Intestinal schistosomiasis and geohelminths of Ukara Island, North-Western Tanzania: prevalence, intensity of infection and associated risk factors among school children

Study area

Ukara is an island located on the Lake Victoria and is part of the Ukerewe district, Mwanza region, North-Western Tanzania. The island has a total population of 34,181 according to the national census of 2012 [13]. It is divided into four wards, namely Bwisya, Bukungu, Nyamanga and Bukiko. There are eight villages: Bwisya, Nyang’ombe, Bukungu, Chifule, Nyamanga, Chibasi, Bukiko and Kome. There are 12 government-owned primary schools. The main socio-economic activities carried out by the inhabitants of the island include fishing, subsistence farming, livestock keeping and small scale businesses. At the time this study was conducted, no control program was in place against intestinal helminth infections.

Study design, population and inclusion criteria

A cross-sectional study was conducted among school going children aged 4-15 years focusing on determining the prevalence of infection with STH and S. mansoni and their associated risk factors. Children were included in the study if parents/guardians had given written informed consented for them to participate in the study and if assent had been obtained from the children. Teachers were involved to educate parents and children on the importance and risks of participating in the study. Children with a history of taking anthelmintic medication in the past three months were excluded from the study.

Sample size determination and sampling procedures

The sample size was calculated as described elsewhere [14],[15], considering the prevalence of S. mansoni infection of 60% in the island of Ukerewe [10], at 95% confidence interval and margin error of 5%. A design effect of 1.5 was considered for the variation in prevalence between schools. A minimum sample of 609 school children was needed for this study.

A simple random sampling method was used to select the villages with a primary school to participate in the study. The number of school children selected from each school was determined by the probability proportional to size of the school and the class population. Systematic sampling, using the class registers as the sampling frame was used, where the names of the children were arranged in alphabetical order. The sampling interval was obtained by dividing the total population in the class with the number of children to be investigated in that class (N/n). After obtaining a start from a table of random numbers, the same interval was kept until the required number of children in each class was obtained.

Data collection

Data analysis

The collected data were entered into a data base using Epi data version 3.1. Data analysis was done using STATA version 12.0 (Stata corp, Chicago). The prevalence of S. mansoni 95% confidence interval (95% CI) was obtained by binomial logistic regression taking into account clustering by schools. The comparison of prevalence by demographic factors for S. mansoni infection was tested for significance using χ² or fisher exact test where appropriate. The age variable was described as mean ± standard deviations. The arithmetic mean of S. mansoni egg counts for each participant was calculated from the counts of two Kato Katz thick smears and multiplied by 24 to obtain individuals’ eggs per gram of faeces [17]. S. mansoni egg counts were over dispersed so were logarithmically transformed prior to analysis. The geometric mean intensity eggs per gram of faeces (GM-epg) of S. mansoni infection were obtained as the antilog of the mean of the transformed egg counts. The comparison of geometric mean egg counts for S. mansoni between various demographic factors was undertaken using t-tests and ANOVA. The intensity of infection was categorized as: 1-99 epg, 100-399 epg, ≥400 epg defined as low, moderate and heavy intensities of infection respectively [17]. To determine the factors associated with S. mansoni infection and intensities, multiple linear and multivariable logistic regression models were used, controlling for other explanatory variables. The model building strategy was to first identify potential factors at bivariate/linear regressions level and include these factors in the multivariable/multiple linear regression level. Factors with P-value < 0.2 were identified at bivariate level and were considered for the final model. Stepwise backward procedures were used to determine whether these variables were independent factors of intensity of S. mansoni infection by using adjusted odds ratios (AOR) for linear models and the 95% confidence interval (CI).

Ethical consideration

Ethical approval was obtained from the joint Ethical and Review Committee of Bugando Medical Centre and Catholic University of Health and Allied Sciences. The study received further clearance from the district administrative authorities (District Education Officer, Medical Officer, Executive Officer) and division authorities at Ukara Island. Kiswahili translated informed assent and consent forms were used to obtain parents/guardians consent for the children to participate in the study and assent of the children. For illiterate parents, a thumb print was used to sign the assent and consent forms after a clear description of the study objectives and their acceptance for their children to participate. All study participants who were infected with S. mansoni and STH were treated with praziquantel (40 mg/kg) and albendazole (400 mg) according to WHO and country guidelines [17].

Demographic characteristics of the study population

Overall prevalence of Schistosoma mansoni and soil transmitted helminth infections

Overall, 66.11% (n = 511/773, 95% CI = 62.76 -69.45) of the study participants were infected with at least one of the parasites investigated in the present study. The overall prevalence of S. mansoni was 63.91% (n = 494/773, 95% CI = 60.51 - 67.30). The overall prevalence of soil transmitted helminths was 6.73% (n = 52/773, 95% CI = 4.95 – 8.49) with the prevalent species of the STH being hookworm 5.69% (n = 44/773, 95% CI = 4.06 –7.32). Other STH helminth observed was A. lumbricoides, 1.03% (n = 8/773, 95% CI = 0.32- 1.75).

