplicates and irrelevant papers were deleted, leaving 1060 citations. Four independent investigators (S.M., S.V., A.A. and A.R.) reviewed the titles, abstracts, and full texts of articles for eligibility. The online tool “Google Translate” (https://translate.google.com/) was used to assist the translation of articles published in languages other than English.

Selection criteria

Studies were considered eligible for inclusion if they were peer-reviewed, observational studies which reported the prevalence of Ascaris infection. Two trained researchers applied inclusion criteria: a sample size of at least 50; published after 1 January 2010; the tested population should be representative of the general population (i.e. randomly-selected people of different ages, socioeconomic status and ethnic backgrounds, without occupations or specific diseases increasing the probability of acquiring ascariasis); and studies that used an internationally accepted coprodiagnostic method with suitable performance characteristics (e.g., Kato-Katz, formalin-ether, McMaster, FLOTAC or Mini-FLOTAC techniques). In order to reduce bias in the estimation of ascariasis prevalence in the general population, studies of the following population groups were excluded: patients with gastrointestinal disorders, patients with diarrhoea, HIV+ patients, patients receiving corticosteroid treatment, patients with any immunodeficiency, haemodialysis patients, patients with tuberculosis, inmates in prisons, pregnant women, mentally retarded patients, immigrants, patients with atopy and other allergies, workers exposed to wastes, wastewater and faecal sludge. In addition, studies were excluded if they were: (1) comparing diagnostic methods; (2) serological investigations; (3) reporting an intervention approach, without the availability of baseline (reference) data; (4) used datasets that overlapped with those of other articles; (5) recorded prevalence after mass drug administration; (6) were performed in non-endemic areas, such as European and North American countries; (7) had a sample size of ≤ 50 individuals; (8) were case reports or series; or (9) were letters, commentaries, reviews or systematic reviews without original data.

Extraction and quality evaluation of data

A data collection form was developed in Microsoft Excel (version 2016; Microsoft Corporation, Redmond, USA); data were extracted independently by five investigators (H.B., S.V., G.D., S.G., S.M. and A.A.), and unanimity was reached on discrepancies after discussion, and also consultation with the senior investigators (C.H. and A.R.). Data were extracted for the following domains: (I) study characteristics (the first author's last name, year of publication, diagnostic methods, country; city, ‘sustainable development goal’ (SDG)-regions or sub-regions [22]); (II) participant characteristics [type of population (children, adult or both), the number of participants, the number of people that tested positive for Ascaris infection, age, sex, residence and intensity of infection (low, moderate, high)]; (III) socio-economic variables (World Bank–income category [23], gross national income per capita [23] and the human development index [HDI] [24]); and (IV) geo-climatic factors (latitude, longitude, mean relative humidity, mean annual precipitation and mean environmental temperature). The data sources for geo-climatic factors were: https://gps-coordinates.org/, https://www.timeanddate.com/ and https://en.climate-data.org/. Furthermore, we recorded total global, regional populations (both sexes) in 2020, estimated by the United Nations [25]. For interventional studies, we extracted only baseline data, and for case-control studies, we only extracted data for healthy people. The data on intensity of Ascaris infection were extracted from studies that assessed and graded intensity according to the criteria defined by WHO [12]. The investigators independently assessed the risk of bias for each study using the Joanna Briggs Institute (JBI) Critical Appraisal Tools for cross-sectional, case-control, cohort and randomized controlled studies, and each study was categorized to have a low, moderate or high risk of bias, as recommended for these tools [26,27,28].

Meta-analysis

All statistical analyses were conducted using Stata statistical software (v.13 Stata Corp., College Station, TX, USA). The prevalence of Ascaris infection in each study was calculated by dividing the number of test-positive cases by the study population. To estimate the pooled prevalence of Ascaris infection, we used a DerSimonian and Laird random-effects model (REM) [29]. REM is capable of incorporating proportions close to, or at the margins (i.e., with a very low or very high prevalence), into conservative pooled prevalence estimates and 95% confidence intervals (CIs). We calculated the pooled prevalence rates at a 95% CI using the ‘metaprop’ command in Stata software. The pooled prevalence in each country was estimated by synthesizing the prevalence rates of all studies from the same country. We stratified estimates into the six endemic SDG regions and the 14 SDG sub-regions (Table 1). We estimated heterogeneity using the χ2 test with Cochran’s Q statistic and quantified with I2; I2 of > 75% was considered to reflect substantial heterogeneity [30]. To calculate the number of people infected with Ascaris, we extrapolated prevalence estimates to the total human population (in 2020) living in a country and/or a region—according to the UN Population Division [25]. Data were entered into ArcGIS 10.2 (ESRI, Redlands, CA, US) to produce maps presenting Ascaris infection prevalence estimates on a country level.

