Advertisement
  • Loading metrics

Effects of a Post-Deworming Health Hygiene Education Intervention on Absenteeism in School-Age Children of the Peruvian Amazon

  • François L. Thériault,

    Affiliation Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal, Quebec, Canada

  • Mathieu Maheu-Giroux,

    Affiliation Department of Global Health & Population, Harvard School of Public Health, Boston, Massachusetts, United States of America

  • Brittany Blouin,

    Affiliation Division of Clinical Epidemiology, Research Institute of the McGill University Health Centre, Royal Victoria Hospital, Montreal, Quebec

  • Martin Casapía,

    Affiliation Asociacíon Civil Selva Amazónica, Iquitos, Perù

  • Theresa W. Gyorkos

    theresa.gyorkos@mcgill.ca

    Affiliations Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal, Quebec, Canada, Division of Clinical Epidemiology, Research Institute of the McGill University Health Centre, Royal Victoria Hospital, Montreal, Quebec

Effects of a Post-Deworming Health Hygiene Education Intervention on Absenteeism in School-Age Children of the Peruvian Amazon

  • François L. Thériault, 
  • Mathieu Maheu-Giroux, 
  • Brittany Blouin, 
  • Martin Casapía, 
  • Theresa W. Gyorkos
PLOS
x

Abstract

Soil-transmitted helminth (STH) infections are a leading cause of disability and disease burden in school-age children of worm-endemic regions. Their effect on school absenteeism, however, remains unclear. The World Health Organization currently recommends delivering mass deworming and health hygiene education through school-based programs, in an effort to control STH-related morbidity. In this cluster-RCT, the impact of a health hygiene education intervention on absenteeism was measured. From April to June 2010, all Grade 5 students at 18 schools in a worm-endemic region of the Peruvian Amazon were dewormed. Immediately following deworming, nine schools were randomly assigned to the intervention arm of the trial using a matched-pair design. The Grade 5 students attending intervention schools (N = 517) received four months of health hygiene education aimed at increasing knowledge of STH prevention. Grade 5 students from the other nine schools (N = 571) served as controls. Absenteeism was measured daily through teachers' attendance logs. After four months of follow-up, overall absenteeism rates at intervention and control schools were not statistically significantly different. However, post-trial non-randomized analyses have shown that students with moderate-to-heavy Ascaris infections and light hookworm infections four months after deworming had, respectively, missed 2.4% (95% CI: 0.1%, 4.7%) and 4.6% (95% CI: 1.9%, 7.4%) more schooldays during the follow-up period than their uninfected counterparts. These results provide empirical evidence of a direct effect of STH infections on absenteeism in school-age children.

Author Summary

Soil-transmitted helminths (STHs) are intestinal parasites that infect over 2 billion people in the world. Children infected with STHs often experience growth delays and perform poorly in school. However, the specific effects of STH infections on school absenteeism in children are not well understood. In this study, the authors followed Grade 5 students from 18 different schools, for four months. Students from all 18 schools were administered drugs at baseline to treat STH infections. Students from nine randomly selected schools then received weekly lectures and were given booklets on STH prevention, in an effort to prevent re-infection. Using a post-trial non-randomized analysis, the authors found that, while this health education intervention had no effect on overall absenteeism rates, Grade 5 students who had been re-infected during the four months following treatment had higher absenteeism rates than students who were uninfected after four months. These results contribute to a better understanding of the negative impact of STH infections on child health, showing that STH infections do increase the risk of absenteeism. These results also show that health education interventions, while powerful tools in the fight against STH infections, likely need to be implemented over longer periods of time to impact overall absenteeism rates. These findings are of interest to decision makers looking to implement and evaluate STH-targeted education programs.

Introduction

Soil-transmitted helminth (STH) infections are the world's most important cluster of neglected tropical diseases and cause significant morbidity in school-age children (i.e. children aged between 5 and 14 years) [1]. It has recently been estimated that the intestinal parasites Ascaris lumbricoides, Trichuris trichiura and hookworms (Ancyclostoma duodenale and Necator americanus), respectively, infect 287 million, 194 million and 135 million school-age children around the world [2]. STH-infected children are at an increased risk of physical and intellectual growth impairments [3]. Globally, school-age children lose 16.7 million disability-adjusted life years (DALYs) to STH infections every year, a morbidity representing over 11% of the total disease burden afflicting this age group [4].

