Prevalence and risk factors of strongyloidiasis among schoolchildren in Sabach Sanjal and Upper Badibou districts in the North Bank East Region of The Gambia

Background Strongyloidiasis is a parasitic disease that mainly affects humans and is caused by a roundworm called Strongyloides stercoralis. It is endemic in humid tropical regions that include Africa, Latin America and Southern Asia. Among the public health important soil-transmitted helminths (STHs) classified as neglected tropical diseases, S. stercoralis is the most neglected. A study of schistosomiasis and STHs mapping was conducted and S. stercoralis larvae were detected using the utilized diagnostic method; thus, this current study described the prevalence and risk factors of S. stercoralis infection in districts of Sabach Sanjal and Upper Badibou in The Gambia. Methods The cross-sectional study enrolled 851 schoolchildren, ages 7 to 14 years old. The participants were enrolled from 17 schools in Sabach Sanjal and Upper Badibou Districts. The WHO random sampling technique n/50 (25 boys and 25 girls) was used. Stool samples were collected from each participant and Kato-Katz smear method was used to screen for S. stercoralis infection. Results Out of the total 851 pupils, 76 pupils (8.9%) were positive for S. stercoralis infection. The mean age of infected persons was 10.1 years (±2.2). The prevalence of infection was higher among females (9.2%) than males (8.7%). Rates of infection for age categories 7–10 years and 11–14 years were 12.4% and 4.2%, respectively. Rates of infection by districts were 12.3% for Sabach Sanjal and 7.1% for Upper Badibou. Schoolchildren from Sabach Sanjal were 1.6 times more likely to have strongyloidiasis compared to those from Upper Badibou (aOR = 1.64, p-value = 0.058). Schoolchildren aged 7–10 years were 3.2 times more likely to have strongyloidiasis infection compared to the 11–14-year-olds (aOR = 3.20, p-value <0.001). Schoolchildren who ‘sometimes’ have water or tissue after defaecation have more infection rate compared to those who ‘always’ have water or tissue after defaecation. However, this difference was not statistically significant (aOR = 1.36, p-value = 0.308). Conclusion The study revealed the prevalence of strongyloidiasis in Sabach Sanjal and Upper Badibou districts of The Gambia. Kato-Katz technique might be inadequate for detecting S. stercoralis; thus, more studies are needed to determine the true prevalence of the disease in these two districts through the combined use of highly sensitive techniques such as Baermann, Koga Agar Culture and polymerase chain reaction.


Study setting
The study was conducted in Sabach Sanjal and Upper Badibou districts located in the eastern part of the North Bank Region of The Gambia. Sabach Sanjal is located on latitude 13 35 ′ 0" N and longitude 15 26 ′ 0" W while Upper Badibou is on latitude 13 34 ′ 25" N and longitude 15 35 ′ 55" W. Both districts have the river Gambia at their southern part and borders with Senegal at their northern part. Sabach Sanjal is about 125 km away from the capital city, Banjul. The two districts have an estimated population of 92, 359 (https:// www.gbosdata.org /downloads/census-2013-8, n.d.), and their main occupations are farming (crop cultivation, cattle rearing, and fishing) and business.

Study design and population
The survey was a cross-sectional study that recruited schoolchildren. It was conducted as part of the 2015 schistosomiasis and STHs mapping survey. While stool samples were examined for S. mansoni and the three major STHs infections, S. stercoralis was also observed and recorded. The NTD mapping protocol recommended by the World Health Organization was used to determine the sample size. The WHO random sampling technique follows the WHO guide for mapping schistosomiasis in the African region which recommends randomly selecting 50 schoolchildren (25 boys, 25 girls) per school/village. The n = 50 is the total number of children sampled per school/village, and the recommended age group is 7-14 years. Based on most similar studies, age was categorised into two subgroups of 7 to 10 years and 11 to 14 years during the analysis in order to determine which group are at higher risk for strongyloidiasis.
The study was conducted in two districts in the North Bank East Region (NBER) (Fig. 1). Approximately 50 students between the ages of 7 and 14 years (25 boys and 25 girls) were randomly selected from each of 6 schools in Sabach Sanjal and 11 schools in Upper Badibou. In Sabach Sanjal, 51 participants (25 boys and 26 girls) were enrolled in one school, and from the remaining 5 schools, 50 participants were enrolled in each giving a total of 301. In Upper Badibou, 50 participants were enrolled in each of the 11 selected schools giving a total of 550 participants. In total, 851 participants from 17 schools were enrolled in the study conducted between May and June 2015. (See Figs. 2 and 3.)

