Prevalence and intensity of Ascaris lumbricoides infections in relation to undernutrition among children in a tea plantation community, Sri Lanka: a cross-sectional study

A cross-sectional study was carried out in Uduwela tea plantation area (7⁰ 13′ − 7⁰ 17’ N and 80⁰ 37′ − 80⁰ 38′E) from January to April 2013. The study area is located in Kandy district and is 120 km away from Colombo, the capital city of Sri Lanka. It covers an area of about 2000 ha which is situated approximately 700 m above sea level, with an estimated population of 10,000 people in 2012. Annual average rainfall is between 2000 and 2500 mm and the average atmospheric temperature and humidity variations between 15 °C to 27 °C and 75% to 85% respectively. Most houses lack appropriate sanitation and portable water resources. As such, residents defecate in nearby jungles and retrieve water from streams and unprotected wells at low altitudes. Majority of residents in the study area are employed in tea plantations as unskilled laborers. This plantation area has been divided into seven sub-administrative divisions. Out of these divisions, three divisions were randomly selected by a lottery method. The target participants for the study were children in 1–12 years age group in the selected divisions. The sample size was calculated using the equation n = t² × p (1-p)/ d² where n = required sample size, p = estimated prevalence, t = 95% confidence level at 95% and d = margin of error at 5%. The estimated prevalence of Ascaris infections was taken as 30% in accordance with a recent survey of the plantation sector reporting a STH infection prevalence of 30% in children. On this basis, the minimal sample size was 323. As cluster sampling used in this study, a design effect was taken as 1.5.Thus the final sample size was 489.

Before the commencement of the study, the objectives of the study were explained to all parents/guardians whose children were selected for the study. An interviewer administrated structured questionnaire was administered to collect socio - demographic characteristics (which included gender, age, number of rooms and family members) and data on de-worming practice. The questionnaire was designed in Sinhala and Tamil. Anthropometric measurements of height and weight were collected following Gibson’s guidelines using anthropometric rods and digital weighing scales to the nearest 0.1 cm and 0.1 kg, respectively in their homes [21]. Children stood on the floor without shoes and made sure that their shoulders, buttocks and the back of the head was in contact with the vertical backboard. Height of the selected children was measured as the maximum distance to the vertical position to the head from heels by using a stadiometer (SECA model 240). Weight was measured for children wearing light weight clothes, ensuring that their trunk was vertical above the waist and the arms and shoulders were relaxed.

A clean and dry wide mouthed container (with the identification number) was given to all the participants for the collection of stool samples. Parents and guardians were instructed to ensure that small proportions of stool were collected and the samples were not contaminated with water or urine. Stool samples were collected the following morning from their homes and transported to the Parasitology laboratory, University of Peradeniya within 2 hours of collection.

Stool examination was conducted by experienced laboratory technicians at University of Peradeniya. Each stool sample was concentrated using formalin-ethyl acetate sedimentation technique [22]. Direct wet mounts were then prepared using Lugol’s iodine as well as in normal saline and examined microscopically under high power (× 40). Positivity of infection was based on the identification of Ascaris eggs. In addition, Ascaris eggs were quantified by Kato Katz technique to determine the intensity of infection [23]. Three slides made from each sample and every slide was read within 30–60 min. The recommended template of 41.7 mg was used and the average number of eggs counted was multiplied by 24 to get the number of eggs per gram of feces (EPG) [24]. Intensity of Ascaris infection was estimated as EPG (eggs per gram of feces) and classified into light, moderate and heavy with a definition of 1–4999 EPG, 5000–49,999 EPG and ≥50,000 EPG, respectively [25].

Data was entered into Microsoft Excel 2007 sheet and cleaned to ensure accuracy (compared with questionnaires). Statistical analyses were performed using SPSS version 17 (SPSS, Chicago, IL, USA) statistical package. The descriptive data of continuous variables and proportions of categorical variables was expressed to obtain an understanding of socio-demographic characteristics of the study community.

Nutritional indicators were classified and standardized into sex-specific Z-scores for height-for-age (HAZ), weight-for-age (WAZ) and weight-for-height (WHZ) by WHO AnthroPlus 1.0.4. In addition, Z values for weight for height of children greater than 9 years old were determined using Epi Info 3.5.1. Children were classified as stunted, underweight and wasted when the Z scores of HAZ, WAZ and WHZ were less than - 2 standard deviation (< -2SD) respectively. This enabled estimation of the prevalence of undernutrition (i.e. underweight and/or wasted and/or stunted) in the study participants and subsequent evaluation of the effect of infection on this estimate. The association between the positivity of Ascaris infectivity with categorical variables (e.g., gender, age group, number of rooms, number of family members, deworming period, stunting, underweight and wasting) and continuous variables (e.g., HAZ, WAZ and WHZ) were calculated using the Chi square test (X²) and student t-test respectively. One-way ANOVA was applied to analyze differences in anthropometric mean Z-scores with the intensity of infections (light, moderate and heavy).

