Prevalence and clinical outcomes of Plasmodium falciparum and intestinal parasitic infections among children in Kiryandongo refugee camp, mid-Western Uganda: a cross sectional study

This study was carried out from October 3, 2017 to May 10, 2018, at Kiryandongo refugee settlement camp in Kiryandongo district, Mid-Western region of Uganda, as shown in Fig. 1 [7]. The Camp is about 255.47 km along Kampala-Gulu road and is surrounded by Lake Albert and Kyoga. The population of Kiryandongo is 280,000 people according to 2014 census and the refugee camp has 52, 613 refugees [8]. The study site was chosen because it falls into a stable malaria transmission zone. Similarly, at 64.9%, the prevalence of anaemia in children aged 6–59 months indicates an increasing trend in Kiryandongo.

The primary target population were divided into two groups aged < 5 and 5–14 years as per the local Uganda Ministry of health guidelines. The secondary study population were the care takers/guardians respectively residing within the refugee resettlement camp.

The study included all children who provided both parental consent and assent in addition to providing blood and stool samples. Children below 3 years and above 14 years were excluded from the study.

Sample size was determined using the formula for estimating single proportion at the level of 95% confidence interval [Zα/2 = 1.96] by taking the proportion of metazoa and protozoa co-infection prevalence in the study area. However, since the prevalence of co-infection is unknown, was used 0.50 to have the largest sample size. The expected margin of error was taken to be 5% (d = 0.05). Using the formula, n = Z² x P x (1-P)/d². A minimum of 437 samples (n) was estimated as shown: n = Z² P (1-P)/d² [9]. Where, n-sample size, P- average prevalence, Z²- value at 95% CI, from table, d – expected marginal error. n = (1.96)² × 0.5x (1–0.50) / (0.05)² = 437. Therefore, once the minimum sample was obtained, a contingency of 8% was added for the sample size to reach 476.

Random sampling of children visiting the Panyadoli Health Centre III (HC III) based on the selection criteria was used to obtain blood film and stool sample by the law of voluntarism. Using roll numbers, random numbers were generated using Microsoft Excel 2013. This technique provided each participant with an equal chance of participating in the study. Along with sample collection, face-to-face interview of parents/guardians was conducted using a questionnaire constituting socio-demographic information. The data collection form has been included as Additional file 1.

A researcher developed structured questionnaire based on the objectives of the study, conceptual framework and guided by the literature review was used to collect data from the respondents. The data collection tool with the questionnaire has been included as Additional file 1.

The choice of diagnostic techniques was dependent on available equipment, reagents, experience of the technician and considerations of cost. The techniques used in this study for analysis of stool and blood samples included malaria microscopic examinations using Giemsa Staining technique, intestinal parasitic detection using Kato Katz and Calorimetric Hb estimation. Microscopy was used for detecting malaria because it is a gold standard putting in mind the limitation of rapid diagnostic kits (RDT) since it detects malaria antigen 4 weeks even after treatment and does not allow differentiation of malaria species.

Completed questionnaires were serialized and coded. Data was entered and cross checked for errors using Microsoft Excel 2013 before running the analysis.

The preliminary data was entered into Microsoft Excel 2013 and exported into STATA 14. Descriptive statistics was done to analyse the background characteristics of participants and the factors associated with Plasmodium falciparum infection, intestinal parasitic infection and co-infection. Chi-square test and Fishers Exact test was used to determine the association between the factors associated with Plasmodium falciparum and intestinal parasites. Statistical significance of the differences in the prevalence of single and co-infection with each factor was determined via Chi-square analysis and Fishers Exact test. Multivariate logistic regression analysis was used to predict the extent of the association with the different infection status as the dependent variable. The study was able to control for confounding factors during analysis by adjusting for age, sex and nutritional status, while the association between helminth and Plasmodium infections, and related outcomes was evaluated.

Table 1 above shows the socio-demographic profile of the 476 participants. Most of the study participants were refugees from the South Sudan, Uganda, DRC, Rwanda and Burundi respectively. Of the 476 participants, 364 (83.3%) wash their hands every time after using the toilet and only 73 (16.4%) participants do not wash their hands after using the toilet. 471 (98.9%) wash their hands before eating; and those who do not wash their hands before eating were 6. Most of the participant use borehole 267 (56.1%) and tap water 201 (42.3) as main sources of drinking water. 432 (90.8%) of the participants had previously been infected with malaria. 454 (95.4%) of the children had ever slept under a mosquito net.

Nutritional status assessment using mid upper arm circumference (MUAC), showed no evidence of malnutrition of 100% of children who had median MAUC of 18.8 cm. Out of the 476 children examined, 300 (63.03%) had a mono-infection, 28(5.88%) were co-infected with malaria and IPIs whereas 148 (31.09%) were not, as shown in Table 2 above. It is worth noting that the overall prevalence of parasitic infections in the study population was 63.03%, of which Plasmodium falciparum accounted for the highest prevalence 262(55.04%).

From table 3, the most infected age group by Plasmodium falciparum was (5–14 years) with the prevalence of 55.7%. The younger age group (< 5 years) had a prevalence of 52.7% and the difference was statistically significant (P = 0.000). A total of 256 (53.8%) males and 220 (46.2%) females were enrolled in the study. Overall, 72.27% (159/220) of the females were slightly more infected with all the intestinal helminths as compared to 55.1% (141/256) of the males but this and the difference was not statistically significant (P > 0.05).

As shown in table 4 above, the most prominent co-infection combinations were of Plasmodium falciparum + Taenia spp. 5 (1.05%), Plasmodium falciparum + Giardia lamblia 2 (0.42%) and Plasmodium falciparum + Schistosoma mansoni 1(0.21%).

The results above further indicate that the odds of developing simple or uncomplicated malaria infection was 14 times higher in individuals double infected with Plasmodium falciparum + Taenia spp. compared to single parasitic infection (Odds = 14.13, P = 0.019).

From table 5 above, the occurrence of co-infection of Plasmodium falciparum + Taenia spp. (P = 0.019, Odds = 14.13) showed a strong predictor of anaemia in the study population. No single infection was a strong predictor of anaemia.

From table 6 above, it is evident that the highest prevalence of anaemia was among the children with no infection at 24.1%.