CWD Model Critique - Chronic Wasting Disease in Deer & Elk

The Science of CWD:
A Chronology of Literature
on Chronic Wasting Disease

by Anthony Grabski, Ph.D.

Journal Review: A Critique of CWD Models - Chronic Wasting Disease in Deer & Elk: a critique of current models and their application.

Eric M. Schauber, Ph.D.
Alan Woolf, Ph.D.

What follows is a layman's digest of the Schauber & Woolf manuscript published in the Wildlife Society Bulletin 2003 31(3): 610-616. The Bulletin is ". . . a scientific quarterly containing original papers relating to wildlife management, conservation law enforcement, conservation education, economics, administration, philosophy, ethics, and contemporary resource problems. . . .All technical information published in the Wildlife Society Bulletin has been selected through peer evaluation. Articles identified as peer refereed are selected from manuscripts that have been subjected to rigorous review by 2-4 independent referees, an Associate Editor, and the Editor." Information on ordering publications may be obtained at www.wildlife.org --Ross Reinhold

An introduction to the researchers
Eric Schauber obtained his MS in Wildlife from the Oregon State University and his Ph.D. in Ecology from the University of Connecticut. Chief among his research interests are epizootiology, ecotoxicology, and population dynamics - with an emphasis on qualitative methods and modeling. He is an assistant professor of Zoology at Southern Illinois University - Carbondale and a wildlife ecologist with the Cooperative Wildlife Research Laboratory.

Alan Woolf is a professor of Zoology at Southern Illinois University - Carbondale and the Director of the Cooperative Wildlife Research Laboratory. He has served the Laboratory for 23 years and been its Director since 1987. His MS degree is from Colorado State University and his Ph.D. from Cornell University. As a long-tenured Director of CWRL, his research interests are necessarily varied, although wildlife disease has been a major component.

Background
People who have attempted to investigate the technical and scientific foundation of the CWD abatement program in Wisconsin are likely familiar with Mike Miller and Elizabeth (Beth) Williams. They are frequent CWD research collaborators and colleagues from Colorado and Wyoming who have served as consultant-advisors to the Wisconsin DNR. When the Wisconsin DNR makes reference to basing its Chronic Wasting Disease Eradication program on the "best available science" the principle conduit for this science has been Miller and Williams.

Miller's most recent and primary contribution to the body of knowledge on CWD has been through the development of complex computer generated models to attempt to ". . . synthesize existing knowledge of CWD" towards the end of generating predictions on future outcomes. Among the conclusions of Miller and his colleagues from these simulation studies are that "CWD can cause extinction of host populations."

At the request of the Governor's Office and the Wisconsin DNR, John Cary - an academic support staff member of the UW Wildlife Ecology Department who possessed advanced computer programming expertise - developed a sophisticated computer model drawing heavily on the groundwork laid by Professor Miller. According to an External Review report of CWD Management in Wisconsin recently released by the Wisconsin DNR (Fischer, et. al. October 18, 2003, p. 18) Mr. Cary's model is "similar to previously published models on CWD management" and has as its "primary assumption" that CWD transmission is ". . . frequency-dependent rather than density-dependent. In a frequency-dependent model, an infected animal transmits disease to a given number of other animals, regardless of how many other animals are in the area."

Professor Miller's models and published research conclusions, coupled with Mr. Cary's replication of his model - complete with multimedia simulations of the extinction effect - were the primary tools to convince the Wisconsin State Legislature, the media, hunters, and the public at large that immediate and drastic action was necessary to avoid the collapse of the State's entire wild herd of white-tail deer.

Motivation for the Current Research Investigation
Aware of the Colorado experience and Wisconsin's "attempted eradication of all white-tailed deer within" the defined eradication zone, Schauber and Woolf reasoned that this is a "prominent management philosophy and a strategy likely to be considered by many agencies responsible for managing populations at risk for CWD."

