Summary
Background
Timor-Leste reduced soil-transmitted helminth (STH) prevalence from 29% to 18% by 2020 through mass drug administration (MDA) in 2015–2019. School-based deworming was interrupted for nearly four years during the COVID-19 pandemic and resumed in 2022 in 3 municipalities and nationwide in 2024. We aimed to estimate STH prevalence, geographical distribution and associated factors following the re-introduction of MDA.
Methods
We conducted a cross-sectional, school-based survey from 26 February to 1 May 2025 across all 14 municipalities of Timor-Leste, sampling two schools per municipality and targeting 50 children between 6 and 14 years of age per school. We employed the Kato-Katz technique for stool sample analysis which is the WHO-recommended method for STH surveillance. Associations between STH infection and risk factors were assessed using univariate and multivariable logistic regression.
Findings
Out of 1960 children, 1236 (63·1%) provided analysable samples (615 [49·8%] girls, 621 [50·2%] boys); median age of the children was 8·7 years (IQR 8–10, range 6–14); ethnicity data was not available. Weighted national STH prevalence was 29·3% (95% CI 24·3–34·2), close to the 2012 baseline (29%) and above the 2020 figure (18%). Prevalence ranged from 2·2% (Oe-Cusse) to 91·5% (Manatuto). Ascaris lumbricoides accounted for 95·8% of infections; moderate-to-heavy infections affected 15·7% of children.
Interpretation
National STH prevalence has returned to near pre-control levels after the pandemic-related interruption of school-based deworming. Restoring and strengthening school-based deworming, with twice-yearly MDA where prevalence is at least 50%, along with WASH measures and community engagement is needed to regain control.
Funding
This study was funded by the Korea International Cooperation Agency (KOICA) and the United Nations Department of Economic and Social Affairs (UNDESA).
Keywords: Soil-transmitted helminths, Timor-Leste, Prevalence, School-going children, Mass drug administration, Neglected tropical diseases
Research in context.
Evidence before this study
We searched PubMed, WHO Institutional Repository for Information Sharing (IRIS), and Ministry of Health databases for records from January 1, 2010 to December 31, 2024 using the terms “soil-transmitted helminths”, “STH”, “Timor-Leste”, “East Timor”, and “prevalence”. Previous national surveys, including the 2012 baseline survey and the 2020 transmission assessment survey (TAS), showed heterogeneous soil transmitted helminthiasis transmission across municipalities in Timor-Leste and an overall decline in prevalence during years of regular mass drug administration. However, no comprehensive national assessment had been conducted following the COVID-19 pandemic, during which school-based deworming was interrupted for nearly four years. Modelling studies and empirical data from other endemic settings suggested that STH prevalence can return to pre-control levels within 2–3 years of treatment cessation, but the extent of any rebound in Timor-Leste was unknown.
Added value of this study
This national survey quantifies STH burden in Timor-Leste after the pandemic-related interruption of school-based deworming, across all 14 municipalities. Weighted national prevalence was 29·3% (95% CI 24·3–34·2), close to the 2012 baseline (29·1%) and well above the 17·9% recorded in 2020–21; moderate-to-heavy-intensity infection rose from 5·6% to 15·7%, and Ascaris lumbricoides accounted for 95·8% of infections. Prevalence ranged from 2·2% (Oe-Cusse) to 91·5% (Manatuto). The increase in STH prevalence was not uniform, it was greatest where deworming resumed only in 2024 (Manatuto 29·6%–91·5%, Ermera 14·6%–50·0%) and absent where it resumed in 2022 (Manufahi 43·0%–21·1%), which points to the speed of programme restart, rather than the pandemic in itself, as what governs the scale of loss. With fewer than two effective rounds delivered, such a survey is better used as a re-baseline for setting treatment frequency than as a post-treatment impact assessment.
Implications of all the available evidence
The findings support the prompt reinstatement and strengthening of school-based deworming in Timor-Leste, with treatment frequency matched to current prevalence, and imply that in high burden settings deworming alone is unlikely to achieve sustainable control. Combining preventive chemotherapy with WASH improvements, health education, and community engagement, together with periodic surveillance and a future formal impact assessment, will be important for regaining progress towards the WHO 2030 targets and for building resilience against future disruptions.
Introduction
Soil-transmitted helminth (STH) infections, which are caused by A. lumbricoides, Trichuris trichiura, and hookworms, affect almost two billion people globally, with the highest burden in tropical and subtropical regions where measures to ensure sanitary practices are often inadequate.1 School-age children bear the greatest morbidity burden, including undernutrition, anaemia, and impaired growth and cognition, perpetuated by poverty.1, 2, 3 The WHO recommends preventive chemotherapy through mass drug administration (MDA) as the primary control strategy, with treatment frequency guided by prevalence among school-age children.4
Timor-Leste introduced national STH control measures soon after independence. Between 2015 and 2019, four annual rounds of MDA targeting lymphatic filariasis and STH consistently achieved the minimum effective coverage threshold, exceeding 65% for lymphatic filariasis and 75% for STH nationally; Dili, the capital city of Timor-Leste, additionally received twice-yearly deworming from 2017.5 The 2020–21 transmission assessment survey (TAS) reported heterogeneous STH prevalence (0–43%) across municipalities and an overall decline from 29% in 2012 to 18% in 2020.5
School-based deworming was then interrupted for about four years. During the COVID-19 pandemic, mass treatment was deprioritised while vaccination and pandemic–response activities were scaled up, and no national STH MDA was delivered between 2020 and 2023. Deworming resumed in October 2022 in three municipalities (Dili, Manufahi, and Lautém), and nationwide school-based deworming resumed in 2024. As a result, fewer than two effective rounds had been delivered in most municipalities by the time of the 2025 survey. Mathematical models and empirical evidence from other endemic settings suggest that STH prevalence can rapidly rebound to pre-control levels within 2–3 years of MDA cessation.6, 7, 8 Understanding the current epidemiological situation is therefore critical for evidence-based programme planning.
