Factors affecting detection and quantification of Schistosoma haematobium eggs in pooled urine samples

Abraham Degarege; Bruno Levecke; Christopher R. Bilder; David M. Brett-Major; Abebe Animut; Yohannes Negash; M. Jana Broadhurst; Tzeyu L. Michaud; Berhanu Erko

doi:10.1371/journal.pntd.0014407

Abstract

Background

Building on our previous work showing that pooled urine testing can rapidly detect Schistosoma haematobium infections but has limited sensitivity in low-intensity settings or with large pool sizes, this study evaluated how urine volume, pool size, infection intensity, and diagnostic method affect the detection and quantification of S. haematobium eggs in pooled samples.

Methods

Between July 2022 and April 2023, 2,134 urine samples from school-age children living in three regional states of Ethiopia were individually examined by deploying urine filtration microscopy (UFM). Subsequently, 5, 10, 20 and 40 individual samples were strategically pooled and examined by deploying UFM and Fluke Catcher (FC) and varying volumes of urine (10, 20 and 30 mL).

Results

UFM was significantly more sensitive than FC for detecting S. haematobium eggs in pooled urine samples ( =0.83, p < 0.001). There were significant interactions between the mean log of urine egg count (UEC) and volume of urine ( =0.01, p = 0.021) or pool size ( =-0.02, p = 0.007) for detecting S. haematobium eggs in pooled samples. The odds of detecting eggs in pooled samples in which the mean UEC of individual samples equaled 10 eggs/mL were 0.78 (95% CI:0.67-0.90) and 0.60 (95% CI:0.51-0.76) times lower when 10 mL rather than 20 mL and 30 mL was examined, respectively. There was a moderate or strong positive correlation between the mean egg count of pooled samples and the average egg count of the individual samples making the pools when pool sizes were 5, 10 or 20 (r ≥ 0.5).

Conclusions

The diagnostic sensitivity of pooled testing for diagnosing S. haematobium infection is affected by a complex interplay between the pool size, the volume of urine examined, the mean UEC and the diagnostic method. The sensitivity of the pooled testing strategy may increase with larger urine volumes, higher infection intensity, and when UFM (rather than FC) is used for testing. However, sensitivity could decrease as the pool size increases. Use of UFM, larger urine volumes and smaller to moderate pool sizes could improve detection of S. haematobium infection in surveillance and mass deworming programs in low-intensity settings.

Author summary

Schistosoma haematobium is a parasite that causes a serious disease affecting millions of people in sub-Saharan Africa and the Middle East. The infection is usually detected by looking for parasite eggs in urine samples under a microscope. However, testing each person individually is time-consuming and costly, especially during large community surveys or mass treatment campaigns. This study explored a faster and more affordable approach called pooled testing, where urine samples from several people are combined and tested together. We collected urine samples from school-aged children in parts of Ethiopia where this infection is common. Each sample was tested individually and in pools of 5, 10, 20, or 40 people, using urine filtration microscopy (UFM) and the Fluke Catcher (FC) methods while varying the urine volume tested (10, 20, or 30 mL). We found that UFM was more accurate than the FC for detecting infection in pooled samples. The pooled testing method worked best when the infected person had more parasite eggs, particularly when larger urine volumes were examined, and smaller pool sizes were used. These results suggest that pooled urine testing could be a practical solution for rapidly monitoring S. haematobium infection in resource-limited communities. Optimizing pool size, urine volume, and testing methods can improve accuracy and support community screening and deworming programs in resource-limited areas.

Citation: Degarege A, Levecke B, Bilder CR, Brett-Major DM, Animut A, Negash Y, et al. (2026) Factors affecting detection and quantification of Schistosoma haematobium eggs in pooled urine samples. PLoS Negl Trop Dis 20(6): e0014407. https://doi.org/10.1371/journal.pntd.0014407

Editor: Luc E. Coffeng, Erasmus MC, University Medical Center Rotterdam, NETHERLANDS, KINGDOM OF THE

Received: October 9, 2025; Accepted: May 25, 2026; Published: June 1, 2026

Copyright: © 2026 Degarege et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All data underlying findings in this submission is fully available, without restriction as supplementary file (S1 Data).

Funding: Research reported in this publication was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under Award Number R21AI164376 awarded to AD, BL, CRB, DBM, AA, JBM, TM and BE. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The National Institute of Allergy and Infectious Diseases of the National Institutes of Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Competing interests: The authors have declared that no competing interests exist.

Background

Urogenital schistosomiasis is a parasitic disease caused by the flatworm Schistosoma haematobium and affects more than 112 million people worldwide, particularly in sub-Saharan Africa and the Middle East [1–3]. In children, the disease can cause hematuria (blood in urine), anemia, malnutrition, and cognitive deficiency, while in adults, complications include kidney damage, fibrosis in the bladder, narrowing of the ureters, genital lesions, and bladder cancer [1–3]. In high endemicity settings, children often suffer more severe disease [4].

To reduce transmission of schistosomiasis and the associated morbidities, school-age children living in endemic regions periodically receive a single oral dose of 40 mg/kg praziquantel through mass drug administration (MDA) programs [5,6]. Decisions on whether to initiate such large-scale deworming programs in a specific area and at which frequency (annually vs. bi-annually) are based on the disease prevalence [5,6]. Once initiated, there is also a need to periodically re-assess the prevalence to verify if scaling down the frequency or even stopping the administration of drugs is justified [5,6]. To this end, the availability of a reliable and sustainable S. haematobium diagnostic strategy remains a top priority in endemic areas. Screening of individual urine samples remains the current diagnostic strategy to detect cases and subsequently draw conclusions. Yet, this strategy is resource intensive [1–4].

A potential cost-saving strategy involves pooling individual samples and applying a cascaded pooled testing (CPT) strategy. In such a testing strategy, samples are first tested as a pool, and samples from positive pools are then tested individually to identify the positive cases [7,8]. Conversely, if a pool tests negative, all individuals within that pool are considered negative [7,8].

CPT has a very long history for a diverse set of applications. While Robert Dorfman [7] is usually given credit as the first to suggest the process, there are records of CPT being used as far back as 1915 [9]. Since then, CPT has been refined and widely used across applications including SARS-CoV-2 testing [10], blood donation screening [11], sexually transmitted infection detection [12], and animal disease testing [13]. General textbook summaries of CPT are available from a variety of sources [9,14,15] and from general overview papers [9,16,17].

CPT has been shown to reduce the number of tests, and hence the operational costs, of testing programs for infectious diseases, including helminthiasis [18–21]. However, the application of CPT to helminth egg detection in urine has been limited. Previously, we evaluated the performance of a CPT strategy (pools of 5, 10 and 20 individual samples) in determining the presence and the intensity of S. haematobium infection (by counting the number of eggs in 10 mL of urine) at the population level applying the urine filtration method (UFM) [22,23]. These studies showed that pooled testing strategy can provide a rapid assessment of infection intensity; however, its sensitivity for estimating population prevalence may be reduced in low intensity settings, particularly as pool size increases [22,23]. Thus, more studies are needed to determine when and how pooling can be optimally implemented to monitor S. haematobium infections in the context of MDA programs.

Potential measures to improve the performance of this strategy are (i) to screen an increased volume of urine and (ii) to apply more sensitive diagnostic tests. UFM is the standard method for the diagnosis of S. haematobium infection. By this method, urine samples are filtered through nylon or polycarbonate membranes followed by microscopic examination of the filter for parasite eggs [1–3]. UFM shows reduced sensitivity for detecting low intensity S. haematobium infection [1–3]. The Fluke Catcher (FC) is an alternative diagnostic method that allows for examining a larger volume of urine. This method has been deployed in veterinary medicine to detect fluke eggs (e.g., Fasciola) in the stool of livestock and consists of a stack of three filters with different mesh sizes [24–26]. Recently, the FC method has shown high sensitivity for detecting Fasciola hepatica eggs in human stool [27], but its performance for detecting and quantifying S. haematobium eggs in urine is not known.

