Introduction

Intestinal parasitosis affects approximately 1.5 billion people worldwide, including 654 million school-age children1. Overall, the most common intestinal parasitic infections (IPIs) are Ascaris lumbricoides, Trichuris trichiura, hookworm, and Strongyloidesspp. which are also referred as soil-transmitted helminths (STHs) due to their transmission through soil. STHs infections are widely distributed in tropical and subtropical areas, with most infected people living in developing countries2. Poor environmental and personal hygiene and sanitation and contamination of food and drinking water from improper disposal of human excreta are suggested to be major contributors to the high prevalence of STHs3,4,5.

A high prevalence of STHs is widely reported in Nepal, particularly among school-aged children6,7,8. These infections can cause malnutrition and poor growth in children9. STHs obtain nutrients needed for survival via the gastrointestinal tract. Associations with malnutrition are possibly mediated through impaired fat digestion, reduced vitamin absorption (particularly vitamin A), and temporary lactose intolerance10. Suggested mechanisms for the effects of undernutrition include appetite suppression, increased nutrient loss, and decreased nutrient absorption and utilization11,12. Physical damage to the gut is also common, e.g., villus atrophy/hyperplasia, disruption of the mucus barrier and chronic inflammation due to increased gut permeability. Undernutrition is widespread in Nepal; one in three adolescents is stunted (height-for-age z score <-2 SD), one in four boys is undernourished (BMI-for-age z score <- 2 SD), and one in five adolescent girls has anaemia13.

According to Engel et al14, approximately 200 million children, predominantly from low- and middle-income countries, do not attain their complete cognitive potential, a circumstance deemed in contravention of the child’s rights (United Nations Convention on the Rights of the Child: General Comment No. 7)14. A large population-based cohort study in India also reported cognitive deficits among undernourished children15. A recent scoping review of 26 articles also indicated the significant impact of being underweight on cognition16. Regarding the pathway linking STHs and cognition, epidemiological research has shown that infected school-age children exhibit poorer health outcomes than their uninfected counterparts. These outcomes include a greater incidence of lethargy17and anaemia18,19. Furthermore, studies have indicated a heightened prevalence of school absenteeism17, along with diminished performance across various cognitive assessments20. For example, studies have suggested that STHs are associated with poor performance at school21,22and low IQ9. Considering the likely impacts of STHs on learning ability via morbidity pathways, the WHO recommends monitoring STHs among school-aged children.

Child neglect is not classified as a crime in Nepal, and while severe child abuse is prohibited, it is not categorized as a state offense23,24. Additionally, minor instances of physical discipline administered to a child by parents or other family members do not constitute a breach of the law25. Conversely, the detrimental impact of exposure to an unfavorable environment has been suggested to have adverse impacts on neurodevelopment26. Therefore, the impact of adverse childhood experiences (ACEs) is anticipated to be comparatively more pronounced in Nepal.

We previously reported the adverse effects of in utero exposure to toxic elements on neurodevelopment in infants27. However, such detrimental effects did not persist at 6, 24 or 36 months of age, while a study showed an association between neurodevelopment and the home environment28,29,30. The effect of STHs on cognition, together with widespread undernutrition and adverse childhood experience (ACE), at a later age was not previously evaluated. Hence, this study aimed to investigate the influence of STHs and nutritional status on the IQ of birth cohort adolescents in the Chitwan Valley, Nepal, while adjusting for ACEs and other sociodemographic characteristics as potential confounders.

Methods

Study overview and timeline

From September to October 2008, two hundred pregnant women who decided to deliver at the hospital were approached at Bharatpur General Hospital in the Chitwan district. The detailed eligibility criteria (in brief, living in the study area for at least 2 years, at term pregnancy, age 18–40 years, singleton pregnancy, and no reports of diabetes, hypertension, or preeclampsia) are delineated elsewhere27,31. Of the potential participants evaluated for inclusion, 119 met the eligibility criteria, and 100 provided informed consent to participate in the Chitwan Birth Cohort study (84% participation rate)27. The study protocol received approval from the ethics committees of the Graduate School of Medicine, University of Tokyo (approval no #2244), and the Bharatpur General Hospital, Chitwan, Nepal.