The primary aim of this study was to quantify current national STH prevalence and infection intensity among school-going children across Timor-Leste after the prolonged interruption of deworming, and to characterise the magnitude and geographic distribution patterns. Secondary aims were to compare current prevalence with the 2012 and 2020 surveys, and to examine the association between infection and child growth and household water, sanitation and hygiene (WASH) factors, to inform the national neglected tropical disease (NTD) strategic plan for 2026–2030.
Methods
We conducted a cross-sectional, school-based sentinel survey from 26 February to 1 May 2025 across all 14 municipalities of Timor-Leste. According to 2025 enrolment data from the Ministry of Education, the school-going population between 6 and 14 years of age was 337,291. The country comprises diverse geographic zones from coastal lowlands to mountainous interior regions, with varying access to water and sanitation infrastructure.
WHO guidance for STH sentinel surveillance suggests selecting one sentinel site per 200,000–300,000 school-age children.9 Timor-Leste with a national school-age population of about 300,000 could in principle be monitored with a single national sentinel site; but to obtain a more informative national picture and coverage of every municipality, we selected two schools in each of the 14 municipalities. In each municipality, one sentinel school was randomly selected from the 2020–21 TAS school list (a subset of the Ministry of Education school register) to allow comparison over time, and one spot-check school was purposively selected from the Ministry of Education database, prioritising schools with >100 enrolled students and geographic representativeness.
Within each school, we aimed to recruit 50 children aged 6–14 years following WHO guidelines for STH surveillance.9 Grade 3 students (typically between 8 and 10 years) were primarily targeted as they are easily accessible and, having been eligible for several previous MDA rounds, makes them a ideal population for assessing programme performance9; when Grade 3 enrolment was insufficient, we expanded sequentially to other grades. To secure at least 50 samples per school accounting for non-responders and guided by the previous participation rates observed in the 2020–21 TAS, we distributed 70 sample containers per school and collected the morning stool samples on the following day.
Field teams comprised laboratory technicians, data collectors, and district public health officers trained in standardised procedures. School visits were coordinated with education authorities and preceded by community sensitisation. Demographic information and school-level WASH indicators were collected using structured questionnaires administered on tablets (KoBoCollect) with built-in validation rules and GPS coordinates. Data on household water sources, sanitation facilities, handwashing practices, and recent deworming history were obtained through paper-based self-administered questionnaires and subsequently entered into Microsoft excel by two data clerks. Sex was recorded as the biological sex held on the school enrolment register and confirmed with the class teacher at recruitment; no data on gender identity were collected, and the term sex is used throughout. Data on race or ethnicity were not collected: the population of Timor-Leste is not administratively categorised by race or ethnicity, these variables are not recorded in the Ministry of Education enrolment register, and their collection was not part of the approved protocol.
Fresh stool samples were collected in pre-labelled containers and processed within 2 h of collection whenever possible, not exceeding 4 h in any case. We employed the Kato-Katz technique, the WHO-recommended method for STH surveillance in moderate-to-high prevalence settings.10 Duplicate slides were prepared from each sample and examined within 30–60 min to optimise hookworm detection. Trained microscopists identified and counted eggs by species. Results were expressed as eggs per gram (EPG) of faeces. Infection intensities were classified using WHO thresholds: light (1–4999 EPG for A. lumbricoides; 1–999 for T. trichiura; 1–1999 for hookworm), moderate (5000–49 999; 1000–9999; 2000–3999, respectively), and heavy (≥50,000; ≥10,000; ≥4000, respectively).10
For quality control, the field supervisor independently re-read every tenth slide examined by the two microscopists. False positives and false negatives were identified, and egg-count agreement was assessed against pre-defined criteria, defined as a discrepancy of more than 10% where 100 or fewer eggs were counted, or of more than 20% where more than 100 eggs were counted. All discordant slides were reconciled and corrected before analysis.
Statistical analysis
Data were cleaned and analysed using Stata/IC version 15·1 (StataCorp, College Station, TX, USA). Municipality-level prevalence was estimated by pooling data from all surveyed schools within each municipality. National prevalence was estimated with a design-based (Horvitz–Thompson) estimator, treating municipalities as strata and schools as clusters and applying post-stratification weights; each school weight was the municipal school-age population multiplied by the school's share of enrolment among sampled schools in that municipality and divided by the number of children examined in that school. The national estimate was obtained as the ratio of weighted positive children to the weighted number examined (Supplementary Table S2). 95% CIs for weighted estimates were obtained by a cluster bootstrap resampling schools within municipality strata. Descriptive counts in the tables are unweighted observed frequencies. Geometric mean infection intensities (EPG) were calculated among positive children.
