Water, Sanitation and Hygiene Practices Associated with Intestinal Parasitic Infections, Diarrhoea, Undernutrition and Clinical Signs of Nutritional Deficiencies among Children in Surkhet, Dailekh and Achham districts, Nepal: a cross-sectional study


 Background: Providing universal access to safe water, sanitation and hygiene (WASH) in remote Nepal remains challenging. Efforts to improve WASH access in these areas are impeded by a dearth of information on the status of WASH conditions and its association with health and nutritional status of children in Nepal. Methods: We investigated WASH practices, nutritional status, and their association with intestinal parasitic infections, diarrhoea, undernutrition and clinical signs of nutritional deficiencies (hereafter health outcomes) during March to May 2018. Data was collected through cross-sectional survey of 1427 households, including questionnaires, observations, stool analysis, anthropometry, water quality measurements and assessment of clinical signs of nutritional deficiencies. Results: We found that 51.1% of children were suffering from intestinal parasitic infections and 55.5% had undernutrition; 52.2% had waterborne illnesses; and 63.9% had clinical signs of nutritional deficiencies. Multivariate mixed logistic regression analysis revealed significant associations between health outcomes and intermittent water supply (adjusted odds ratio (AOR)=2.72, 95% confidence intervals (CI)=1.18-6.31, P =0.02), washing hands when they look dirty (AOR=0.47, 95% CI=0.32-0.71, P =0.01), handwashing after going to toilet (AOR=0.37, 95% CI=0.13-1.02, P =0.06), cleanliness of the toilet (AOR=0.68, 95% CI=0.47-0.98, P =0.04), presence of earthen floor (AOR=2.29, 95% CI=1.20-4.37, P =0.01), caregivers who can read (AOR=4.07, 95% CI=1.00-16.55, P =0.05), no toilet (AOR=6.12, 95% CI=1.08-14.25, P =0.04), cleanliness of caregivers hand (AOR=0.61, 95% CI=0.41-0.89, P =0.01), regular deworming (AOR=0.44, 95% CI=0.20-0.94, P =0.03), own food production (AOR=0.67, 95% CI=0.46-0.97, P =0.03), animals inside the house overnight (AOR=1.71, 95% CI=1.17-2.51, P =0.01), source water quality (AOR=10.44, 95% CI=1.61-67.4, P =0.01), better socioeconomic status (AOR=0.43, 95% CI=0.25-0.75, P =0.01), and food supplements (AOR=0.57, 95% CI=0.38-0.84, P =0.01). Conclusions: Findings suggest improvements in WASH services, along with household hygiene and nutritional interventions, may together reduce child morbidity and mortality in Nepal.

Conclusions: Findings suggest improvements in WASH services, along with household hygiene and nutritional interventions, may together reduce child morbidity and mortality in Nepal.

Background
Globally, an estimated 785 million people lack basic water services and at least 2 billion people use a water source contaminated with faeces [1]. In addition, 4.2 billion people lack safely managed sanitation and 673 million practice open defecation [1]. Only 28% of the people in low-income countries have access to soap and water for handwashing on their premises [1]. A total of 297,000 WASH attributable diarrhoea deaths occurred yearly among children under 5 years, representing 5.3% of all deaths in this age group [2].
Likewise, about 2.4 million deaths annually are due to diarrhoea, subsequent malnutrition, and their consequences [3]. Children in low-income settings are particularly vulnerable to WASH related diseases [4].
Insufficient WASH, inadequate nutrition, and subsequent infections, as well as growth and development impairments, are interrelated problems among children in low-income countries [5]. Globally, every fifth child under-five is stunted, one of 13 children underfive is wasted and every seventh child under-five is underweight. Nearly 90% of these cases occurred in South Asia and Sub-Saharan Africa over the past decade [6][7][8]. An association between improved WASH and better growth outcomes in children has been reported in various studies. For instance, a study from Peru conducted among children two years of age found that the surveyed children living under the worst WASH conditions were 1.0 cm shorter than their counterparts living under the best conditions [9].
Furthermore, a study conducted among children less than four years old in Bangladesh provided evidence that children living with better WASH conditions, including handwashing facilities, had a height-for-age score (HAZ) 0.54 greater than children living under worse WASH conditions [10].
Two billion people worldwide were infected with intestinal parasites, with the highest burden of disease among children and their caregivers living in resource-poor settings [11][12][13]. Studies have shown that infections with intestinal parasites among children were associated with stunting, physical weakness and low educational performance [12,14,15]. Nutrition is also closely interlinked with such multifactorial determinants [16,17].