The birth cohort was tracked after a span of 14 years, utilizing the available addresses of participants’ households and their contact details. The follow-up study protocol received approval from the ethics committees of the National Health Research Council (approval no. NHRC#198–2022) situated in Kathmandu, Nepal, and the University of Tokyo (approval no. 2022260NI-(1)). Between September and October 2022 and 2023, seventy-four participants were visited in their respective homes following prior appointments. The study’s background and objectives were explained to participants and their parents before obtaining informed consent and assent. Informed assent and consent were duly signed by all participants (n = 74) and their legal guardians. All procedures conducted in this study with human participants adhered to the principles of the Helsinki Declaration.

Evaluation of STHs

Participants were instructed to use a clean stick to access a sample of feces about the size of the thumb from the first, middle, and last parts of their stools early in the morning into the given sterile vial, and care was taken not to contaminate the sample with urine or soil. The collected samples were preserved in 2.5% potassium dichromate solution to preserve the parasites and transported to the Central Department of Zoology, Tribhuvan University, for further investigation.

Stool samples were analyzed using the Kato–Katz method, which provides the intensity of infection (an accurate measure of the number of eggs present per gram of stool)32. The total number of parasite-stained eggs was counted under a microscope to calculate the number of eggs per gram (epg) in a standard amount of sieved fecal sample. The Kato template was used to extract around 41.7 mg of fecal matter for each smear. To determine the quantitative results, the egg count per smear was multiplied by 24, converting the data into eggs per gram (epg) of fecal material33. For quality control, 10% of the smears were randomly selected and evaluated twice. There was no significant difference (paired t test; all p > 0.05) between the two evaluations, and the correlation between egg density and age was 0.88.

Anthropometry of participants

Body weight was accurately recorded to the closest 0.1 kg by employing a portable digital scale (Model BF-046 WH; Tanita, Tokyo, Japan). Height measurements were taken with a precision of 0.1 cm using a stadiometer. The body mass index (BMI) was subsequently computed by dividing the weight (in kilograms) by the square of the height (in square meters). The z scores for BMI (BMIZ) were determined based on age and sex according to the World Health Organization (WHO) growth reference BMI-for-age scores for 5–19 years. Participants were divided into groups of normal weight ( > − 2 SD) and underweight ( < − 2 SD) according to WHO guidelines34.

Interviews with parents and participants during home visits

Structured questionnaires were employed to thoroughly gather a comprehensive array of information through interviews conducted with both parents and participants. This included the current educational attainment of mothers and fathers in terms of years, the annual family income, educational grades, and the most recent grade point average (GPA) of participants. Furthermore, participants were requested to provide self-assessment scores for the quality of their school. Similarly, mothers were asked to provide self-assessment scores for their perceived intelligence and quality of the growth environment. All self-assessments or perceived scores were assigned values ranging from 1 to 5, with 1 representing the lowest and 5 indicating the highest in each category.

Postnatal growth environment

The authors (RPP and SB) visited the residences of each mother/child pair 14 years after childbirth (with an average time span of 14.04 ± 0.4 years from the baby’s birth). During these visits, the postnatal growth environment was assessed through the application of the Adverse Childhood Experiences Questionnaire (ACE-Q)35. The ACE-Q is a 10-item instrument (with potential scores ranging from 0 to 10) designed to quantify occurrences of adverse or traumatic experiences that an individual has encountered prior to reaching the age of 18 years. The ACE-Q assesses an individual’s exposure to various forms of childhood adversity, encompassing psychological, physical, and sexual abuse, along with instances of household dysfunction such as domestic violence, substance misuse, and incarceration35,36. The ACE-Q was individually administered by RPP to male participants and by SB to female participants, employing a self-report format to ensure privacy (owing to the potentially triggering nature of some questions related to trauma). Each affirmative response to the questions was tabulated, yielding a cumulative ACE-Q score (ranging from 0 to 10). A higher score signifies a greater number of adverse childhood experiences encountered by the individual, thereby indicating an elevated risk for issues pertaining to social, mental, or general well-being. A score of 4 or more was considered clinically significant37.