The aim of our study was to explore how pooled urine testing can be optimally implemented for monitoring S. haematobium infections in the context of an MDA program. For this, we designed a set of laboratory experiments to examine the impact of (i) volume of urine (10, 20 and 30 mL), (ii) pool size (5, 10, 20, and 40 individual samples) (iii) intensity of infection, and (iv) diagnostic method (UFM vs. FC) on the ability of a pooled testing strategy to detect and quantify S. haematobium eggs in pooled urine samples.

Methods

Ethics statement

This study was approved by the institutional review boards of the University of Nebraska Medical Center, USA (Ref. No. IRB #0875-21-EP) and Aklilu Lemma Institute of Health Research, Ethiopia (Ref. No. ALIPB IRB/63/2014/2021). The district health office, school authorities, and teachers were informed about the purpose and procedures of the study. Written informed consent without a signature was obtained from parents or guardians of children who assent to participate in the study. As the study population was members of a distinct cultural group or community in which signing forms is not the norm, the institutional review board approved a waiver of signed consent for this study. Accordingly, parents or legal guardians received written information describing the study objectives, procedures, risks, and benefits. Verbal informed consent was obtained in lieu of written signatures, with a study team member serving as witness to confirm that the information was explained, questions were answered, and participation was voluntary. A praziquantel treatment (40 mg/kg body weight) was provided free of charge to all children infected with S. haematobium (based on UFM using 10 mL of urine). Praziquantel was administered by participating health workers from the local area.

Study design

We conducted an experimental study using urine specimens collected from Ethiopian children between June 2022 and April 2023. Urine samples were collected at baseline and one month after praziquantel was administered to children infected with the parasite based on urine testing results. All urine samples were individually examined for S. haematobium eggs by UFM. Urine samples with S. haematobium eggs detected were pooled with negative samples and examined using UFM and FC at the Medical Parasitology Laboratory of Aklilu Lemma Institute of Health Research of Addis Ababa University.

Study population and study area

The study was conducted among school-age children (5–15 years of age) living in the three Ethiopian Regional States (Afar, Benishangul-Gumuz and Gambella) where urogenital schistosomiasis is endemic (prevalence of the disease in Afar: 2.3% to 52.0%; in Benishangul-Gumuz: 35.9% to 57.8%; in Gambella: 35.9% to 43.8%) [22,23,28–30]. The Global Positioning System coordinates of the study areas were as follows: Afar 11.8167, 41.4167; Benishangul Gumuz 10.6390, 35.7330; Gambella 7.6184, 34.6893. A total of 15 villages were selected for this study, including seven villages from the Afar region, seven from the Benishangul-Gumuz region, and one from the Gambella region. These locations were chosen based on reports of S. haematobium transmission or expectations of its presence as recommended by local community leaders. Children living in these study regions are sporadically treated with praziquantel as part of the MDA program. However, the children who participated in this study did not receive PZQ within three months of the data collection period.

Sample size

The sample size for each study region was estimated using published sample size tables from a review article that summarize the minimum number of samples required for sensitivity and specificity analyses [31]. These estimates were calculated using Power Analysis and Sample Size (PASS) software, incorporating specified confidence interval levels, statistical power, and disease prevalence assumptions [32]. We previously found a 51.7% sensitivity of the hierarchical pooling technique for detecting S. haematobium infection using UFM with 10 mL urine (H_o) [23]. In order to test the hypothesis that the sensitivity of the pooled testing is at least 60% (H_a) when the volume of urine examined is > 10 mL with a power of 80% (type II error of 20%) and type I error level of 5%, we needed to screen a minimum of 995 children in Afar and 995 children in Benishangul-Gumuz where the prevalence of S. haematobium infection was estimated to be 20%, and 498 in Gambella where the prevalence of S. haematobium infection was estimated to be 40% [22,23,28–30]. Due to the low prevalence rate in some villages in the Afar region, we proposed collecting samples from an additional 100 (10% of 995) children in Afar to ensure the collection of an adequate number of positive samples to assess sensitivity. The total sample size estimated for this study was 2,588.

Urine sample collection and processing

We first briefed the school directors and village officials on the study’s objectives. Subsequently, the administrators of the villages informed the villagers about the aim of the research. Then, the study team recruited children who agreed to participate, and whose parents or guardians approved their participation (see ethics statement). Eligible children were recruited using consecutive (convenience) sampling from predefined community collection points, including nearby health posts, schools, and open field stations within the villages. Consent discussion was conducted privately in nearby health posts, schools, or designated open fields within the villages, where the study objectives, procedures, potential benefits, and risks were explained. Following consent and assent, participating children were asked to provide at least 80 mL in labeled 200 mL plastic containers between 10:00 am and 3:00 pm. Urine samples were collected before recording demographic data such as age and gender to minimize bias. From each 80 mL urine sample, 10 mL was filtered using a polycarbonate filter membrane (13 mm diameter and 12–20 μm pore size) and examined for S. haematobium eggs at the collection sites or the nearby health posts using UFM. Prior to subsampling for filtration and egg quantification, the individual urine samples were thoroughly homogenized by gentle inversion and agitation to minimize sedimentation and ensure an even distribution of eggs within each aliquot. UFM results were shared confidentially with participating children and their parents or guardians at the sample collection sites, with support from local health personnel or nurses and school staff or village officials when needed. Children whose urine tested positive for S. haematobium eggs were treated with Praziquantel by participating health workers. The remaining 70 mL of urine was transferred to a vial containing 0.7 mL formalin (37% formaldehyde) and transported to the Medical Parasitology Laboratory of Aklilu Lemma Institute of Health Research of Addis Ababa University.

Experimental pooling of urine samples was carried out in the laboratory, approximately 1–3 weeks after their collection in the field. We first re-tested 10 mL of the formalin-fixed urine samples from each child using UFM after the field data collection. An experienced microscopist with over 15 years of expertise in using UFM for S. haematobium diagnosis performed the examination. Based on the UFM results obtained at the collection site and in the laboratory, each individual urine sample was declared either positive (any egg was found at least once) or negative for S. haematobium eggs (no eggs were found on both occasions). Then, plastic vials containing formalin-fixed individual urine samples were arranged in groups of 5, 10, 20, and 40 samples, respectively (Fig 1). In each group, one positive sample was included, while the remaining samples were negative. The positive sample included in each pool was selected at random from the set of known positive specimens and was not chosen based on infection intensity. Then, an aliquot of each sample within a group was transferred into a new vial, resulting in pooled samples with a fixed total volume of 140 mL (pools of 5: 28 mL per sample; pools of 10: 14 mL per sample; pools of 20: 7 mL per sample and pools of 40: 3.5 mL per sample). Finally, after shaking/mixing the pooled samples to ensure uniform distribution of the eggs of the parasite, an aliquot of 10 mL, 20 mL, and 30 mL of each pooled urine sample were tested deploying UFM [33,34] and FC [24]. The pooled samples were mixed by gentle inversion and agitation immediately before aliquots were taken for filtration and egg quantification. One month after treatment, children who tested positive for S. haematobium infection using UFM in the field and were treated with praziquantel (40 mg/kg body weight) were re-examined using the same methodology as described above. During the second-round survey, children who tested negative for S. haematobium using UFM in the field but were subsequently found positive in laboratory testing received treatment.

Download:

Fig 1. The applied strategy to experimentally make pools of 5, 10, 20 and 40 individual samples.

Formalin-fixed individual urine samples were arranged in groups of 5 (Panel A), 10 (Panel B), 20 (Panel C) and 40 (Panel D) individual urine samples containing one positive sample, while the remaining samples were negative. Then, an aliquot, proportional to the number of individual samples, was transferred into a new vial, resulting in pooled samples with a fixed total volume of 140 mL. Created in BioRender. Mengist, A. (2026) https://BioRender.com/ad0khd9.

https://doi.org/10.1371/journal.pntd.0014407.g001

Data management and statistical data analysis

The raw data were first entered into an Excel spreadsheet and carefully checked for errors, incorrect values, or missing data. Since missing values accounted for less than 1% and were related to age, sex, or individual sample test results, incomplete entries were removed. Inaccurate or inconsistent values were corrected by cross-referencing with the source documents. After addressing any inaccuracies, inconsistencies, or missing data, the cleaned dataset was transferred to R for analysis [35].