While malnutrition is directly associated with insufficient dietary intake, underlying contributing factors such as lack of access to safe water and sanitation result in recurrent infectious diseases, e.g. diarrhea and intestinal worms, which interfere with the digestive process by competing with the host for nutrients, inhibiting the absorption of nutrients and leading to compromised immunity. It is estimated that up to 45% of global malnutrition related child deaths annually could be prevented by improving WASH conditions and practices [8,18,19].
Nepal is a low-income country with a population of 28 million people, out of which 80% reside in rural areas [20,21]. The Government of Nepal set a national target to provide basic levels of water services, as well as access to improved sanitation for all by 2017 [22]. Even though, 89% of the total population currently has access to at least basic water supply services and 62% to basic sanitation facilities, numerous challenges to sustain these achievements and to make WASH services available to the currently unserved people exist. In addition, various studies revealed that water quality at the point of distribution is frequently poor [23][24][25]. Even if the water is safe at the source, it is likely to be subject to recontamination due to inadequate hygiene during transportation and storage. Nevertheless, few households consistently practice water treatment [23,26].
Also, open defecation remains prevalent and sanitation solutions mainly consist of simple pit latrines that pose a significant risk of faecal contamination in the household environment [8]. Risks related to poor hygiene are manifold; foods are unhygienically prepared, stored, and consumed in environments contaminated with flies [27]. Floors are mostly made of mud painted with cow dung. Kitchen wares and clothes are unhygienically handled and washing hands is rarely practiced [27]. Unsafe water is used for household chores and latrine surfaces are not frequently washed, increasing the presence of pathogens that may affect the health of the children directly or indirectly [28]. This results in an increased risk to different waterborne diseases, especially among children [22]. The Demographic and Health Survey (DHS) of 2016 showed that one out of 25 children in Nepal die before reaching the age of five and almost 3,500 die yearly from preventable causes [29,30]. Diarrhoea is one of the most common illnesses among children and continues to be a major cause of childhood morbidity and mortality [29].
In a recent study, 31.5% of children in the Eastern region of Nepal were found to be infected with intestinal parasites. Not using soap after defecation, the habit of thumb sucking and not wearing sandals were found to be significantly associated with parasitic infections [31]. The DHS in 2016 revealed that 36.0% of children under-five in Nepal are stunted, 10% wasted and 27% underweight. Stunting was found to be more common in rural children (40%) and the poorest households (49.0%) [29]. Fifty three percent of children aged 6-59 months are anaemic, a condition which is more common (56.0%) among rural children [29]. Only one out of four children sees a health care provider for the treatment of illnesses [32]. The treatment costs of the WASH-associated disease burden and consequences therefrom may further increase the risks, by affecting households' budgets and thereby limiting the amount of food available, exacerbating the risks of insufficient nutrient intake and undernutrition among children [33].
The "Sustainable Development Goals (SDGs)" recognised unsafe water and inadequate sanitation and hygiene as a major cause of child morbidity and mortality, and SDG 6 was developed to address these challenges [34]. The data required to monitor the SDGs is lacking in many countries [1] and this is also the case in Nepal where there is a dearth of information on the status of WASH conditions in remote areas. Likewise, the health and nutritional status among children in remote hilly areas and their associations with WASH has not been investigated in this locale. As a result, efforts to combat health and nutritional problems among children in these settings do not effectively incorporate WASH interventions. The aim of this study, therefore, is to assess the influence of WASH practices on the health and nutritional status among children aged 6 months to 10 years of age in three districts in the rural hilly areas of Nepal. Findings from this study are crucial to support and inform efforts to deliver WASH, health and nutritional interventions, including hygiene behavioural change initiatives, in the study areas.

Study areas
The study areas were located in the Districts of Surkhet, Achham, Dailekh in the Karnali For our study, all areas in the selected districts were very remote, difficult to reach and lacked access to proper infrastructure, such as an electrical grid.
Study design, sample population, sample size and sampling methods The study sites were located within the Helvetas Swiss Intercooperation Nepal, Water Resources Management Programme (WARM-P) sites. The sites were selected by the local teams of an organisation following a set of criteria: (a) extreme remote location (b) is an area with the availability of a piped water supply scheme, that has not yet received the WARM-P training intervention (c) not having access to products for household water treatment. The database of the households having children's aged 6 months up to 10 years in the targeted communities were prepared. We then conducted a cross-sectional study from March to May 2018 that involved 1427 households. A minimum of 345 households were randomly sampled from each of the districts (Surkhet A, Surkhet B, Dailekh and Achham) in the Karnali province of Nepal. If a household declined to participate, then the neighbouring household having children in the targeted age ranges were included. Sample size and statistical power were calculated using G*Power 3.1. A sample size of 300 households were required in each of the four areas to detect an effect in Cohen's f2 at one-tailed alpha of 0.05 and a statistical power of 90% with multiple logistic regression and 15 predictor variables adjusting for clustering [38,39]. We therefore enrolled a minimum of 345 households from each of the three districts, for a total of 1427 households.