It has long been argued that STH infections could also increase school absenteeism [5], but little research has adequately investigated this association. A randomized controlled trial in Kenya found that a deworming intervention reduced absenteeism by at least 7 percentage points – a reduction of 25% of the baseline absenteeism rate – in school-age children [6]. This trial had a large sample size and strong methodology, and is often cited as evidence that STH infections have an important effect on absenteeism in school-age children [3], [7]. However, to date, such positive strong results have not been replicated. Only two other trials have measured the impact of deworming interventions on school absenteeism, and neither found significant treatment effects [8], [9]. These latter two trials had important methodological limitations, such as high attrition rates, selection bias and no sample size calculations, which might have led to underpowered analyses. An extensive review of all three trials has recently been published [10]. Unfortunately, the pooled effect measures reported in this meta-analysis are difficult to interpret because these trials measured absenteeism differently, and because these trials were conducted in student populations with different baseline STH infection prevalences. In addition, a systematic literature search found only seven observational studies investigating the effects of STH infections on absenteeism in school-age children [11][17]. Summarizing these results is difficult, as study quality was variable and results were discordant.

Deworming interventions have clearly been shown to be successful at reducing STH burden in school-age children [3]. A single dose of albendazole can reduce the Ascaris, Trichuris and hookworm burdens, measured in eggs per gram of feces, by up to 99%, 77% and 97% of their respective baseline values [18]. Similarly, a single dose of mebendazole can reduce these parasite burdens by up to 99%, 83% and 84% of their respective baseline values [18]. In regions where over 50% of school-age children are infected with at least one STH species, the World Health Organization (WHO) recommends treating all school-age children with a single dose of albendazole or mebendazole, once or twice a year, through school-based programs [19]. Health hygiene education is also recommended by WHO as an essential component of any school-based deworming intervention [20]. By promoting hygienic behaviours, health hygiene education programs could further reduce the risk of STH re-infection following deworming, and thus maximize the health impact of deworming interventions. Indeed, we have shown previously that a health education intervention was effective in increasing STH knowledge and in reducing the intensity of Ascaris infection [21]. The present study had two main objectives: 1) to assess the effect of a health hygiene education intervention aimed at reducing STH re-infection on absenteeism, and 2) to estimate the relationship between STH infection intensity and absenteeism in school-age children.

Methods

Ethics statement

This study was approved by the Research Ethics Board of the Research Institute of the McGill University Health Centre in Montréal, Canada and the Comité Institucional de Bioética of the Asociación Civil Impacta Salud y Educación in Lima, Peru. The trial is registered with clinicaltrials.gov (Registration number: NCT01085799). Written informed consent was obtained from each child's parent or guardian and written child assent was obtained from each child.

Study location and study participants

This trial was conducted in Belén, a community of extreme poverty located on the banks of the Itaya River in the Peruvian Amazon. Belén is home to approximately 65,000 people, most of whom have limited access to potable drinking water. Lacking sewers and sanitation systems, and subjected to seasonal floods, Belén is highly STH-endemic. Recent surveys have found that up to 86% of Belén Grade 5 students are infected with at least one STH species [22]. Grade 5 children were selected as the study population because: 1) their mean age was expected to be 10 years, the mid-range of the highest STH risk group of school-age children (5–14 years of age), in terms of STH prevalence, intensity and burden of disease; 2) they are able to provide informed assent to participate in the study, to complete interviewer-administered questionnaires, and to provide adequate stool specimens on the day of the interview; 3) previous experience suggested a high response rate; and, 4) they would still be in primary school the following year for potential follow-up study.

Trial design

A detailed description of the study design is available elsewhere [21]. Briefly, 18 elementary schools were paired, based on Grade 5 enrolment and geographical zone. In each of the nine pairs, one school was randomly assigned to the intervention arm of the trial, while the other served as a control. Between April and June 2010, Grade 5 students from intervention and control schools were dewormed with a single oral dose of 400 mg albendazole. In addition, a pre-tested interviewer-administered questionnaire was used to collect information on sociodemographic indicators and STH-related knowledge. Immediately following deworming, Grade 5 students attending intervention schools received four months of enhanced health hygiene education, while Grade 5 students attending control schools followed their usual curriculum, which did not include the health hygiene education intervention. The health hygiene education intervention consisted of booklets, posters, interactive activities and weekly lectures aimed at increasing students' knowledge of STH transmission and prevention. Grade 5 teachers from intervention schools also participated in workshops on STH transmission and prevention. Participating students provided single stool specimens at two timepoints: 1) before deworming and 2) four months post-deworming. Stool specimens were analyzed using the Kato-Katz method [23], and the intensity of STH infection was categorized according to species-specific WHO threshold values [20]. Moderate and heavy infections were combined into a single category to increase statistical power. The health education intervention was shown to be effective in increasing STH knowledge and in reducing Ascaris intensity of infection [21]. The effect size measures for the intervention on Trichuris and hookworm re-infection did not reach statistical significance.