The data analysis consisted of three consecutive steps. First, we provided the overall prevalence and intensity of S. haematobium infections and their variation across age, sex, and geographical area. Second, we examined the impact of (i) the volume of urine, (ii) the pool size, (iii) the intensity of infection, and (iv) the diagnostic method on the diagnostic sensitivity (the ability to detect a single infected individual in pooled samples) and egg count. Finally, we assessed and compared the infection intensity at a population level between an individual and a pooled testing strategy.

The overall prevalence of S. haematobium is the proportion of the screened children that tested positive at least once either in the field or at the laboratory. To assess factors associated with the prevalence of S. haematobium infection, we fitted logistic regression model. To assess factors associated with the intensity of S. haematobium infection, we fitted zero-inflated negative binomial (ZINB) regression models to account for overdispersion and excess zero egg counts. Age, gender, and village of residence were included as covariates. Stata was used to find the zero-inflated negative binomial model [36].

The overall diagnostic sensitivity was determined by calculating the proportion of the pooled urine samples that tested positive for S. haematobium eggs. A line graph was used to examine the changes in the sensitivity of the pooled test across the volume of urine examined; pool size; diagnostic methods; and intensity levels of infection (level 1: mean urine egg count (UEC) ≤ 25th percentile, level 2: 25^th percentile < mean UEC ≤ 50^th, level 3: 50^th percentile < mean UEC ≤ 75^th percentile, levels 4: > 75th percentile). A logistic regression model was fit to the data to estimate the sensitivity. Included in this model were: urine volume (10, 20 and 30 mL), pool size (5, 10, 20 and 40), UEC, and diagnostic method (UFM and FC) as main effects, a urine volume and UEC interaction, a pool size and UEC interaction, and random effect for pool ID. Pool ID was included as a random effect because urine samples from the same individual were examined using 10 mL, 20 mL, and 30 mL urine volumes, as well as in pools of 5, 10, 20, and 40 samples. This model was chosen through a model building process. To interpret the interactions, we performed pairwise comparisons across different combinations. To correct for multiple comparisons, we deployed the Tukey method. Based on this model, we determined the combinations of the volume of urine, diagnostic method and pool size that would allow for an estimated diagnostic sensitivity of at least 95% across different scenarios of infection intensity.

Egg counts from pooled urine samples were modeled using ZINB regression to account for overdispersion and excess zeros. Pool size (5,10,20,40), the mean egg count of constituent individual samples, diagnostic method (UFM vs FC) and their interaction were included as covariates to estimate factors associated with egg count/intensity in pooled urine samples. Stratified analyses by urine volume (10, 20, and 30 mL) were conducted to evaluate differences in dispersion across strata. Estimates/results from ZINB are reported as incidence rate ratios (IRRs), with 95% confidence intervals.

A paired t-test was used to assess the differences in the average UEC between the pooled samples and members of the pooled samples (i.e., the individual samples making the pools). We also explored the relationship between the UEC in the pool and the average UEC across the corresponding individual samples using a scatter plot and a linear regression model to fit the data to understand the relationship. Pearson correlation was used to examine the strength of agreement in UEC in the pool and mean UEC across the corresponding individual samples [37]. The data were analyzed using the R statistical software package [38].

Results

Prevalence and intensity of S. haematobium infections

A total of 2,634 children (ages ranging from 5 to 15 years, 41.7% girls) were sampled at baseline across three regions and 15 villages (Afar: nine villages; Benishangul-Gumuz: five villages and Gambella: one village). Overall, 14.05% out of the 2,634 children tested positive for S. haematobium eggs on at least one occasion (i.e., when tested in the field or at the laboratory) when deploying UFM. Most of the infection intensities were light (1–49 eggs per 10 ml of urine) in nature (83.50%). Overall, infections were more prevalent in older children (5–10 years: 12.15% vs. 11–15 years: 16.19%), boys (boys: 15.04% vs. girls: 12.66%) and in Gambella (Gambella: 24.37% vs. Benishangul-Gumuz: 12.36% vs. Afar: 11.64%). Infections were observed in all villages, where the prevalence ranged from 0.55% (Ambash in Afar) to 38.00% (Gabole in Afar). Fig 2 summarizes the prevalence and the mean UEC of S. haematobium infection across age, sex, regions, and villages. S1 Fig shows distribution of S. haematobium egg counts (log eggs/10 mL urine) among study participants, stratified by study region.

Download:

Fig 2. The prevalence and arithmetic mean UEC of S. haematobium infection among 2,634 Ethiopian school-aged children by regions, sex, age groups and villages (1 Admegala; 2. Ambash 3. Andade; 4: Buri; 5: Erimberte; 6: Gabole; 7: Haldebe; 8: Hassoba; 9: Kusra; 10 Akandayu; 11: DuleHode; 12: Kumruk; 13: Salima; 14 Shetalo; 15: Village.

https://doi.org/10.1371/journal.pntd.0014407.g002

In the logistic regression model, an increase in age was associated with an increase in the odds of S. haematobium infection (OR = 1.06, 95% CI: 1.02–1.10, p = 0.002), whereas village and gender were not associated with the odds of S. haematobium infection (S1 Table 1). In zero-inflated negative binomial models, there was moderate evidence for village being associated with S. haematobium infection intensity (IRR = 1.09, 95% CI: 1.01–1.17, p = 0.025), whereas there was not sufficient evidence that age and gender were associated with infection intensity (S1 Table).

Factors affecting the diagnostic sensitivity

Overall, the diagnostic sensitivity tends to increase as a function of the intensity level of infection and the volume of urine examined. The sensitivity tends to be higher when using the UFM method compared to the FC method. Fig 3 summarizes the diagnostic sensitivity across the volume of urine, the pool size, the diagnostic method, and the intensity of infection.

Download:

Fig 3. The observed sensitivity by diagnostic methods (urine filtration microscopy and Fluke Catcher), volume of urine examined (10, 20 and 30 mL), pool size (5, 10, 20 and 40) and mean UEC across the individual samples making the pools.

The red line, representing sensitivity based on the examination of 10 mL using UFM for a pool size of 10, overlaps with other lines.

https://doi.org/10.1371/journal.pntd.0014407.g003

These observations were confirmed by the logistic regression model (Table 1). UFM was significantly more sensitive than FC for detecting S. haematobium eggs in pooled urine samples ( =0.83, p < 0.001). There were significant interactions between the mean log of urine egg count (UEC) and volume of urine ( =0.01, p = 0.021) or pool size ( =-0.01, p = 0.007) for detecting S. haematobium eggs in pooled samples. The odds of detecting eggs in pooled samples when 10 mL was examined and when the mean UEC of individual samples equaled 10 eggs/mL were 0.78 (95% CI: 0.67-0.90) and 0.60 (95% CI:0.51-0.76) times lower than when 20 mL and 30 mL are examined, respectively.

Download:

Table 1. Factors affecting the predicted diagnostic performance of pooled testing based on the logistic regression model.

https://doi.org/10.1371/journal.pntd.0014407.t001

A subgroup analysis showed that the effects of both urine volume and pool size on the sensitivity of UFM for detecting S. haematobium in pooled samples varied by infection intensity. The effect of urine volume was greater in heavy-intensity infections (≥50 eggs/10 mL; β = 0.06, SE = 0.018, p < 0.001) than in light-intensity infections (<50 eggs/10 mL; β = 0.03, SE = 0.006, p < 0.001). Similarly, the negative effect of increasing pool size on sensitivity was more pronounced in heavy-intensity infections (β = −0.06, SE = 0.010, p < 0.001) compared with light-intensity infections (β = −0.03, SE = 0.006, p < 0.001).

Based on the model predictions, we further explored which combination of urine volume, pool size and diagnostic method would allow for a diagnostic sensitivity of at least 95% across different scenarios of infection intensity (Fig 4).

