Disentangling the effects of intermittent faecal shedding and imperfect test sensitivity on the microscopy-based detection of gut parasites in stool samples

Abstract

Background: Gut-parasite transmission often involves faecal shedding, and detecting parasites in stool samples remains the cornerstone of diagnosis. However, not all samples drawn from infected hosts contain parasites (because of intermittent shedding), and no test can detect the target parasites in 100% of parasite-bearing samples (because of imperfect sensitivity). Disentangling the effects of intermittent shedding and imperfect sensitivity on pathogen detection would help us better understand transmission dynamics, disease epidemiology, and diagnostic-test performance. Using paediatric Giardia infections as a case-study, here we illustrate a hierarchical-modelling approach to separately estimating the probabilities of host-level infection ([Formula: see text]); stool-sample-level shedding, given infection ([Formula: see text]); and test-level detection, given infection and shedding ([Formula: see text]).

Methods/findings: We collected 1-3 stool samples, in consecutive weeks, from 276 children. Samples (413 overall) were independently examined, via standard sedimentation/optical microscopy, by a senior parasitologist and a junior, trained student (826 tests overall). Using replicate test results and multilevel hierarchical models, we estimated per-sample Giardia shedding probability at [Formula: see text] and observer-specific test sensitivities at [Formula: see text] and [Formula: see text]. Gender-specific infection-frequency estimates were [Formula: see text] and [Formula: see text]. Had we used a (hypothetical) Perfect Test with 100% narrow-sense sensitivity ([Formula: see text]), the average probability of detecting Giardia in a sample drawn from an infected child ([Formula: see text]) would have been [Formula: see text]. Because no test can be >100% sensitive, [Formula: see text] (which measures clinical sensitivity) can only be brought above ~ 0.44 by tinkering with the availability of Giardia in stool samples (i.e., [Formula: see text]); for example, drawing-and-pooling 3 replicate samples would yield [Formula: see text].

Conclusions: By allowing separate estimation (and modelling) of pathogen-shedding probabilities, the approach we illustrate provides a means to study pathogen transmission cycles and dynamics in unprecedented detail. Separate estimation (and modelling) of true test sensitivity, moreover, may cast new light on the performance of diagnostic tests and procedures, whether novel or routine-practice.

Fig 1. Study area.

Location of the eight kindergartens participating in the study in four urban sectors of the Federal District, Brazil, 2019. The map was created using QGIS 3.36.1 (www.qgis.org) and publicly available shape files from the Instituto Brasileiro de Geografia e Estatística (IBGE; https://www.ibge.gov.br/geociencias/organizacao-do-territorio/malhas-territoriais.html).

Fig 2. Detecting Giardia in children: sampling-testing strategy and outcome coding.

Each child provided up to three stool samples, and each sample was tested twice (once by each of two independent observers). The result of each test (the reading of three microscope slides) was coded as “1” when at least one Giardia cyst or trophozoite was unambiguously identified in at least one slide, and “0” otherwise; when a child did not provide a sample, the results of the missing tests were coded “-”. We depict two hypothetical examples. In panel A, Child 1 provided three serial samples (Samples 1 to 3). Giardia was detected in one of the tests run on Sample 2; this true-positive result shows that the child was infected and that there was shedding in Sample 2, although one test (by Observer 2) failed to detect the parasites (a false-negative result). None of the four tests run on Samples 1 and 3 yielded detections; for each of those samples, this can be due to either (i) no parasite shedding or (ii) false-negative test results. Therefore, the results of the four tests run on Samples 1 and 3 are ambiguous (highlighted in blue), and whether there was parasite shedding in those two samples remains uncertain. In panel B, Child 2 provided just two samples (Samples 1 and 2), and all four test results were ambiguous; child-level infection and sample-level shedding are therefore uncertain. Also in panel B, the red dots in Child 2 and Sample 1 highlight the possibility that infection is present and shedding occurs—but the parasites then go undetected by the (imperfect) tests.

Fig 3

Predictions from the top-ranking multilevel site-occupancy model: probabilities of child-level Giardia infection ($\mathbf{\Psi }$); sample-level Giardia shedding, given infection ($\mathbf{\theta }$); and diagnostic test-level Giardia detection, given shedding (i.e., narrow-sense test sensitivity, $\mathit{p}$).

Fig 4. Model-averaged predictions and unconditional standard errors (SEs).

A, model-averaged probabilities of child-level Giardia infection; naïve gender-specific frequencies (and approximate SEs; see Table 2) are also plotted for comparison. B, model-averaged probabilities of stool-sample–level Giardia shedding, given infection. C, model-averaged probabilities of test-level Giardia detection, given shedding (i.e., narrow-sense test sensitivity).

Fig 5. Pathogen shedding/availability (θ): routes, examples, and importance.

The diagram stresses the complexity of pathogen shedding (or availability), the diversity of major pathogens depending on shedding for transmission, and the importance of measuring shedding to develop a rigorous understanding of both (i) the biological underpinnings of pathogen transmission cycles and dynamics and (ii) the observation process underpinning pathogen diagnosis and surveillance. The case-study described in this report (Giardia) belongs in the gut protozoa, and is highlighted in bold typeface. *Pathogens may be ‘available’ (with probability θ) in either vector bloodmeals or blood samples drawn from infected hosts. **For respiratory transmission of Cryptosporidium, see, e.g., [42]. ***Female sandflies pick Leishmania parasites when they ingest infected dermal macrophages together with a bloodmeal (see, e.g., [43]).