Using Auxiliary Information to Improve Wildlife Disease Surveillance When Infected Animals Are Not Detected: A Bayesian Approach

Heisey, Dennis M.; Jennelle, Christopher S.; Russell, Robin E.; Walsh, Daniel P.

doi:10.1371/journal.pone.0089843

Cited by 22 publications

(54 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These data represent the “learning dataset” of Heisey et al. (). We classified animals into 16 discrete surveillance classes based on age, sex, and mortality source.…”

Section: Methodsmentioning

confidence: 99%

“…Following (Heisey et al., ), we used a discrete proportional hazards model to estimate the relative CWD infection rate associated with the surveillance class of interest based on the Wisconsin data:

E false[y_{italicij} false] = 1 - exp false(- exp (false[μ_{ref} + x_{italicij} bold-italicβ + τ \times t_{i} + α_{j} false]) false)

where y ij is the binary outcome ( y ij = 1 indicates a positive CWD sample) for deer i located in the j th spatial unit; μ ref is an intercept term for the reference class against which other surveillance classes are compared; x ij is a row vector indicating to which surveillance class animal i belongs and its location in the j th spatial unit; β is a column vector of surveillance class log infection rate ratios; τ accounts for a linear time trend in overall CWD infection rate; t i is the year the i th deer was collected; and α j is a random effect for the j th spatial unit. We specified the reference class ( μ ref ) as hunter‐harvested yearling males because it is a significant portion of the annual harvest and thus a demographic class for which CWD presence is particularly relevant (Heisey et al., ; Walsh & Miller, ). The inverse of the complimentary log–log link ( inv‐cloglog ) function maps parameters from the complimentary log–log scale to the prevalence scale such that π ref = inv‐cloglog ( μ ref ) is prevalence of the reference class (hunter‐harvested yearling males) in our surveillance stream, and π l = inv‐cloglog ( μ ref + β l ) is prevalence of the l th surveillance class.…”

Section: Methodsmentioning

confidence: 99%

“…Thus, w l = exp( β l ) estimates the CWD infection ratio of the l th surveillance class and represents the propensity of CWD to occur in the l th surveillance class relative to the reference class. When the number of disease cases is relatively large (as in our data), w l are essentially equivalent to the infection risk ratio (Heisey et al., ). It is worth noting that π ref refers to apparent prevalence in the reference class.…”

Section: Methodsmentioning

confidence: 99%

“…Prior to estimating the potential contribution of each surveillance class (Table ) to detecting the presence of CWD (i.e., “real weights” of Heisey et al. ), a detection surveillance goal must be determined. It is composed of two quantities; a disease detection threshold or reference class population prevalence ( π ref ) below which the entire population is effectively disease free, and a detection threshold probability ( p det ; termed β in Heisey et al.…”

Section: Methodsmentioning

confidence: 99%

See 3 more Smart Citations

Applying a Bayesian weighted surveillance approach to detect chronic wasting disease in white‐tailed deer

Jennelle

Walsh

Samuel

et al. 2018

Journal of Applied Ecology

Self Cite

View full text Add to dashboard Cite

Surveillance is critical for early detection of emerging and re‐emerging infectious diseases. Weighted surveillance leverages heterogeneity in infection risk to increase sampling efficiency. Here, we apply a Bayesian approach to estimate weights for 16 surveillance classes of white‐tailed deer in Wisconsin, USA, relative to hunter‐harvested yearling males. We used these weights to conduct a surveillance programme for detecting chronic wasting disease (CWD) in white‐tailed deer at Shenandoah National Park (SHEN) in Virginia, USA. Generally, for surveillance, risk of infection increased with age and was greater in males. Clinical suspect deer had the highest risk, with weight estimates of 33.33 and 9.09 for community‐reported and hunter‐reported suspect deer, respectively. Fawns had the lowest risk with an estimated weight of 0.001. We used surveillance weights for Wisconsin deer to determine sampling effort required to detect a CWD‐positive case in SHEN if prevalence in yearling males ≥0.025. The sampling required to detect CWD was 37–91 adult deer, depending on the adult male:female ratio in the surveillance stream. We collected rectal biopsies from 49 female and 21 male adult deer, and 10 additional samples from vehicle‐killed deer. CWD was not detected and we concluded with 95% probability that prevalence in the reference population (yearling males) was between 0.0% and 3.6%. Synthesis and applications. Our approach allows managers to estimate relative surveillance weights for different host classes and quantify limits of disease detection in real time when only a sample of animals from a population can be tested, resulting in considerable cost savings for agencies performing wildlife disease detection surveillance. Additionally, it provides a rigorous means of estimating prevalence limits when a disease/pathogen is not detected in a sample set. It is therefore applicable to other wildlife, domestic animal and human disease systems, which can be characterized by surveillance classes with heterogeneous probability of infection. This methodology is also extendable to other disciplines such as invasive species, environmental toxicology, and generally, any ecological question seeking to efficiently use scarce financial and human resources to maximize the detection probability of a rare event.