Download:

Fig 4. The predicted sensitivity across different combinations of volume, pool size, infection intensity and diagnostic method.

These predicted values for diagnostic sensitivity were based on the logistic regression model in Table 2.

https://doi.org/10.1371/journal.pntd.0014407.g004

Download:

Table 2. The lowest possible value of mean urine egg counts across the individual samples that can be detected with a probability of at least 95% for each combination of volume, pool size and diagnostic method.

https://doi.org/10.1371/journal.pntd.0014407.t002

Overall, the lowest possible value of mean UEC across the individual samples that can be detected with an estimated probability of at least 95% for each combination of volume and pool size was lower when UFM was employed for testing than the FC. Table 2 summarizes the lowest possible value of mean UEC across the individual samples (expressed as eggs in 10 mL urine) that can be detected with an estimated probability of at least 95% for each combination of volume, pool size and diagnostic method. These values are based on the logistic regression model described in Table 1.

Zero-inflated negative binomial regression model was fitted to account for overdispersion and excess zeros in pooled egg counts. In the negative binomial egg count, across all urine volumes (30, 20, and 10 mL), egg count or intensity of infection in pooled urine samples decreased consistently with an increase in pool size (IRR ≈ 0.98, p < 0.001) but increased with an increase in mean individual egg counts making the pools (IRR range: 1.05–1.06, p < 0.001) and when UFM used for the diagnosis than FC (IRR range: 1.29–1.49, p < 0.001) (S2 Table).

Comparison of infection intensity at population level between the individual and pooled testing strategy

As would be expected, the mean UEC estimates of pooled samples tended to decrease with an increase in pool size because a smaller volume of the positive urine sample is examined. This is true for both UFM and FC. The average UEC per 10 mL in pooled samples tends to be lower than the average UEC per 10 mL of the individual samples making the pools regardless of the

pool size and diagnostic method. Table 3 summarizes the mean UEC at the population level based on an individual and pooled testing strategy across the diagnostic method, different pool sizes and urine volume.

Download:

Table 3. Comparison of the urine egg count of individual and pooled samples.

https://doi.org/10.1371/journal.pntd.0014407.t003

There was a significant positive correlation between the mean UEC of pools of 5, 10, or 20 samples and the mean UEC of the corresponding individual urine samples making the pools for both UFM (r: 0.47-0.76) and FC (r: 0.30-0.58) regardless of the volume of urine (Fig 5). For pools of 40, the correlation coefficient was not significantly different from zero for all combinations of diagnostic methods and volume of urine.

Download:

Fig 5. The agreement in egg counts of pooled urine samples for different pool sizes, volume of urine and diagnostic methods.

The scatter plots represent the agreement in egg counts of pooled urine samples (pools of 5, 10, 20 and 40) and the corresponding mean egg counts per 10 mL of individual urine samples when 10, 20 and 30 mL of urine is examined with urine filtration microscopy (Panel A) and Fluke Catcher (Panel B). ‘R’ represents Pearson’s’ correlation coefficient, ‘p’ the corresponding p-value. The dashed line represents the line of equivalence.

https://doi.org/10.1371/journal.pntd.0014407.g005

Discussion

This study assessed the impact of diagnostic method, volume of urine examined, pool size, and intensity of infection on the detection and quantification of S. haematobium eggs in pooled urine samples. The sensitivity of the pooled testing strategy increased significantly when UFM was deployed compared to the FC method. The estimated sensitivity of the pooled testing strategy also increased with an increase in the volume of urine examined when the mean UEC equaled 10 eggs per 10 mL of urine. Although the increase in diagnostic sensitivity as a function of increasing urine volume and intensity of infection is not unexpected and has been discussed in other settings [22,23,39], it has practical applications in how pooled strategies may be reconciled with surveillance needs in various collection and endemicity settings. The increased sensitivity of UFM for detecting heavy intensity infections, [3,4] and the reduced dilution effect when negative samples are added to the pool, due to higher S. haematobium egg counts in positive samples, could have enhanced the sensitivity of the pooled testing strategy as infection intensity increased. Similarly, since S. haematobium eggs are unevenly distributed in urine samples [3,4], using a larger volume may help minimize these variations and increase the probability of finding positive samples, making it easier to detect low-intensity infections in pooled samples. This study tested a low prevalence scenario by introducing single S. haematobium-positive specimens into each pool. A study using porcine fecal samples also showed a decreased infection detectability in pooled samples containing only one infected sample compared to pooled samples made by combining multiple samples with light infection, even when a highly sensitive PCR tool was used for examination [40].

The increased sensitivity of the pooled testing strategy seen when UFM is deployed is consistent with previous literature [40–43]. Such advantages may be attenuated over time with newer technologies. For instance, one study reported a higher sensitivity of the point-of-care circulating cathodic antigen cassette test for detecting S. mansoni infection in pooled urine samples than the Kato Katz thick smear technique [39]. The current study was the first study to apply FC for detecting S. haematobium eggs in pooled urine samples. The best way to apply FC in this setting is not yet clear, but its evaluation is an important step, recognizing the limitations in broadly distributing and sustaining high quality microscopy as an approach to community surveillance.

The present study also highlighted a trend of decreased sensitivity with an increase in the size of the pools despite lack of significance when adjusted for volume of urine, diagnostic test and egg count in the logistic regression model. Previous studies also reported an inverse relationship between the sensitivity of a pooled testing strategy and the number of individual samples pooled [14,27]. The amount of urine contributed by the individuals to the pools decreases as a function of the size of the pool. Hence, the probability of the transfer of eggs from the individual samples to the pooled samples may decrease as the pool size increases, particularly when the infection intensity is low. In addition, as the pools were made by mixing one positive individual sample and the remaining negative samples, the dilution effect in pooling the samples will be more pronounced as the size of the pool increases. This can result in false negatives, reducing overall sensitivity. If a pool is too large and the initial test fails to detect a weak positive due to dilution, that individual may not be identified in follow-up testing. However, the distribution of sourcing egg burden was not even among the pools. This lack of continuity in egg density across specimens assessed could have introduced a differential selection bias in the estimated sensitivity across diagnostic modalities.

Moreover, there was a strong positive correlation between the UECs obtained by examining pooled samples and the mean UEC of the corresponding individual samples making the pools, suggesting that an optimization approach exists to minimize the consequence of higher pool size and approach greater than minimal pool size in practice. The average UECs of the pools of 5 samples per 10 mL urine, 20 samples per 20 mL urine, and 40 samples per 20 mL urine were also comparable to the mean egg counts of the corresponding individual samples making the pools. Previous studies also have reported discernible analytic linkages between estimates of the intensity of soil-transmitted helminth infections [21,44,45] and S. haematobium [22,23] infections by pooled and individual testing strategies.

The present study generally reinforced the value of high-quality microscopy and influence of urine volume and source urine egg burden. Additional studies are needed that evaluate the application of such a sampling and pooling strategy on the detection of S. haematobium in varying scenarios of endemicity. Microsimulation studies could be used as a complement when impractical to field test the scenarios (e.g., level of endemicity, program phase, and sampling efforts) [46]. Although UFM was used to classify individual samples for pool construction, this step did not influence the diagnostic comparison, as both UFM and FC were applied independently to the same pooled specimens. Any individual-level misclassification would be non-differential and would likely bias results toward the null rather than favor either method.

Another strength of this study is the use of experimentally diluted/standardized egg concentrations to evaluate pooled testing performance under controlled conditions. The controlled single-positive design was used as a necessary first step to establish performance characteristics. By recreating low-intensity infections through dilution, we were able to isolate the effects of pool size, urine volume, and diagnostic method on sensitivity and make direct comparisons across testing strategies. Relying solely on naturally occurring infection intensities would have introduced substantial variability and limited comparability between pools, particularly given the relatively small number of heavy-intensity infections in our field samples. Nevertheless, because experimental dilution may not fully capture the biological heterogeneity of naturally infected urine, we complemented these analyses with stratified evaluations using observed field infection intensities to confirm real-world applicability (Fig 3). Future studies should further validate pooled testing strategies under entirely field-based conditions in which samples are pooled randomly regardless of infection status to evaluate the operational performance and real-world applicability of the optimized pooling strategy.