Absenteeism data

Absenteeism data for each school day from the date of deworming to the date of follow-up stool specimen collection was abstracted from Grade 5 teachers' attendance logs. On any given day, students with justified or unjustified absences were classified as absentees, while students who attended class for at least part of the school day were classified as attendees. Absenteeism was defined as the percentage of school days a given student was absent among the total number of days the school was open during the four-month follow-up period.

Socioeconomic status

Each student was categorized in a socioeconomic status (SES) quartile, using an asset-based proxy constructed with principal components analysis [24] that explained 36% of the variance. This proxy included the following assets: house made of cement/bricks, electricity-wired house, a gas stove, a radio, and a television. The 1st quartile was comprised of students with the highest SES scores while the 4th quartile was comprised of students with the lowest SES scores.

Objective 1: Effect of a health hygiene education intervention on absenteeism

The mean absenteeism rates of girls and boys were compared using t-tests. Similarly, the mean absenteeism rates of students in each of the four SES quartiles were compared using analysis of variance (ANOVA). Scatter plots and univariate linear regression were used to assess the relationship between age and absenteeism.

The mean absenteeism rates of students attending intervention schools were compared to those of students attending control schools using t-tests. To measure the effect of the health hygiene education intervention on school-level absenteeism while accounting for clustering by school, two hierarchical models were fitted. Both models are specified below:where Absenteeismij is the absenteeism rate of student i in intervention school j; µ0j is a vector of school-specific random effects for the nine intervention schools; εij are the residuals in the intervention schools; Absenteeismic is the absenteeism rate of student i in control school c; υ0c is a vector of school-specific random effects for the nine control schools; and ξic are the residuals in the control schools. Therefore, the intercepts of the first model (αInt) and second models (αCtl) are the mean absenteeism rates in intervention and control schools, respectively. The effect of the education intervention on absenteeism rates was estimated by calculating the difference between the average absenteeism rates in intervention and control schools (i.e., αInt - αCtl). Confidence intervals for this effect size measure were obtained from 9,999 bootstrap replicates. The clustered nature of our observations was taken into account by defining the resampling unit at the school level.

In the days immediately following deworming, the STH burden was expected to be very similar in both arms of the study. The health hygiene education intervention was hypothesized to delay re-infection in the intervention arm of the study. Therefore, the effect of the health hygiene education intervention on absenteeism rates was expected to become increasingly stronger over time (since deworming). To explore the sensitivity of our results, we repeated the analysis described above three times, excluding 1) the first 30 days immediately following deworming; 2) the first 60 days immediately following deworming; and 3) the first 90 days immediately following deworming.

Objective 2: Effect of STH infection intensity on absenteeism

Three hierarchical linear regression models were built to estimate the effect of infection intensity on absenteeism, for each STH species separately (i.e. Model 1 for Ascaris, Model 2 for Trichuris and Model 3 for hookworm). The models used fixed effects for infection intensity, SES quartiles and age, and random effects at the school level. The three models had the following general formulation:where absenteeismij is the percentage of schooldays absent of student i from school j; βk is a vector of coefficients for light and moderate-to-heavy infections; ψs is a vector of coefficients for the second highest, second lowest and lowest SES quartiles; δ is a coefficient for age; and φ is a school-level random effect. Similar models were also used to measure the effect of any STH infection and of STH infection by multiple species (i.e. polyparasitism) on the percentage of schooldays absent during the follow-up period.