Questionnaire survey
A quantitative structured questionnaire was administered to the caregivers (male or female) in the households. The questionnaire contained mostly closed ended, multiple choices questions with Likert scale answer categories. The interviews that took around 15 minutes in on average were complemented by structured observations. The questionnaire was coded in "Open Data Kit (ODK)" software [40] on tablets (Samsung Galaxy note 10.1 N8010) and contained questions on the use of water sources, psychological factors concerning water handling and hygiene practices, observations of WASH infrastructure, information on WASH promotion activities received, and children's history of waterborne illnesses in the past seven days, including potential confounders, such as socio-demographic and economic factors. The questionnaire was developed using standardized questions in line with international guidelines, incorporating necessary amendments as per our study areas [41,42]. The questionnaire was developed in English and translated into Nepali, with a back-translation. Interviewers were intensively trained for two days on data collection procedures. Pretesting was done in early March 2018 with households in an area with similar socioeconomic conditions as the study areas. Detailed interviews were conducted with caregivers during the pretesting to ascertain the reliability of the questions used in the final survey instrument [43]. Adaptations to the questionnaire were made after pretesting.
Child diet and household food security Child dietary information was assessed following the guidelines of the Food and Agricultural Organisation (FAO) [42]. The caregivers were requested to recall whether or not the nine different food groups were consumed within the past seven days and, if consumed, the frequency of consumption. The survey methods were adapted by conducting detailed interviews during pre-testing to ascertain accuracy. In addition, supervisors re-interviewed a subset of surveyed households to assess reproducibility [43].
Household food security was assessed by questions relating to the adequacy and availability of food during the entire year.

Anthropometric measurements
Trained paramedics collected anthropometric measurements, adhering to the standard procedures [44]. Supine lengths were obtained for children younger than two years old (using a Seca BabyMat 210). For children aged more than 24 months to 10 years, a height measuring board and a digital scale (Seca 877; Hamburg, Germany) were used with a precision of 0.1 cm and 0.1 kg, respectively. Before each measurement, scales were calibrated with a standardized weight and the children were measured barefoot [43].
Anthropometric indices were calculated using the AnthroPlus (WHO; Geneva, Switzerland) in accordance with the World Health Organisation (WHO) [45,46]. Age in total months was recorded for each child and confirmed with an inspection of the birth certificate. Three anthropometric indices: (a) weight-for-age (WAZ, underweight); (b) height-for-age (HAZ, stunting); and (c) body mass index-for-age (BMIZ, thinness) were expressed as z-scores (i.e. differences from the median in standard deviations) [8]. Z-scores of ≥ -2 were regarded as normal, those between < -2 and ≥ -3 as indicating moderate undernutrition and those below < -3 as severe undernutrition [44,45]. Children were considered to be undernourished if at least one of the anthropometric indices indicated undernutrition.

Parasitological survey
On the day following the household survey visit, caregivers were requested to provide a fresh morning stool sample without urine contamination from the participating child. The samples were processed on the same day by two experienced laboratory technicians. Each stool sample was analysed using direct wet-mount and formalin-ether concentration techniques to detect intestinal protozoa and helminths following standard guidelines [47][48][49]. In addition, a duplicate Kato-Katz thick smear was prepared for the diagnosis of Signs and symptoms of nutritional deficiencies During the household survey, the children were screened for clinical signs and symptoms indicating nutritional deficiencies, such as wasted appearance (protein deficiency), loss of hair pigment and easy "pluckability" (protein deficiency), bitot's spot (vitamin A deficiency), dry and infected cornea (vitamin A deficiency), oedema (thiamine deficiency), several types of dermatitis (niacin deficiency), spongy, bleeding gums (vitamin C deficiency), pale conjunctiva (iron-deficiency anaemia), red inflamed tongue (riboflavin deficiency), sub-dermal haemorrhage (vitamin C deficiency), bowed legs (vitamin D deficiency), and goitre (iodine deficiency) [53].