Moreover, comparing the observed egg count distributions with theoretical Poisson expectations could further inform mixing efficiency and support simulation-based optimization of pooling strategies. Although this formal modeling was beyond the scope of the present study, all samples were thoroughly homogenized before aliquoting to minimize sedimentation and ensure random egg allocation. In addition, statistical approaches are needed that allow for the estimation of the prevalence of S. haematobium infection in a population based upon the results of the pooled examination. While some statistical methods have been described for soil-transmitted helminths, S. mansoni, and different infections in vector populations, these approaches need validation for S. haematobium [46–48]. Moreover, there is a lack of time and other cost-benefit analyses to delineate when pooling urine samples during a large-scale epidemiological survey of S. haematobium infection is indeed cost-saving. This is particularly relevant for the FC, which would require more time, personnel effort, and water for testing compared to conventional UFM, potentially limiting its feasibility for field deployment in resource-limited settings. This will support program managers and healthcare decision-makers in creating the most economical survey for rapidly monitoring MDA initiatives targeted at S. haematobium infection control. Additionally, samples with known positive and negative results for S. haematobium eggs were mixed using a hierarchical strategy. The sensitivity of the UFM or FC could vary when applied to urine samples that are randomly mixed without checking for the presence of S. haematobium eggs. Furthermore, alternative pooling strategies, such as the array technique, may provide better sensitivity in detecting infections compared to the hierarchical approach. Additionally, the use of more sensitive molecular techniques could improve the likelihood of detecting light infections in pooled samples. Future studies could employ highly sensitive molecular diagnostic tools to analyze randomly pooled samples using both cascaded and array strategies. These studies could also explore the optimal pool sizes that minimize the expected number of tests for both array and hierarchical pooled urine testing while maintaining reliable sensitivity. Such pool sizes could then be utilized in large-scale epidemiological surveys in endemic regions. Moreover, although urine samples were also collected one month after praziquantel treatment; the number of positive samples available for pooling after treatment was relatively small. This limited our ability to perform a separate, powered analysis of post-treatment pooling performance. Pre- and post-treatment data were therefore combined, with infection intensity included in the models to account for these differences. Future studies with larger post-treatment sample sizes will be important to specifically assess the performance of pooled urine testing strategies in low-intensity, post-treatment settings and to further refine optimal pooling volumes under these conditions.

Conclusions

In conclusion, the diagnostic sensitivity of a pooled sample examination strategy is influenced by the intricate interaction between pool size, urine volume examined, urine egg count, and the diagnostic method used. The sensitivity of the pooled testing strategy may increase with larger urine volumes, higher infection intensity, and when UFM (rather than FC) is used for testing. However, sensitivity could decrease as the pool size increases. Choosing a pooling strategy for diagnosing S. haematobium infection should consider the pool size, urine volume, urine egg count and diagnostic method.

Supporting information

S1 Fig. Distribution of S. haematobium egg counts (log eggs/10 mL urine) among study participants, stratified by study region.

https://doi.org/10.1371/journal.pntd.0014407.s001

(TIF)

S1 Table. Factors associated with the prevalence and intensity of S. haematobium infection from a logistic and Zero-inflated negative binomial regression models.

https://doi.org/10.1371/journal.pntd.0014407.s002

(DOCX)

S2 Table. Zero-inflated negative binomial regression of pooled egg counts.

https://doi.org/10.1371/journal.pntd.0014407.s003

(DOCX)

S1 Data. Complete anonymized dataset underlying the results reported in this manuscript.

https://doi.org/10.1371/journal.pntd.0014407.s004

(XLSX)

Acknowledgments

We extend our gratitude to the study participants for providing urine samples. We also wish to thank the health professionals at the local health posts, as well as the health offices at the district and regional levels, and the community leaders for their invaluable assistance in facilitating participant recruitment and data collection.

