Seroepidemiologic Survey of Potential Pathogens in Obligate and Facultative Scavenging Avian Species in California

Straub, Mary H.; Kelly, Terra R.; Rideout, Bruce A.; Eng, Curtis; Wynne, Janna; Braun, Josephine; Johnson, Christine K.

Seroprevalence for each pathogen was estimated along with 95% confidence intervals. The two-sided Fisher's exact test was used to compare seroprevalences between free-flying condor release locations, between the three scavenger bird species, and between captive and free-flying condors. West Nile virus serostatus among free-flying California condors was evaluated for associations with sex, age (years), rearing site, time since last vaccination (never vaccinated, 3 years and > 3 years), number of previous vaccines (range: 0-4), and release location using logistic regression analysis (n = 85 condors). Five condors had been free-flying initially and were then brought into captivity for medical treatment for various reasons including lead exposure, microtrash, or injury. Because these individuals might differ from the others in terms of their overall health, which may affect their response to WNV vaccination, they were excluded from the analysis. Six other individuals had insufficient sample volume for WNV serology. Rearing site was defined as the location where the individual spent the majority of its first year of life: the Los Angeles Zoo; San Diego Zoo Safari Park; sites outside of California including the Oregon Zoo and the World Center for Birds of Prey in Boise, Idaho; and the wild (reference category). Each of the variables was first assessed in univariable logistic regression models. Predictor variables with p-values of 0.2 in the univariable models, in addition to potential confounding variables, were included in the multivariable analysis. Biologically plausible two-way interactions were also assessed for significance. Multivariable models were evaluated using Akaike Information Criterion (AIC), and multicollinearity among the predictor variables was assessed using variance inflation factors (VIF). Overall model fit was evaluated using the Hosmer Lemeshow test [88] . Spearman's rank correlation was used to test the association between time since previous vaccination and age, as well as time since previous vaccination and number of vaccinations. Statistical analyses were performed using R (version 2.15.1, R-development Core team, http://www.r-project.org).