Multiple linear regression was used to determine the association between anthropometric indicators (HAZ, WAZ and WHZ) and infection status and intensity. Multiple logistic regression analysis was applied to determine the predictive effect of stunting, being underweight and wasting on the intensity and infection status. Ascaris infection status and the intensity were considered as dependent variables while socio demographic characteristics were considered as independent variables for both logistic and linear regressions. The differences were considered to be statistically significant when the p value was less than 0.05 at 95% confidence interval (CI) for the multivariate regression analysis.

The ethical approval for the study was obtained from the Ethical Review Committee (ERC), Faculty of Allied Health Sciences, University of Peradeniya, Sri Lanka. Permission to conduct the study was given by the administrative authorities of the tea plantations and written informed consent was obtained from the parents/ guardian of the children who formed part of this study. Laboratory reports were issued with the signature of the Head of the department (Department of Parasitology) with recommendations. After the examination of stool samples, all children infected with ascariasis were administered anthelmintic treatments (single dose of 400 mg Albendazole) under the supervision of administrative and medical authorities.

Four hundred eighty nine children with the mean age of 6.2 (SD = 3.4) years were recruited into the study and 249 (50.6%) of the participants were females. A total of five parasites were identified with Ascaris lumbricoides (38.4%, 188/489) followed by Entamoeba coli (16.6%), Enterobius vermiularis (2.0%), Endolimax nana (1.8%) and Giardia lamblia (0.2%). Children who lived in houses with one or two bedrooms and had more than five family members showed higher prevalence of Ascaris infection. Children who had taken anthelmintic treatments more than 6 months before conduction of the study were 1.6 times more likely to be infected than those were treated within last 6 months (p = 0.019). Approximately half of the infected children (51%, 96/188) were suffering from light intensity infections while 30% (57/188) and 19% (35/188) of children had moderate and heavy intensity infections respectively (Table 1).

Prevalence of undernutrition among children in this community was 61.7% (302/489). The most common type of undernutrition in this community was being underweight (45%, 220/489) while 23.3% (n = 114) and 20% (n = 98) of children were stunted and wasted respectively (Fig. 1). However, no significant association was found between Ascaris infection status and undernutrition. Non-infected children had higher mean values of HAZ, WAZ and WHZ than infected children (Fig. 2). Results of the one-way ANOVA analysis revealed that mean values of nutritional indicators were not significantly associated with infection intensities of Ascaris infections (Table 2).

Multiple logistic regression revealed that age and gender was significantly associated with being underweight. Female children were 1.8 times more likely to be underweight (OR = 1.78; CI = 1.22–2.58, p = 0.003) than male children. Children in the age group of 7–12 year were 1.5 times (OR = 1.50; CI = 1.04–2.16, p = 0.030) more likely to be underweight. Children living with more than 5 family members were 1.6 times more likely to be stunted (OR = 1.61; CI = 1.02–2.53, p = 0.040) than those living with less than 4 family members. However, no demographic factors were found to be significantly associated with wasting in this study group. Infection intensity and/or infection status were not found to be significantly associated with nutritional indicators.

Estimated coefficients (β) from multiple linear regression models are shown in Table 3. Heavy infections with Ascaris lumbricoides were significantly correlated with a decrease in WHZ scores (β = − 0.20, 95% CI = − 0.31 to − 0.03, p = 0.020). However, the intensity was not correlated with HAZ and WAZ scores. Children of the female sex and those aged between 7 to 12 years showed significant negative correlations between WAZ (β = − 0.14, 95%CI = − 0.23 to − 0.05, p = 0.003and β = − 0.11, 95% CI = − 0.19 to − 0.02, p = 0.021 respectively). Conversely, HAZ and WHZ scores were not correlated with age and gender. Children who lived in houses with more than 2 rooms were positively correlated with WHZ and children with more than five family members were negatively correlated with HAZ (β = 0.08, 95% CI = 0.01 to − 0.15, p = 0.070 and β = − 0.09, 95% CI = − 0.16 to − 0.01, p = 0.046 respectively).