References

1. Mutapi F, Maizels R, Fenwick A, Woolhouse M. Human schistosomiasis in the post mass drug administration era. Lancet Infect Dis. 2017;17(2):e42–8. pmid:27988094
- View Article
- PubMed/NCBI
- Google Scholar
2. World Health Organization. Schistosomiasis. https://www.who.int/news-room/fact-sheets/detail/schistosomiasis 2023 December 30.
3. McManus DP, Dunne DW, Sacko M, Utzinger J, Vennervald BJ, Zhou XN. Schistosomiasis. Nat Rev Dis Primers. 2018;4(1):13.
- View Article
- Google Scholar
4. Verjee MA. Schistosomiasis: still a cause of significant morbidity and mortality. Res Rep Trop Med. 2019;31(10):153–63.
- View Article
- Google Scholar
5. WHO. Helminth control in school-age children: a guide for managers of control programs. Geneva: World Health Organization. 2011.
6. WHO. Accelerating work to overcome the global impact of neglected tropical diseases - A roadmap for implementation. Geneva: World Health Organization. 2012.
7. Dorfman R. The Detection of Defective Members of Large Populations. Ann Math Statist. 1943;14(4):436–40.
- View Article
- Google Scholar
8. Tebbs JM, McMahan CS, Bilder CR. Two-stage hierarchical group testing for multiple infections with application to the infertility prevention project. Biometrics. 2013;69(4):1064–73. pmid:24117173
- View Article
- PubMed/NCBI
- Google Scholar
9. Hughes-Oliver JM. Pooling Experiments for Blood Screening and Drug Discovery. Screening. 2006;:48–68.
- View Article
- Google Scholar
10. Abdalhamid B, Bilder CR, McCutchen EL, Hinrichs SH, Koepsell SA, Iwen PC. Assessment of Specimen Pooling to Conserve SARS CoV-2 Testing Resources. Am J Clin Pathol. 2020;153(6):715–8. pmid:32304208
- View Article
- PubMed/NCBI
- Google Scholar
11. American Red Cross. Infectious disease testing. https://www.redcrossblood.org/biomedical-services/blood-diagnostic-testing/blood-testing.html 2026. 2026 February 18.
12. Ando N, Mizushima D, Watanabe K, Takano M, Shiojiri D, Uemura H, et al. Modified self-obtained pooled sampling to screen for Chlamydia trachomatis and Neisseria gonorrhoeae infections in men who have sex with men. Sex Transm Infect. 2021;97(5):324–8. pmid:33082236
- View Article
- PubMed/NCBI
- Google Scholar
13. Valas S, Ngwa-Mbot D, Stourm S, Mémeteau S, Tabouret M. A retrospective evaluation of pooled serum ELISA testing in the frame of the French eradication program for infectious bovine rhinotracheitis. Prev Vet Med. 2023;214:105890. pmid:36947947
- View Article
- PubMed/NCBI
- Google Scholar
14. Norman LJ, Kotz S, Wu X. Inspection errors for attributes in quality control. CRC Press. 1991.
15. Christopher B, Loughin T. Analysis of categorical data with R. Chapman and Hall/CRC, 2024.
16. Grobe N, Cherif A, Wang X, Dong Z, Kotanko P. Sample pooling: burden or solution?. Clin Microbiol Infect. 2021;27(9):1212–20. pmid:33878507
- View Article
- PubMed/NCBI
- Google Scholar
17. Bilder C. Group testing for identification. Wiley StatsRef: Statistics Reference. 2019. https://doi.org/10.1002/9781118445112.stat08227
18. van Zyl GU, Preiser W, Potschka S, Lundershausen AT, Haubrich R, Smith D. Pooling strategies to reduce the cost of HIV-1 RNA load monitoring in a resource-limited setting. Clin Infect Dis. 2011;52(2):264–70. pmid:21288854
- View Article
- PubMed/NCBI
- Google Scholar
19. Edouard S, Prudent E, Gautret P, Memish ZA, Raoult D. Cost-effective pooling of DNA from nasopharyngeal swab samples for large-scale detection of bacteria by real-time PCR. J Clin Microbiol. 2015;53(3):1002–4. pmid:25552360
- View Article
- PubMed/NCBI
- Google Scholar
20. Bogere N, Bongomin F, Katende A, Ssebambulidde K, Ssengooba W, Ssenfuka H, et al. Performance and cost-effectiveness of a pooled testing strategy for SARS-CoV-2 using real-time polymerase chain reaction in Uganda. Int J Infect Dis. 2021;113:355–8. pmid:34757007
- View Article
- PubMed/NCBI
- Google Scholar
21. Leta GT, French M, Dorny P, Vercruysse J, Levecke B. Comparison of individual and pooled diagnostic examination strategies during the national mapping of soil-transmitted helminths and Schistosoma mansoni in Ethiopia. PLoS Negl Trop Dis. 2018;12(9):e0006723. pmid:30199526
- View Article
- PubMed/NCBI
- Google Scholar
22. Degarege A, Erko B, Mekonnen Z, Legesse M, Negash Y, Vercruysse J, et al. Comparison of individual and pooled urine samples for estimating the presence and intensity of Schistosoma haematobium infections at the population level. Parasit Vectors. 2015;8:593. pmid:26572138
- View Article
- PubMed/NCBI
- Google Scholar
23. Degarege A, Levecke B, Negash Y, Animut A, Erko B. Clinical sensitivity and time-to-result of a cascaded pooled testing approach for assessing the prevalence and intensity of Schistosoma haematobium infection. PLoS Negl Trop Dis. 2024;18(8):e0012435. pmid:39173069
- View Article
- PubMed/NCBI
- Google Scholar
24. Flukefinder Diagnostic system. Easy Detection & Diagnosis of Liver Flukes. https://flukefinder.com/ 2023 December 30.
25. Bosco A, Ciuca L, Maurelli MP, Vitiello P, Cringoli G, Prada JM, et al. Comparison of Mini-FLOTAC, Flukefinder and sedimentation techniques for detection and quantification of Fasciola hepatica and Calicophoron daubneyi eggs using spiked and naturally infected bovine faecal samples. Parasit Vectors. 2023;16(1):260. pmid:37533114
- View Article
- PubMed/NCBI
- Google Scholar
26. Reigate C, Williams HW, Denwood MJ, Morphew RM, Thomas ER, Brophy PM. Evaluation of two Fasciola hepatica faecal egg counting protocols in sheep and cattle. Vet Parasitol. 2021;294:109435. pmid:33946031
- View Article
- PubMed/NCBI
- Google Scholar
27. Zárate-Rendón DA, Vlaminck J, Levecke B, Briones-Montero A, Geldhof P. Comparison of Kato-Katz Thick Smear, Mini-FLOTAC, and Flukefinder for the Detection and Quantification of Fasciola hepatica Eggs in Artificially Spiked Human Stool. Am J Trop Med Hyg. 2019;101(1):59–61. pmid:31074405
- View Article
- PubMed/NCBI
- Google Scholar
28. Chala B, Torben W. An Epidemiological Trend of Urogenital Schistosomiasis in Ethiopia. Front Public Health. 2018;6:60. pmid:29556490
- View Article
- PubMed/NCBI
- Google Scholar
29. Geleta S, Alemu A, Getie S, Mekonnen Z, Erko B. Prevalence of urinary schistosomiasis and associated risk factors among Abobo Primary School children in Gambella Regional State, southwestern Ethiopia: a cross sectional study. Parasit Vectors. 2015;8:215. pmid:25886292
- View Article
- PubMed/NCBI
- Google Scholar
30. Degarege A, Mekonnen Z, Levecke B, Legesse M, Negash Y, Vercruysse J, et al. Prevalence of Schistosoma haematobium Infection among School-Age Children in Afar Area, Northeastern Ethiopia. PLoS One. 2015;10(8):e0133142. pmid:26252615
- View Article
- PubMed/NCBI
- Google Scholar
31. Bujang MA, Adnan TH. Requirements for Minimum Sample Size for Sensitivity and Specificity Analysis. J Clin Diagn Res. 2016;10(10):YE01–6. pmid:27891446
- View Article
- PubMed/NCBI
- Google Scholar
32. Power Analysis & Sample Size. List of Sample Size Procedures – Over 1200 Scenarios. List of Sample Size Procedures | Sample Size Software | PASS. 2026 February 2.
33. World Health Organization. Basic laboratory methods in medical parasitology. Geneva: WHO. 1991.
34. Hong S-T. Review of Recent Prevalence of Urogenital Schistosomiasis in Sub-Saharan Africa and Diagnostic Challenges in the Field Setting. Life (Basel). 2023;13(8):1670. pmid:37629527
- View Article
- PubMed/NCBI
- Google Scholar
35. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. 2023.
36. StataCorp. Stata Statistical Software: Release 16. College Station, TX: StataCorp LLC. 2019.
37. Meloun M, Militký J. Correlation. 2011.
38. Wickham H. ggplot2: Elegant Graphics for Data Analysis. 2nd ed. Springer. 2016.
39. Lo NC, Coulibaly JT, Bendavid E, N’Goran EK, Utzinger J, Keiser J, et al. Evaluation of a Urine Pooling Strategy for the Rapid and Cost-Efficient Prevalence Classification of Schistosomiasis. PLoS Negl Trop Dis. 2016;10(8):e0004894. pmid:27504954
- View Article
- PubMed/NCBI
- Google Scholar
40. Pedersen KS, Johansen M, Jorsal SE, Nielsen JP, Bækbo P, Angen O. Pooling of porcine fecal samples for quantification of Lawsonia intracellularis by real-time polymerase chain reaction. J Vet Diagn Invest. 2014;26(3):342–5. pmid:24621847
- View Article
- PubMed/NCBI
- Google Scholar
41. Ochodo EA, Gopalakrishna G, Spek B, Reitsma JB, van Lieshout L, Polman K, et al. Circulating antigen tests and urine reagent strips for diagnosis of active schistosomiasis in endemic areas. Cochrane Database Syst Rev. 2015;2015(3):CD009579. pmid:25758180
- View Article
- PubMed/NCBI
- Google Scholar
42. Degarege A, Animut A, Negash Y, Erko B. Performance of Urine Reagent Strips in Detecting the Presence and Estimating the Prevalence and Intensity of Schistosoma haematobium Infection. Microorganisms. 2022;10(10):2062. pmid:36296338
- View Article
- PubMed/NCBI
- Google Scholar
43. Feleke DG, Alemu Y, Bisetegn H, Debash H. Accuracy of Diagnostic Tests for Detecting Schistosoma mansoni and S. haematobium in Sub-Saharan Africa: A Systematic Review and Meta-Analysis. Biomed Res Int. 2023;2023:3769931. pmid:37621699
- View Article
- PubMed/NCBI
- Google Scholar
44. Kure A, Mekonnen Z, Dana D, Bajiro M, Ayana M, Vercruysse J, et al. Comparison of individual and pooled stool samples for the assessment of intensity of Schistosoma mansoni and soil-transmitted helminth infections using the Kato-Katz technique. Parasit Vectors. 2015;8:489. pmid:26400064
- View Article
- PubMed/NCBI
- Google Scholar
45. Mekonnen Z, Meka S, Ayana M, Bogers J, Vercruysse J, Levecke B. Comparison of individual and pooled stool samples for the assessment of soil-transmitted helminth infection intensity and drug efficacy. PLoS Negl Trop Dis. 2013;7(5):e2189. pmid:23696905
- View Article
- PubMed/NCBI
- Google Scholar
46. Levecke B, Anderson RM, Berkvens D, Charlier J, Devleesschauwer B, Speybroeck N, et al. Mathematical inference on helminth egg counts in stool and its applications in mass drug administration programmes to control soil-transmitted helminthiasis in public health. Adv Parasitol. 2015;87:193–247. pmid:25765196
- View Article
- PubMed/NCBI
- Google Scholar
47. Speybroeck N, Williams CJ, Lafia KB, Devleesschauwer B, Berkvens D. Estimating the prevalence of infections in vector populations using pools of samples. Med Vet Entomol. 2012;26(4):361–71. pmid:22486773
- View Article
- PubMed/NCBI
- Google Scholar
48. Mitchell S, Pagano M. Pooled testing for effective estimation of the prevalence of Schistosoma mansoni. Am J Trop Med Hyg. 2012;87(5):850–61. pmid:22964721
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Mutapi F, Maizels R, Fenwick A, Woolhouse M. Human schistosomiasis in the post mass drug administration era. Lancet Infect Dis. 2017;17(2):e42–8. pmid:27988094
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. World Health Organization. Schistosomiasis. https://www.who.int/news-room/fact-sheets/detail/schistosomiasis 2023 December 30.