Table 1 Global and regional prevalence of Ascaris infection, and estimated numbers of infected people (results from 758 datasets performed in 81 countries)
Full size table

We undertook several subgroup, as well as univariate and multivariate meta-regression, analyses to explore possible sources of heterogeneity and effects of socio-economic, study characteristics, geo-climatic parameters on the prevalence of Ascaris infection in people. These analyses were performed using the ‘metareg’ command in Stata [31] in relation to study participants (children, adults and total population); diagnostic method used; age, sex and residence; country income level (low, lower-middle, upper-middle or high) and country HDI (low, medium, high or very high); risk of bias levels for each study (low, moderate or high); geographical latitude and longitude; mean annual environmental temperature, relative humidity and precipitation; year of publication; and the start and end dates of sampling. Results were considered as statistically significant if the P value was < 0.1.

Results

Study characteristics

Our search of electronic databases identified a total of 11,245 articles; following the removal of duplicate articles and a critical appraisal of article titles and abstracts, 1060 potentially relevant articles were identified for full-text evaluation (Fig. 1). After applying the eligibility criteria, 616 articles containing 758 datasets were included in the quantitative synthesis (Fig. 1); these datasets represented 4,923,876 people from 81 countries in six SDG regions and 14 SDG sub-regions. Of these datasets, 353 were from sub-Saharan Africa, 162 from Eastern & South-eastern Asia, 110 from Latin America and the Caribbean, 92 from Central and Southern Asia, 37 from Northern Africa and Western Asia, and four from Oceania. Considering SDG sub-regions; Eastern Africa (n = 184), Western Africa (n = 117), South-Eastern Asia (n = 97), Southern Asia (n = 89), and South America (n = 86) had highest eligible datasets, while Northern Africa (n = 7), Southern Africa (n = 6), Caribbean region (n = 4), Oceania (n = 4) and Central Asia (n = 3) had the lowest eligible datasets. The main characteristics of the studies included are provided in Additional file 1: Table S1. In this study, it should be noted that only populations in endemic areas were considered in the analyses; populations in North America, Europe and Australia & New Zealand were not considered.

Fig. 1
figure 1

The literature search strategy and study-selection process, indicating numbers of studies excluded and included

Full size image

Global and regional prevalence of Ascaris infection

Overall, 164,879 people from a general population of 4,923,876 tested positive for Ascaris infection, indicating a pooled prevalence of 11.01% (95% CI: 10.27–11.78%). Significant heterogeneity (I2 = 99.8%, P < 0.001) was found between studies. The highest pooled prevalence estimates were for Central & Southern Asia (12.91%, 10.01–16.13%), Latin America & the Caribbean (12.75%, 10.75–14.88%) and sub-Saharan Africa (11.66%, 10.56–12.81%), while the lowest prevalence was from Northern Africa & Western Asia (2.04%, 1.47–2.70%) (Table 1). Considering the SDG sub-regions, the highest prevalence rates were: 28.77% (7.07–57.66%) in Melanesia (Oceania), 19.06% (15.51–22.87%) in South-Eastern Asia, 18.83% (13.04–25.42%) in Central America and 15.31% (11.39–19.69%) in Middle Africa, and the lowest prevalences were 1.39% (1.07–1.74%) in Eastern Asia, 1.71% (1.14–2.38%) in Western Asia, and 4.01% (0.85–9.20%) in Northern Africa (Table 1). Countries (for which ≥ 3 eligible studies were available) with the highest prevalence rates were Chad (48.19%), Liberia (42.45%), Ecuador (39.58%), Madagascar (37.16%), the Philippines (34.69%), Solomon Islands (33.96%), Indonesia (32.02%), Rwanda (31.22%) and Sri Lanka (30.81%) (Table 2). Figure 2 shows the Ascaris infection prevalence estimates for individual countries.