Associations between STH infection and risk factors (age, sex, municipality, household water source, toilet type, handwashing station, soap availability, water at the handwashing station, shoe use, utensil use, handwashing before meals and after defecation, and deworming in the past 3 years) were assessed using univariate and multivariable logistic regression with municipality entered as a fixed effect and multicollinearity assessed using variance inflation factors. Because the household questionnaire was subject to item non-response, entering all WASH indicators simultaneously would have restricted the analysis to 207 children and left the reference municipality with no infections. Age, sex, and municipality were therefore estimated in a single model fitted to all 1236 children examined, and each household WASH factor was estimated in a separate model adjusted for age, sex, and municipality on that factor's own complete-case sample. No imputation was performed; all analyses were complete-case, and denominators for household WASH variables therefore vary by item. Because prevalence was extreme in both the highest- and lowest-burden municipalities, a Firth penalised logistic model was fitted as a sensitivity analysis. For temporal comparison, 2025 prevalence data were compared with the 2012 baseline and 2020–21 TAS results, and the association between the MDA interruption duration and prevalence change were assessed using Spearman's rank correlation coefficient.
Geographic distribution was mapped using QGIS software version 3·14 (QGIS Development Team) with municipality boundary shapefiles obtained from the Timor-Leste Health Information System portal.
Ethics statement
The study protocol was approved by the Timor-Leste National Public Health Institute (05/INSP-TL/UEPD-AF/I/2025). Written informed consent was obtained from parents or guardians, along with child assent. Participation was voluntary with the right to withdraw at any time. All infected children received a single 400 mg dose of albendazole during the subsequent national deworming round conducted in July–August 2025, in accordance with national guidelines.
Role of the funding source
The funder had no role in study design, data collection, analysis, manuscript preparation, or the decision to submit the paper for publication.
Results
Out of the 28 selected schools, 27 (96·4%) were surveyed across all 14 municipalities; one school in Bobonaro municipality was inaccessible due to deteriorated road conditions during the rainy season (Fig. 1). Among 1960 enrolled students, 1697 (86·6%) were present on survey days; 1563 (79·7%) provided consent and 1238 (63·2%) submitted stool samples (Fig. 1; Supplementary Table S1). Two samples had insufficient quantity for duplicate examination, leaving 1236 samples for analysis (Fig. 1). The median age of participants was 8·7 years (IQR 8–10); 615 (49·8%) were female and 621 (50·2%) male (Table 1). Stool submission among consented children varied by municipality (48–95%) and was lowest in Ermera (48%) but differed little by sex (female 80·0%, male 78·2%) or age. On quality-control re-reading of every tenth slide, there were no false positives and 11 false negatives, and 20 slides showed egg-count discrepancies exceeding the pre-defined agreement limits; all discordant slides were reconciled and corrected before analysis. Prevalence in the sentinel and spot-check schools was broadly similar (31·2% verus 26·1%; within-municipality paired comparison p = 0·10).
Fig. 1.
Flow of participants through the survey. Legend–Percentages are of the enrolled total (n = 1960) unless indicated.
Table 1.
Soil-transmitted helminth prevalence, infection intensity, and co-infections by municipality and sex.
| Municipality | No. of stool examined |
Any STH positive |
Moderate-to-heavy intensity infection |
Co-infection | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Total (N) | Female | Male | Total | Female | Male | Total | Female | Male | ||
| Manatuto | 94 | 48 (51·1) | 46 (48·9) | 86 (91·5) | 42 (87·5) | 44 (95·7) | 78 (83·0) | 38 (79·2) | 40 (87·0) | 0 (0·0) |
| Bobonaro | 98 | 44 (44·9) | 54 (55·1) | 49 (50·0) | 21 (47·7) | 28 (51·9) | 27 (27·6) | 12 (27·3) | 15 (27·8) | 0 (0·0) |
| Ermera | 44 | 22 (50·0) | 22 (50·0) | 22 (50·0) | 13 (59·1) | 9 (40·9) | 13 (29·6) | 8 (36·4) | 5 (22·7) | 0 (0·0) |
| Lautém | 103 | 57 (55·3) | 46 (44·7) | 44 (42·7) | 24 (42·1) | 20 (43·5) | 32 (31·1) | 19 (33·3) | 13 (28·3) | 0 (0·0) |
| Ainaro | 69 | 34 (49·3) | 35 (50·7) | 29 (42·0) | 18 (52·9) | 11 (31·4) | 10 (14·5) | 5 (14·7) | 5 (14·3) | 0 (0·0) |
| Liquiçá | 66 | 21 (31·8) | 45 (68·2) | 24 (36·4) | 8 (38·1) | 16 (35·6) | 7 (10·6) | 1 (4·8) | 6 (13·3) | 5 (7·6) |
| Dili | 84 | 52 (61·9) | 32 (38·1) | 24 (28·6) | 15 (28·9) | 9 (28·1) | 9 (10·7) | 6 (11·5) | 3 (9·4) | 4 (4·8) |
| Aileu | 103 | 59 (57·3) | 44 (42·7) | 28 (27·2) | 16 (27·1) | 12 (27·3) | 7 (6·8) | 4 (6·8) | 3 (6·8) | 0 (0·0) |
| Manufahi | 95 | 46 (48·4) | 49 (51·6) | 20 (21·1) | 10 (21·7) | 10 (20·4) | 2 (2·1) | 1 (2·2) | 1 (2·0) | 0 (0·0) |
| Baucau | 109 | 52 (47·7) | 57 (52·3) | 14 (12·8) | 3 (5·8) | 11 (19·3)a | 4 (3·7) | 1 (1·9) | 3 (5·3) | 0 (0·0) |
| Covalima | 80 | 44 (55·0) | 36 (45·0) | 6 (7·5) | 1 (2·3) | 5 (13·9) | 4 (5·0) | 1 (2·3) | 3 (8·3) | 0 (0·0) |
| Viqueque | 79 | 35 (44·3) | 44 (55·7) | 4 (5·1) | 0 (0·0) | 4 (9·1) | 0 (0·0) | 0 (0·0) | 0 (0·0) | 0 (0·0) |
| Atauro | 120 | 55 (45·8) | 65 (54·2) | 3 (2·5) | 2 (3·6) | 1 (1·5) | 0 (0·0) | 0 (0·0) | 0 (0·0) | 0 (0·0) |
| Oe-Cusse | 92 | 46 (50·0) | 46 (50·0) | 2 (2·2) | 1 (2·2) | 1 (2·2) | 1 (1·1) | 1 (2·2) | 0 (0·0) | 0 (0·0) |
| Total | 1236 | 615 (49·8) | 621 (50·2) | 355 (28·7) | 174 (28·3) | 181 (29·2) | 194 (15·7) | 97 (15·8) | 97 (15·6) | 9 (0·7) |
Data are n (%) unless otherwise indicated. Counts are unweighted observed frequencies; the population-weighted national prevalence, estimated with municipality strata and school clusters, was 29·3% (95% CI 24·3–34·2). Municipalities are ordered by STH prevalence (highest to lowest). F, female; M, male. Moderate-to-heavy intensity infection defined by WHO thresholds: Ascaris lumbricoides ≥5000 eggs per gram (EPG); Trichuris trichiura ≥1000 EPG; hookworm ≥2000 EPG. Co-infection refers to simultaneous infection with two or more STH species (Ascaris, Trichuris, Hookworm); a further 8 children were co-infected with the non-STH cestode Hymenolepis nana.