The effect of Ascaris infections on absenteeism was also measured using a hierarchical logistic regression model. As previously stated, the single oral dose of 400 mg albendazole administered to the students is particularly efficacious against Ascaris, with up to 99% egg reduction rates [18]. It was therefore assumed that most students were nearly free of Ascaris infections in the days following deworming. Students who had heavier Ascaris infections at follow-up likely acquired these infections in the months following deworming. If Ascaris infections have a causal effect on absenteeism, the daily odds of absenteeism should increase with time in students who were infected after four months of follow-up, as these students gradually acquire their infections. Absenteeism was therefore redefined as a binary variable for each day, from the date of deworming to the date of follow-up stool specimen collection. The logit of absenteeism was modelled as a function of the number of days since deworming, Ascaris infection intensity at follow-up and SES quartile. Product terms between time since deworming and both Ascaris infection intensity at follow-up and SES quartile were added to the model. Random intercepts were also added to the model to account for school clustering and for within-student correlation in daily observations. Scatter plots were used to visually assess the assumption of a linear relationship between the logit of absenteeism and the number of days since deworming. The model is defined as:where Absenteismijt is the probability of being absent for student i from school j on day t;

βk is a vector of coefficients for light and moderate-to-heavy infections; ψs is a vector of coefficients for the second highest, second lowest and lowest SES quartiles; λ is a coefficient for time since deworming; ωk is a vector of coefficients for the interactive effect of Ascaris infection intensity and time; υs is a vector of coefficients for the interactive effect of SES quartile and time; φ is a school-level random effect; and η is a student-level random effect.

Unless otherwise stated, all statistical analyses were performed using the R statistical software, version 2.11.1 [25]. The hierarchical models were fitted using the lme4 library [26]. When applicable, histograms and residual plots were used to visually assess the distributional assumptions of the models used.

Results

Description of study population

A total of 1,486 students were officially enrolled in Grade 5 at the 18 participating schools. Parental informed consent and child assent was provided for 1,285 students. Complete parasitological and absenteeism data were available for 1,088 students

(N = 517 in intervention schools and N = 571 in control schools) (Figure 1). The baseline characteristics of Grade 5 students did not differ significantly between intervention and control schools (Table 1). A detailed comparison of students in both arms of the trial is available elsewhere [21].

thumbnail
Table 1. Summary data showing the comparability in baseline characteristics of intervention and control study schools and children (adapted from [21]).

https://doi.org/10.1371/journal.pntd.0003007.t001

The 18 participating schools were open for a total of 85,107 student-days between the date of deworming and the date of follow-up stool specimen collection for the 1,088 Grade 5 students included in the final sample. Absenteeism data were available for 73,098 (85.9%) of these student-days. On average, students were absent for 11.3% (95% CI: 10.7%, 12.0%) of school days during the four-month follow-up period.

Sex

Sex was not associated with absenteeism. On average, girls and boys were respectively absent for 11.3% and 11.3% of schooldays during the follow-up period (p = 0.97).

Socioeconomic status

SES was significantly associated with absenteeism. On average, students in the highest, second, third and lowest SES quartiles were, respectively, absent for 9.4%, 11.1%, 12.8% and 13.5% of schooldays during the follow-up period (p<0.001).

Age

Absenteeism tended to increase with age. Students' ages ranged from 8 to 16 years, with an average of 10.34 years. A hierarchical linear regression model that accounted for clustering at the school level estimated that absenteeism increased by 1.7% (95% CI: 1.1%, 2.3%) for every one-year increase in age.

Objective 1: Effect of a health hygiene education intervention on absenteeism

The health hygiene education intervention had no statistically significant effect on absenteeism in Grade 5 children. On average, students attending intervention and control schools were respectively absent for 11.4% and 11.3% of schooldays during the follow-up period (p = 0.93). The mean absenteeism rate of intervention and control schools was estimated at 12.9% (95% CI: 10.2%, 15.6%) and 12.1% (95% CI: 9.1%, 15.0%), respectively. Attending an intervention school was not associated with a significant school-level decrease in absenteeism during the four-month follow-up period (difference  = 0.8%; 95% CI: −3.1%, 4.8) %).