Neurodevelopmental indicators such as WASI-II IQ scores

The Wechsler Abbreviated Scale of Intelligence (WASI), introduced in 1999, offers an assessment range spanning from 6 to 89 years. The WASI II comprises four subtests—vocabulary, block design, similarities, and matrix reasoning—requiring approximately 35 min for completion. The WASI-II yields scores for the full-scale intelligence quotient with 4 subsets (FSIQ-4)38. The WASI-II INDIA (hereinafter referred to as WASI II), which furnishes a succinct yet dependable assessment of cognitive ability for application in research settings, was selected for this study. The decision to adopt the WASI-II was based on its standardization study conducted on a nationally representative sample encompassing approximately 1,540 individuals aged 6 to 90 years. Notably, this sample is reflective of India’s population and is anticipated to closely correspond to the population and norms of Nepal. The WASI-II is frequently used in the field of neurotoxicology. The WASI-II is considered one of the measures of choice for brief evaluation of intelligence39. The psychometric properties, including reliability and validity, have been well-established39in prior research using this tool in diverse populations40,41.

The WASI II was administered to the children within a span of 4 weeks from the target age group (i.e., 14 years ± 1 month, as stipulated in the WASI II manual), with the age of the children recorded in months and days. The WASI II assessment for all participants was conducted by a skilled researcher (RPP), who maintained a blinded approach to the participants’ exposure status. Rigorous adherence to standardized assessment conditions was upheld, encompassing factors such as providing advanced information about the assessment, minimizing the presence of other individuals, adopting a friendly manner while interacting with the children, ensuring moderate lighting at the assessment location, and considering the participants’ well-being (e.g., absence of hunger or sleepiness). The assessment of participants was conducted within the familiar confines of their own homes, chosen for convenience for both the caregivers and the cohort children (thus avoiding any potential hesitancy due to a less familiar setting).

Translation and training

All tests and interviews were meticulously translated (and subsequently back-translated) between Nepali and English while ensuring the inclusion of culturally appropriate adjustments. The Research Assistant (coauthor SB) underwent comprehensive training (conducted by Principal Investigator RPP) and engaged in supervised practice sessions spanning a period of two weeks. Prior to analysis, a thorough review was conducted to ascertain the validity of the written test responses, encompassing assessments of adherence to acceptable ranges and the presence of any missing data prior to data entry.

Statistical analysis

First, the normality of the distribution of all the variables was examined and skewed variables were log-transformed.

Differences in the sociodemographic, socioeconomic, growth environmental and rated characteristics were investigated using the independent t test for continuous variables, by infection status or by IQ cutoff status (Tables 1 and 3). Similarly, differences in behavioral and lifestyle characteristics were examined by Fisher’s exact test or the chi-square test, which are used to analyze categorical variables (Tables 2 and 3). Sociodemographic, socioeconomic, behavioral and lifestyle characteristics were examined for associations with STH incidence using univariate and multivariate logistic regression models for mutual adjustment (Table 4). Similarly, STHs prevalence, and nutritional status were examined for associations with developmental delay (IQ < 69) using univariate and multivariate logistic regression models for mutual adjustment (Table 3).

Bivariate (Model 1) and multivariate (Model 2) analyses were also conducted to examine the (unadjusted and fully adjusted associations, respectively) associations between WASI II FSIQ scores and the presence of A. lumbricoides9,42,43, undernutrition15,16, maternal education41,44,45,46, annual family income22,47,48,49and the ACE-Q score50,51,]adjusting for participant sex and age at the time of the WASI II assessment52, which are known to correlate with neurodevelopmental indicators such as IQ (Table 4).

