Host heterogeneity in disease transmission is commonly seen across host-pathogen systems and identifying individuals who contribute disproportionately to pathogen transmission (i.e. superspreaders) is key to understanding disease dynamics and managing outbreaks. It is often assumed that shedding intensity is directly proportional to infection intensity. However, theory predicts that co-infection might modulate the relationship between infection intensity and shedding, promoting increased onward transmission. Here we quantify the relative importance of infection intensity and co-infection on shedding in Heterakis gallinarum, a gastrointestinal helminth of gamebirds. We found that infection intensity was a poor predictor of shedding intensity. Instead, increased shedding was linked to co-infections with other endoparasites. Our results show that shedding intensity is not simply explained by infection intensity, but rather is the result of complex host-parasite and parasite-parasite interactions. This highlights the importance of considering such interactions in understanding disease emergence and persistence in natural populations. 

Study system and sample collection

Ring-necked pheasants (Phasianus colchicus) were collected postmortem between November and December 2023 from 8 recreational pheasant shoots across south-west England. Pheasants were sexed and body size (beak - cloaca length, to the nearest 0.5cm) and body mass (to the nearest 0.5g) were recorded. Scaled mass index (SMI) was calculated as a measure of body condition following (Peig and Green, 2009). We calculated SMI separately for males and females to account for sexual size dimorphism.

A faecal sample was collected directly from the cloaca and refrigerated at 4°C within 10 hours of collection. The lower digestive tract (small intestine, large intestine, and caeca) was removed and immediately stored in 10% formalin. In total, faecal samples and intestines from 58 pheasants were obtained.

This research was conducted under the approval of the Ethical Committee of the University of Exeter (ethical approval number 513904). The study was conducted in accordance with all relevant ethical regulations and principles. All pheasants were released under license and killed under the Game Act 1831.

Quantification of shedding intensity

A modified McMaster method was used to quantify shedding intensity (Levecke et al., 2011). 0.5g of faecal matter was suspended in 7ml of sodium nitrate flotation solution (specific gravity 1.2 +- 0.05) (VetLab Supplies). The suspended sample was homogenised and strained to remove any large debris. 0.5ml aliquots were added to two slide chambers on a McMaster slide. Slides were visually examined using light microscopy at 40x magnification. The number of *H. gallinarum *eggs on each slide was recorded. Slides were also examined for evidence of secondary helminth infections and the presence of Eimeria spp. oocytes. Eggs of Syngamus trachea and *Capillaria spp. *are challenging to visually distinguish, so were recorded only the presence or absence of co-infecting helminths. Published keys assisted in the morphological identification of helminth eggs and Eimeria spp. oocytes (Deviyanti et al., 2023, Goldová et al., 2006, Metwally et al., 2020). Counts were expressed as eggs per gram (EPG) and oocytes per gram (OPG), for helminths and Eimeria spp., respectively. This was obtained by multiplying the sum of both chambers by 50.

Quantification of infection intensity

Intensity of infection was quantified as the number of helminths present in the lower digestive tract. For each sample, the digestive tract was opened longitudinally and flushed with running water over a fine mesh sieve (aperture of 150 mic). Helminths collected in the sieve were retained, identified (Tanveer et al., 2015) and counted. The lining of the digestive tract was also examined. *H. gallinarum and Capillaria spp. *were observed in the sampled pheasants. Infection intensity quantification was conducted blind with respect to faecal egg counts.

Statistical analyses

We used a generalised linear model with a negative binomial error structure to identify predictors of *H. gallinarum shedding. A *negative binomial error structure was used to account for overdispersion. *H. gallinarum *eggs per gram (EPG) of faeces was included as the response variable and counts of *H. gallinarum *adults found in the digestive tract, host sex, host body condition, sampling location and co-infection status (co-infection / no co-infection) were included as explanatory variables. The duration of faecal sample storage was included as an additional covariate to account for possible sample degradation over time (Crawley et al., 2016).

Second, we explored the role of co-infection on *H. gallinarum *shedding in more detail using generalised linear models with a negative binomial error structure. All models included the same variables as the initial model but varied in their measure of co-infection: a) co-infection with another helminth (yes / no), b) co-infection with Eimeria spp. (yes / no), and c) the number of detected co-infections (none, one, two).