Results were not statistically significantly different when the first, first two or first three months of data immediately following deworming were excluded from analysis. However, these results are consistent with the hypothesis that the magnitude of the health hygiene education intervention on absenteeism rates increases with time since deworming (Table 2). After removing the first three months of data immediately following deworming, attending an intervention school was associated with a slight, non-significant decrease in absenteeism (difference  = −0.8%, 95% CI: −4.5%, 2.9%).

thumbnail
Table 2. Sensitivity analysis: Mean percentage of schooldays lost to absenteeism in intervention (n = 9) and in control (n = 9) schools in the first 125 days following mass deworming, as estimated by the hierarchical models, which each exclude a different number of days post-deworming.

https://doi.org/10.1371/journal.pntd.0003007.t002

Objective 2: Effect of STH infections on absenteeism

STH infections were associated with a significant increase in the percentage of schooldays absent during the four-month follow-up period (Figure 2). After adjusting for SES quartiles, age, and clustering by schools, students who had moderate and heavy Ascaris infections at follow-up had missed 2.4% (95% CI: 0.1%, 4.7%) more schooldays than students with no Ascaris infections. Similarly, students with light hookworm infections at follow-up had missed 4.6% (95% CI: 1.9%, 7.4%) more schooldays than students with no hookworm infections. Trichuris infections at follow-up were also associated with an increase in absenteeism, but the effect was not statistically significant at α = 0.05 (Table 3). Having any STH infection, or having multiple STH infections, were both associated with an increase in absenteeism, but the effect was not statistically significant (Table 4).

thumbnail
Figure 2. Effect of STH infections on the percentage of schooldays lost.

STH infection status was measured four months after deworming, and absenteeism was measured from the day following deworming to the day of follow-up stool specimen collection. Mean (± standard error) percentage of schooldays lost presented for each STH infection category.

https://doi.org/10.1371/journal.pntd.0003007.g002

thumbnail
Table 3. Effect of STH infection intensity, SES and age on the percentage of schooldays absent during a four-month follow-up period, as estimated by three different hierarchical linear regression models, for each STH infection.

https://doi.org/10.1371/journal.pntd.0003007.t003

thumbnail
Table 4. Effect of infection by any STH species and by multiple STH species on the percentage of schooldays absent during a four-month follow-up period, as estimated by two different hierarchical linear regression models.

https://doi.org/10.1371/journal.pntd.0003007.t004

The daily odds of absenteeism slightly increased as the number of days since deworming increased in students who had light and moderate-and-heavy Ascaris infections at follow-up. A hierarchical logistic regression model estimated that the adjusted Odds Ratio (aOR) comparing the odds of absenteeism between students with light Ascaris infections and no Ascaris infections at follow-up increased from 0.98 (95% CI: 0.81, 1.78) on the day following deworming to 1.11 (95% CI: 0.93, 1.34) on the 125th day following deworming (approximately four months post-deworming). Similarly, the aOR comparing the odds of absenteeism between students with moderate-to-heavy Ascaris infections and no Ascaris infections at follow-up increased from 1.11 (95%CI: 0.83, 1.49) on the day following deworming to 1.19 (95% CI:0.90, 1.59) on the 125th day following deworming (Figure 3). The same model found significant interaction effects between SES quartiles and the number of days since deworming on the odds of absenteeism. The aOR comparing the odds of absenteeism between students in the lowest and highest SES quartiles increased from 1.05 (95% CI: 0.82, 1.33) on the day following deworming to 1.73 (95% CI: 1.36, 2.19) on the 125th day following deworming (Figure 4).

thumbnail
Figure 3. Effect of Ascaris infection on the daily odds of absenteeism, following deworming.

The daily odds of absenteeism in students with A) light and B) moderate-to-heavy Ascaris infections four months after deworming, compared to the odds in students with no Ascaris infection four months after deworming, as estimated by a hierarchical logistic regression model. Black lines represent the estimated adjusted odds ratio (aOR), and shaded areas represent 95% confidence intervals.

https://doi.org/10.1371/journal.pntd.0003007.g003

thumbnail
Figure 4. Effect of socioeconomic status (SES) on the daily odds of absenteeism, following deworming.

The daily odds of absenteeism in students of A) the second lowest and B) the lowest SES quartile, compared to those in students of the highest SES quartile, as estimated by a hierarchical logistic regression model. Black lines represent the estimated adjusted odds ratio (aOR), and shaded areas represent 95% confidence intervals.