p < 0·05 compared to female.
The weighted national STH prevalence was 29·3% (95% CI 24·3–34·2) (Supplementary Table S2); the unweighted prevalence was 28·7% (355/1236; 95% CI 25·0–32·0). Municipality-level prevalence showed extreme variation, from 2·2% in Oe-Cusse to 91·5% in Manatuto (Table 1; Supplementary Table S3).
A. lumbricoides dominated, accounting for 95·8% (340/355) of infections. T. trichiura represented 6·5% (23/355) and hookworm only 0·3% (1/355). Co-infections with two or more STH species occurred in 2·5% (9/355) of positive cases, predominantly A. lumbricoides with T. trichiura. Other intestinal parasites detected included Hymenolepis nana (3·6%; 45/1236) and Enterobius vermicularis (0·4%; 5/1236).
Among A. lumbricoides-positive samples, the median egg count was 6522 eggs per gram (EPG) (IQR 1446–15 030). , for T. trichiura, the median was 156 EPG (IQR 96–264), and geometric mean 169 EPG and a single hookworm case had 132 EPG. The grading of infection intensity across all species in various municipalities are summarised in Supplementary Table S3.
For comparison over time, we drew on two earlier national surveys: the 2012 baseline survey (2214 children in 30 schools; Ministry of Health, unpublished) and the 2020–21 TAS (8623 children in 234 schools).5 National prevalence was 29·1% (644/2214) in the 2012 baseline survey, 17·9% (1542/8623) in the 2020–21 TAS, and 28·7% (355/1236, unweighted) in 2025 (Supplementary Table S4). Moderate-to-heavy-intensity infection followed the same pattern: 5·6% (482/8623) of children examined in the 2020–21 TAS had moderate-to-heavy infections, compared with 15·7% (194/1236) in 2025; comparable intensity data are not available for the 2012 baseline survey. The change between 2020 and 2025 was not uniform across municipalities. The change was most pronounced in Bobonaro (9·8%–50·0%), Ermera (14·6%–50·0%), and Manatuto (29·6%–91·5%), all of which resumed deworming only in 2024, whereas municipalities that resumed earlier or had low baseline prevalence remained stable or declined, including Manufahi (43·0%–21·1%), Dili (24·1%–28·6%), Oe-Cusse (3·8%–2·2%), Viqueque (17·1%–5·1%) (Fig. 2; Supplementary Table S4). Across municipalities, the association between the duration of MDA interruption and the magnitude of post-2020 prevalence was positive but not statistically significant (Spearman's ρ = 0·31; p = 0·30).
Fig. 2.
Municipality-level STH prevalence across three time points (2012–2025) with the school-based deworming schedule. Legend–For each municipality, the upper panel shows any-STH prevalence at the 2012 baseline, the 2020–21 TAS, and the 2025 survey; the dotted line marks the 20% threshold. The shaded band marks the municipality-specific period without national mass deworming, ending at the year deworming resumed—2022 in Dili, Lautém, and Manufahi, and 2024 in the other municipalities. Line colour reflects the 2025 prevalence band (red ≥20%, orange 10–<20%, blue <10%). The strip below each panel, shows the number of mass drug administration (MDA) rounds delivered per year from 2012 to 2024 (grey 0, light blue 1, dark blue 2 rounds). Atauro has no 2012 baseline because it was then part of Dili municipality and only later became a separate municipality; accordingly, in line with Dili, Atauro received twice-yearly MDA during 2015–2019 (shown as 2 rounds per year in the strip). Municipality estimates (two schools each) illustrate the pattern of change rather than precise municipal values; see limitations.
Infected and uninfected children differed in mean height and weight, but these differences were attributable to age; after adjustment for age and sex, overall infection was not associated with height (adjusted coefficient −0·61 cm, 95% CI −1·36 to 0·14; p = 0·11) or weight (0·04 kg, −0·4 to 0·5; p = 0·88). Analysed by intensity, however, height was lower among children with moderate (−1·05 cm, 95% CI −2·03 to −0·08; p = 0·034) and heavy (−2·97 cm, −5·87 to −0·06; p = 0·045) infections than among uninfected children which was statistically significant.