Inclusion and exclusion criteria
Schoolchildren aged 7 and 14 and residing in the districts were selected for the study. Participants' parents gave consent for their children to be recruited in the study. All participants with complete samples and data were included. Exclusion criteria includes students who were absent or sick during the study period.

Sample collection
Pre-labelled screw-capped plastic stool containers (bearing barcode and date of sample collection), tissue and wooden spatula were given to the participants to produce stool samples. Each participant was escorted to the toilet by a member of the research team and  asked to provide a three-quarter filled container of stool. The samples were obtained between 8:00 am and 12:00 noon.

Laboratory procedures
The stool samples were processed at ambient temperature and transported within 2 h of collection to a regional laboratory at Farafenni District Hospital. Samples were analysed using Kato-Katz method (Katz et al., 1972). For each sample, spatula was used to collect portion of the stool sample and placed on a tissue paper. After this, a screen was used to press on top of the sample. This allows the sample to sieve through the gauze and accumulate above the gauze. The spatula was used to scrape the sieved sample and transfer about 0.0147 g of the sieved sample to a hole of 6 mm on a 1.5 mm thick template placed on a microscope slide. The template was then diligently removed leaving the sample on the slide intact. The pre-soaked cellophane strip was used to cover the faecal material and the slide was inverted and pressed firmly on a smooth hard surface. The slide with the cellophane was placed upwards to allow water to evaporate while glycerol cleared the faeces. The prepared Kato-Katz slides were then examined under a light microscope for S. stercoralis larvae. Two independent trained microscopists with over 10 years of experience examined the slides. The microscopy results were recorded on a laboratory reporting sheet.

Laboratory results
Parents of schoolchildren with positive results were notified, and treatments were prescribed by a doctor. The researchers gave an awareness of talk/health education on preventing strongyloidiasis in schoolchildren.
Field survey data was collected using the Task Force for Global Health (Atlanta, USA) LINKS ® data collection system. This is an android phone-based programme with an in-built questionnaire that helps the various teams to standardise epidemiological data collection. The questions asked were on water and latrine use practices including source of drinking water in the school and at home, type of water source, latrine availability in school and type, and conditions of existing latrine. The teams collected the GPS coordinates, the total number of pupils, the total number of boys and girls in school, toilet availability, water supply and handwashing facilities at the school level. All the survey data were managed in the LINKS system. Study participants' age and gender were also entered in the inbuilt LINKS system, while laboratory results on S. stercoralis infection were recorded on lab forms. Raw data were downloaded and then curated from LINKS framework server into Microsoft Excel spreadsheets (Microsoft Corporation, Redmond, WA, USA) for analysis (Sedionoto et al., 2019).

Data analysis
A total of 851 participants were registered in the study, all of whom had complete data and were included in the final analysis. The statistical analysis conducted for this study were descriptive statistics and binary logistic regression analysis. The descriptive analysis focused on mean, frequencies, percentages, and standard deviation. For binary logistic regression, simple analysis was performed first to screen for important variables (p-value <0.25), and then multiple logistic regression analysis was performed to obtained adjusted Odds Ratios (aOR). Analysis of logistic regression was used to classify the significant factors associated with the outcome of interest, S. stercoralis infection. Adjusted odds ratios (aOR) with 95% Cl were estimated to assess the strength of associations and statistical significance. All statistical analyses were carried out using SPSS version 26. S. stercoralis infection distribution maps were created using Health Mapper version 4.3.2 (WHO Public Health Information and Geographic Information Systems, Geneva, Switzerland).