https://doi.org/10.1371/journal.pntd.0003007.g004

Discussion

Objective 1: Effect of a health hygiene education intervention on absenteeism

Four months of a health hygiene education intervention was not enough to reduce absenteeism in Grade 5 students. There was virtually no difference in the mean percentage of schooldays missed among students attending intervention and control schools. However, this result is not surprising given the short follow-up period. Four months might be insufficient for Grade 5 students attending intervention schools to 1) learn about STH prevention; 2) translate their newly acquired knowledge into behavioural changes; 3) have lower STH re-infection rates because of their newly developed healthy behaviours; 4) have fewer illness episodes because of their lower STH re-infection rates; and 5) attend school more frequently because of a lower STH burden. The health hygiene education intervention reduced Ascaris infection intensity [21], suggesting that points 1–3, as outlined above, were successfully achieved during the follow-up period. Similarly, a recent trial found that health hygiene education significantly reduced STH infections in Chinese schoolchildren [27]. However, only 34 (6.6%) students from the current study's intervention schools and 57 (10.0%) students from the current study's control schools had moderate-and-heavy Ascaris infections (i.e. infection intensity ≥5,000 epg) four months after deworming. This infection intensity threshold is usually associated with an increased risk of Ascaris-related morbidity [20]. Too few students from intervention and control schools were therefore at risk of experiencing Ascaris-related illness episodes during the four-month follow-up period for a significant effect of the health hygiene education intervention to be observed on absenteeism.

Objective 2: Effect of STH infections on absenteeism

Absenteeism rates during the follow-up period were higher in students who had STH infections than in those who were still uninfected, four months after deworming. Ascaris and hookworm infections were more strongly associated with absenteeism than Trichuris infections, which is consistent with earlier findings [13], [14], [17].

The effect measures reported in this study might underestimate the true impact of STH infections on absenteeism for a number of reasons. First, the 18 study schools were beneficiaries of a school lunch program. On any given day, students present at the study schools therefore received a nutritious snack of fortified bread. Such programs offer strong incentives for students to attend school [28]. Infected children who might otherwise stay home because of STH-related illness episodes might therefore attend school to benefit from the lunch program. Second, the single oral dose of 400 mg albendazole used to deworm all students at baseline is very effective against Ascaris infections. In fact, a parasitological survey of 267 Grade 5 students, conducted two weeks after deworming, found an Ascaris egg reduction rate of 98.8% [29]. It is thus almost certain that students with moderate or heavy Ascaris infections four months after deworming did not harbour such intense infections during the entire follow-up period (i.e. they were free of moderate or heavy Ascaris infections for part of the four-month follow-up period). The percentage of schooldays absent during follow-up for this group of students might therefore underestimate the typical absenteeism rate of school-age children who are infected over an entire four-month period.

The high likelihood that students with Ascaris infections at follow-up were free of infection immediately after deworming also allowed for a novel approach to the analysis of deworming data. After adjusting for appropriate potential confounders, it was found that the daily odds of absenteeism increased more rapidly with time since deworming in students with Ascaris infections than in those with no Ascaris infections at follow-up (i.e. time since deworming was modifying the effect of Ascaris infection intensity on the daily odds of absenteeism). This result, while not statistically significant, nevertheless suggests that Ascaris infections are related to absenteeism in school-age children. As Grade 5 students gradually acquired Ascaris infections following deworming, they would increasingly be more likely to be absent from school than students who did not acquire Ascaris infections during the four-month follow-up period.

Effect of age on absenteeism

Age was strongly associated with absenteeism. Under normal grade progression in the Peruvian school education system, Grade 5 students should be, on average, 10 or 11 years old. Students older than 11 years are therefore likely to have repeated one or more grades. Grade repetition is strongly associated with poor academic performance, which in turn is strongly associated with high absenteeism rates [5]. In the study population where Grade 5 students' ages ranged from 8 to 16 years, it was therefore not surprising to observe a strong positive relationship between increasing absenteeism and increasing age.

Effect of socioeconomic status on absenteeism

SES quartile was also strongly associated with absenteeism. Students from the lowest SES quartiles were absent for a greater percentage of schooldays during the four-month follow-up period. Furthermore, the daily odds of absenteeism increased significantly faster in students of the lowest SES quartiles compared to students of the highest SES quartiles, as time since deworming increased. This statistically significant effect modification can be explained by differential patterns of family livelihood. Families of lower SES might call upon their school-age children to work the family's agricultural plot (locally called “chacra”) more frequently than families of high SES. Because deworming occurred near the end of the rainy season, the weather became increasingly favourable to outdoor manual labour as time since deworming increased. It is therefore possible that students of lower SES quartiles spent an increasing number of days at work as the follow-up period progressed.