In Model 3, the explanatory variables (presence and egg density of A. lumbricoides infections) were adjusted solely for the “significantly associated covariates” identified (i.e., z scores for BMI, maternal education and log annual income) from the univariate model (Model 1). This adjustment aimed to ascertain the “minimal” impact of explanatory variables and covariates on the response variables. Model 2 involved the implementation of a multivariate regression model, thereby evaluating the comprehensive (i.e., full) adjusted influence of the explanatory variables and covariates on the response variables.

Among the 100 initially recruited mother–infant pairs within the cohort, follow-up data spanning 14 years were available for 74 participants. A p value less than 0.05 indicated statistical significance. The statistical analyses were performed utilizing SPSS version 25 (SPSS, IBM Corporation).

Results

Table 1 summarizes the characteristics of the mother-adolescent pairs at the 14-year follow-up. None of the characteristics differed significantly according to the status of parasitic infection. Individuals with an extremely low IQ indicating delayed development (i.e., FSIQ < 69) were considered to have a low height, inferior nourishment, low maternal education, and low annual family income, and their mother had an inferior growth environment score compared to that of the children with an FSIQ > 69. As anticipated, participants with extremely low IQs (i.e., < 69) exhibited low GPA and low mother-rated academic performance. There was no significant difference between infected and uninfected participants in nutritional status (i.e., BMI categories) or behavioral practices. The majority of participants did not trim their nails regularly, did not consume anthelmintics within the past 6 months, played in soil, and did not drink boiled or filtered water. Positive health behaviors (i.e., handwashing with soap before eating) were more frequently observed among uninfected participants (Supplementary Table 1).

Table 1 Characteristics of study participants (n = 74).
Full size table

Table 2 shows the prevalence of STHs among the study participants. Only 12 samples (16.2%) demonstrated shedding of eggs from A. lumbricoides. There was a greater prevalence of A. lumbricoides in male participants than in female participants. Egg density (epg) was greater among females than among males, but this difference was not statistically significant. Out of the 74 cohort participants followed up, 25—which was more than one-third (i.e., 33.78%) of participants—indicated an extremely low intelligence quotient (IQ) evaluated by the Wechsler Abbreviated Scale of Intelligence (WASI). There was no difference in roundworm incidence or egg density when participants were compared based on neurodevelopmental performance (i.e., IQ < 69 vs. > 70, the cut-off for delayed development). The prevalence of A. lumbricoides was not associated with any of the behavioral characteristics evaluated. However, participants who reported not using soap to wash their hands before eating, participants who reported a nail-biting habit, and participants who reported playing in soil had increased odds of contracting ascariasis compared to their counterparts. Again, none of the evaluated associations reached statistical significance (Supplementary Table 2).

Table 2 Prevalence and egg density of A. lumbricoides in Chitwan birth cohort participants (n = 74).
Full size table

Table 3 shows the associations of A. lumbricoides infection and nutritional status with developmental delay. The prevalence of A. lumbricoides and sex were not associated with neurodevelopmental delay according to either the univariate or multivariate models. Participants with BMIZ <-2SD (i.e., underweight) had higher odds ratios of neurodevelopmental delays than did their counterparts according to both the univariate and multivariate models.

Table 3 Odds of developmental delay with respect to STHs infection and nutritional status (n = 74).
Full size table

Table 4 summarizes the associations between A. lumbricoides egg density and WASI-II full-scale intelligence quotient (FSIQ) scores. The FSIQ scores were normally distributed (Supplementary Fig. 1). None of the associations (Model 1: bivariate; Model 2: fully adjusted; Model 3: minimally adjusted multivariate model) achieved statistical significance (i.e., p < 0.05). Among the contextual covariates, mother’s education level and reported annual family income were significantly (p < 0.05) positively associated with IQ scores, with consistent significance and direction in all 3 models.