Drinking water quality examination
Water samples were collected from the household's main drinking water source (point of collection) and from the container used for drinking water transport and storage (point of use). For the water quality examination at the point of use, the caregivers were requested to bring fresh drinking water in situ from the point of collection to the household in the same container they used to fetch their drinking water daily [54]. The sample at the source was taken after first, letting the water run for 60 seconds from the tap, and then filling a sterile Nasco Whirl-Pak bag. Back at the household, a second water sample was taken by pouring water from the transport container into another sterile Nasco Whirl-Pak bag. All water samples were stored inside cooler bags and immediately analysed at the field site using the membrane filtration technique, as follows: 100 ml water samples were passed through sterile 0.45 µm millipore cellulose membrane filters, using sterilized filtration equipment. The filter pads were plated on Nissui Compact Dry Coli-scan plates and incubated for 24 hours at 35 +/-2°C. A solar energy based electrical power supply system was set-up to run a low-power incubator at the field-site. Colony forming units of total coliforms and E. coli were counted after 24 hours of incubation.

Data management and statistical analysis
Data cleaning was performed daily during the survey and if there was any missing value or inconsistency, the respective household was consulted the following day. Readings of intestinal parasites and nutritional deficiencies screening results were double entered into an Excel 2010 spread sheet (Microsoft; Redmond, USA) and cross-checked. Numerical variables were described by means and standard deviations if normally distributed and by medians and interquartile range, otherwise. Categorical variables were described by absolute and relative frequencies. We employed χ 2 statistics to assess the differences in distribution for categorical variables between the study areas. Household socioeconomic status was characterised by factor analysis, which included the reported household assets, such as the availability of electricity, radio, television, solar panel, mobile phone, bicycle, motorbike, fridge, watch, own house, and land ownership. Three factors reflecting three different socioeconomic domains were retained, and divided into three categories: (a) low, (b) medium and (c) high, using the k-means procedure [55]. The median of the score from factor analysis was categorised as poor, medium or better depending on whether the scores were below, at or above the medium. The same procedure was applied to create one variable for the hygiene of drinking water transport and water storage containers, latrine hygiene, the household environment, kitchen and the personal hygiene. Of these, three factors were retained and categorised, reflecting poor, moderate and better status.
We assessed four outcome variables: (a) intestinal parasitic infection, (b) diarrhoea, (c) undernutrition including stunting, thinness and underweight, and (d) clinical signs and symptoms of nutritional deficiencies. The associations between the outcome variables and risk factors were assessed first by conducting univariate mixed logistic regression analyses with random intercepts of the study areas. Since only a few undernutrition cases were of a severe nature, the cases were pooled into a binary variable of stunted/nonstunted, and underweight/non-underweight, for the subsequent analysis. Similarly, there was a low prevalence of parasites, such as T. trichiura, E. vermicularis and hookworm, so all reported intestinal parasitic infections were pooled into a binary variable of parasite infection/no infection to maximize statistical power. The clinical signs of nutritional deficiencies and diarrhoea outcomes were already coded into binary variables and were used as such for the subsequent comparative analysis.
We assessed associations between the binary outcome variables and hypothesized risk factors using a multivariate mixed logistic regression model with random study area intercepts and controlling for demographic variables, such as age, sex and socioeconomic status, as potential confounders. The variables with P<0.2 in univariate analysis (using likelihood ratio test) were included in the final multivariate model. Odds ratios were reported and the associations were considered as statistically significant if p-values were <0.05. The statistical analyses were done with STATA version 14 (STATA Corporation, College Station, TX, USA).

Socio-demographic characteristics of the study participants
The details of the socio-demographic and socioeconomic characteristics of the interviewed households are provided in Table 1. Interviewees were roughly equally distributed across the four areas (Table 1). Among the 1427 interviewees interviewed, 25.0% were from Surkhet A, 25.1% from Surkhet B, 23.3% from Dailekh and 26.6% from Achham District.
Caregivers aged 25-40 years of age constituted the largest group (57.9%), whereas those aged >40 years constituted the smallest group (13.9%). More than 80% of the caregivers could both read and write. More than half (64.5%) of the survey participants had 5-10 people residing within their household. Agriculture was the main (60.6%) occupation of the household heads. A total of 59.7% of the households kept animals inside the home overnight. A majority (84.1%) of the households had earthen floors. The majority of children (99.1%) included in the study were between 6 months and 5 years (<5 years) of age, while 0.9 % were age 6 to 10 years (>5 years). The number of children per household in the age range surveyed varied between one and six. Around 52.7% of the households across the study areas had access to electricity. The study areas significantly differed in terms of the distribution of sex, age, education, number of people residing in the household, animals kept inside the household, and the type of floor materials used in the household (P = 0.01).
