Background

Malnutrition is a major public health problem leading to child morbidity and an underlying cause for more than half of child deaths worldwide, particularly in low socioeconomic communities in developing countries [1, 2]. Insufficient dietary consumption and infectious diseases are major contributing factors for poor nutritional status of children [3, 4]. Young children are more prone to malnutrition because of high protein and energy needs and vulnerability of infections than adults [5]. Undernourished children are less productive of physical and mental development and increase susceptibility to infections and the risk of early deaths [6,7,8]. Malnutrition commonly has long lasting effects among children and adversely influences their adult logical and work capacity [9]. Anthropometric measurements are a cost-effective and reliable method to evaluate the health and nutrition status of the individuals and in a society [10]. To identify the actual priorities of child health within a society, it is necessary to determine the associations between individual variables and health-related factors in a child population.

Previous studies in Sri Lanka reported that undernutrition among children is still a major health problem associated with poor sanitation and personal hygiene, low socioeconomic status, overcrowding and low educated parents [11,12,13,14]. The plantation system was introduced by the British governors around the mid of the nineteenth century with tea and rubber. They brought Tamil community from South India to Sri Lanka for the purpose of working as laborers in the plantation sector and they now comprise about 5 % of the total population in Sri Lanka [15]. One of the major features of the plantation sector is laborers residing and working in the same plantations for generations. Therefore, Tamil community in the plantation sector has been isolated socio-economically and geographically from the other parts of Sri Lanka. In the twenty-first century, plantation community showed the lowest educational levels, health and nutrition conditions and the highest mortality levels in Sri Lanka [16].

The demographic and Health Survey (DHS) in 2006 reported that 30% of children less than 5 years in the plantation sector of Sri Lanka were underweight (< −2SD weight-for-age), 42% were stunted (< −2SD height-for-age) and 13% were wasted (< −2SD BMI-for-age) [17]. The Literacy rate in the plantation sector of Sri Lanka was 86.1%, while the national average was 95.7% [17]. Infant mortality and low birth weight (less than 2.5 kg) in the plantation sector were 29% and 31%, respectively, while the rates of Urban and Rural sectors were less than 20% [17]. Access to safe water in the plantation sector is 52.4% and the corresponding percentages for Urban and Rural sectors were 98.9% and 90.9% respectively [18]. Nearly 73% of households in the plantation sector had latrine facilities while the national average was 86.7% in 2012 [18]. It is important to determine the true extent and underlying causes of child nutritional status in the plantation sector of Sri Lanka to commence proper intervention strategies to decrease the levels of malnutrition in childhood. However, limited data are available on nutritional status of children and its association with socio-economic characteristics in the plantation sector. Therefore, this study was designed to determine socio-economic characteristics associated with the nutritional status of children in the plantation sector, Sri Lanka.

Methods

Study setting and design

A community based cross-sectional study was conducted in three villages surrounded by tea plantations located at Gangawata Korale, Patha Hewaheta and Galaha divisional secretariat divisions in Sri Lanka from January to August 2014. Geographically, these study sites are mountainous areas at the jungle fringes and the mean annual rainfall and temperature ranges between 2000 and 2500 mm and 15 °C to 27 °C respectively. The majority of the inhabitants were employed in tea plantations and factories as unskilled laborers. Although 60% of houses had poured flush toilets, defecation is often done in nearby jungles and no proper water supplements methods were observed in these study areas. Therefore, unprotected water streams close to the residents were used as water sources for their domestic requirements. Generally the education level of the adult population was very low and the majority of people were living in small houses known as line rooms, constructed by the colonial planters about one hundred years ago. The line rooms have small sq. feet with little ventilation, and overcrowding due to the large number of children and their dependent parents.

Children aged 1 to 15 were included in this study and the sample size was calculated using a formula of N = Z2Pd/d2 with 5% margin of error, 95% confidence interval, and 40% malnourished children in the plantation sector [15]. The calculated minimum sample size was 369. The design effect was taken as 1.5 due to the cluster sampling method was used in this study. Thus the final sample size was 553. A List of names of all children aged between 1 to 15 years in selected tea plantation areas was made with the help of regional administrative and health officers and 553 children were selected randomly using a lottery method for the study. Selected children were categorized into three groups as preschool (1–5 years), children attending to primary schools (6–10 years) and secondary schools (11–15 years) according to their ages and educational status. Children who did not return consent forms, not resident in a study area at least three consecutive years and those suffering from chronic diseases such as heart diseases, kidney and liver failures were excluded from the study.