In multivariate logistic regression adjusted for age and sex and fitted to all 1236 children examined (Table 2), municipality was the strongest predictor of infection. Compared with Oe-Cusse (reference), the adjusted odds ratio of infection (aOR) was substantially higher in several other municipalities. No substantial multicollinearity was detected, with a maximum variance inflation factor of 1·96. A survey-weighted sensitivity analysis (municipality strata, school clusters, post-stratification weights) yielded materially similar estimates, with municipality remaining the dominant correlate. Municipality prevalence is extreme in both directions (Manatuto 91·5%, Oe-Cusse 2·2%), so a Firth penalised logistic model was fitted as a sensitivity analysis; it gave the same ranking of municipality effects and a less extreme estimate for Manatuto (adjusted OR 274·23, 95% CI 76·26–1511·77, versus 362·28, 74·35–1765·19 by maximum likelihood). Age was associated with infection (aOR 1·23 per year, 1·11–1·37; p < 0·001) but sex was not (1·03, 0·77–1·39; p = 0·83).
Table 2.
Univariate and multivariate logistic regression analysis of risk factors for soil-transmitted helminth infection.
| Risk factors | Any STH positive |
||
|---|---|---|---|
| n/N (%) | Univariate OR (95% CI); p value | Multivariate OR (95% CI); p value | |
| Panel A. Demographic and geographic factors (single model, n = 1236) | |||
| Age | 355/1236 (28·7) | 1·43 (1·32–1·56); <0·001 | 1·23 (1·11–1·37); <0·001 |
| Sex | |||
| Female | 174/615 (28·3) | Ref | Ref |
| Male | 181/621 (29·2) | 1·04 (0·81–1·33); 0·74 | 1·03 (0·77–1·39); 0·83 |
| Municipality | |||
| Aileu | 28/103 (27·2) | 16·80 (3·87–72·84); <0·001 | 17·56 (4·04–76·30); <0·001 |
| Ainaro | 29/69 (42·0) | 32·63 (7·42–143·39); <0·001 | 25·50 (5·76–112·82); <0·001 |
| Atauro | 3/120 (2·5) | 1·15 (0·19–7·05); 0·88 | 1·14 (0·19–6·96); 0·89 |
| Baucau | 14/109 (12·8) | 6·63 (1·47–30·00); 0·01 | 5·50 (1·21–25·02); 0·03 |
| Bobonaro | 49/98 (50·0) | 45·0 (10·49–193·01); <0·001 | 38·87 (9·02–167·45); <0·001 |
| Covalima | 6/80 (7·5) | 3·65 (0·72–18·61); 0·12 | 3·60 (0·70–18·40); 0·12 |
| Dili | 24/84 (28·6) | 18·0 (4·10–79·0); <0·001 | 18·35 (4·17–80·71); <0·001 |
| Ermera | 22/44 (50·0) | 45·0 (9·83–205·90); <0·001 | 41·69 (9·08–191·49); <0·001 |
| Lautém | 44/103 (42·7) | 33·56 (7·84–143·73); <0·001 | 30·48 (7·10–130·83); <0·001 |
| Liquiçá | 24/66 (36·4) | 25·71 (5·81–113·90); <0·001 | 17·07 (3·79–76·91); <0·001 |
| Manatuto | 86/94 (91·5) | 483·75 (99·90–2342·54); <0·001 | 362·28 (74·35–1765·19); <0·001 |
| Manufahi | 20/95 (21·1) | 12 (2·72–53·01); 0·001 | 11·88 (2·68–52·56); 0·001 |
| Oe-Cusse | 2/92 (2·2) | Ref | Ref |
| Viqueque | 4/79 (5·1) | 2·40 (0·43–13·47); 0·32 | 2·16 (0·38–12·14); 0·38 |
| Panel B. Household WASH factors (each factor in a separate model adjusted for age, sex, and municipality) | |||
| WASH in house | |||
| Water source (n = 512) | |||
| No water | 5/19 (26·3) | Ref | |
| Unsafe water | 44/174 (25·3) | 0·95 (0·32–2·78); 0·92 | 0·80 (0·24–2·64); 0·71 |
| Safe water | 99/319 (31·0) | 1·26 (0·44–3·59); 0·67 | 0·92 (0·28–2·97); 0·88 |
| Toilet (n = 565) | |||
| Open defecation | 6/30 (20·0) | Ref | Ref |
| Unimproved toilet | 9/36 (25·0) | 1·33 (0·41–4·30); 0·63 | 1·12 (0·25–5·15); 0·88 |
| Improved toilet | 144/499 (28·9) | 1·62 (0·65–4·05); 0·30 | 1·70 (0·48–5·99); 0·41 |
| Handwash (HW) station at toilet (n = 588) | |||
| No | 42/145 (29·0) | Ref | Ref |
| Yes | 128/443 (28·9) | 1·00 (0·66–1·51); 0·99 | 0·81 (0·48–1·36); 0·43 |
| Soap at HW station (n = 359) | |||
| No | 8/18 (44·4) | Ref | Ref |
| Yes | 93/341 (27·3) | 0·47 (0·18–1·22); 0·12 | 0·51 (0·17–1·54); 0·23 |
| Water available at HW station (n = 440) | |||
| No | 34/106 (32·1) | Ref | Ref |
| Yes | 89/334 (26·7) | 0·77 (0·48–1·24); 0·28 | 0·67 (0·37–1·21); 0·18 |
| WASH practice | |||
| Mostly wear shoes (n = 665) | |||
| No | 51/188 (27·1) | Ref | Ref |
| Yes | 143/477 (30·0) | 1·15 (0·79–1·68); 0·47 | 1·00 (0·63–1·58); 0·99 |
| Mostly use utensil for meals (n = 665) | |||
| No | 121/396 (30·6) | Ref | Ref |
| Yes | 73/269 (27·1) | 0·85 (0·60–1·19); 0·34 | 0·90 (0·59–1·36); 0·61 |
| Mostly wash hands before meal (n = 665) | |||
| No | 62/231 (26·8) | Ref | Ref |
| Yes | 132/434 (30·4) | 1·19 (0·83–1·70); 0·33 | 1·12 (0·73–1·74); 0·60 |
| Mostly wash hands after defecation (n = 665) | |||
| No | 61/206 (29·6) | Ref | Ref |
| Yes | 133/459 (29·0) | 0·97 (0·68–1·39); 0·87 | 1·03 (0·66–1·60); 0·90 |
| Taken deworming tablet in last 3 years (n = 665) | |||
| No | 45/162 (27·8) | Ref | Ref |
| Yes | 149/503 (29·6) | 1·09 (0·74–1·62); 0·65 | 0·97 (0·61–1·55); 0·90 |
OR, odds ratio; CI, confidence interval; Ref, reference category; WASH, water, sanitation, and hygiene; VIF, variance inflation factor.