Water handling, water quality, sanitation, hygiene and WASH promotion Tables 2 and 3 describe water handling, water quality, sanitation and hygiene practices, WASH infrastructure, WASH promotion and WASH related psychological factors in the households stratified by the four study areas. 75.5% of the respondents depend on a communal village tap for drinking purposes, 20.7% have access to piped water in the house or yard, and the remaining respondents use open water sources (1.7%), unmanaged pipe systems (0.3%) and the river (0.3%). More than half (54.4%) of the respondents were quite confident about the safety of the available drinking water. Only 16.5% of the households reported treating their water at the point of use. One third (33.7%) of the respondents reported disliking the taste of treated water. Virtually all of the respondents (99.0%) cleaned their containers used for drinking water transport and storage at least every other day. Soap or ashes were used for cleaning these containers (42.2% vs. 47.4%, respectively) ( Table 2).
Regarding water quality, we found that the vast majority of water samples from the point of collection and point of use were contaminated with E. coli (93.6% and 95.3%, respectively) and total coliform bacteria (99.4% and 98.8%, respectively). 5% of the water samples at the point of consumption confirmed with WHO's guidelines for microbial safety of drinking water (<1 CFU E. coli/100 mL), 16% were in the low risk category (1-10 CFU E. coli/100 mL), 51% in the intermediate risk category (10-100 CFU E. coli/100 mL) and 28% in the high or very high risk categories (>100 CFU E. coli/100 mL) [56]. Further details of the microbial water quality at the source and the point of use are presented in Table 2.
Concerning sanitation and hygiene, 6.3% of the households did not have latrines, and 93.7% of the households had traditional pit latrines. Almost half of the latrines (48.7%) were in a poor hygienic state (lowest category). 76.0% of the respondents reported having washed their hands with soap less than five times per day prior to the day of survey. The overall hygiene conditions were very low or low in 64.0% of the surveyed households. In addition, 42.8% of the caregivers and their children had a low personal hygiene score (Table 3).
Only 10.9% of the respondents reported having received information on water treatment and hygiene. Among those, 89.7% reported that this information changed their WASH behaviour, e.g. using soap more often for washing hands. About 52.2% of the respondents reported door to door visits as an effective means of hygiene literacy promotion (Table 3).
Child health, health-seeking behaviours and KAP survey Table 4 shows the percentage distribution of child health records, health-seeking behaviours, and risk awareness of the causes of diarrhoea and of perceived vulnerability stratified by study areas. A total of 49.9% of the children from under age 6 months to five years were reported to be sick within seven days prior to the survey. Respiratory illnesses and fevers were most common (both 40.4%) followed by diarrhoea (16.5%). For respondents not seeking health care services, the main reason was a feeling of non-requirement (61.9%), followed by financial constraints (21.2%), preference of selftreatment (7.1%), and lack of accessibility (1.8%) (  Tables 4 and 5 show the prevalence of diarrhoea and associated risk factors, respectively, in the study sample. A total of 16.5% <5 years children's suffered from diarrhoea within the seven days prior to the survey. The results from the multivariate regression analysis demonstrated that children above the age of five showed significantly lower odds of diarrhoea (adjusted odds ratio (AOR) = 0.39; 95% CI: 0.26-0.57; P = 0.01) compared to their younger counterparts. The children from the households experiencing a service interruption of more than one week of their main drinking water source at the time of visit had 2.87 higher odds of suffering from diarrhoea (AOR = 2.87; 95% CI: 1.24-6.61; P = 0.01) than children from households not experiencing an interruption. The children of caregivers who were aware of the need for handwashing during critical times, such as after using the latrine, were significantly protected against diarrhoea (AOR = 0.37; 95% CI: 0.14-1.10; P = 0.05) compared to children whose caregivers were unaware. Households with clean latrines were significantly protected against diarrhoea (AOR = 0.66; 95% CI: 0.45-0.95; P = 0.03) compared to households with dirty latrines. Similarly, the children with visually clean hands without spots of dirt were significantly protected against diarrhoea compared to children with dirty hands (AOR = 0.63; 95% CI: 0.41-0.98; P = 0.04). Children living in households with a floor made of "earth painted with cow dung" had 2.29 times higher odds of suffering from diarrhoea than children living in households with a cement floor (AOR = 2.29; 95% CI: 1.20-4.37; P = 0.01).