Table 4 Association of A. lumbricoides egg density, nutritional and contextual variables with FSIQ scores of WASI II at 14 years (n = 74).
Full size table

Discussion

Neither the presence nor the egg density of A. lumbricoides infections was associated with the cognitive performance (WASI IQ) score of the cohort participants according to univariate or multivariate models. Associations remained insignificant in both analyses [i.e., categorical according to logistic regression and continuous variable evaluation via linear regression]. However, A. lumbricoidesegg density (epg) exhibited a nonsignificant but consistent inverse association in both the univariate and multivariate models (Models 2 and 3) after adjustment for these variables (Table 4). Consistent insignificant findings were reported among primary school pupils in Ile-Ife, Osun State, Nigeria53,] with a comparable prevalence (22.1%) of A. lumbricoides but almost double the percentage of participants (65.4%) who experienced cognitive deficits. In contrast, the presence of A. lumbricoidesinfections was associated with significantly lower cognitive scores in school-age children in Java, Indonesia43. Improved cognitive function in primary schoolchildren infected with A. lumbricoideswas reported after five months of receiving this intervention in the northern region of Jakarta, Indonesia42. In addition, pooled results from a systematic review and meta-analysis of 36 observational and intervention studies indicated the cognitive benefit of deworming for helminth infections9. Out of 25 participants who indicated delayed development (IQ < 69), only 3 (12%) participants reported A. lumbricoides infections. Hence, these nonsignificant associations may also be due to the small sample size, which resulted in a lack of statistical power.

This study also revealed that undernutrition (BMIZ<–2SD) increased the odds of a developmental deficit according to both the univariate and the adjusted multivariate models. The negative impact of underweight status on the cognitive neurodevelopment of children is well documented. For example, a scoping review of 26 articles indicated a significant impact of underweight on cognitive neurodevelopment16. A recent population-based cohort study in India with a large sample size (n= 41,554) also reported cognitive deficits among undernourished children15. However, the lack of an association between BMI and IQ according to linear regression models urges careful interpretation of the association. Yet, Ascariasis may lead to issues such as impaired fat digestion, reduced vitamin absorption, and temporary lactose intolerance10. These problems can result in decreased appetite, increased nutrient loss, and reduced nutrient uptake11,12, collectively contributing to low BMIZ. When BMIZ falls below a certain threshold (i.e., <-2SD), individuals who are underweight may experience weakness or lethargy17due to anemia18,19or severe weight loss, which can lead to higher rates of school absenteeism17and cognitive deficits9. This illustrates a complex relationship between ascariasis, malnutrition and cognitive decline. The scenario of chance finding cannot be discounted.

This study also investigated the prevalence of STHs among Chitwan birth cohort participants followed up after 14 years of birth. The prevalence of A. lumbricoidesin our study (i.e., 16.2%) was slightly lower than the prevalence reported in various studies in Nepal, such as 20% among schoolchildren in Kathmandu, Nepal54; 21.8% among public and private schoolchildren in Kathmandu, Nepal55; and 22.63% among schoolchildren in the Bhaktapur district, Nepal56. A lower prevalence of A. lumbricoides was also recently reported among school-age adolescents—8.33% among primary school children in Bhaktapur, Nepal57; 5.5% among private school-going pupils in the Dharan Sub metropolitan City, Nepal58; 2.3% among rural area school children in the Lokhim VDC, Nepal59; and 1.8% among school children in the Saptari District, Nepal60. In contrast, a high prevalence of A. lumbricoides has also been reported in many previous studies. For example, a 50.92% prevalence of ascariasis was reported in schoolchildren in the Rangeli Municipality of Morang District in Eastern Nepal6. Similarly, a recent study among schoolchildren in Kritipur Municipality, Kathmandu, reported a 34.2% prevalence of A. lumbricoides8. Another study among children from five schools in Bhairahawa, Nepal, reported a 29.1% overall prevalence of A. lumbricoides7. This discrepancy in the prevalence of STHs might be due to differences in climatic conditions and microecology61, population characteristics, and awareness of steps of prevention62.