Collection of data

Members of the research group visited each selected child and parents or adults in their households and the objectives of the study were explained briefly before the beginning of the study. Written consents were then obtained from parents or guardians of each selected child. Then each parent/guardians who signed the informed written consent were interviewed by members of the research group to elicit information on the educational and employment status of parents, sanitary and household conditions (i.e., type of drinking water, latrine system, number of rooms and housing constructions) and demographic data (i.e., age, gender. Number of siblings and family members) using an interviewer administered structured questionnaire (Additional file 1). Ages of children were obtained from their birth certificates. The questionnaire was prepared in both Sinhala and Tamil languages in order to make the questions clear to the participants.

Assessment of nutritional status in children

Anthropometric measurements were collected to assess the nutritional status of children. Body weight and height were obtained from each participant at the household level. All children stood barefooted against a vertical wall and height was obtained using a stediometer to the nearest 0.1 cm. Body weight of children with lightweight clothes was measured to the nearest 0.1 kg using a digital balance which was validated before starting the measurement of weight. All measurements were taken by two different research members to minimize the intra personal errors and calculated the mean of these measurements to evaluate final anthropometrical values. Body weight and height were standardized into age specific Z - scores for height-for-age (HAZ), weight-for-age (WAZ) and BMI-for-age (BAZ) by WHO Anthro Plus 1.0.4 and Epi Info 3.5.1 software. Children with scores of WAZ, HAZ and BAZ below than - 2.00 were defined as underweight, stunted and wasted respectively, and more than 2.00 WAZ and BAZ were defined as overweight and obese respectively [19].

Statistical analysis

Data were entered into Microsoft Excel 2007 and verified with the questionnaire. Data were grouped into three categories: (1) individual characteristics, (2) education and employment status of parents and (3) household characteristics and analyzed statistically using SPSS version 20 (IBM. Somers, NY). Confidence intervals were used in descriptive statistics to describe socio-economic characteristics of the study subjects. Inter observer error of height and weight measurements were determined using a reliability test. The chi-square test was used to analyze the proportions of categorical nutritional indicators (underweight, stunting, wasting, overweight and obese).

Anthropometric measurements of WAZ, HAZ and BAZ were expressed as mean and standard deviations and the variance between three groups (preschool, primary school and secondary school) were analyzed by one-way ANOVA method. Then pairwise comparison of these categories was analyzed by post-hoc Tukey test to determine categories which significantly differed from each other. Multivariate logistic regression using forward elimination model was applied to determine the strength of association of socio-economic variables with under nutritional indicators (stunting, underweight and wasting). For these applications, nutritional indicators were considered as dependant variables while the socio-economic factors were considered as independent variables. Odds ratios with 95% confidence interval were calculated for each factor in logistic regression model. The statistical significant difference was declared if p value was less than 0.05.

Results

Demographic and socio-economic characteristics

A total of 547 children with a mean age of 7.0 (SD = 3.6) years were participated and 289 (52.8%) of them were male. Out of total participants, 206 (37.7%) were preschool children, 180 (32.9%) were primary school children and 161 (29.4%) were secondary school children. Table 1 shows their demographic, household and parent’s education and employment status of the study subjects. In preschool children, 53% of study subjects were females and 47% were males. In primary and secondary school children, 50% and 55% were females respectively. The number of siblings of study children had ranged from 0 to 6 with the average of 2.6 ± 0.7 and the mean number of family members was 6.8 ± 2.2 (Range 2–12). Overall, the level of education and monthly income of the majority of their parents were very low. However, most of the fathers were employed while the majority of mothers were in homes as housewives. More than 75% children were living in small attached houses made by cement floors, walls and metal laminate roofs. When considering sanitary conditions, more than half of children had latrine facilities. However, most of them had no proper treated drinking water sources.