Ethical issues
The Scientific Coordinating Committee of the Joint Gambia Government and MRC Ethics Committee reviewed and approved the study (document no: SCC 1415). School head teachers were informed and before the commencement of the study, parents of the participating children gave consent on behalf of their children.

Socio-demographic characteristics of the study participants
The study enrolled a total of 851 schoolchildren and all of them completed the survey (Table 1). They comprised of 425 boys (49.9%) and 426 girls (50.1%) between the age bracket of 7-14 years from the two study districts ( Table 2). The mean (± 2SD) age of infected individuals was 10.1 years (±2.2) ( Table 2). Participants' ages were grouped into 7-10 years and 11-14 years.

Prevalence and risk factors of strongyloidiasis
Out of the total 851 pupils (Sabach Sanjal: 301, Upper Badibou: 550), 76 pupils (8.9%) were positive for S. stercoralis. Of the positive cases, 37 (49%) were from Sabach Sanjal and 39 (51%) from Upper Badibou ( Table 3). The prevalence of infection was higher among females (9.2%) than males (8.7%). The prevalence by age category was 12.4% among 7-10 years category and 4.2% among 11-14 years category (Table 3). The prevalence by districts was 12.3% in Sabach Sanjal and 7.1% in Upper Badibou (Table 3). Table 1 shows the prevalence of S. stercoralis infection among the 17 schools. More than half (54.0%) of the tested population in Maka Farafenni school had positive stool samples, thus constituting the school with the highest disease burden in the two districts. Meanwhile, Kerr Biram School recorded the lowest prevalence of 2.0% (Table 1). In terms of risk factors of infection, schoolchildren from Sabach Sanjal were 1.6 times more likely to have S. stercoralis infection compared to those from Upper Badibou (aOR = 1.64, p-value = 0.058). Schoolchildren aged 7-10 years were 3.2 times more likely to have S. stercoralis infection compared to the 11-14-year olds (aOR = 3.2, p-value <0.001) ( Table 3). Schoolchildren who sometimes have water or tissue after defaecation have more infection rate compared to those who always have water or tissue after defaecation. However, this difference was not statistically significant (aOR = 1.36, p-value = 0.308). No significant associations were found between S. stercoralis infection and variables such as gender, latrine in school, availability of drinking water in school, latrine type, latrine condition, provision of handwashing material after defaecation and facility available after defecation (p-value >0.05) ( Table 3). Association of district and age group interaction was also assessed in the regression model and found not to be significant (p-value = 0.529). Apart from S. stercoralis, 4.3% and 2.5% of the studied children were found infected with schistosomiasis and STH, respectively.

