Assessment of Chronic Wasting Disease Prion Shedding in Deer Saliva with Occupancy Modeling

Davenport, Kristen A.; Mosher, Brittany A.; Brost, Brian M.; Henderson, Davin M.; Denkers, Nathaniel D.; Nalls, Amy V.; McNulty, Erin; Mathiason, Candace K.; Hoover, Edward A.

doi:10.1128/jcm.01243-17

Cited by 32 publications

(34 citation statements)

References 47 publications

(79 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this case,

π_{italicij}

(Equation ) does not represent a true detection probability, but rather the “net” rate at which positive samples are correctly classified either due to the real detection process or due to a false‐positive error (Royle & Link, ). Consequently, the application of models that assume false‐positive errors only occur in the unoccupied state is not suitable for addressing questions regarding the efficacy of alternate sampling designs and laboratory procedures (Table , Figure ; e.g., Davenport et al., ; Pilliod, Goldberg, Arkle, & Waits, ; Wilcox et al., , ; Williams, Huyvaert, & Piaggio, ). By letting false‐positive errors arise in both occupancy states,

p_{italicij}

in the model we present (Equation ) is a “correct” detection probability, thereby providing accurate inference concerning detection and enabling reliable optimization of diagnostic procedures (Table , Figure ).…”

Section: Discussionmentioning

confidence: 99%

“…We anticipate similar scenarios where data are contaminated not only with false‐negative errors, but also false positives (e.g., Guillera‐Arroita et al., ). The model we propose (Equation ) could be extended to each of these scenarios and used, for example, to examine temporal dynamics in disease occurrence or its occurrence at multiple scales (e.g., Davenport et al., ).…”

Section: Discussionmentioning

confidence: 99%

“…We present a model‐based approach that allows false‐positive errors in both negative and positive samples, thereby providing unbiased inference concerning the occurrence and detection processes. The modelling framework is general and broadly applicable because it only requires repeat sampling and trials on negative controls (i.e., known negative samples), data that are already collected in many diagnostic protocols (e.g., Carey et al., ; Davenport et al., ; Lachish et al., ; Mosher et al., ). We first describe occupancy models that address imperfect sensitivity and specificity, which we subsequently extend to fully accommodate false‐positive errors.…”

Section: Introductionmentioning

confidence: 99%

“…These "occupancy" models rely on replicate surveys to estimate the true but partially unobserved (i.e., latent) occurrence of the species while controlling for the detrimental effect of false-negative errors (MacKenzie et al, 2002;Tyre et al, 2003). Because laboratory protocols often include multiple replicates on the same sample (e.g., molecular assays performed in triplicate), occupancy models have also been used to account for false-negative errors in laboratory studies (Davenport et al, 2018;Lachish, Gopalaswamy, Knowles, & Sheldon, 2012;McClintock, Nichols et al, 2010;Mosher et al, 2017;Schmidt, K ery, Ursenbacher, Hyman, & Collins, 2013).…”

mentioning

confidence: 99%

See 3 more Smart Citations

A model‐based solution for observational errors in laboratory studies

Brost

Mosher

Davenport

2018

Molecular Ecology Resources

Self Cite

View full text Add to dashboard Cite

Molecular techniques for detecting microorganisms, macroorganisms and infectious agents are susceptible to false-negative and false-positive errors. If left unaddressed, these observational errors may yield misleading inference concerning occurrence, prevalence, sensitivity, specificity and covariate relationships. Occupancy models are widely used to account for false-negative errors and more recently have even been used to address false-positive errors, too. Current modelling options assume false-positive errors only occur in truly negative samples, an assumption that yields biased inference concerning detection because a positive sample could be classified as such not because the target agent was successfully detected, but rather due to a false-positive test result. We present an extension to the occupancy modelling framework that allows false-positive errors in both negative and positive samples, thereby providing unbiased inference concerning occurrence and detection, as well as reliable conclusions about the efficacy of sampling designs, handling protocols and diagnostic tests. We apply the model to simulated data, showing that it recovers known parameters and outperforms other approaches that are commonly used when confronted with observation errors. We then apply the model to an experimental data set on Batrachochytrium dendrobatidis, a pathogenic fungus that is implicated in the global decline or extinction of hundreds of amphibian species. The model-based approach we present is not only useful for obtaining reliable inference when data are contaminated with observational errors, but also eliminates the need for establishing arbitrary thresholds or decision rules that have hidden and unintended consequences.

show abstract

“…In this case,

π_{italicij}

p_{italicij}

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

A model‐based solution for observational errors in laboratory studies

Brost

Mosher

Davenport

2018

Molecular Ecology Resources

Self Cite

View full text Add to dashboard Cite

show abstract

“…A key feature of CWD is the long incubation period (Davenport et al, 2018). The time from infection to death is in the range of 1.5-2.5 years in mule deer (Fox, Jewell, Williams, & Miller, 2006).…”

Section: Introductionmentioning

confidence: 99%

A method that accounts for differential detectability in mixed samples of long‐term infections with applications to the case of chronic wasting disease in cervids

et al. 2018

View full text Add to dashboard Cite

Surveillance of wildlife diseases is logistically difficult, and imperfect detection is a recurrent challenge for disease estimation. Using citizen science can increase sample sizes, but it is associated with a cost in terms of the anatomical type and quality of the sample. Additionally, biological tissue samples from remote areas lose quality due to autolysis. These challenges are faced in the case of emerging chronic wasting disease (CWD) in cervids. Here, we develop a stochastic scenario tree model of diagnostic sensitivity, allowing for a mixture of tissue sample types (lymph nodes and brain) and qualities while accounting for different detection probabilities during the CWD infection, lasting 2–3 years. We apply the diagnostic sensitivity in a Bayesian framework, enabling estimation of age‐class‐specific true prevalence, including the prevalence in latent, recently infected stages. We provide a simulation framework to estimate the sensitivity of the surveillance system (i.e., the probability of detecting the infection in a given population), when detectability varies among individuals due to different disease progression. We demonstrate the utility of our framework by applying it to the recent emergence of CWD in a European population of reindeer. We estimated apparent CWD prevalence at 1.2% of adults in the infected population of wild reindeer, while the true prevalence was 1.6%. The sensitivity estimation of the CWD surveillance was performed in an adjacent small (c. 500) and a large (c. 10,000) reindeer population, demonstrating low certainty of CWD absence. Our method has immediate application to the mandatory testing for CWD in EU countries commencing in 2018. Similar approaches that account for latent stages and a serial disease progression in various tissues with a temporal pattern of diagnostic sensitivity may enhance the estimation of the prevalence of wildlife diseases more generally.

show abstract