Table 2 Prevalence estimates of Ascaris infection, and estimated numbers of infected people in 81 countries
Full size table
Fig. 2
figure 2

Estimated Ascaris infection prevalence rates in the general human population in different countries using the geographic information system (GIS)

Full size image

An extrapolation to the 2020 world population indicated that ~ 731.8 million (range: 682,619,992 to 782,985,736) people harboured Ascaris infection. The sub-regions with the highest burden of Ascaris infection were Southern Asia (~ 254.1 million people), South-eastern Asia (~ 127.4 million people), South America (~ 49.8 million people), Eastern Africa (~ 49.5 million people) and Western Africa (~ 44.4 million people). More detail on the global and regional Ascaris infection prevalences and burdens are given in Table 1.

Intensity of Ascaris infection

In total, 137 of 758 datasets included information on the intensity of Ascaris infection (Additional file 2: Table S2). According to WHO guidelines, we appraised these data sets and estimated the prevalence of low, moderate and high intensities among infected people in each SDG region (for which ≥ 3 eligible studies were available). Our analyses showed that the highest prevalence of high intensity of Ascaris infection among infected people was found in Latin America and the Caribbean regions (8.35%, 3.88–14.11%), followed by Eastern and South-Eastern Asia (3.80%, 1.07–7.59%), Central and Southern Asia (0.62%, 0.01–0.63%), and Sub-Saharan Africa (0.26%, 0.01–0.97%). Our findings indicated that the prevalence of high intensity infection was lower in 2018–2021 (1.43%, 0.44–2.79%) compared with 2010–2013 (3.32%, 1.11–6.31%). Detail on low and moderate infection intensities are presented in Table 3.

Table 3 Intensity of Ascaris infection based upon numbers of infected people with cut-offs for low, medium or high intensity infection
Full size table

Prevalence of Ascaris infection according to sex, age, residence and population

According to the type of population studied, 494 datasets related to children, 256 to both children and adults (all age groups) and eight studies to adults. In subgroup analysis, the global Ascaris infection was 13.50% (12.39–14.65%) in studies of children, 6.89% (6.19–7.63%) in studies of all age groups and 8.85% (1.53–20.25%) in studies recruiting only adults (Table 4).

Table 4 Prevalence estimates of Ascaris infection, according to a priori-defined subgroups and socio-demographic geographic parameters
Full size table

Of the 616 studies included, 138 reported separate, pooled prevalence rates for males and females. Subgroup analysis revealed pooled prevalence of 14.67% (12.61–16.86%) in males and 14.61 (12.64–16.68%) in females. A total of 71 studies reported pooled prevalence rates for different age specific groups; a subgroup analysis revealed the lowest and highest prevalence rates of 14.14% (11.67–16.80%) and 25.82% (23.59–28.12%) in people of 12–18 and ≥ 51 years of age, respectively (Table 4). In addition, 143 studies reported prevalence rates for people in urban or rural regions; a subgroup analysis revealed that people living in rural areas had a higher prevalence (17.33%, 14.21–20.69%) of Ascaris infection than those living in urban areas (10.02%, 7.29–13.12%) (Table 4).

Prevalence of Ascaris infection in relation to diagnostic methods used and risk of bias

Various diagnostic methods were used in the studies included. We stratified these diagnostic methods into three categories: (1) the Kato-Katz method for 368 datasets (2) direct wet mount, formalin-ether and/or other routinely used sedimentation and/or concentration methods for 353 datasets; (3) other techniques, including McMaster, Lumbreras rapid sedimentation, Lutz sedimentation, flotation methods (FLOTAC and Mini-FLOTAC), Ritchie, Stoll’s dilution egg count and/or polymerase chain reaction (PCR) for 37 datasets. If more than one diagnostic test was used in a study and total positivity was recorded, we utilised this value; however, if this was not the case, we selected data relating to the diagnostic method that recorded the higher prevalence. Subgroup analysis revealed pooled prevalence rates of 12.68% (11.40–14.01%), 9.22% (8.45–10.01%), and 11.90% (7.29–17.43%) using Kato-Katz, routine parasitological methods, and other non-routine diagnostic methods, respectively (Table 4).