Prevalence of intestinal parasites and associated risk factors
A table 6 and 7 shows the prevalence of intestinal parasitic infections in the study population and the associated risk factors. The overall prevalence of intestinal parasitic infection is 51.1%.The prevalence of intestinal parasitic infections, stratified by sex, age group and study area, are summarised in Table 6. The predominant helminth species infecting the children was Ascaris lumbricoides (21.1%), followed by Hymenolepsis nana (4.6%), hookworm (3.2%), Enterobius vermicularis (2.7%), and Trichuris trichiura (0.7%).
Polyparasitism and co-infection were not common. Significant differences across the study areas were observed for all recorded helminths except for Trichiuris trichiura (P = 0.23).
Infections with all types of helminths were of light intensity. About 23.4% of the children were infected with Giardia intestinalis (Table 6).
Multivariate analysis showed that the odds of infection with intestinal parasites among children in households where the caregivers could only read were four times higher than the odds among children whose caregivers could both read and write (AOR = 4.07; 95% CI: 1.00-16.55; P = 0.05). Children in households relying on a simple pit latrine for defecation had six times higher odds of being infected with intestinal parasites than those in households with a pour flush pit latrine (AOR = 6.12; 95% CI:1.57-23.9; P = 0.01).
Children with caregivers having clean hands were significantly protected from intestinal parasitic infection (AOR = 0.61; 95% CI: 0.41-0.89; P = 0.01) compared to children of caregivers with dirty hands (Table 7). Child-feeding practices and household food security Almost all caregivers (99.6%) reported having breastfed the participating child. Only 1.3% of the children were breastfed less than six months. About 97.0% of the children received weaning food at the age of five months for girls and at the age of six months for boys (during the traditional rice feeding ceremony in Nepal). About 43.3% of the children received food supplements in addition to regular meals. The dietary diversity scores were low with only 11.2% of the households having consumed all listed nine food groups in the past seven days (Table 8 and Additional file 1: Supplementary Table 1). The consumption of milk or milk products and eggs at least once per week was low (9.2% and5.3%, respectively) (Additional file 1: Supplementary Table 1). About 60% of the households never produced their own food and among the households that did produce their own food, 20.8% reported self-sufficiency throughout the entire year. All the analysed variables were significantly different across the study areas.
Prevalence of undernutrition and associated risk factors Table 9 shows the percentage distribution of undernutrition in the study sample stratified by sex, age groups and study areas. The prevalence of undernutrition was high at 55.5%, while the prevalence of stunting was 44.5%, thinness 11.2%, and underweight 29.9%. Table 10 provides an overview of the association between undernutrition and the caregiver's socioeconomic status, nutrition and health awareness, attitude and practices, intestinal parasitic infections, hygiene status, and water quality of the households Prevalence of nutritional deficiencies and associated risk factors A total of 63.9% of the children in the study suffered from at least one sign of a nutritional deficiency. About one third (35.9%) of the children suffered from pale conjunctiva) a sign of iron deficiency and anaemia; followed by Bitot's spot (19.8%), a sign of vitamin A deficiency; red inflamed tongue (18.3%), a sign of riboflavin deficiency (Vitamin B2); spongy bleeding gums (16.3%), a sign of vitamin C deficiency; wasted appearance (13.8%), a sign of protein deficiency; dry and infected cornea (13.2%), a sign of vitamin A deficiency; loss of hair pigment (10.7%), a sign of protein deficiency; sub-dermal haemorrhage (4.6%), a sign of vitamin C deficiency; oedema (2.7%), a sign of thiamine deficiency (Vitamin B1); bowed legs (2.6%), a sign of vitamin D deficiency; and goitre (0.6%), a sign of iodine deficiency (Table 9). The children >5 years had twice the odds of having nutritional deficiencies compared to their younger counterparts (AOR = 1.84; 95% CI: 1.30-2.62; P = 0.01). Children whose caregivers washed their hands after cleaning a baby's bottom were at lower odds of having nutritional deficiencies (AOR = 0.60; 95% CI = 0.40-0.92; P = 0.02) compared to children whose caregivers did not follow this practice. Children whose caregivers were aware of the benefit of wearing shoes were at lower odds of having nutritional deficiencies (AOR = 0.24; 95% CI: 0.09-0.62; P = 0.01) compared to those whose caregivers were unaware. Children from households producing their own food were significantly protected against nutritional deficiencies (AOR = 0.51; 95% CI: 0.35-0.76; P = 0.01) compared to children from households without their own food production. On the other hand, children from households in the higher category of latrine hygiene were at higher odds of nutritional deficiencies than latrines without any hygiene (AOR = 0.61; 95% CI: 0.41-0.91;  (Table 11).

