Translator Disclaimer
5 July 2021 Inferring Chronic Wasting Disease Incidence from Prevalence Data
Michael W. Miller, Lisa L. Wolfe
Author Affiliations +

Incidence of chronic wasting disease infection showed strong, positive correlation (r≥0.944) with apparent prevalence among female and male mule deer (Odocoileus hemionus) in seven herds previously studied in Colorado and Wyoming, US. With attention to monitoring method consistency and context, inferring that observed prevalence trends reflect underlying epidemic dynamics in mule deer herds appears justifiable.

Measures of disease occurrence—especially prevalence and incidence—provide a foundation for understanding trends and effects of intervention. Wildlife disease studies tend to report apparent prevalence, the readily measured proportion of animals detected as infected among those sampled at a point in time, despite its epidemiologic limitations. Apparent prevalence serves as a tangible approximation for the more elusive measure of the true proportion infected. Detection method and sampling framework affect the exact relationship between apparent and true prevalence. Incidence gauges disease dynamics more directly than prevalence does, by measuring the rate of new infections during a reference period. However, meeting the general requirement for at least two sampling occasions per individual presents challenges in natural systems. Although not interchangeable, prevalence (P) and incidence (I) do represent “two sides of the same coin,” related through the disease's expected duration in infected individuals (D), such that P'ID in a steady-state system (Freeman and Hutchison 1980).

Measuring and comparing prevalence trends within and among affected cervid herds has yielded arguably valuable insights into the epidemiology of chronic wasting disease (CWD; Williams and Young 1980). This infectious prion disease occurs primarily in deer (Odocoileus spp.) and wapiti (Cervus canadensis) in multiple foci, including some in Colorado and neighboring Wyoming, US, that have been monitored for more than two decades (Williams and Young 1992; Miller et al. 2000, 2020). Early analyses of prevalence data revealed differences among host species and among affected herds, likely patterns of temporospatial expansion, epidemic trends over time, and, unexpected but consistently observed, differences between mule deer (Odocoileus hemionus) sex and age classes (Miller et al. 2000; Miller and Conner 2005). Recently, prevalence trends have been used in assessing the potential effectiveness of measures for controlling CWD in natural systems (Conner et al. 2007; Manjerovic et al. 2014; Wolfe et al. 2018; Miller et al. 2020).

An assumed relationship between CWD prevalence and incidence underpins practical application of such findings. Changes in apparent prevalence have been inferred as reflecting parallel changes in underlying CWD incidence. Responding to a reviewer's fair question about the basis for such assumptions, Miller et al. (2020) posited that “chronic wasting disease prevalence and incidence show positive correlation across a wide range of values in field studies where both parameters were measured in the same mule deer herd. . .” (p. 787). This Letter expands upon that summation, detailing the observed relationship between CWD prevalence and incidence using published and unpublished field data from infected mule deer herds.

Chronic wasting disease prevalence and incidence were reported previously in six mule deer herds located in northcentral Colorado or southeast Wyoming (Miller et al. 2008; Geremia et al. 2015; DeVivo et al. 2017; Table 1). Although Wolfe et al. (2018) reported prevalence but not incidence in a seventh herd, data from 264 adult deer sampled on two or more occasions during their field study were available to calculate incidence (no. positive/total [deer×yr]) for the 3-yr periods before (n=217 deer×yr) and after (n=360 deer×yr) a management intervention (Table 1). All seven herds had data from female deer; three had contemporary paired observations for both sexes, and one included data from both sexes over multiyear periods before and after a management intervention (Table 1).

Table 1

Chronic wasting disease prevalence and incidence among mule deer (Odocoileus hemionus) herds previously studied in Colorado (CO) and Wyoming (WY), USA. nd indicates not done.


The incidence of new CWD infections showed strong, positive correlation (r=0.944) with prevalence among female mule deer across the seven herds (Fig. 1A). The correlation was equally strong (r=0.947) among males over a wider prevalence range (Fig. 1A). Prevalence tended to underestimate incidence slightly (∼0.8×) among females and to overestimate incidence slightly (∼1.4×) among males (Fig. 1A). The observation-based relationships appear somewhat closer than the 1.5–2.6× overestimate derived from the foregoing equation using an expected duration of natural infection of 1.5-2.6 yr (Miller et al. 2008).

Figure 1

A. The incidence of new chronic wasting disease infections showed strong, positive correlation with prevalence among female (circles) and male (triangles) mule deer (Odocoileus hemionus) in herds previously studied in Colorado and Wyoming, USA. B. An observed decline in prevalence among male, but not female, deer after management intervention in the Estes Park, Colorado, study area (denoted by solid symbols on both panels; Wolfe et al. 2018), was accompanied by a measurable decline in incidence among males (arrow), but not females. See Table 1 for data sources.


As a rule, prevalence exceeds incidence for chronic diseases without recovery, because cases linger in the sampled population. Following this general rule, CWD prevalence—a direct measure of active infections by virtue of diagnostics and disease traits—tended to be equal to, or greater than, incidence, especially among males when prevalence values were high (Table 1 and Fig. 1). It follows that using prevalence among adult males as an index for underlying incidence offers a conservative approach for monitoring CWD epidemic trends and responses to control efforts in mule deer. Given consistent monitoring approaches over time, increasing prevalence suggests epidemic growth within the represented demographic and decreasing prevalence suggests lowered incidence. The observed decline in CWD incidence among sampled males in apparent response to local disease management also reflected in prevalence data from the Wolfe et al. (2018) study supports such inference (Table 1 and Fig. 1B).