A critical appraisal using JBI tools showed that 378 datasets had a low risk of bias (score: 7–9/9), 269 datasets had a moderate (4–6/9), and 111 studies had a high risk of bias (≤ 3/9). Moreover, the prevalence for studies with a low, moderate and high risks of bias were 7.81% (6.94–8.73%), 14.28% (12.68–15.96%), and 16.23% (12.77–20.02%), respectively (Table 4).

Prevalence of Ascaris infection in relation to socio-demographic variables

For the 758 datasets included, 153, 385, 178 and 42 datasets represented countries with low, lower middle, upper middle and high income levels, respectively. Subgroup analysis (Table 4), according to income level, revealed the highest prevalence rates of Ascaris infection in countries with low (13.23%, 11.64–14.91%) and lower middle (12.17%, 10.89–13.52%) income levels, with the lowest prevalence estimated for high income countries (0.88%, 0.33–1.64%). On the other hand, totals of 273, 210, 190 and 85 datasets represented countries with low, medium, high and very high HDI levels. Subgroup analysis according to HDI level, indicated that the highest and lowest prevalence rates were estimated for countries with low (12.31%, 11.01–13.67%) and very high (5.01%, 3.90–6.25%) HDI levels (Table 4). Random-effects meta-regression analyses showed a significant, decreasing trend in prevalence with increasing income (coefficient [C] = − 4.37e−06; P-value = 0.0001) and HDI (C = − 0.145; P-value = 0.0001) levels (Additional file 4: Fig. S2A, B).

Prevalence of Ascaris infection over time

Another subgroup analysis was conducted to explore the prevalence of Ascaris infection over time. For this analysis, we stratified time into three categories: year of publication; year of the start of sampling; and year of the end of sampling. Subgroup analysis based on publication year showed prevalence rates of 9.18% (7.87–10.58%), 12.61% (11.15–14.15%), and 10.73% (9.57–11.95%) for studies published 2010–2013, 2014–2017 and 2018–2021, respectively (Table 4). Subgroup analyses according to the beginning and end dates of sampling, revealed that studies conducted between 2010 and 2015 reported the highest prevalence rates. Random-effects meta-regression analysis showed a non-significant, increasing trend in prevalence rates over time (C = 0.001; P-value = 0.4; Additional file 4: Fig. S3). Considering the start and end dates of sampling, subgroup analyses showed that studies between 2010 and 2015 reported higher prevalence rates than those performed before 2010 or between 2016 and 2021 (Table 4).

Relationship between the prevalence of Ascaris infection and geographical location/climate

According to geographical parameters, the highest and lowest prevalence rates were at latitudes 0–20° (13.41%, 12.34–14.51%), and 40–60° (0.53%, 0.17–1.04%); as well as longitudes at 60–80° (16.72%, 14.23–19.36%) and 40–60° (4.55%, 3.54–5.69%), respectively (Additional file 3: Table S3). Meta-regression analyses showed significant decreasing trends in prevalence with increasing geographical latitude (C = − 0.003, P-value < 0.001), while a non-significant increasing trend was observed with increasing geographical longitudes (C = 0.0001, P-value = 0.07) (Additional file 4: Fig. S4A, B).

Considering climate parameters, the highest prevalences were in regions with mean relative humidity of ≥ 80 (i.e. 18.60%, 15.54–21.86%), mean annual precipitation of > 200 mm (15.26%, 11.04–20.01%), and a mean annual temperature of 25–29 °C (14.81%, 13.24–16.46%), while the lowest prevalences were estimated for regions with a mean relative humidity of ≤ 40 (5.61%, 4.48–6.85%), a mean annual precipitation of 0–50 mm (4.72%, 4.02–5.48%), and a mean annual temperature of 9–13 °C (0.92%, 0.51–1.43%) (Additional file 3: Table S3). There was a significant, increasing prevalence trend with increasing relative humidity (C = 0.002; P-value < 0.001), precipitation rate (C = 0.000; P-value < 0.001) and environmental temperature (C = − 0.0006; P-value < 0.001) (Additional file 4: Fig. S5A–C).