Therefore they concluded that it would be "useful to critically examine the premises and empirical support of published CWD models." They maintain that science-based wildlife management can only advance ". . . if competing models and management alternatives are carefully explored in a decision-theoretic framework" as ". . . . all scientific knowledge is tentative and provisional, and science advances by repeatedly confronting hypotheses and models with logic and data."

Review of the Literature and Examination of Data
The authors reviewed published models of CWD epizootiology and concluded they share a "common assumption" of frequency-dependent transmission. "The idea of frequency-dependent transmission is based on the premise that opportunities for contact between an infectious individual and susceptible individuals are unaffected by population size (de Jong et. Al. 1995)." [This assumption of independence from population size or density serves to explain why the Wisconsin DNR initially determined that ALL deer in the affected area must be exterminated]

"The presumption of frequency-dependent versus density-dependent is critical to the predicted outcome of an epizootic: host-pathogen extinction versus host-pathogen coexistence." Witness a conclusion of Gross & Miller in one of their published reports: " . . . a disturbing result of this modeling exercise was our inability to identify a set of parameters that permits sustained coexistence of CWD in a wild deer population." [Layman's interpretation: this model predicts that once let loose in a population CWD marches on until it kills every deer!] Schauber & Woolf hasten to emphasize that "this dire outcome of CWD models is entirely a predictable consequence of the frequency-dependent assumption and does not stem from any particular known characteristic of CWD." [Emphasis mine]

The authors then proceed to review the characteristics of disease and animal behavior that determine whether a frequency-dependent model, a density-dependent model, or a combination of the two is most appropriate. They cite the CWD transmission hypothesis via bodily fluids through direct and indirect contact as a case that suggests density does make a difference. This combined with the tendency of animals to congregate on winter ranges argues that "exudates of an infected animal potentially can contact a larger number of animals" than if the population were more sparsely distributed. The authors also note that CWD has been found much more prevalent in captive cervid herds maintained at high densities than in free living herds. After outlining additional evidence for including at least some role for population density in CWD models, the authors note that in their examination of the model's mathematical structure, "even weakly density-dependent transmission may enable host-pathogen coexistence." [Layman interpretation: the deer survive in the face of CWD]. Their conclusion: "The many unknown aspects of CWD transmission prohibit robust prediction of the population impact." [Emphasis mine]

Empirical Evidence
The acid test of any theoretical model - no matter how intelligent is its construction - is the accuracy of its predictions. When the weatherman's model predicts a storm is coming, how often is it correct? In examining the empirical data offered by Miller et. al., the authors were surprised to find an error! Their field data contradicted their model output. They thus conclude that ". . . . the empirical age-prevalence relationships of CWD in free-living mule deer, particularly its rarity in older male deer, cannot be explained by the model and indicates that some important biological processes are missing from the model."

The authors also express concern with the 2001 version of the model that lumped together the data between the sexes, obscuring potential differences. Other difficulties in verifying the accuracy of the model's predictions were cited.

Management Implications
The frequency-dependent assumption that generates the dire consequence of extinction places a heavy burden on management action. It implies that Eradication will not successfully control the spread of the disease - "unless nearly 100% of the hosts are eliminated." The authors warn that embarking on Eradication and then failing to accomplish it 100% may simply "hasten the extinction of the host population without preventing disease spread to other populations." [Interesting to note that two Ph.D. scientists come to a similar conclusion as my 10th grade education neighbor-farmer: "there's no way they'll kill ALL those deer, just a waste of time and money!"]

The authors note that frequency-dependent CWD models "represent a small set of possible outcomes in wild populations. Other outcomes are also plausible and their actuality depends on the true (but unknown) relationships between transmission and population density, sex and age structure, and spatial structure." The risk of placing unswerving faith in an untested model is that any "modeler with understanding of the fundamentals of ecological theory and no consideration for the validity of assumptions [Emphasis mine]" could produce models to justify whatever management action is deemed most desirable.