That said, a few provisos do apply. Whether similar relationships extend to white-tailed deer (Odocoileus virginianus), wapiti, or other susceptible cervid species remains to be determined. Inferences seem most appropriate for time steps of one to a few years. Moreover, care should be taken to assure observed prevalence “trends” are not merely an artifact of sampling. Comparisons made over time require careful and consistent definition of the denominator in prevalence estimation. Sampling bias should be avoided or minimized (Conner et al. 2000; Walsh 2012; WAFWA 2017). For example, given the twofold difference in prevalence among male vs. female mule deer (Miller and Conner 2005), including both sexes in estimates derived from harvest samples could significantly misrepresent trends if one sampling period included robust female harvest and the other did not. Similarly, a shift from random sampling at check stations to sampling only taxidermy submissions might misrepresent the true prevalence trend among males if mature males were only a small (but oversampled) portion of a herd's male demographic (Conner et al. 2000; Miller and Conner 2005). Changes in hunting regulations also could alter the underlying age structure of harvested animals, thereby influencing trends in apparent prevalence (Miller et al. 2020). Spatial context merits consideration as well—the data analyzed here were from relatively localized study areas, and prevalence trends measured over larger spatial scales could obscure local patterns (e.g., Miller et al. 2020). Comparing prevalence trends between herds, populations, or jurisdictions may present added challenges given greater opportunity for confounding factors (e.g., variable regulations or sampling approaches) that may influence apparent prevalence (Walsh 2012; WAFWA 2017).

With attention to monitoring methods and the context of application, inferring that observed CWD prevalence trends also reflect underlying epidemic dynamics appears justifiable. Chronic wasting disease prevalence does appear to be a reasonable index for incidence in mule deer herds across the range of values spanning those encountered in reported data sets. It follows that either declining or relatively flat prevalence trends may indeed offer encouraging evidence of short-term epidemic suppression (Manjerovic et al. 2014; Wolfe et al. 2018; Miller et al. 2020).

Supported by the Colorado Division of Parks and Wildlife. M. Conner, M. Wood, an anonymous reviewer, and the Assistant Editor provided helpful reviews of earlier drafts.



Conner MM, McCarty CW, Miller MW. 2000. Detection of bias in harvest-based estimates of chronic wasting disease prevalence in mule deer. J Wildl Dis 36:691–699. Google Scholar


Conner MM, Miller MW, Ebinger MR, Burnham KP. 2007. A meta-BACI approach for evaluating management intervention on chronic wasting disease in mule deer. Ecol Appl 17:140–153. Google Scholar


DeVivo MT, Edmunds DR, Kauffman MJ, Schumaker BA, Binfet J, Kreeger TJ, Richards BJ, Schätzl HM, Cornish TE. 2017. Endemic chronic wasting disease causes mule deer population decline in Wyoming. PLoS One 12:e0186512. Google Scholar


Freeman J, Hutchison GB. 1980. Prevalence, incidence and duration. Am J Epidemiol 112:707–723. Google Scholar


Geremia C, Miller MW, Hoeting JA, Antolin MF, Hobbs NT. 2015. Bayesian modeling of prion disease dynamics in mule deer using population monitoring and capture-recapture data. PLoS One 10:e0140687. Google Scholar


Manjerovic MB, Green ML, Mateus-Pinilla N, Novakofski J. 2014. The importance of localized culling in stabilizing chronic wasting disease prevalence in white-tailed deer populations. Prev Vet Med 113: 139–145. Google Scholar


Miller MW, Conner MM. 2005. Epidemiology of chronic wasting disease in free-ranging mule deer: Spatial, temporal, and demographic influences on observed prevalence patterns. J Wildl Dis 41:275–290. Google Scholar


Miller MW, Runge JP, Holland AA, Eckert MD. 2020. Hunting pressure modulates prion infection risk in mule deer herds. J Wildl Dis 56:781–790. Google Scholar


Miller MW, Swanson HM, Wolfe LL, Quartarone FG, Huwer SL, Southwick CH, Lukacs PM. 2008. Lions and prions and deer demise. PLoS One 3:e4019. Google Scholar


Miller MW, Williams ES, McCarty CW, Spraker TR, Kreeger TJ, Larsen CT, Thorne ET. 2000. Epizootiology of chronic wasting disease in free-ranging cervids in Colorado and Wyoming. J Wildl Dis 38: 676–690. Google Scholar


Walsh DP, editor. 2012. Enhanced surveillance strategies for detecting and monitoring chronic wasting disease in free-ranging cervids. US Geological Survey Open-File Report 2012-1036, Madison, Wisconsin, 42 pp. . Accessed January 2021. Google Scholar


WAFWA (Western Association of Fish and Wildlife Agencies). 2017. Recommendations for adaptive management of chronic wasting disease in the west. WAFWA Wildlife Health Committee and Mule Deer Working Group. Edmonton, Alberta, Canada and Fort Collins, Colorado, 20 pp. . Accessed January 2021. Google Scholar


Williams ES, Young S. 1980. Chronic wasting disease of captive mule deer: A spongiform encephalopathy. J Wildl Dis 16:89–98. Google Scholar


Williams ES, Young S. 1992. Spongiform encephalopathies in Cervidae. Rev Sci Tech 11:551–567. Google Scholar


Wolfe LL, Watry MK, Sirochman MA, Sirochman TM, Miller MW. 2018. Evaluation of a test and cull strategy for reducing chronic wasting disease prevalence in mule deer (Odocoileus hemionus). J Wildl Dis 54:511–519. Google Scholar
© Wildlife Disease Association 2021
Michael W. Miller and Lisa L. Wolfe "Inferring Chronic Wasting Disease Incidence from Prevalence Data," Journal of Wildlife Diseases 57(3), 718-721, (5 July 2021).
Received: 30 November 2020; Accepted: 29 January 2021; Published: 5 July 2021

Back to Top