Skip to main content

High prevalence does not necessarily equal maintenance species: Avoiding biased claims of disease reservoirs when using surveillance data

Cite this dataset

Wilber, Mark; DeMarchi, Joseph; Fefferman, Nina; Silk, Matthew (2022). High prevalence does not necessarily equal maintenance species: Avoiding biased claims of disease reservoirs when using surveillance data [Dataset]. Dryad.


Many pathogens of public health and conservation concern persist in host communities. Identifying candidate maintenance and reservoir species is therefore a central component of disease management. The term maintenance species implies that if all species but the putative maintenance species were removed, then the pathogen would still persist. In the absence of field manipulations, this statement inherently requires a causal or mechanistic model to assess.  However, we lack a systematic understanding of i) how often conclusions are made about maintenance and reservoir species without reference to mechanistic models ii) what types of biases may be associated with these conclusions, and iii) how explicitly invoking causal or mechanistic modeling can help ameliorate these biases. Filling these knowledge gaps is critical for robust inference about pathogen persistence and spillover in multihost-parasite systems, with clear implications for human and wildlife health. To address these gaps, we performed a literature review on the evidence previous studies have used to make claims regarding maintenance or reservoir species. We then developed multihost-parasite models to explore and demonstrate common biases that could arise when inferring maintenance potential from observational prevalence data. Finally, we developed a new theory to show how model-driven inference of maintenance species can minimize and eliminate emergent biases. In our review, we found that 83% of studies used some form of observational prevalence data to draw conclusions on maintenance potential and only 6% of these studies combined observational data with mechanistic modeling. Using our model, we demonstrate how the community, spatial, and temporal context of observational data can lead to substantial biases in inferences of maintenance potential. Importantly, our theory identifies that model-driven inference of maintenance species elucidates other streams of observational data that can be leveraged to correct these biases. Model-driven inference is an essential, yet underused, component of multidisciplinary studies that make inference about host reservoir and maintenance species. Better integration of wildlife disease surveillance and mechanistic models is necessary to improve the robustness and reproducibility of our conclusions regarding maintenance and reservoir species.


We used a two-step search process to compile a list of studies that made claims or statements regarding `maintenance' or `reservoir' species. First, we performed a search on Web of Science Core Collection using the search string TS=((``maintenance species'' OR ``reservoir species'' OR ``reservoir population'' OR ``maintenance population'' OR ``reservoir* of infection'' OR ``maintenance host'' OR ``reservoir host'' OR ``multihost'') AND (``disease'' OR ``pathogen'' OR ``parasit*'')). The search was initiated on September 4, 2021. The search string returned 1,819 papers ranging from 1991 to 202, which we reviewed given the criteria described below. However, the terms `reservoir*' and 'maintenance*' are used broadly in disease ecology and the 1,819 papers returned above represented a sub-sample of all papers that make inference regarding reservoir and maintenance species. To ensure that our choice of search terms did not lead to biased results, we generalized our search to  TS=((``reservoir'' OR ``maintenance'' OR ``multihost'') AND (``pathogen'' OR ``parasit*'')), which returned 12,512 results when the search was initiated on Nov. 1, 2021. From these results, we randomly chose 700 papers, which based on our previous search were predicted to yield approximately 100 studies that fit our review criteria.  We compared our results between the two searches to determine whether the proportion of papers in each category was consistent.  We included studies in our analysis that drew some conclusions on the maintenance or reservoir potential of one or more non-human animal species.  This included studies where testing for maintenance potential was a central question and studies that drew post-hoc conclusions on maintenance potential. We excluded literature reviews, meta-analyses, and human-focused studies from our analysis. However, we included studies that looked at non-human reservoir hosts that could contribute to infection in humans (e.g., non-human mammals contributing to schistosomiasis).  In total our first search yielded 346 papers for inclusion and the second 86. For each article that met our criteria we recorded the study type, which we divided into four broad categories: field study, laboratory study, mathematical modeling, and life history extrapolation. Articles could fit into multiple categories. For example, some studies used field-collected data to parameterize mathematical models. Therefore, those studies would be classified as both a field study and mathematical model. We did not count statistical models, such as generalized linear models that were fit to prevalence data or correlation analyses, as mathematical modeling. We further classified studies by the sub-category "methods and measurements". This sub-category included prevalence surveys, challenge assays, models, transmission assays, histopathology, genetics, and species removals. Methods and measurements varied widely between studies and individual studies often used multiple "methods and measurements".

Usage notes

Refer to the README.txt file for usage notes.


United States Department of Agriculture, Award: 1012932