Finally, we used a generalized linear model with a negative binomial error structure to identify predictors of H. gallinarum infection intensity.* *Counts of *H. gallinarum *in the digestive tract were included as the response variable and host sex, host body condition, sampling location, and co-infection status were included as explanatory variables.

All models were fitted using an ordinary least squares framework and inspected for homogeneity of variance, normality of error structures, linearity and overdispersion. Significance of factors was obtained by comparing two nested models, with or without variables of interest, using likelihood ratio tests. Host body condition and H. gallinarum infection intensity were scaled to aid model conversion.

All analyses were conducted in R version 4.3.0 (R Core Team, 2013) using the packages lme4 (Douglas Bates et al., 2015), tidyverse (Wickham et al., 2019), , DHARMa (Hartig, 2018), ggplot2 (Wickham, 2016), MASS (Venables and Ripley, 2013), patchwork (Pedersen, 2019), and gg.gap (Jiacheng Lou, 2019).

References

Crawley, J. A. H., S. N. Chapman, V. Lummaa and C. L. Lynsdale (2016). "Testing storage methods of faecal samples for subsequent measurement of helminth egg numbers in the domestic horse." Veterinary Parasitology 221: 130-133.

Deviyanti, F., P. Hastutiek, Arimbi, M. Mufasirin, D. Sari and A. Sunarso (2023). "Culling Layer Hen Gastrointestinal Helminth Identification at Wonokromo Market Surabaya." Journal of Parasite Science 7: 66-70.

Douglas Bates, M., B. Bolker and S. Walker (2015). "Fitting linear mixed-effects models using lme4." Journal of Statistical Software 67(1): 1-48.

Goldová, M., V. Paluš, V. Letková, A. Kočišová, J. Čurlík and J. Mojžišová (2006). "Parasitoses in pheasants (Phasianus colchicus) in confined systems." Veterinarski Arhiv 76: 83-89.

Hartig, F. (2018). "DHARMa: residual diagnostics for hierarchical (multi-level/mixed) regression models." R Package version 020.

Jiacheng Lou, J. Z., Yizhu Lvy, Zhi Jin (2019). "Define Segments in y-Axis for 'ggplot2'."

Levecke, B., J. M. Behnke, S. S. R. Ajjampur, M. Albonico, S. M. Ame, J. Charlier, S. M. Geiger, N. T. V. Hoa, R. I. Kamwa Ngassam, A. C. Kotze, J. S. McCarthy, A. Montresor, M. V. Periago, S. Roy, L.-A. Tchuem Tchuenté, D. T. C. Thach and J. Vercruysse (2011). "A Comparison of the Sensitivity and Fecal Egg Counts of the McMaster Egg Counting and Kato-Katz Thick Smear Methods for Soil-Transmitted Helminths." PLOS Neglected Tropical Diseases 5(6): e1201.

Metwally, D., T. Al-Otaibi, S. Albasyouni, M. El-Khadragy and R. Alajmi (2020). "Prevalence of eimeriosis in the one-humped camels ( Camelus dromedarius ) from Riyadh and Al-Qassim, Saudi Arabia." PeerJ 8: e10347.

Pedersen, T. L. (2019). "Package ‘patchwork’." R package http://CRAN. R-project. org/package= patchwork. Cran.

Peig, J. and A. J. Green (2009). "New perspectives for estimating body condition from mass/length data: the scaled mass index as an alternative method." Oikos 118(12): 1883-1891.

R Core Team, R. (2013). "R: A language and environment for statistical computing."

Tanveer, S., S. Ahad and M. Z. Chishti (2015). "Morphological characterization of nematodes of the genera *Capillaria, Acuaria, Amidostomum, Streptocara, Heterakis, *and *Ascaridia *isolated from intestine and gizzard of domestic birds from different regions of the temperate Kashmir valley." Journal of Parasitic Diseases 39(4): 745-760.

Venables, W. N. and B. D. Ripley (2013). Modern applied statistics with S-PLUS, Springer Science & Business Media.

Wickham, H. (2016). ggplot 2: Elegant Graphics for Data Analysis, Springer-Verlag New York.

Wickham, H., M. Averick, J. Bryan, W. Chang, L. D. A. McGowan, R. François, G. Grolemund, A. Hayes, L. Henry and J. Hester (2019). "Welcome to the Tidyverse." Journal of open source software 4(43): 1686.