The outcome is infection with any soil-transmitted helminth species.
Household WASH indicators, each estimated in a separate model adjusted for age, sex, and municipality (Supplementary Fig. S2a and b; Supplementary Tables S5 and S6), showed no significant association with infection (Table 2).
Discussion
This national survey quantifies the STH prevalence among school-going children in Timor-Leste following a prolonged, pandemic-related interruption of school-based deworming. An increase in prevalence following the cessation of deworming was anticipated; the contribution of these data lies in quantifying the magnitude of that increase and its spatial heterogeneity. The weighted national prevalence of 29·3% is close to the 2012 baseline (29%) and well above the 2020 figure (18%), indicating an almost complete reversal of the gains recorded during years of regular treatment, and moderate-to-heavy infections affected more than half of infected children. These findings are best interpreted at the national level; because only two schools were sampled per municipality, municipality-specific estimates are indicative and are used here to describe the pattern of change rather than to classify individual municipalities.
The national increase was not uniform, and this heterogeneity is informative about its likely cause. The largest rebounds occurred in municipalities where deworming resumed only in 2024 (Manatuto, Ermera, Bobonaro), whereas municipalities that resumed earlier (Dili, Manufahi, and Lautém) or that had low baseline prevalence (Oe-Cusse, Atauro, and Viqueque) remained stable or continued to decline. Although the correlation between the duration of interruption and the magnitude of rebound did not reach statistical significance, which is unsurprising given only 14 municipalities and the limited variation in interruption duration, the contrast between early- and late-resuming municipalities is consistent with the interruption of preventive chemotherapy, rather than the pandemic itself, being the proximate driver of the increase. This pattern is consistent with the modelling and empirical studies6, 7, 8 and provides real-world evidence of the indirect effect of COVID-19 on NTD control programmes.
The scale of the rebound in the highest-transmission municipalities is also notable. Modelling studies have generally predicted a return towards pre-control levels within a few years of interruption,6, 7, 8,11, 12, 13, 14 but in several municipalities, most strikingly Manatuto, where prevalence reached 91·5% against a 2012 baseline of approximately 30%, the 2025 estimates appear to exceed earlier baselines rather than simply returning to them. Although this pattern could suggest an intense-transmission foci, a pattern that deterministic models calibrated to lower-transmission settings may underestimate, this interpretation should, however, be treated cautiously. Each municipality was represented by only two schools, the 2012 and 2025 surveys used related but not identical sampling frames, and the confidence intervals around municipal estimates are wide. The apparent overshoot is therefore best regarded as a hypothesis for confirmation in a future adequately powered impact assessment rather than an established finding.
Because national deworming was interrupted between 2020 and 2023 and resumed nationally only in 2024, fewer than two effective rounds preceded the 2025 survey. This survey was therefore a national prevalence assessment rather than a post-treatment impact assessment conducted after the five effective rounds envisaged in the WHO monitoring and evaluation framework, and the four-category decision framework applied after such an assessment (prevalence <2%, 2%–<10%, 10%–<20%, and ≥20%) is not yet applicable.4 We instead treat the survey as a re-baseline and use current prevalence to set the frequency of school-based deworming as it is re-established: twice-yearly treatment where prevalence is at least 50% (Manatuto, Ermera, and Bobonaro) and annual treatment where it is 20% to less than 50% (Lautém, Ainaro, Liquiçá, Dili, Aileu, and Manufahi). In the remaining municipalities, all below 20%, we suggest maintaining the school-based platform at a reduced frequency rather than withdrawing it, considering the impact of MDA interruption this should be a programmatic judgement rather than a WHO requirement (Supplementary Table S4). After five years of regular treatment, an impact assessment should be conducted and the four-category framework applied. Applying the accompanying morbidity criterion to the present data illustrates why that assessment cannot be brought forward: no municipality has an upper 95% confidence bound for moderate-to-heavy-intensity infection below 2%, including Atauro and Viqueque, where no such infections were detected (upper bounds 3·0% and 4·6%), because two schools per municipality do not provide sufficient precision.
A. lumbricoides accounted for almost all infections, indicating that albendazole or mebendazole remain appropriate for preventive chemotherapy.15 The very low hookworm prevalence may reflect improved sanitation, reduced barefoot exposure, or residual effects of previous MDA. The morbidity relevance of infection intensity was supported by the growth analysis: after adjustment for age and sex, height was significantly lower among children with moderate and heavy infections but not among those with light infections. However, detection of H. nana (3·6%) underscores broader parasitological challenges requiring attention. While albendazole has limited efficacy against H. nana, praziquantel could be considered in high-prevalence foci such as Atauro, where 24·4% (11/45) of positive cases reside.