Univariate and mulitivariate meta-regression analyses to identify source of heterogeneity

Additional file 3: Table S4 shows the results of the univariate and multivariate meta-regression analysis of study characteristics exploring the source of heterogeneity of the prevalence estimates Ascaris infection. As depicted in Additional file 3: Table S4, univariate analysis revealed that all covariates were significantly associated with heterogeneity of prevalence estimates, except for longitude (β, 0.05; 95% CI: − 0.00008–0.0005, P-value = 0.139), implementation year start (β, 0.05; 95% CI: − 0.001–0.004, P-value = 0.280), and implementation year start (β, 0.05; 95% CI: 0.03–0.06, P-value = 0.709). In the final multivariate meta-regression, the following variables including income level (β, 0.05; 95% CI: − 0.00007 to − 0.00002, P-value < 0.001), geographical longitude (β, 0.05; 95% CI: 0.0006–0.001, P-value < 0.001), temperature (β, 0.05; 95% CI: 0.001–0.006, P-value = 0.001), humidity (β, 0.05; 95% CI: 0.0003–0.002, P-value = 0.013), and; year of the start of sampling (β, 0.05; 95% CI: − 0.014 to − 0.0007, P-value = 0.030) remained significantly associated with heterogeneity of the prevalence of Ascaris infection.

Discussion

Ascariasis continues to be an NTD of major public health significance worldwide, causing substantial morbidity in endemic regions [7]. Here, we performed a global systematic review and a meta-analysis of published studies of Ascaris infection in endemic regions. Our findings indicate that, globally, 11% of the ~ 6.6 billion people living in endemic regions, representing the ~ 732 million people, harbour Ascaris. When compared with the estimate by Pullan et al. [4], derived from data collected up to 2010, our estimates indicate a 3.5% reduction in prevalence (14.5% vs 11.0%) and a 10% reduction in the number of people with Ascaris infection (819 vs 730 million infected people). These findings are consistent with other recent studies [4, 32,33,34], reporting a reduction in the prevalence of STH infections globally. Pullan et al. [4] reported a substantial reduction in the prevalence of Ascaris infection in all endemic regions between 1990 (~ 32%) and 2010 (14.5%). Our findings from subgroup analyses also indicated a lower prevalence between 2016 and 2021 than between 2010 and 2015, although meta-regression analyses showed a non-significant, increasing trend over time. The likely explanation for this reduction in the prevalence of Ascaris infection in endemic regions is mass drug administration (MDA) within the context of control programmes [35].

Our results indicated that the prevalence of Ascaris infection varies among SDG regions and sub-regions, with the lowest prevalences (< 4%) in countries in Eastern Asia (China and the Republic of Korea) and Western Asia (Iraq, Iran, Palestine, Turkey and the United Arab Emirates) and the highest prevalences (> 20%) in some countries in Oceania (Papua New Guinea and the Solomon Islands), South-eastern Asia (Indonesia, Laos PDR, Malaysia, Myanmar, the Philippines, Sri Lanka and Vietnam), the Latin and Caribbean region (Brazil, Colombia, Ecuador, Honduras and Nicaragua), sub-Saharan Africa (including Angola, Burundi, Central African Republic, Chad, DR Congo and Madagascar) and South Asia (Afghanistan, Bangladesh, India and Pakistan). Lower prevalences in countries such as the Republic of Korea, China, Iran and Turkey are likely due to socio-economic improvements that have occurred over the past three decades [4, 36]. In tandem with community-based MDA, substantial improvements in personal and public hygiene (including the availability of toilets; hand washing before eating and after defaecation; consumption of washed vegetables and filtered water) have been achieved [37,38,39]. Improved water quality, sanitation and hygiene have been shown to be significantly associated with a lower prevalence of STHs, particularly Ascaris [39]. In contrast, a higher prevalence of Ascaris infection in South Asia, sub-Saharan Africa and Latin America might reflect less improvement in these measures. Furthermore, our results showed that the prevalence of Ascaris infection is higher in countries with lower levels of income and HDI.