Prevalence of Schistosoma mansoni stratified by demographic characteristics

The prevalence of S. mansoni in relation to the demographic characteristics of the study participants is shown in Table 2. The prevalence of S. mansoni did not differ by sex of the study participants (P = 0.23) but varied significantly by age groups, with the youngest age group (4 - 10 years) having the highest prevalence (P < 0.014) (Figure 1B). The prevalence decreased with increased age of the study participants, with the age group aged 11-15 years having the lowest prevalence. Similarly, the prevalence of S. mansoni varied significantly by school location, with schools located along the shorelines of Lake Victoria having the highest prevalence (P < 0.0001) (Table 2). The prevalence of the disease also varied significantly by reported parental occupation, with children reporting their parents to be involved in fishing activities having the highest prevalence (P < 0.0001) compared to those from farming families. Reported history of regularly visiting (for bathing/swimming/washing clothes or utensils more than three times a week) Lake Victoria was associated with significantly higher prevalence of the disease (P < 0.0001) (Table 2).

Table 2

Prevalence of Schistosoma mansoni stratified by demographic characteristics of the study participants

Variable	No examined	Prevalence (%)	95% CI	χ2	P–value
Sex
Female	421	277 (65.80)	46.79 – 92.51	1.43	0.23
Male	352	217 (61.65)	41.90 – 90.69
Age (in years)
4 – 7	234	165 (70.51)	57.68 – 86.19	9.2117	<0.010
8 – 10	370	234 (63.41)	40.94 – 98.22
11 – 15	120	95 (55.88)	39.91 – 78.22
Schools
Chifule	190	145 (76.32)	69.70 – 83.56
Mubule	155	98 (63.23)	56.97 – 70.17
Kome	149	144 (96.64)	93.72 – 99.57	190.18	<0.001
Nyamanga	155	81 (52.26)	44.70 – 61.09
Kumambe	124	26 (20.97)	12.59 – 34.91
Parents occupation
Peasants	561	327 (58.29)	52.65 – 64.53	27.9905	<0.001
Fishing	212	167 (78.77)	76.96 – 80.63
Presence of toilet at home/school
Yes	675	427 (63.26)	56.91 – 70.32	0.97	0.33
No	98	67 (68.37)	65.15 – 71.73
Lake visit
Always	653	442 (67.69)	64.04 – 70.62	26.07	<0.001
Sometimes	120	52 (43.33)	26.59 – 70.62
Paddy cultivation
Always	128	80 (62.5)	53.99 – 71.00	0.13	0.717
Sometimes	645	414 (64.19)	60.47 – 67.89

Open image in new window

Intensity of Schistosoma mansoni and soil-transmitted helminths

Of all the children found infected with any of the STH observed in the study, the majority had a light to moderate intensity of infection.

Factors associated with Schistosoma mansoni infection and intensity

The results of multiple linear regression revealed that fishing as parents occupations (AOR = 1.20, 95% CI; 1.02-1.42, P < 0.03), reportedly involved in paddy cultivation (AOR = 1.45,95% CI: 1.004-2.10, P < 0.048) and the location of the schools at Nyamanga (AOR = 1.75,95% CI; 1.37 – 2.24, P < 0.0001), Kome (AOR = 7.52, 95% CI; 5.78-9.75, P < 0.0001), Mubule (AOR = 2.22,95% CI; 1.73 – 2.86, P < 0.0001) and Chifule (AOR = 3.91,95% CI; 3.03-5.04, P < 0.0001) villages P < 0.0001) along the shorelines of Lake Victoria remained significantly associated with the intensity of S. mansoni infection (P < 0.0001) Additional file 1. The prevalence of STH was very low to allow the analysis of factors associated with STH in the present study population.

Prevalence of Schistosoma mansoni and soil-transmitted helminth infection

The overall prevalence of S.mansoni observed in the present study was similar to the prevalence of 64.3% reported among school children along the Lake shore in Sengerema district, North-Western Tanzania [8], but slightly higher compared to studies in Mbita Island in Western Kenya (60.5%) [18] and Sesse islands on Lake Victoria in Uganda (58.1%) [19]. In addition, the prevalence and intensity of S. mansoni infections varied significantly by demographic characteristics of the study participants. The prevalence and intensity of infection varied by age of the study participants, gender, village of residence and parental occupation. In endemic areas, it is usually acknowledged that S. mansoni intensities of infection show a peak at the age group 6-19 years and, thereafter, the intensities decline gradually with an increased age [20]-[22]. Similarly, in the present study, the youngest age groups had the highest intensity of infection. Our study further showed that in the study area, infection with S. mansoni starts at a young age (probably < 4 years of age). If these children are not treated in time [22], by the time they start school, they may have developed significant morbidities [10].