Despite coverage of the study area by the national deworming campaign in Nepal since 200463, the 16.2% prevalence of A. lumbricoides in the study area is unexpected and relatively high. This may be partially explained by prevailing unhygienic behaviors and lifestyles (Supplementary Table 1) among participants. For example, most of the participants (i.e., 90%) reported that they did not consume anthelmintic drugs that are distributed freely in schools. In addition, more than two-thirds of the participants (67.6%) did not use soap to wash their hands before eating food by hand; 92% of the participants reported inadequate hand hygiene behavior and presented with ascariasis infections. Furthermore, approximately two-thirds of the participants reported that they played in soil but did not trim their nails regularly. In addition, the high prevalence of A. lumbricoides infection may also be attributed to the extremely resistant nature of Ascaris eggs, which contributes significantly to transmission dynamics. However, the small sample size limits our ability to generalize the findings.

The prevalence and egg density of A. lumbricoides were similar between males and females and did not differ significantly based on IQ status. None of the behavioral, lifestyle, socioeconomic, or demographic factors were associated with the prevalence of A. lumbricoides. Even hand-washing habits, which differed significantly by infection status, lost statistical significance in logistic regression according to both the univariate and multivariate models. Limitations of the small sample size might be behind this inconsistency due to the lack of statistical power. Hence, future studies with larger sample sizes need to be conducted to confirm these associations.

As anticipated, maternal education and reported annual income were consistently significantly associated with participants’ cognitive performance according to both univariate and multivariate models. Mothers’ education is an established contributor to their children’s IQ41,44,45,46. This study also confirmed the positive association with full-scale IQ. Although a positive association between reported annual income and IQ has been suggested by many other studies22,47,48,49, the direction of the association and the large effect size, even after full adjustment for other variables, such as maternal education, BMIZ, ACE, and other relevant postnatal environment variables, are noteworthy. Participants with educated mothers and greater annual family income may have a more enriched environment, which might ultimately contribute to better IQ scores. Social determinants of health are nonmedical social and environmental factors such as level of education, income, relationships, housing conditions, and social status that impact an individual’s health status. Social determinants contribute to health inequities, as individuals who consistently experience sociodemographic disadvantages demonstrate worse health outcomes than their counterparts. The WHO estimates that factors considered to be social determinants may account for 30–55% of health outcomes64.

A few limitations of this study should be considered. The first and main methodological limitation of the study is the use of single Kato-Katz smear assessments, which may be inadequate for fully assessing long-term parasitic burden and can be sensitive to day-to-day and within-stool variations in eggs per gram (epg) of stool. However, the Kato–Katz method is the method of choice for measuring the number of eggs present32. Second, the limited sample size, especially in subgroup analyses, could have heightened the risk of type II error, reducing the likelihood of identifying more subtle associations. Consequently, the study’s statistical power might be inadequate for detecting certain effects. Moreover, the generalizability of the findings could be constrained, as the sample may not accurately reflect the wider population. Further studies with larger and more diverse groups are necessary to validate and expand upon these results. Third, while IQ was assessed using the WASI II by a trained researcher, and key measures such as the ACE score were also conducted by research staff, some subjective assessments provided by parents or participants (e.g., growth environment, school quality, and academic performance) may not fully align with objective measures. Additionally, no interim health assessments, such as diarrheal episodes or breastfeeding practices, were conducted, which may have influenced cognitive outcomes during early childhood.

Conclusion

Despite the coverage of an ongoing national deworming program since 2004 in the study area, this cohort of participants indicated a relatively high prevalence (i.e., 16%, (12/74)) of A. lumbricoides. Developmental deficits among one-third of participants might have stemmed from their morbidity associated with undernutrition (16%, (12/73)). Although none of the lifestyle, hygiene or behavioral factors explained the increase (infection of A. lumbricoides), the high prevalence of unhealthy lifestyles and inadequate hygiene behaviors highlights the need for, and importance of, “in-depth” health and nutritional education.