Discussion
The move to control schistosomiasis and other NTDs has gained momentum after the London Declaration of 30th January 2012, which had advocated for the elimination of NTDs including Strongyloides stercoralis in some countries (Abah and Arene, 2015; Wiria et al., 2015). This call (2012 WHA Resolution 65.21) has intensified global efforts to reduce the disease burden and achieve elimination targets. The distribution of S. stercoralis infection worldwide varies enormously from one country to another and even within the same country, depending on socio-economic and ecological conditions (Olsen et al., 2009;Schär et al., 2013;Ngui et al., 2016). To our knowledge, this is the first school-based survey in The Gambia that assessed the prevalence of S. stercoralis infection using a traditional parasitological technique, Kato Katz. The study showed that 8.9% of the schoolchildren had S. stercoralis infection. This value is higher compared to the prevalence of S. stercoralis infection found in other countries such as 2.1% in Madagascar (Hakami et al., 2019), 2.2% in Zanzibar (Knopp et al., 2008a), 5.8% in Ethiopia (Legesse and Erko, 2004), 0.2% in Angola (Mirante et al., 2016) and 2.7% in Thailand (Sedionoto et al., 2019). However, our prevalence is comparable to the 9.7% prevalence reported in the Eastern Caribbean Sea (Kurup and Hunjan, 2010) and the 10.8% prevalence in Zanzibar (Knopp et al., 2008b). The prevalence is lower compared to studies from Malaysia, Ethiopia, Angola, Cambodia, India, Nigeria and Cote d'Ivoire, where considerable higher prevalence of 15.8%, 20.7%, 21.4%, 12.8%, 31.6%, 36%, 13.8%, 4-48% respectively were reported (Amor et al., 2016;Dacal et al., 2018;Forrer et al., 2017;Maikenti et al., 2020;Rajkumari and Jayaseelam, 2016;Glinz et al., 2010). However, the use of different techniques makes a comparison of findings difficult. Apart from that, the studied population/group, the different culture, type of food eaten, personal hygiene and sanitary facilities of each country would also influence the study outcome. There was striking differences in the rate of infection between the various schools in this study. The study results showed that the prevalence of strongyloidiasis was higher in Sabach Sanjal than in Upper Badibou district. Upper Badibou is slightly more developed than Sabach Sanjal. In majority, the occupation of people in Sabach Sanjal is farming and they often miss routine reproductive child clinics and vaccination campaign including deworming activities, while people in Upper Badibou do business and partly engage in some farming activities. The study also found comparable prevalence of strongyloidiasis between males and females. A previous study found that females were more infected with strongyloidiasis than their male counterparts (Panda et al., 2012). Other studies, however, found the contrary that males were more infected with S. stercoralis than females (Kola et al., 2013;Terefe et al., 2019;Obeta, 2019;Khieu et al., 2014b). This difference in infection rate between males and females could be attributed to the socioeconomic and cultural factors that exist in the study settings. For example in Cambodia, it is attributed to the main economic activity that is subsistence-level rice farming (Khieu et al., 2014b).
This current study found that S. stercoralis infection had significant association with only age group and district out of all the potential risk factors examined. Age group 7-10 years were 3.2 times more likely to develop strongyloidiasis compared with the age group 11-14 years. This is consistent with previous observations by Khieu and other (Khieu et al., 2014b). and Agbo and colleagues (Oche Agbo et al., 2019) who found that age is a risk factor for strongyloidiasis. In the study by Agbo and colleagues, children of age group 8-12 years in Katsina, Nigeria has significantly higher risk of infection due to walking bare footed, playing with sand, eating food without washing hands and personal hygiene. However, in our study, we found no correlation between S. stercoralis infection and the availability of facilities after defaecation as well as water or tissue availability. This, however, contrasts with findings of Schär et al.  who found an association between the absence of sanitary facilities and strongyloidiasis.