The authors conclude that the current state of knowledge of CWD points to an "urgent need for research into the transmission dynamics of CWD to firmly base management decisions on the best available science." At present lacking this adequate scientific base, they recommend a decision-theoretic framework be used to analyze the potential benefits and costs of alternative management actions.

Given the uncertain guidance available from the current base of science, the authors raise a number of practical concerns with the "eradication solution."

It remains an open question whether such extreme culling programs will be logistically or politically feasible, particularly if CWD introduction is not a one-time occurrence or the CWD agent persists in the environment.

Complete elimination of CWD from all North American deer and elk herds is unlikely, despite the best efforts of humans, suggesting that it could be reintroduced relatively frequently into disease-free populations.

Even if necessary and successful, defeating CWD via host eradication would come at cost, not only economic but also in terms of public perception of wildlife resources, acceptance of management paradigms, and interruption of the hunting tradition.

Editor Comments
The prime justification for the Eradication Program (as well as the total statewide ban on all baiting & feeding) is the probability that Wisconsin would lose its entire white-tailed deer herd to CWD. This dire consequence is based entirely on computer modeling exercises that have as a fundamental assumption that CWD is entirely independent of population density and dependent on frequency of contact.

If the frequency-dependence assumption is true (and therefore the loss of the entire state herd - as well as neighboring states - is credible), a significant reduction of the infected population that is less than 100% is unlikely to curtail expansion of the disease. Put in simple language, a B+ effort ain't gonna do the job.

So it is instructive to weigh the Probability that all the infected hosts could be killed. The barriers to achievement seem insurmountable. 1) The geographic area of known infection has expanded at least four fold, 2) a ring of counties surrounding the current Intensive Harvest zone counties were severely undersampled during the 2002 CWD testing - leaving plausible the possibility that the area of infection is much more widely dispersed than currently understood, 3) a significant outbreak has been discovered in Northern Illinois, 4) the state budget for eradication has been dramatically cut, 5) the DNRs available manpower for the CWD project has been severely strained, and 6) landowner opposition in the form of protected refuges continue to exist.

It is also instructive to identify and weigh the Costs of persisting against all odds. 1) What other programs suffer and to what degree because money and staff time is being siphoned off to fight the CWD war? 2) What are the costs of the tarnished reputation of the department due to the program making so many enemies? 3) What are the long term consequences to managing CWD when the bulk of money and resources is put into exterminating deer as opposed to expanded research? 4) What are the consequences to the sport of hunting? 5) To what degree is the goal of significantly reducing deer population across the state being compromised?6) What are the consequences of tainting the consumption of venison? 7) What are the consequences of risking human conflict as the battle stakes are raised and people become more polarized over this issue?

With the low probability of success and high costs (that rise as the effort becomes more dedicated), it would seem that to persist against the odds and bear the cost would logically require a high degree of confidence in the frequency-dependent assumption and in the overall predictive accuracy of the Miller-Cary theoretical modeling. One wonders, given the modest body of knowledge about CWD and the highly technical and complex nature of modeling, whether the managers and leadership of the DNR have the capacity to weigh the merits and competing scientific arguments?

When well-credentialed scientists like Schauber & Woolf come forward - in a respected, refereed journal - with a persuasive case that the known base of scientific knowledge is inadequate to justify a wild deer eradication project, the case for bold and decisive action is materially weakened. I believe taking in all that is NOW known about the particular case here in Wisconsin and the uncertainties of CWD knowledge in general argues strongly for a reassessment and selection of an alternative course of action. DNR Secretary Hassett and his staff would be wise to follow the advise of Hippocrates in Epidemics, Bk. I, Sect. XI. "As to diseases, make a habit of two things—to help, or at least to do no harm."

At present the "Harm" is painfully clear; the "Helping" is not.

--Ross Reinhold

15 year Harvest Data in Unit 70A measure effectiveness of CWD Management Plan (published April 1, 2004)