Another explanation for the variation in prevalence of Ascaris infection in SDG regions is differences in environmental factors, such as temperature, humidity, rainfall, soil moisture, and contamination [40]. Environmental factors have an important impact on the embryonation of Ascaris eggs; consequently, egg development, viability and infectivity depend on a particular environmental temperature (≥ 25 °C) and humidity (≥ 55%) [41]. Some laboratory studies have demonstrated that higher temperature and humidity facilitates larval development, and other investigations have indicated that extreme humidity, desiccation and/or temperature can lead to compromised embryonation and larval development within eggs [41,42,43]. The present findings showed an increased prevalence of Ascaris infection in areas with higher environmental temperature, rainfall and humidity, such as in countries within South-East and Southern Asia, SubSaharan Africa and Latin America.

Another factor that may influence the observed variation in prevalence could be differential specificity and sensitivity of the diagnostic methods employed in individual studies. Diagnostic methods for the detection of Ascaris eggs include: direct microscopy of faecal smears, the Kato-Katz [44], McMaster [45], formol-ether concentration [46], FLOTAC [47] and mini-FLOTAC [48] techniques. The present investigation showed that 48.5% of studies included used the Kato-Katz method, as recommended by WHO. However, this latter method performs poorly when egg numbers in faecal samples are low [7, 49]. The remainder of the studies used a range of established methods. Variation in the standardisation and performance of diagnostic methods undoubtedly contributes to variability in estimates of prevalence and/or intensity of Ascaris infection. Molecular methods are now also being used or evaluated, including PCR-based detection of Ascaris DNA in faecal samples [50]. Some such approaches utilise a multiplex or a multi-parallel approach to detect and/or distinguish different parasite species [51, 52]. However, such methods are not yet in routine use around the world, as they require specialised equipment and trained personnel. Medley et al. [53] discussed some concerns about some coprodiagnostic methods employed for STHs, including Ascaris.

Despite the large number of studies reporting the global prevalence of Ascaris in humans, a major knowledge gap remains in relation to infection intensity. Only 19% of the datasets used in this systematic review contain information on infection intensity. “Intensity” was reported in a number of ways, but, most commonly, the proportion of individuals recorded to have low, moderate or high intensity was based upon EPG “cut-off”, in accord with 2002 WHO guidelines [12]. Without knowledge of infection intensity, it is challenging to assess the public health impact of a parasite such as Ascaris [1, 5]. The paucity of data on infection intensity is particularly problematic when (i) the prevalence is low; (ii) MDA is no longer implemented; and (iii) there is a need to identify high risk individuals who remain predisposed to heavy infection [6]. In tandem with the observed reduction in prevalence over time, we also observed a substantial reduction in the proportion of individuals with high intensity infection over time. The prevalence of moderate or high intensity infections is recommended as a key indicator of the success of whether STHs have been eliminated as a public health problem [54]. Therefore, this reduction of high intensity infections is of major epidemiological importance, and reflects, at least in part, the success of control strategies/programmes. Montresor et al. [55] also raised some issues regarding the implementation and effectiveness of preventative MDA in endemic regions. These authors reported that, of 96 countries endemic for ascariasis, four did not implement preventative MDA, 23 countries implemented preventative MDA without effective coverage, and 42 and 28 countries implemented MDA for < 5 and > 5 years, respectively.