Concerning the two districts, we observed significant heterogeneity between the schools; three schools in the Sabach Sanjal district and four schools in the Upper Badibou district had S. stercoralis infection. The highest and lowest prevalence were both found in Upper Badibou district. We could not distinguish any differences between the participants of the two districts or schools in the demographic data, latrine facilities or personal hygiene conditions. Therefore, we are unable to determine why participants from certain schools had higher infection rates. A recent assessment report conducted by the environment unit of the Ministry of Health, The Gambia found that most communities in both districts lack access to potable drinking water and eleven communities are still practicing open defaecation (CLTS, 2019). However, in this study, the detection of S. stercoralis larvae is part of a bigger schistosomiasis and STHs mapping which uses an AFRO questionnaire capturing data on schistosomiasis and STHs, thus the environmental variables and other potential risk factors for S. stercoralis such as walking barefooted, open defaecation, playing without shoes and eating fruits and vegetables that are not well washed within the two districts' school climate were not evaluated. Previous studies have reported that S. stercoralis infection is correlated with temperature, rainfall, open defaecation, inadequate portable water, farming, walking barefooted, playing without shoes, consuming fruits and vegetables which are not well washed (Zeehaida et al., 2011;Kola et al., 2013;CDC, 2013;Beknazarova et al., 2016;Forrer et al., 2019). Differences between the two districts in socioeconomic demography, environmental factors and children's activities may explain our observations.
The study revealed a significant prevalence of S. stercoralis among schoolchildren in Sabach Sanjal and Upper Badibou districts. However, the study utilized only Kato-Katz method to detect S. stercoralis larvae in stool samples. More suitable approaches to detect S. stercoralis larvae such as larva recovery method, agar plate culture and polymerase chain reaction could be used to increase the detection rate. For example, the examination of multiple stool samples per participant coupled with the use of a combination of more sensitive techniques such as Baermann method and Koga agar plate can increase the sensitivity up to 70% or 93% . The prevalence of the disease could have been higher in these two districts of North Bank East Region if more sensitive diagnostic techniques were employed in the present study as Kato-Katz is not a suitable technique for detection and diagnosis of S. stercoralis (WHO, 2020). A study that compared Kato-Katz and APC among Peruvian Amazon communities found out infection rate of 0% with Kato-Katz technique and 22% with APC (Machicado et al., 2012). On the other hand, a prevalence of 24.4% among schoolchildren in Cambodia was obtained using combined two methods, Baermann and Koga agar plate culture techniques . Additionally, a recent systematic review of 82 studies in Africa found Kato-Katz to be a less popular tool for diagnosis of S. stercoralis with just 2 of these studies utilising the technique for detection of S. stercoralis (Hailu et al., 2020). The study also found that Kato-Katz detected S. stercoralis larvae in only 8 out of 810 participants (Hailu et al., 2020). These are further evidence that S. stercoralis is not an ideal method for diagnosis of S. stercoralis. Given the evidence highlighted by the previous studies, our findings may be an underestimation of the true prevalence estimates of S. stercoralis in the study area. Further studies that will employ more suitable and sensitive techniques such as Baermann method, Koga agar plate and polymerase chain reaction are required to elucidate the true status and prevalence of S. stercoralis in Upper Badibou and Sabach Sanjal. Despite the low prevalence, our study revealed the occurrence of S. stercoralis in Sabach sanjal and upper Badibou for the first time.
Our study examined double Kato-Katz slides prepared from a single stool sample per participant to maximise the S. stercoralis larvae's excretion in stool specimens. This is part of a strategy to increase the yield of larvae. Another study revealed that larva output in faecal specimens ranged from 0.003 larvae per gram to 151.2 larvae per gram . However, the use of a Kato-Katz technique only in this study is a limitation and cannot efficiently detect all S. stercoralis infection (Machicado et al., 2012;Hailu et al., 2020;WHO, 2020). Thus, the result of our study should be interpreted cautiously. Questionnaires that capture more relevant risk factors such as walking barefooted, contact with the soil, contact with human waste or sewage, contact with farm animals, eating unwashed fruits and vegetables, farmer's children, awareness about worm infections and open defaecation should also be used.
We conducted awareness campaigns on the prevention of the disease and treated the affected schoolchildren with albendazole, a widely available antihelminth. Another treatment option is ivermectin which is superior in treating strongyloidiasis than albendazole. The effectiveness of ivermectin is documented in many studies and its cure rate is better than albendazole (Henriquez-Camacho et al., 2016).

Conclusion
The study revealed notable prevalence of strongyloidiasis in Sabach Sanjal and Upper Badibou districts of The Gambia. Kato-Katz technique might be inadequate for detecting S. stercoralis; thus, more studies are needed to determine the true prevalence of the disease in these two districts through the combined use of highly sensitive techniques such as Baermann, Koga Agar Culture and polymerase chain reaction.