Table 1 Descriptive characteristics of children in the study group
Full size table

Determinants of child nutritional status

The reliability coefficient of height was 0.958 and weight was 0. 972. Nutritional status of the study subjects is presented in Table 2. Undernutrition was more common among these children. Overall, high number of children were observed with underweight (n = 195, 35.6%), stunting (n = 147, 26.9%) and wasting (n = 180, 32.9%). However, small numbers of children showed overweight (n = 15, 2.7%) and obese (n = 17, 3.1%). Of the undernourished children, 53%, 51% and 48% of underweight, stunted and wasting children were males, respectively. More than half of undernourished children lived with large number of family members and most of their parents’ had low educational background. Monthly family income of many undernourished (> 70%) children was less than 20,000 Sri Lankan rupees (133 USD) and lived in small attached houses consisting one or two rooms. In addition, children had not house own latrine facilities showed high prevalence of undernutrition compare to children lived in houses which had latrine facilities (Table 2). When considering the preschool children, undernutrition comparatively more common in males than females. In contrast, the level of undernutrition among school female children was higher than male children. Figure 1 shows that secondary school children had the highest percentage of one type of undernutrition conditions (Underweight or stunting or wasting) while two or more than two types of undernutrition conditions were more common among primary school children. Underweight was the commonest undernutrition condition among preschool and primary school children while wasting was the commonest in secondary school children (Fig. 2). With respect to children’s groups, undernutrion was more common in primary school children while secondary school children showed the highest proportions of overnutrition conditions comparable to other groups. Underweight, wasting and all Z score nutritional indicators were significantly associated with these children’s groups (Table 3). Primary school children showed lowest HAZ and BAZ mean Z score while the lowest mean Z score for WAZ remained among secondary school children. According to Tukey test, mean WAZ and mean BAZ in primary school children were significantly lower from the other two groups (WAZ: Pre and primary school child p < 0.001, secondary and primary school children p = 0.033; BAZ: Pre and primary school children p = 0.001, secondary and primary school children p = 0.035) and Mean HAZ was significantly lower than preschool children (p = 0.005).

Table 2 Demographic and socioeconomic characteristics with nutritional status in children
Full size table
Fig. 1
figure 1

Venn diagrams representing of undernutrition of study subjects

Full size image
Fig. 2
figure 2

Prevalence of undernutrition by age categories

Full size image
Table 3 Nutritional characteristics of the study sample
Full size table

Risk factors of child nutritional status

The risk factors associated with nutritional status in relation to socio-economic characteristics among these communities were analyzed by multivariate logistic regression method. Table 4 shows that female children were 2.9 times more likely to be underweight (OR = 2.89; CI = 1.50–5.55) and 2.8 times more likely to be stunted (OR = 2.75; CI = 1.39–5.45) in preschool children. Wasting was significantly correlated with birth order (OR = 6.39; CI = 1.23–33.08), number of siblings (OR = 3.16; CI = 1.11–8.97) and the employment of mothers were (OR = 2.12; CI = 1.05–4.28). In primary school children, family members were 2.3 times (OR = 2.31; CI = 1.13–4.74) and one or two room houses were 2.2 times (OR = 2.15; CI = 1.11–4.15) more likely to be underweight. Similarly, children had occupied and educated mothers were 2 times (OR = 2.03; CI = 1.02–3.63) and 4.1 times (OR = 4.09; CI = 1.86–9.00) more likely to be underweight and wasting respectively (Table 5).

Table 4 Multivariate analysis of risk factors for nutritional status of preschool children
Full size table
Table 5 Multivariate analysis of risk factors for nutritional status of primary school children
Full size table

In secondary school children, family members and low monthly income were 2.9 times (OR = 2.93; CI = 1.15–7.41 and OR = 2.90; CI = 1.05–8.06) more likely to be underweight. However, paternal education (OR = 0.33; CI = 0.15–0.76) and employment (OR = 0.37; CI = 0.14–0.99) were predictive factors for wasting (Table 6). Household latrine facilities, drinking treated water and the types of wall, roof and floor were not significantly associated with any group of child nutritional status.

Table 6 Multivariate analysis of risk factors for nutritional status of secondary school children
Full size table

Discussion

Malnutrition causes children more susceptible to infections. Those frequent infections negatively affect their physical and cognitive abilities locking them into a vicious cycle. This study confirms that undernutrition is still a considerable health problem in the plantation sector in Sri Lanka. This is much similar to the prevalence of malnutrition reported by the DHS of Sri Lanka in 2006 [17]. However, National Nutrition and Micronutrient Survey in Sri Lanka reported that 15.1%, 27.3% and 21.9% of children aged between 6 and 59 months had stunting, underweight and wasting respectively in 2012 [20]. It is commonly believed that these differences result from a low standard of living, high levels of infectious diseases, and low purchasing ability, all of which have an effect on feeding practices [21].