Discussion
Our findings highlight several important health and WASH issues in remote areas of rural Nepal. Household and personal hygiene, including adequate handwashing, appropriate sanitation and access to regular and safely managed drinking water supply in the surveyed area, were inadequate. These poor WASH conditions were associated with diarrhoea, parasitic infections and nutritional deficiencies, which were highly prevalent among the surveyed children. More than half of the children in the study area suffered from undernutrition. This alarming health outcome was mainly linked to the low socioeconomic status of the household, poor nutrition, and to the lack of regular deworming activities, but undernutrition was not significantly associated with poor WASH conditions. The high prevalence of intestinal parasitic infections among children (51.1%) found in our study is similar to or higher than the rate reported in studies conducted in other parts of Nepal [30, 31, 57]. The higher infection rates could be explained by the fact that our study areas were located in extremely remote rural and hilly areas where accessibility was difficult and there was a lack of infrastructure, which together result in a low level of access to basic health and WASH services [30, 31, 57]. Our analysis revealed that children from households without latrines developing a parasitic infection relative to those with latrines were four times likely to have infected with an intestinal parasites. This indicates that the poor hygienic status observed in the pit latrines used by more than 94% of the households and lack of water to flush the toilets is a major risk factor in the transmission of parasites. The cleanliness of caregivers' hands was identified as a significant risk factor for children's parasitic infections, suggesting that caregivers' hands play a critical role in transferring parasites from the household environment to their children. There is strong evidence that a high load of pathogens in the household environment and inadequate handwashing cause an increased pathogen load on caregivers' hands [60]. We observed poor handwashing conditions, with 59.9% of households not having adequate infrastructure to wash hands and a limited presence of soap/water at the handwashing stations. Handwashing practices were largely insufficient, with 76% of the caregivers stating that they wash their hands with soap less than five times per day. The importance of clean hands to prevent parasitic infections is in agreement with previous studies Surprisingly, despite the low quality of the water consumed, we did not find any association between drinking water quality and parasitic infections in the study households. We hypothesize that the transmission of pathogens through drinking water was minor compared to the high loads of pathogens existing in the household environment due to unhygienic sanitation, the practice of painting the earthen floors with cow dung and the presence of animal feces in and around the house. Contrary to this, two other recently conducted studies in Nepal identified the consumption of untreated water as a risk factor for infection with Giardia species [30,65]. These studies, however, may have been conducted in a different, hygienically less challenging context.
We also examined the association between intestinal parasitic infections and different socio-demographic variables. Children belonging to households where caregivers could read but not write, or had not received any education, were at higher odds of intestinal parasitic infections compared to children whose caregivers could both read and write. Interruptions of the water supply were associated with diarrhoea. Caregivers who reported more frequent interruptions to their drinking water systems that lasted more than a week, had 2.7 times higher odds of having children with diarrhoea. Underlying reasons could be the subsequent lack of water for hygienic practices or it could be that intermittent water services exhibit an increased risk of bacterial contamination in the system or it could be that the risk of pathogen infiltration is greater during such low pressure events in the piped network [77,78]. Past research has reported that the households with the presence of E.coli in their drinking water were 3.6 times more likely to have children with diarrhoea than those that had water in compliance with WHO's guidelines for drinking water quality Our study found a high prevalence of undernutrition (55.5%) with stunting alone being at 44.5%. A reason for this could be that the study areas are located in food-scarce regions where agricultural activity has been affected by the impact of climate change, i.e. untimely heavy rainfall, droughts and hailstorms. Rural households in these areas are generally poor and their resilience and coping abilities to deal with reduced agricultural productivity is low. The vulnerability of low-income households towards increasing prices for food items has been demonstrated by Green et al [80]. In our multivariable mixed logistic regression analyses, undernutrition was significantly associated with a household's lower socioeconomic status, lower production of one's own food, lower provision of supplemental food to children and irregular deworming activities.
Although the relation between undernutrition and parasitic infections is not well understood, undernutrition may be caused by recurring infections in the gut, which limit the proper absorption of calories and nutrients [81,82]. Our findings on the association between undernutrition and intestinal parasitic infection are in agreement with studies conducted elsewhere [82,83]. However, contrary to findings from a study conducted in Bangladesh, our study did not identify recent diarrhoea infection as a risk factor for undernutrition [84]. As we collected data during a cross-sectional survey, we do not have any longitudinal information on the frequency and severity of diarrhoea cases occurring in the study population. We think that there is a high chance that chronic diarrhoea, as well as environmental enteropathy, are linked with undernutrition in our study site, but our data only refers to one point in time and, thus, cannot be confirmed [85-87]. WASH indicators were only significantly related with undernutrition in univariate analysis, but not in the multivariable logistic regression models. In univariate analysis, we found that low hygiene score in the kitchen, low personal hygiene of the caregiver and the child, and the presence of E.coli or total coliforms in the water source were associated with undernutrition.