show abstract

“…These data represent the “learning dataset” of Heisey et al. (). We classified animals into 16 discrete surveillance classes based on age, sex, and mortality source.…”

Section: Methodsmentioning

confidence: 99%

E false[y_{italicij} false] = 1 - exp false(- exp (false[μ_{ref} + x_{italicij} bold-italicβ + τ \times t_{i} + α_{j} false]) false)

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Applying a Bayesian weighted surveillance approach to detect chronic wasting disease in white‐tailed deer

Jennelle

Walsh

Samuel

et al. 2018

Journal of Applied Ecology

Self Cite

View full text Add to dashboard Cite

show abstract

“…Surveillance is often based on the assumption of a simple random sample, but this can almost always be improved upon if there is auxiliary information available about disease risk factors. 15 Owing to the fact that conducting an unbiased surveillance in free-ranging mammal populations is often more challenging, the passive opportunistic case identi¯cation is, thus, generally more often used for the detection of disease events in wild animals. 16 Government-supported active and passive wildlife disease survey programs have long been carried in Taiwan, however, zoo animals were often the major target simply because the majority of these exhibited animals were imported and the varieties and population of native wild-ranging wild animals are very limited.…”

Section: Introductionmentioning

confidence: 99%

Disease Surveillance in Rescued and Road-Killed Wild-Ranging Carnivores in Taiwan

et al. 2015

View full text Add to dashboard Cite

The objective of this study was to investigate the causes of death and potential diseases carried by the wild-ranging carnivores in Taiwan through a government-supported disease survey program. During the period of August 2011 to January 2015, a total of 51 carcasses from rescued but dead or road-killed carnivores were necropsied for histopathology, molecular and immunological assays, microbiology, and parasitology. The cases included 31 Taiwan ferret badgers (TWFBs) (Melogale moschata subaurantiaca), 12 masked palm civets (MPCs) (Paguma larvate taivana), 5 small Chinese civets (SCCs) (Viverricula indica pallida), and 3 crab-eating mongooses (CEMs) (Herpestes urva). Zoonotic rabies and fatal canine distemper were diagnosed in four TWFBs and three MPCs, respectively. A high prevalence rate of lungworm infestation (23/31; 74.2%) was observed in TWFBs. In addition, a unique fatal Staphylococcus hyicus pneumonia and a fatal heavy systemic sarcopic mange infestation were diagnosed in a TWFB and a suckling MPC kid, respectively. Road tra±c accidents and stray dog-associated killing were the most common etiologies for the death of wild-ranging carnivores.

show abstract

Detection error influences both temporal seroprevalence predictions and risk factors associations in wildlife disease models

Tabak

Pedersen

Miller

2019

Ecology and Evolution

View full text Add to dashboard Cite

Understanding the prevalence of pathogens in invasive species is essential to guide efforts to prevent transmission to agricultural animals, wildlife, and humans. Pathogen prevalence can be difficult to estimate for wild species due to imperfect sampling and testing (pathogens may not be detected in infected individuals and erroneously detected in individuals that are not infected). The invasive wild pig (Sus scrofa, also referred to as wild boar and feral swine) is one of the most widespread hosts of domestic animal and human pathogens in North America. We developed hierarchical Bayesian models that account for imperfect detection to estimate the seroprevalence of five pathogens (porcine reproductive and respiratory syndrome virus, pseudorabies virus, Influenza A virus in swine, Hepatitis E virus, and Brucella spp.) in wild pigs in the United States using a dataset of over 50,000 samples across nine years. To assess the effect of incorporating detection error in models, we also evaluated models that ignored detection error. Both sets of models included effects of demographic parameters on seroprevalence. We compared our predictions of seroprevalence to 40 published studies, only one of which accounted for imperfect detection. We found a range of seroprevalence among the pathogens with a high seroprevalence of pseudorabies virus, indicating significant risk to livestock and wildlife. Demographics had mostly weak effects, indicating that other variables may have greater effects in predicting seroprevalence. Models that ignored detection error led to different predictions of seroprevalence as well as different inferences on the effects of demographic parameters. Our results highlight the importance of incorporating detection error in models of seroprevalence and demonstrate that ignoring such error may lead to erroneous conclusions about the risk associated with pathogen transmission. When using opportunistic sampling data to model seroprevalence and evaluate risk factors, detection error should be included.

show abstract

Using Auxiliary Information to Improve Wildlife Disease Surveillance When Infected Animals Are Not Detected: A Bayesian Approach

Cited by 22 publications

References 23 publications

Applying a Bayesian weighted surveillance approach to detect chronic wasting disease in white‐tailed deer

Applying a Bayesian weighted surveillance approach to detect chronic wasting disease in white‐tailed deer

Disease Surveillance in Rescued and Road-Killed Wild-Ranging Carnivores in Taiwan

Detection error influences both temporal seroprevalence predictions and risk factors associations in wildlife disease models

Contact Info

Product

Resources

About