We observed an inverse relationship between the proximity to the lake and S. mansoni prevalence and intensity of infection in the study areas, with schools located closest to Lake Victoria having the highest prevalence compared to schools which were located away from the Lake shores. Our observations were consistent with the results of Handzel et al who observed the decrease in prevalence of S. mansoni with increasing distance from the Lake Victoria shore [23]. The school located nearest the lake (750 meters) had a mean prevalence of 80% which decreased to 20% at a distance of 4-13 km from the lake shore [23]. Similarly, in Mbita and its adjacent islands, schools located in close proximity to Lake Victoria had the highest prevalence of S. mansoni[18],[24],[25].

The overall prevalence of soil-transmitted helminths observed in the present study population was very low compared to 12.4% [18], 16.2% [26] and 42.5%% [23] reported from Western Kenya. Of all the soil-transmitted helminths observed in the present study, hookworms were the most predominant species, although their prevalence was very low as compared to data reported by previous studies in North-Western Tanzania which quoted a prevalence of 38% [8] and 37% [27] and in Western Kenya (42.5%) [23]. Along the Lake Victoria shores in Western Kenya and North-Western Tanzania, hookworms appear to be the predominant species and other soil-transmitted helminths areas are rarely found. Two previous studies in North-Western Tanzania reported a prevalence of <1% of A. lumbricoides, T. trichiura and E. vermicularis[8],[27]. The high tolerance of hookworms’ eggs and larval stages to the variation of the soil temperature has been described as a key factor for the high transmission and prevalence of these parasites in the area [12].

The majority of the infected children had a light to moderate infection intensity of these helminths infections. This observation was consistent with data from previous studies which observed that most of the individuals infected with soil-transmitted helminths normally excrete a low number of eggs [18],[24]-[26].

Intensity of S. mansoni infections

Epidemiological surveys along the Lake Victoria have shown that school children are carrying the highest intensity of infection as compared to other age groups [9],[18] and male individuals are more heavily infected [9],[28]. In the present study, the majority of the study participants infected with S. mansoni had light to moderate intensities and only a few were heavily infected with S. mansoni. Male individuals had the highest infection intensity as compared to female individuals. The difference in intensity of infection between sexes is mainly associated with the variation of exposure to risk areas and the time spent in water sources [29]. Male individuals tend to spend more time in water sources compared to female individuals [30]. A high exposure is associated with swimming and sometime fishing in male children and can result in the maintenance of a high prevalence and intensity of infection into adulthood [9],[30].

The focal nature of S. mansoni transmission along the present study area appears to influence intensity of infection. We have observed variations in the intensity of infection between schools, with schools located along the lake shores having the highest intensity of infection as compared to schools which were located away from the lake. Similar findings have been reported in Ssese island on the Lake Victoria shore in Uganda [19] and in Western Kenya [18],[23]-[25].

Risk factors associated with intensity and S. mansoni infection

In S. mansoni endemic areas, gender, age group, geographical location and occupation are some of the well-described demographic factors reported to be associated with infection and intensities [29],[31]-[33]. Similarly, our findings showed that S. mansoni infection was mainly associated with the younger age group (4 - 10 years). In addition, parental occupation, especially involvement in fishing, the location of the school and a reported history of visiting the lake regularly were significantly associated with S. mansoni infection. However, on a multivariable analysis, only the location of schools remained associated with S. mansoni infection. As explained above, the proximity to the lake shores was associated with an increased risk of high infection in the present study area. In Sesse Island, in Uganda, a combination of malacological and parasitological surveys revealed that S. mansoni infection was only occurring in certain parts of the island [19]. The same applied in Western Kenya [18],[34]. The present study did not include malacological surveys and these surveys are recommended in future studies in the area.

Conversely, a multiple linear regression model revealed that being male and the location of the schools along the shorelines of Lake Victoria remained significantly associated with the intensity of S. mansoni infection. Similar results have been described elsewhere in sub-Saharan Africa [10],[35]. A heavy intensity of infection was mainly seen in male individuals. The schools location along the shorelines of Lake Victoria mainly defines the level of exposure and transmission of the disease in the study area. Children attending schools located in villages which had close proximity to the lake, had highest intensities and appeared to be more exposed to cercariae infested water as compared to those who were living away from the lake shore [18],[24],[25].