Even though WHO recommends to start providing infants food in addition to breast milk from six months onwards [88], 97% of the children participating in the study received weaning food before this age. Multivariable mixed logistic regression analyses revealed considerably higher odds of undernutrition among children who were introduced with weaning food at an age of less than 6 months; however, the association was not significant. Nepal has a culture of starting to wean girls at five months, while boys usually receive weaning food at six months of age. The weaning practices before reaching six months of age may have caused some food borne infections and environmental enteropathy, resulting in nutritional deficiencies and undernutrition since the food provided to the infants was likely to be contaminated [82,85]. We observed unhygienic handling and improper storage of food in the study areas-96% of the households did not have a refrigerator. Unsafe water was used to wash feeding and storage containers, unhygienic kitchen clothes were used to dry children's utensils and caregivers did not wash their hands with soap while preparing and feeding children. . We found a significant protective association between children (six months to ten years) who received food supplements in addition to regular meals. This is in line with other studies that found a significant impact of nutrition on growth [92, 93].
However, surprisingly, we did not find a significant association between dietary diversity scores and undernutrition, which is in contrast to a study conducted in Indonesia that reported of higher dietary diversity scores associated with lower likelihood of child stunting [94]. We assume that several confounding factors, such as household environment could mediate the effect of dietary diversity on undernutrition status, indicating the need for more depth research in this study population.
The prevalence of having at least one clinical sign for a nutritional deficiency was high (63.9%). Because there is a dearth of studies conducted on children having clinical signs of nutritional deficiencies in Nepal and other similar countries in terms of geography, income level of its population, literacy level and life expectancies, our results cannot be compared with other studies. The most frequently encountered nutritional deficency, pale conjunctiva (Niacine deficiency), which was found in 36.0% of children, can be related to the lack of animal proteins in diets.
Contrary to our results regarding risk factors associated with undernutrition, clinical signs of nutritional deficiencies were significantly associated with WASH conditions. Our analysis identified a significant protective association with handwashing, improved latrine cleanliness and lower number of total coliforms in the drinking water source. In contrast, signs of nutritional deficiencies were positively associated with keeping animals inside the house and the low personal hygiene of caregivers and of the children. Further in-depth research is required to provide more insight into these issues.
Our study has several limitations. First, this study is a cross-sectional survey conducted was not identified as a risk factor for undernutrition. To assess this association, we propose to implement a controlled longitudinal study design that involves the collection of data on recurring chronic diarrhoea and which would measure markers of environmental enteropathy. Second, we consider some of our outcome variables such as diarrhoea to be subject to a seasonal effect [8,95,96]. Third, the findings presented in our study cannot be generalised to other rural areas of Nepal, as the study was conducted in an extremely remote setting characterized by exceptionally low access to basic services. Third, an anthropometric survey has certain limitations related to the inaccuracy of children's dates of birth [95]. Although, we checked available birth certificates to validate the reported ages, these certificates were not always forthcoming. Fourth, only one stool sample per participating child was examined by a double Kato-Katz thick smear, which likely led to an underestimation of the true prevalence of parasitic infections due to the low sensitivity of the Kato-Katz technique. Fifth, the information we obtained during the interviews was selfreported by the caregivers and may be subject to recall and respondent bias [8,97].

Conclusion
This study provides new insight into the burden of intestinal parasites, childhood waterborne illnesses (i.e. diarrhoea), undernutrition and nutritional deficiencies among children living in extremely remote hilly areas of Nepal. Our study shows that these health measures were all are highly prevalent in the study areas and that drinking water quality is poor. We found that all health outcome measures (intestinal parasitic infections, childhood waterborne illnesses, undernutrition and nutritional deficiencies) showed strong negative associations with household hygiene, such as appropriate sanitation, adequate handwashing, good personal hygiene, and access to safely manage drinking water supply, as well as the provision of adequate nutrition. In view of these findings, concerted efforts are needed to improve nutrition and household hygiene, e.g., by improving infrastructure, water quality and behaviour supporting appropriate sanitation, adequate handwashing, and good personal hygiene. We also suggest that greater attention should be given to