An important issue that was not possible to evaluate in the present study, but should be highlighted, is the potential for cross transmission of Ascaris between pigs and humans (and vice versa). There has been considerable debate, over many years, as to species status of A. lumbricoides (humans) and A. suum (pigs), and whether they represent distinct species or operational taxonomic units that are reproductively isolated or are capable of interbreeding and producing viable offspring [17, 56]. Although their status is controversial, there is clear evidence that cross transmission of Ascaris occurs between pigs and humans living in close proximity, particularly in non-endemic regions [18, 57]. Some studies indicated that 4–7% of Ascaris from people in Guatemala and China were A. lumbricoides × A. suum ‘hybrids’ [19, 58]; other investigations have identified three main haplotype clusters (A, B and C) for Ascaris, employing part of the mitochondrial (mt) cytochrome c oxidase 1 (cox-1) gene (383 bp) as a marker [59,60,61,62]. There is genetic divergence between clusters: worms in clusters A and B (representing humans or pigs) were more closely related to each other, whereas worms in cluster C (from pigs or humans cross-infected with pig Ascaris) were genetically more distinct [16, 59, 60, 62]. It has been shown that haplotypes belonging to clusters A and B are geographically distributed worldwide in both host species (pig or human), but in different proportions [17]; the majority of worms studied in China belonged to cluster B [61], whereas most worms isolated from Uganda belonged to cluster A [59]; however, there was no clear geographical association. Moreover, important new information has recently emerged from a genome analysis of Ascaris worms recovered from human subjects in Kenya where pig husbandry is rare [16]. Most of these worms had mitochondrial genomes that clustered more closely to A. suum than to A. lumbricoides, suggesting that pig-Ascaris infection in people in Kenya might have been acquired previously from infected humans who had lived closely with pigs or non-human primates or another host animal harbouring pig-Ascaris, rather than from pigs [16, 63]. Phylogenetic analyses also revealed evidence to support a highly interbred Ascaris species genetic complex [16]. In many endemic regions, a close association between people and pigs is maintained and, therefore, may be one of the reasons for the persistence of Ascaris infection over time. However, it is not possible using the current coprodiagnostic methods to determine whether Ascaris hybrids do indeed exist.

Easton et al. [16] recommended that a One Health approach be employed for the control of human ascariasis, because pigs can serve as a reservoir for human Ascaris infection, also with potential implications for the spread of anthelmintic resistance. Evidence provided by these authors [16] also signals a need to utilise genome-based approaches to unravel the complexity of the enigmatic and controversial questions regarding the species status of human- and pig-Ascaris and host affiliation(s), and the genetic composition of Ascaris populations in both humans and pigs in different countries around the world.

The present study has a number of strengths. By including 616 studies and results for ~ 5 million people from 81 countries, this is one of the most comprehensive reviews on the epidemiology of Ascaris to date. We only included data for the ‘general population’ and excluded high risk groups, with the intent of minimising prevalence overestimates. However, this may have led to underestimates of Ascaris infection prevalence; therefore, our estimates might not be entirely representative of individuals in all communities. We also pooled data to highlight differences within and between different regions around the world, and assessed the impact of geographical, climatic and socio-economic factors on the prevalence of Ascaris infection.

However, some limitations of the present study are acknowledged. First, we included only studies listed in PubMed and Scopus; therefore we may have missed some studies, such as those published in non-indexed journals. However, we searched Google Scholar to identify additional, potentially relevant studies; nonetheless, given the large number of datasets studied, the inclusion of any potentially (unintentionally) omitted studies would not have significantly altered the overall findings. Second, as mentioned above, in some cases, our estimates might not be representative of national prevalences or of all communities in a country, particularly for those with small numbers of eligible studies. Data were not available for many countries, and there was insufficient data for some important endemic regions, including Central Asia, sub-Saharan Africa and Oceania. Moreover, estimates for countries such as Malaysia and Indonesia might not be truly representative of their whole population, because the majority of studies conducted in these countries focused on high-risk regions or communities. Although, in the present study, we included only studies published after 2010, in a substantial number of studies, sampling dates were before 2010 (see Table 4); therefore, as a result of changes in social and environmental conditions over time, the prevalence estimates from earlier studies may not have been entirely reflective of the infection status at the time. Third, few studies reported data on infection intensity or age/sex-specific prevalences. A fourth limitation is the existence of high heterogeneity recorded across all regions and sub-groupings, which is expected in global prevalence estimates across time periods and locations [64,65,66]; the sources of this heterogeneity are explained in our results.

Conclusions

The present findings on the prevalence and intensity of Ascaris infection should assist health policy makers in designing and supporting ascariasis intervention/control programmes that improve public health and reduce the burden of infection and disease. Our findings indicate that, despite a significant reduction in prevalence and intensity of Ascaris infection over past decades, a substantial portion of the world’s human population (i.e. ~ 732 million) is still infected with Ascaris, calling for enhanced efforts. Furthermore, despite its central epidemiological importance, many routine surveys fail to assess the intensity of Ascaris infection. This study calls for continued global efforts to control and prevent human ascariasis, to work toward achieving key SDGs by reducing the burden of STHs and increasing human health and wellbeing.