The very high prevalence in some municipalities has implications for how control is delivered. In the highest-burden settings, where almost all children were infected and majority had moderate-to-heavy infections, school-based deworming alone is unlikely to interrupt transmission, because a large environmental reservoir and infection among untreated preschool children and adults can sustain reinfection. In such settings, deworming should be complemented by WASH improvements, health education, and community-based treatment to achieve sustainable control15, 16, 17, 18, 19; these measures constitute essential rather than optional components of a durable response. At the same time, previous analyses have noted that community-wide MDA is substantially more resource-intensive than school-based delivery,6,20,21 and in a setting such as Timor-Leste, where even school-based deworming has been difficult to maintain, a pragmatic approach is to first restore and strengthen school-based deworming, increasing frequency where prevalence is highest, while progressively adding WASH and community components. The relative stability of municipalities that maintained partial school-based activity during the pandemic supports the value of protecting this delivery platform. However, it should be seen as a foundation and not as a substitute for the broader measures needed in high transmission settings.
The absence of an independent association between household WASH indicators and infection after adjustment for municipality should be interpreted cautiously. WASH data were self-reported, came from a caregiver questionnaire with substantial item non-response that required each indicator to be modelled separately, and were strongly patterned by municipality, which limits the power to detect independent effects; the finding does not imply that WASH is unimportant for STH control.15, 16, 17, 18, 19
This study has several limitations. First, and most importantly, only two schools were sampled per municipality; the survey was designed and powered for national estimation, and municipality-level estimates (of prevalence, intensity, and WASH indicators) have wide uncertainty and should not be used for municipality-specific decisions without confirmatory sampling. Second, the stool submission rate was 63% (of enrolled children) and varied between municipalities (lowest in Ermera), which could introduce bias if non-response were related to infection status. However, this rate was not unusually low for the setting: attendance on the collection day (86·6%) was close to the national attendance rate for children aged 5–17 years (84%),22 and submission among children who were present (1238/1697, 72·9%) was similar to that in the 2020–21 TAS (8623/11,677, 73·8% with an adequate sample). Submission differed little by sex or age, but municipalities with few analysed children (for example, Ermera, n = 44) yielded imprecise estimates. Third, single-day sampling with a single stool per child may misclassify light infections and underestimate intensity. Fourth, the comparison across time points relies on earlier surveys that used related but not identical sampling frames, so temporal changes should be read as programmatic trends rather than exact like-for-like differences. Fifth, WASH and behavioural data were self-reported and incomplete. Finally, as a cross-sectional survey it cannot establish causation, including for the association between deworming interruption and rebound.
Future work should include a formal impact assessment once several effective rounds of deworming have been delivered, to classify municipalities according to the WHO framework; longitudinal studies of reinfection and seasonal variation; and environmental and molecular studies in persistently high-prevalence settings to characterise transmission and monitor for reduced drug efficacy.
STH prevalence among school-going children in Timor-Leste has increased to near pre-control levels following the interruption of school-based deworming during the COVID-19 pandemic, with the largest increases where treatment resumed latest. Restoring and strengthening school-based deworming, with frequency matched to current prevalence and complemented by WASH strategies, health education, and community engagement, is needed to regain progress towards the WHO 2030 targets and to build resilience against future disruptions.
Contributors
HJ, AY, and MHC conceptualised and designed the study. HJ, MMAAN, FP, JS, AM and MC coordinated field implementation. HJ, AM, and MC supervised data collection. HJ performed statistical analyses and draughted the initial manuscript. AY provided technical oversight and WHO regional perspective. All authors contributed to data interpretation, critically revised the manuscript for important intellectual content, and approved the final version. HJ, FP and MC had full access to all data. HJ and MC accessed and verified all the data reported in the study and take responsibility for data integrity and analysis accuracy.
Data sharing statement
De-identified participant data, data dictionary, and study protocol will be made available upon reasonable request to the corresponding author (MHC). Requests should include a brief proposal outlining intended use and will require data access agreement. Data will be available beginning 3 months after publication, with no end date specified.
Editor note
The Lancet Group takes a neutral position with respect to territorial claims in published maps and institutional affiliations.
Declaration of interests
All authors report grants from the Korea International Cooperation Agency (KOICA) and the United Nations Department of Economic and Social Affairs (UNDESA), paid to their institutions, which supported the Timor-Leste school deworming programme and survey from which the data reported in this study were derived. The funders provided no support for the analysis or preparation of this manuscript. The authors declare no other competing interests.
Acknowledgements
We thank the Ministry of Health leadership, particularly the Department of Communicable Diseases Control; municipal health teams across all 14 municipalities; WHO Country Office Timor-Leste; Universidade Nacional Timor Lorosa'e; National Health Laboratory; field teams; school directors and teachers; and children and families for their support. We acknowledge the WHO regional office for technical support. This study was funded by the Korea International Cooperation Agency (KOICA) and the United Nations Department of Economic and Social Affairs (UNDESA). The findings, interpretations, and conclusions expressed in this work represent the personal opinions of the authors and do not necessarily reflect the views of the World Health Organization, the Ministry of Health of Timor-Leste, or any other affiliated organizations. During the preparation of this work Claude (Anthropic) was used to assist with grammar checking and language refinement. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the final content of the publication.