[ref3] 3. McManus DP, Dunne DW, Sacko M, Utzinger J, Vennervald BJ, Zhou XN. Schistosomiasis. Nat Rev Dis Primers. 2018;4(1):13.
View Article
Google Scholar

[7] View Article

[8] Google Scholar

[ref4] 4. Verjee MA. Schistosomiasis: still a cause of significant morbidity and mortality. Res Rep Trop Med. 2019;31(10):153–63.
View Article
Google Scholar

[10] View Article

[11] Google Scholar

[ref5] 5. WHO. Helminth control in school-age children: a guide for managers of control programs. Geneva: World Health Organization. 2011.

[ref6] 6. WHO. Accelerating work to overcome the global impact of neglected tropical diseases - A roadmap for implementation. Geneva: World Health Organization. 2012.

[ref7] 7. Dorfman R. The Detection of Defective Members of Large Populations. Ann Math Statist. 1943;14(4):436–40.
View Article
Google Scholar

[15] View Article

[16] Google Scholar

[ref8] 8. Tebbs JM, McMahan CS, Bilder CR. Two-stage hierarchical group testing for multiple infections with application to the infertility prevention project. Biometrics. 2013;69(4):1064–73. pmid:24117173
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref9] 9. Hughes-Oliver JM. Pooling Experiments for Blood Screening and Drug Discovery. Screening. 2006;:48–68.
View Article
Google Scholar

[22] View Article

[23] Google Scholar

[ref10] 10. Abdalhamid B, Bilder CR, McCutchen EL, Hinrichs SH, Koepsell SA, Iwen PC. Assessment of Specimen Pooling to Conserve SARS CoV-2 Testing Resources. Am J Clin Pathol. 2020;153(6):715–8. pmid:32304208
View Article
PubMed/NCBI
Google Scholar

[25] View Article

[26] PubMed/NCBI

[27] Google Scholar

[ref11] 11. American Red Cross. Infectious disease testing. https://www.redcrossblood.org/biomedical-services/blood-diagnostic-testing/blood-testing.html 2026. 2026 February 18.

[ref12] 12. Ando N, Mizushima D, Watanabe K, Takano M, Shiojiri D, Uemura H, et al. Modified self-obtained pooled sampling to screen for Chlamydia trachomatis and Neisseria gonorrhoeae infections in men who have sex with men. Sex Transm Infect. 2021;97(5):324–8. pmid:33082236
View Article
PubMed/NCBI
Google Scholar

[30] View Article

[31] PubMed/NCBI

[32] Google Scholar

[ref13] 13. Valas S, Ngwa-Mbot D, Stourm S, Mémeteau S, Tabouret M. A retrospective evaluation of pooled serum ELISA testing in the frame of the French eradication program for infectious bovine rhinotracheitis. Prev Vet Med. 2023;214:105890. pmid:36947947
View Article
PubMed/NCBI
Google Scholar

[34] View Article

[35] PubMed/NCBI

[36] Google Scholar

[ref14] 14. Norman LJ, Kotz S, Wu X. Inspection errors for attributes in quality control. CRC Press. 1991.

[ref15] 15. Christopher B, Loughin T. Analysis of categorical data with R. Chapman and Hall/CRC, 2024.

[ref16] 16. Grobe N, Cherif A, Wang X, Dong Z, Kotanko P. Sample pooling: burden or solution?. Clin Microbiol Infect. 2021;27(9):1212–20. pmid:33878507
View Article
PubMed/NCBI
Google Scholar

[40] View Article

[41] PubMed/NCBI

[42] Google Scholar

[ref17] 17. Bilder C. Group testing for identification. Wiley StatsRef: Statistics Reference. 2019. https://doi.org/10.1002/9781118445112.stat08227

[ref18] 18. van Zyl GU, Preiser W, Potschka S, Lundershausen AT, Haubrich R, Smith D. Pooling strategies to reduce the cost of HIV-1 RNA load monitoring in a resource-limited setting. Clin Infect Dis. 2011;52(2):264–70. pmid:21288854
View Article
PubMed/NCBI
Google Scholar

[45] View Article

[46] PubMed/NCBI

[47] Google Scholar

[ref19] 19. Edouard S, Prudent E, Gautret P, Memish ZA, Raoult D. Cost-effective pooling of DNA from nasopharyngeal swab samples for large-scale detection of bacteria by real-time PCR. J Clin Microbiol. 2015;53(3):1002–4. pmid:25552360
View Article
PubMed/NCBI
Google Scholar

[49] View Article

[50] PubMed/NCBI

[51] Google Scholar

[ref20] 20. Bogere N, Bongomin F, Katende A, Ssebambulidde K, Ssengooba W, Ssenfuka H, et al. Performance and cost-effectiveness of a pooled testing strategy for SARS-CoV-2 using real-time polymerase chain reaction in Uganda. Int J Infect Dis. 2021;113:355–8. pmid:34757007
View Article
PubMed/NCBI
Google Scholar

[53] View Article

[54] PubMed/NCBI

[55] Google Scholar

[ref21] 21. Leta GT, French M, Dorny P, Vercruysse J, Levecke B. Comparison of individual and pooled diagnostic examination strategies during the national mapping of soil-transmitted helminths and Schistosoma mansoni in Ethiopia. PLoS Negl Trop Dis. 2018;12(9):e0006723. pmid:30199526
View Article
PubMed/NCBI
Google Scholar

[57] View Article

[58] PubMed/NCBI

[59] Google Scholar

[ref22] 22. Degarege A, Erko B, Mekonnen Z, Legesse M, Negash Y, Vercruysse J, et al. Comparison of individual and pooled urine samples for estimating the presence and intensity of Schistosoma haematobium infections at the population level. Parasit Vectors. 2015;8:593. pmid:26572138
View Article
PubMed/NCBI
Google Scholar

[61] View Article

[62] PubMed/NCBI

[63] Google Scholar

[ref23] 23. Degarege A, Levecke B, Negash Y, Animut A, Erko B. Clinical sensitivity and time-to-result of a cascaded pooled testing approach for assessing the prevalence and intensity of Schistosoma haematobium infection. PLoS Negl Trop Dis. 2024;18(8):e0012435. pmid:39173069
View Article
PubMed/NCBI
Google Scholar

[65] View Article

[66] PubMed/NCBI

[67] Google Scholar

[ref24] 24. Flukefinder Diagnostic system. Easy Detection & Diagnosis of Liver Flukes. https://flukefinder.com/ 2023 December 30.