Limitations

It must be noted that, while intervention status was randomly allocated, infection status was not. Students who acquired infections during the follow-up period were inherently different from students who were still uninfected, four months after deworming. Although analyses adjusted for potential confounders, it is possible that residual confounding biased the effect measures of STH infections on absenteeism. It must also be noted that this trial was powered to detect a statistically significant effect of the health hygiene education intervention on STH re-infection rates, rather than the relationship between STH infections and absenteeism [21]. Analyses examining outcomes other than re-infection might therefore have been underpowered. Light Ascaris infections and both light and heavy Trichuris infections had positive but statistically non-significant effects on absenteeism. Effect modification between Ascaris infection intensity and time since deworming on the daily odds of absenteeism was also positive but statistically non-significant. With a larger sample size, these effects might have been statistically significant. Other studies [8], [14] have also attempted to measure absenteeism's association with STH infections despite being powered to detect statistically significant differences in health-related outcomes more directly related to STH infections on the hypothetical causal chain. No study has ever been designed and adequately powered for the sole purpose of measuring the effects of STH infections on absenteeism in school-age children. Finally, it must be noted that attendance was not measured directly, and that using the teachers' attendance logs might have introduced measurement bias in the analyses. Assuming that this type of measurement error was random, the reported estimates could have been affected by regression dilution bias and be biased towards the null [30], although it should be noted that such measurement error would be non-differential (i.e. that the same degree of measurement error would be expected to have occurred in both intervention and control schools).

Conclusion

In conclusion, results of the present study suggest that Ascaris infections are related to absenteeism in school-age children. Because the health hygiene education intervention was successful in reducing Ascaris infections following deworming [21], it is plausible that this intervention would eventually have lowered absenteeism and also have had a positive impact on future health, had the follow-up period been longer. Incorporating health hygiene education in school-based deworming programs has been found to have positive effects on Ascaris-related morbidity [20], but this study found no impact on absenteeism rates. Our results also highlight the importance of allowing a sufficiently long follow-up period before evaluating the health impact of STH-specific health hygiene education interventions in school-age children, especially in areas where STH infection (and subsequent STH-attributable morbidity) is gradually acquired over time.

Acknowledgments

The authors are indebted to the school principals, teachers and students for their collaboration in this study. The authors are also indebted to their local study coordinator, interviewers and laboratory technicians.

Author Contributions

Conceived and designed the experiments: FLT MMG BB MC TWG. Performed the experiments: FLT MMG BB MC TWG. Analyzed the data: FLT MMG BB TWG. Wrote the paper: FLT MMG BB MC TWG. Assisted in Data Interpretation: FLT MMG BB MC TWG.