Footnotes
Supplementary data related to this article can be found at https://doi.org/10.1016/j.lansea.2026.100870.
Appendix A. Supplementary data
References
- 1.WHO . World Health Organization; Geneva: 2017. Guideline: Preventive Chemotherapy to Control Soil-Transmitted Helminth Infections in at-Risk Population Groups. [PubMed] [Google Scholar]
- 2.Brooker S., Clements A.C.A., Bundy D.A.P. Global epidemiology, ecology and control of soil-transmitted helminth infections. Adv Parasitol. 2006;62:221–261. doi: 10.1016/S0065-308X(05)62007-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Miguel E., Kremer M. Worms: identifying impacts on education and health in the presence of treatment externalities. Econometrica. 2004;72(1):159–217. [Google Scholar]
- 4.WHO . World Health Organization; Geneva: 2024. Assessing Schistosomiasis and Soil-Transmitted Helminthiases Control Programmes: Monitoring and Evaluation Framework. Licence: CC BY-NC-SA 3.0 IGO. [Google Scholar]
- 5.WHO . World Health Organization; Dili: 2021. Transmission Assessment Survey in Timor-Leste, 2020–2021: Final Report. [Google Scholar]
- 6.Malizia V., Giardina F., Vegvari C., et al. Modelling the impact of COVID-19-related control programme interruptions on progress towards the WHO 2030 target for soil-transmitted helminths. Trans R Soc Trop Med Hyg. 2021;115(3):253–260. doi: 10.1093/trstmh/traa156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Borlase A., Le Rutte E.A., Castaño S., et al. Evaluating and mitigating the potential indirect effect of COVID-19 on control programmes for seven neglected tropical diseases: a modelling study. Lancet Glob Health. 2022;10(11):e1600–e1611. doi: 10.1016/S2214-109X(22)00360-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Truscott J.E., Turner H.C., Farrell S.H., Anderson R.M. Soil-transmitted helminths: mathematical models of transmission, the impact of mass drug administration and transmission elimination criteria. Adv Parasitol. 2016;94:133–198. doi: 10.1016/bs.apar.2016.08.002. [DOI] [PubMed] [Google Scholar]
- 9.WHO . 2nd ed. World Health Organization; Geneva: 2011. Helminth Control in School-Age Children: A Guide for Managers of Control Programmes. [Google Scholar]
- 10.WHO . 2nd ed. World Health Organization; Geneva: 2019. Bench Aids for the Diagnosis of Intestinal Parasites. [Google Scholar]
- 11.Truscott J.E., Turner H.C., Anderson R.M. What impact will the achievement of the current WHO targets for anthelmintic treatment coverage in children have on the intensity of soil-transmitted helminth infections? Parasit Vectors. 2015;8:551. doi: 10.1186/s13071-015-1135-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Clarke N.E., Clements A.C.A., Doi S.A.R., et al. Differential effect of mass deworming and targeted treatment on intensity of soil-transmitted helminth infection. Lancet Glob Health. 2017;5(7):e682–e693. [Google Scholar]
- 13.Anderson R., Truscott J., Hollingsworth T.D. The coverage and frequency of mass drug administration required to eliminate transmission of soil-transmitted helminths. Phil Trans R Soc B. 2014;369(1645) doi: 10.1098/rstb.2013.0435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Pullan R.L., Smith J.L., Jasrasaria R., Brooker S.J. Global numbers of infection and disease burden of soil-transmitted helminth infections in 2010. Parasit Vectors. 2014;7:37. doi: 10.1186/1756-3305-7-37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Keiser J., Utzinger J. Efficacy of current drugs against soil-transmitted helminth infections: systematic review and meta-analysis. JAMA. 2008;299(16):1937–1948. doi: 10.1001/jama.299.16.1937. [DOI] [PubMed] [Google Scholar]
- 16.Freeman M.C., Garn J.V., Sclar G.D., et al. The impact of sanitation on infectious disease and nutritional status: a systematic review and meta-analysis. Int J Hyg Environ Health. 2017;220(6):928–949. doi: 10.1016/j.ijheh.2017.05.007. [DOI] [PubMed] [Google Scholar]
- 17.Strunz E.C., Addiss D.G., Stocks M.E., Ogden S., Utzinger J., Freeman M.C. Water, sanitation, hygiene, and soil-transmitted helminth infection: a systematic review and meta-analysis. PLoS Med. 2014;11(3) doi: 10.1371/journal.pmed.1001620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Freeman M.C., Ogden S., Jacobson J., et al. Integration of water, sanitation, and hygiene for the prevention and control of neglected tropical diseases: a rationale for inter-sectoral collaboration. PLoS Negl Trop Dis. 2013;7(9) doi: 10.1371/journal.pntd.0002439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Campbell S.J., Savage G.B., Gray D.J., et al. Water, sanitation and hygiene (WASH): a critical component for sustainable soil-transmitted helminth and schistosomiasis control. PLoS Negl Trop Dis. 2014;8(4) doi: 10.1371/journal.pntd.0002651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Anderson R.M., Truscott J.E., Pullan R.L., Brooker S.J., Hollingsworth T.D. How effective is school-based deworming for the community-wide control of soil-transmitted helminths? PLoS Negl Trop Dis. 2013;7(2) doi: 10.1371/journal.pntd.0002027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Montresor A., Gabrielli A.F. Finding realistic solutions to NTD target delays. Lancet Glob Health. 2022;10(11):e1541–e1542. doi: 10.1016/S2214-109X(22)00427-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.ILO . International Labour Organization; Geneva: 2019. Timor-Leste National Child Labour Survey 2016: Analytical Report. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.