In the present study, primary school children are more likely to be undernourished than preschool and secondary school children. Preschool children, mainly depend on breastfeeding and may be protected by the mother’s immune system at birth [22]. Breast milk protects small children from infectious diseases and finally affects their nutritional status. With the increasing of age, children start to take complementary food in addition to breastfeeding and get nutrition from many food sources. Some studies have reported that the deviation of normal growth rates in children starts from several months after birth because breastfeeding is no longer enough to fulfill the requirements of growing children [23]. Inadequate complementary foods also might be a reason of increasing trends of malnutrition in older children. This might be attributed by stopping or reducing breastfeeding and insufficient intake of complementary food.

Underweight was the most common undernutrition indicator among these children. WAZ reflects body mass relative to age and it fluctuates with time and therefore it has a role in acute as well as chronic malnutrition [24]. High prevalence of stunting and wasting also reveal that most of the children in the study groups were suffering from long term chronic malnutrition, which negatively effect on both mental and physical improvements in childhood. Decreasing food purchasing ability, dietary changes and poor personal hygiene of mothers as well as infectious diseases may have a role in the nutritional status of children [21]. The present study revealed that primary school children showed the lowest mean Z scores of WAZ and HAZ compare to preschool and secondary school children. Since undernutrition increases with age, it reflects longer exposure to chronic malnutrition is common among children in this community. Knowledge of food nutrients and improved hygiene are a few of the factors that could be involved in reducing the prevalence of undernutrition among secondary school children.

Social customs and norms in communities and families describe the roles of members in households. In Asia, a number of the restriction process consists regarding the health of small children [25]. This study showed that preschool female children had significantly higher rates of underweight and stunting than male in this community. This is inconsistent with the studies in India and Bangladesh reported that moderate and severe undernutrition was higher in female [26, 27]. Gender inequality of male preference over female in care and feeding practices could be a major reason for this finding and it is commonly seen in many cultures and social beliefs in Asia [28]. In addition, Religious influences also discourage the consumption of some healthy foods. Therefore, child feeding practices of mothers and caregivers may affect dietary intake of children in this community. This indicates that more attention on the maternal characteristics at the household level should be paid for the improvement of child health particularly in preschool stage.

Education generates knowledge and favorable environment that positively affects for child health. In the present study, a declined trend of child malnutrition is observed where the educational level of mothers increased. Similar findings were reported in numerous studies showing children with low educated mothers had higher prevalence of undernutrition than highly educated mothers [29,30,31,32]. Education makes the mothers inform nutritional values of food and get more understandings of child physical and mental growth. Underage marriages are still common in the plantation sector and most of young low educated, married girls are neither mentally nor physically ready for motherhood. They often have a greater chance of giving low birth weights children [33]. This could be another factor for child undernutrition in the plantation sector. National Family Health Survey in India revealed that underage marriage mothers were a contributing factor for undernutrition of their children [34].

The main source of income for most families in these communities is working as laborers in tea plantations on a day-to-day basis. Because of the uncertainty in these occupations, both parents prefer to work in the industrial sector or migrate to Middle Eastern countries. Therefore, parents, particularly mothers have very limited time to prepare meals and feed children at home. Most of the children in the plantation sector spend their day time with grandparents or other relatives and this could be disadvantaged on children’s physical and mental growth and development particularly in preschool and primary school children [33]. Children of unemployed mothers were better nourished than the children whose mothers were employed [35, 36]. In addition, employment of mothers outside of their homes creates many social problems in family level and it negatively affect for the physical and mental development of children [37]. In many cultures in Asia, father is the money earner and the decision maker at the household level. Highly educated fathers more care of their children and provide health facilities when they need. However, in this study, high educated and occupied fathers play a significant role in decreasing the rate of wasting among secondary school children. This could be an ability to make decisions about getting good quality food for children [38, 39]. Further studies are needed to determine how parents’ educational influences to the nutritional status of children.