References

  1. 1. Hotez PJ, Brindley PJ, Bethony JM, King CH, Pearce EJ, et al. (2008) Helminth infections: the great neglected tropical diseases. J Clin Invest 118: 1311–1321.
  2. 2. de Silva NR, Brooker S, Hotez PJ, Montresor A, Engels D, et al. (2003) Soil-transmitted helminth infections: updating the global picture. Trends Parasitol 19: 547–551.
  3. 3. Bethony J, Brooker S, Albonico M, Geiger SM, Loukas A, et al. (2006) Soil-transmitted helminth infections: ascariasis, trichuriasis, and hookworm. Lancet 367: 1521–1532.
  4. 4. WHO (2002) World Health Report 2002: Reducing Risks, Promoting Healthy Lives Geneva: World Health Organization.
  5. 5. Pollitt E (1990) Malnutrition and Infection in the Classroom. Paris: UNESCO.
  6. 6. Miguel E, Kremer M (2004) Worms: Identifying Impacts on Education and Health in the Presence of Treatment Externalities. Econometrica 72: 159–217.
  7. 7. WHO (2005) The Evidence is in: Deworming helps meet the Millenium Development Goals. World Health Organization.
  8. 8. Simeon DT, Grantham-McGregor SM, Callender JE, Wong MS (1995) Treatment of Trichuris trichiura infections improves growth, spelling scores and school attendance in some children. J Nutr 125: 1875–1883.
  9. 9. Watkins WE, Cruz JR, Pollitt E (1996) The effects of deworming on indicators of school performance in Guatemala. Trans R Soc Trop Med Hyg 90: 156–161.
  10. 10. Taylor-Robinson DC, Maayan N, Soares-Weiser K, Donegan S, Garner P (2012) Deworming drugs for soil-transmitted intestinal worms in children: effects on nutritional indicators, haemoglobin and school performance. Cochrane Database Syst Rev 7: CD000371.
  11. 11. De Carneri I, Garofano M, Grassi L (1967) [The role of trichocephaliasis in infantile physical and mental retarded development studied by a quantitative survey in Basso Lodigiano]. Riv Parassitol 28: 103–122.
  12. 12. Nokes C, Bundy DA (1993) Compliance and absenteeism in school children: implications for helminth control. Trans R Soc Trop Med Hyg 87: 148–152.
  13. 13. Simeon D, Callender J, Wong M, Grantham-McGregor S, Ramdath DD (1994) School performance, nutritional status and trichuriasis in Jamaican schoolchildren. Acta Paediatr 83: 1188–1193.
  14. 14. Hutchinson SE, Powell CA, Walker SP, Chang SM, Grantham-McGregor SM (1997) Nutrition, anaemia, geohelminth infection and school achievement in rural Jamaican primary school children. Eur J Clin Nutr 51: 729–735.
  15. 15. Raj SM, Sein KT, Anuar AK, Mustaffa BE (1997) Effect of intestinal helminthiasis on school attendance by early primary schoolchildren. Trans R Soc Trop Med Hyg 91: 131–132.
  16. 16. Williams J, Leherr K (1998) Children's Health and Nutrition as Educational Issues: A Case Study of the Ghana Partnership for Child Development's Intervention Research in the Volta Region of Ghana. Agency for International Development, Bureau for Africa.
  17. 17. Ahmed A, Al-Mekhlafi HM, Azam MN, Ithoi I, Al-Adhroey AH, et al. (2012) Soil-transmitted helminthiasis: a critical but neglected factor influencing school participation of Aboriginal children in rural Malaysia. Parasitology 139: 802–808.
  18. 18. Steinmann P, Utzinger J, Du ZW, Jiang JY, Chen JX, et al. (2011) Efficacy of single-dose and triple-dose albendazole and mebendazole against soil-transmitted helminths and Taenia spp.: a randomized controlled trial. PLoS One 6: e25003.
  19. 19. WHO (2006) Preventive Chemotherapy in Human Helminthiasis. Geneva: World Health Organization.
  20. 20. WHO (2012) Eliminating soil-transmitted helminthiases as a public health problem in children: Progress report 2001-2010 and strategic plan 2011-2020. Geneva: World Health Organization.
  21. 21. Gyorkos TW, Maheu-Giroux M, Blouin B, Casapia M (2013) Impact of Health Education on Soil-Transmitted Helminth Infections in Schoolchildren of the Peruvian Amazon: a cluster-randomized controlled trial. PLoS Negl Trop Dis 7: e2397.
  22. 22. Casapia M, Joseph SA, Nunez C, Rahme E, Gyorkos TW (2006) Parasite risk factors for stunting in grade 5 students in a community of extreme poverty in Peru. Int J Parasitol 36: 741–747.
  23. 23. WHO (2002) Prevention and control of schistosomiasis and soil-transmitted helminthiasis. WHO Technical Report Series. Geneva: World Health Organization.
  24. 24. Filmer D, Pritchett LH (2001) Estimating Wealth Effects without Expenditure Data-or Tears: An Application to Educational Enrollments in States of India. Demography 38: 115–132.
  25. 25. R Development Core Team (2010) R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.
  26. 26. Bates D, Maechler M, Bolker B, Walker S (2014) lme4: Linear mixed-effects models using Eigen and S4. R package version 1.0-6.
  27. 27. Bieri FA, Gray DJ, Williams GM, Raso G, Li YS, et al. (2013) Health-education package to prevent worm infections in Chinese schoolchildren. N Engl J Med 368: 1603–1612.
  28. 28. Bundy DA, Drake LJ, Burbano C (2012) School food, politics and child health. Public Health Nutr: 1012-9.
  29. 29. Gyorkos TW, Maheu-Giroux M, Blouin B, Saavedra L, Casapia M (2013) Efficacy of a single dose of Albendazole for soil-transmitted helminth infections in school children of a village in Iquitos, Peru. Rev Peru Med Exp Salud Publica 30: 601–607.
  30. 30. Hutcheon JA CA, Hanley JA (2010) Random measurement error and regression dilution bias. BMJ 340: c2289.