[ref25] 25. Bosco A, Ciuca L, Maurelli MP, Vitiello P, Cringoli G, Prada JM, et al. Comparison of Mini-FLOTAC, Flukefinder and sedimentation techniques for detection and quantification of Fasciola hepatica and Calicophoron daubneyi eggs using spiked and naturally infected bovine faecal samples. Parasit Vectors. 2023;16(1):260. pmid:37533114
View Article
PubMed/NCBI
Google Scholar

[70] View Article

[71] PubMed/NCBI

[72] Google Scholar

[ref26] 26. Reigate C, Williams HW, Denwood MJ, Morphew RM, Thomas ER, Brophy PM. Evaluation of two Fasciola hepatica faecal egg counting protocols in sheep and cattle. Vet Parasitol. 2021;294:109435. pmid:33946031
View Article
PubMed/NCBI
Google Scholar

[74] View Article

[75] PubMed/NCBI

[76] Google Scholar

[ref27] 27. Zárate-Rendón DA, Vlaminck J, Levecke B, Briones-Montero A, Geldhof P. Comparison of Kato-Katz Thick Smear, Mini-FLOTAC, and Flukefinder for the Detection and Quantification of Fasciola hepatica Eggs in Artificially Spiked Human Stool. Am J Trop Med Hyg. 2019;101(1):59–61. pmid:31074405
View Article
PubMed/NCBI
Google Scholar

[78] View Article

[79] PubMed/NCBI

[80] Google Scholar

[ref28] 28. Chala B, Torben W. An Epidemiological Trend of Urogenital Schistosomiasis in Ethiopia. Front Public Health. 2018;6:60. pmid:29556490
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref29] 29. Geleta S, Alemu A, Getie S, Mekonnen Z, Erko B. Prevalence of urinary schistosomiasis and associated risk factors among Abobo Primary School children in Gambella Regional State, southwestern Ethiopia: a cross sectional study. Parasit Vectors. 2015;8:215. pmid:25886292
View Article
PubMed/NCBI
Google Scholar

[86] View Article

[87] PubMed/NCBI

[88] Google Scholar

[ref30] 30. Degarege A, Mekonnen Z, Levecke B, Legesse M, Negash Y, Vercruysse J, et al. Prevalence of Schistosoma haematobium Infection among School-Age Children in Afar Area, Northeastern Ethiopia. PLoS One. 2015;10(8):e0133142. pmid:26252615
View Article
PubMed/NCBI
Google Scholar

[90] View Article

[91] PubMed/NCBI

[92] Google Scholar

[ref31] 31. Bujang MA, Adnan TH. Requirements for Minimum Sample Size for Sensitivity and Specificity Analysis. J Clin Diagn Res. 2016;10(10):YE01–6. pmid:27891446
View Article
PubMed/NCBI
Google Scholar

[94] View Article

[95] PubMed/NCBI

[96] Google Scholar

[ref32] 32. Power Analysis & Sample Size. List of Sample Size Procedures – Over 1200 Scenarios. List of Sample Size Procedures | Sample Size Software | PASS. 2026 February 2.

[ref33] 33. World Health Organization. Basic laboratory methods in medical parasitology. Geneva: WHO. 1991.

[ref34] 34. Hong S-T. Review of Recent Prevalence of Urogenital Schistosomiasis in Sub-Saharan Africa and Diagnostic Challenges in the Field Setting. Life (Basel). 2023;13(8):1670. pmid:37629527
View Article
PubMed/NCBI
Google Scholar

[100] View Article

[101] PubMed/NCBI

[102] Google Scholar

[ref35] 35. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. 2023.

[ref36] 36. StataCorp. Stata Statistical Software: Release 16. College Station, TX: StataCorp LLC. 2019.

[ref37] 37. Meloun M, Militký J. Correlation. 2011.

[ref38] 38. Wickham H. ggplot2: Elegant Graphics for Data Analysis. 2nd ed. Springer. 2016.

[ref39] 39. Lo NC, Coulibaly JT, Bendavid E, N’Goran EK, Utzinger J, Keiser J, et al. Evaluation of a Urine Pooling Strategy for the Rapid and Cost-Efficient Prevalence Classification of Schistosomiasis. PLoS Negl Trop Dis. 2016;10(8):e0004894. pmid:27504954
View Article
PubMed/NCBI
Google Scholar

[108] View Article

[109] PubMed/NCBI

[110] Google Scholar

[ref40] 40. Pedersen KS, Johansen M, Jorsal SE, Nielsen JP, Bækbo P, Angen O. Pooling of porcine fecal samples for quantification of Lawsonia intracellularis by real-time polymerase chain reaction. J Vet Diagn Invest. 2014;26(3):342–5. pmid:24621847
View Article
PubMed/NCBI
Google Scholar

[112] View Article

[113] PubMed/NCBI

[114] Google Scholar

[ref41] 41. Ochodo EA, Gopalakrishna G, Spek B, Reitsma JB, van Lieshout L, Polman K, et al. Circulating antigen tests and urine reagent strips for diagnosis of active schistosomiasis in endemic areas. Cochrane Database Syst Rev. 2015;2015(3):CD009579. pmid:25758180
View Article
PubMed/NCBI
Google Scholar

[116] View Article

[117] PubMed/NCBI

[118] Google Scholar

[ref42] 42. Degarege A, Animut A, Negash Y, Erko B. Performance of Urine Reagent Strips in Detecting the Presence and Estimating the Prevalence and Intensity of Schistosoma haematobium Infection. Microorganisms. 2022;10(10):2062. pmid:36296338
View Article
PubMed/NCBI
Google Scholar

[120] View Article

[121] PubMed/NCBI

[122] Google Scholar

[ref43] 43. Feleke DG, Alemu Y, Bisetegn H, Debash H. Accuracy of Diagnostic Tests for Detecting Schistosoma mansoni and S. haematobium in Sub-Saharan Africa: A Systematic Review and Meta-Analysis. Biomed Res Int. 2023;2023:3769931. pmid:37621699
View Article
PubMed/NCBI
Google Scholar

[124] View Article

[125] PubMed/NCBI

[126] Google Scholar

[ref44] 44. Kure A, Mekonnen Z, Dana D, Bajiro M, Ayana M, Vercruysse J, et al. Comparison of individual and pooled stool samples for the assessment of intensity of Schistosoma mansoni and soil-transmitted helminth infections using the Kato-Katz technique. Parasit Vectors. 2015;8:489. pmid:26400064
View Article
PubMed/NCBI
Google Scholar

[128] View Article

[129] PubMed/NCBI

[130] Google Scholar

[ref45] 45. Mekonnen Z, Meka S, Ayana M, Bogers J, Vercruysse J, Levecke B. Comparison of individual and pooled stool samples for the assessment of soil-transmitted helminth infection intensity and drug efficacy. PLoS Negl Trop Dis. 2013;7(5):e2189. pmid:23696905
View Article
PubMed/NCBI
Google Scholar

[132] View Article

[133] PubMed/NCBI

[134] Google Scholar

[ref46] 46. Levecke B, Anderson RM, Berkvens D, Charlier J, Devleesschauwer B, Speybroeck N, et al. Mathematical inference on helminth egg counts in stool and its applications in mass drug administration programmes to control soil-transmitted helminthiasis in public health. Adv Parasitol. 2015;87:193–247. pmid:25765196
View Article
PubMed/NCBI
Google Scholar

[136] View Article

[137] PubMed/NCBI

[138] Google Scholar

[ref47] 47. Speybroeck N, Williams CJ, Lafia KB, Devleesschauwer B, Berkvens D. Estimating the prevalence of infections in vector populations using pools of samples. Med Vet Entomol. 2012;26(4):361–71. pmid:22486773
View Article
PubMed/NCBI
Google Scholar

[140] View Article

[141] PubMed/NCBI

[142] Google Scholar

[ref48] 48. Mitchell S, Pagano M. Pooled testing for effective estimation of the prevalence of Schistosoma mansoni. Am J Trop Med Hyg. 2012;87(5):850–61. pmid:22964721
View Article
PubMed/NCBI
Google Scholar

[144] View Article

[145] PubMed/NCBI

[146] Google Scholar

Figures

Abstract

Background

Methods

Results

Conclusions

Author summary

Background

Methods

Ethics statement

Study design

Study population and study area

Sample size

Urine sample collection and processing

Data management and statistical data analysis

Results

Prevalence and intensity of S. haematobium infections

Factors affecting the diagnostic sensitivity

Comparison of infection intensity at population level between the individual and pooled testing strategy

Discussion

Conclusions

Supporting information

S1 Fig. Distribution of S. haematobium egg counts (log eggs/10 mL urine) among study participants, stratified by study region.

S1 Table. Factors associated with the prevalence and intensity of S. haematobium infection from a logistic and Zero-inflated negative binomial regression models.

S2 Table. Zero-inflated negative binomial regression of pooled egg counts.

S1 Data. Complete anonymized dataset underlying the results reported in this manuscript.

Acknowledgments

References