Generally the majority of residents in the plantation sector live in the same places where their previous generations were lodged called “Line houses” without major modifications. In this community, 68% children lived in long terraces of one or two room houses attached to each other under minimal socio-economic conditions. Further, few subjects were living in quarters and detached houses. The high prevalence rate of underweight among children in attached houses indicates that poverty and poor sanitary conditions were common in this study group. Similarly, a survey in India found that the prevalence of undernutrition among children in low standard households increases two times than the children living in high standard households [40]. In the present study, we revealed that sanitary and treated water facilities were inadequate in this community. This causes the high prevalence of repeated infectious diseases still remain in the plantation sector [11–13, 41–43]. Inadequate dietary intake weakens the responses of the immune system in children and increases vulnerability to infections. Several studies have been confirmed that infections were commonly diagnosed in children suffering from undernutrition [44, 45]. Another study identified that soil of the plantation sector of Sri Lanka has been contaminated with helminth eggs [46]. When handling contaminated soil, water or food those infective stages of micro-organisms can enter into the body and initiate diseases. This association between undernutrition and infection generates a cycle of repeated illness and deteriorating nutritional status. Critical nutrition specific interventions include promoting micronutrient supplementation, breastfeeding practices, access to clean drinking water and good sanitation practices and reducing the incidence of underage marriage of girls and delivering children with low birth weights can break this cycle and enhance their nutrition status [47, 48]. Although, hospitals and dispensaries in the plantation sector provide basic medical services to the estate inhabitants, lack of qualified health staffs and medical facilities are the major difficulties faced by the health service in the plantation sector [49].

In the present study, majority of undernourished children has more than five family members. Poor allocation of household resources cannot fulfill the nutritional requirements of each member equally. Particularly, children in large family size have poor access to sufficient and quality food [50]. This is consistent with several studies in India that have identified that the large family size negatively affect for child nutritional status [51, 52]. In addition, the presence of many siblings is also a risk factor for undrenutrition due to limited availability of time for care and feeding of each child [53]. Another study has reported that the food availability of children in large families were lower than the children in small families [54]. Birth order affects the attention of mothers to intra household activities and child care. In the present study, higher birth order has a negative impact on wasting in preschool children. This reflects the nutritional status of the first children to be better than that of last children. However, in the wide confidence interval (CI = 1.23–33.08) means that the data on wasting are too inconstant to make a reliable estimate. Families with more children have a high economic burden for the consumption of food and thus majority of young children are likely to be undernutrition [47]. Similarly, children in higher birth order (>3) were prone to undernutrition than those with low birth order in India [55]. Family planning programs promoting the reduction of the size and increasing the gap between childbirths will reduce the numbers of malnourished children from higher birth rates.

More than half of children use common latrine facilities in this community and shows a higher prevalence of underweight than those used household latrines. This might be explained by the poverty and lack of good quality latrines. Similarly, several studies have revealed a large number (>50%) of children living in tea plantations in Sri Lanka used poorly defined public toilets and open ground for their defecation [11–13]. Water purification centers were not available in this area and water from unprotected storage tanks were used for domestic requirements (i.e., drinking and washing clothes). Studies conducted in Sudan and Philippines revealed that improved water and better quality sanitation facilities positively affect, for good health conditions of children [56, 57]. Some studies have been reported that untreated water causes diarrheal diseases and positively correlated with the malnutrition of children [58, 59]. Therefore, we suggest that sanitary facilities affect the nutritional growth irrespectively of social demographic characteristics.

High household income of the family offers the opportunity to provide good quality food and more health services. Another study shows the links paternal education is a strong determinant for enhancing the household income in developing countries [60]. Household Income and Expenditure Survey in 2012/2013 reported that mean income per household in the estate sector were 30,220 (USD 201) Sri Lankan Rupees (SLR) while the national level was SLR 45878 (USD 305). However the majority of this community had less than 20,000 rupees (USD 133) per month income [18]. In this study, Underweight was more common in lower income groups than children in higher income households. This is similar to the studies conducted in India showed that children in low income groups had a high rate of undrenutrition [61]. This indicates that poverty is not the only factor responsible for undernutrition but inadequate dietary intake, infection, poor hygienic habits and the environment and low education status might contribute to child undernutrition among low income groups. Access to higher education for women and family planning programs, improvement of sanitary facilities and the socioeconomic conditions are major factors to enhance the child nutritional status of this community. However, an obvious limitation is this study was not powered for many of the comparison made and it was only powered to estimate an overall prevalence for all children 1–15 years old.

Conclusions

Parents’ education and occupation, birth order, number of siblings, family income and gender are main attributes to the undernutrition in this community. Government as well as non-government organizations need to prioritize to implement community based programs to enhance the knowledge of child health and decrease the level of undernutrition. In addition, future detailed studies are required to determine factors affecting the nutritional status of children in this study area.