Since January 2007, a total of 11 rabid deer from 4 deer farms were identified in the neighboring counties of Lancaster and Chester, Pennsylvania, USA. These farms were located with a 45-mile radius in an area of high prevalence for rabies among all animals (excluding bats) (Figure 1). We investigated this cluster to identify factors that might have contributed to disease transmission, assess the risk to humans, and provide recommendations for prevention.

We conducted a case–control study among deer farms in Pennsylvania. A case farm was defined as a registered deer farm with >1 laboratory-confirmed case of rabies in a deer. A control farm was defined as a registered deer farm that did not report any laboratory-confirmed rabies in deer. Ten control farms were chosen by referral from farmers and from the Pennsylvania Deer Farmers Association website. Control farms were limited to the affected counties and 1 noncontiguous county to account for local ecologic diversity. Four farms (farms A–D) met the case definition. Staff from the Centers for Disease Control and Prevention, state and local health departments, and the Pennsylvania Department of Agriculture conducted site visits to each affected farm. A standardized questionnaire was administered in person or by telephone to all study farms during August 2–5, 2010. Proportions were compared with Fisher exact test by using SAS version 9.2 (SAS Institute Inc., Cary, NC, USA).

Farmers’ ages, levels of education, and number of years farming did not differ between case and control farms (Table 1). Control farms tended to be larger than case farms (27.4 vs. 15.3 acres; odds ratio [OR] 0.99, 95% CI 0.95–1.03) with more deer per acre (36.8 vs. 34.7; OR 0.99, 95% CI 0.96–1.04), although neither difference was significant (Table 1). Farming practices did not differ between case and control farms (Table 1). Trough and bottle feeding of deer were common practices, and most (71%) farms reported using sweet deer feed containing molasses, which might attract rabies reservoir species, such as raccoons. Deer were moved infrequently between pens, and few outside deer were brought onto the farms (mean 2.5 deer/year); however, interstate travel of deer was reported. Most (71%) farms reported vaccinating deer against at least 1 disease, but deer were vaccinated on case farms only in response to previously rabid deer (Table 1). A low perceived risk for rabies was cited as the primary barrier to rabies vaccination among control farms. In contrast, witnessed contact between deer and rabies reservoir species was relatively common (43% of farms reported contact with skunks, and 36% reported contact with raccoons) (Table 2).

The primary method of preventing rabies in animals is vaccination, which in turn reduces the risk for transmission to humans. Although not licensed for use in deer, rabies vaccine is expected to have a similar safety and efficacy profile in deer as it does in other ruminants. The Compendium of Animal Rabies Prevention and Control recommends that rabies vaccination be considered in livestock that have frequent contact with humans or in livestock that are particularly valuable (6). Captive deer would meet both criteria on the basis of the human exposures reported in the survey and their economic value (individual deer can cost hundreds to hundreds of thousands of dollars). Interstate transportation of deer is state regulated and generally prohibits importation of deer from states with endemic chronic wasting disease. However, the survey did document interstate travel of captive deer, which raises concern for the possible translocation of a raccoon rabies virus variant across oral rabies vaccination boundaries. To decrease this risk, rabies vaccination should be strongly considered in any deer transported between farms or across state borders.

Although no specific management practices were identified as major risk factors among affected deer farms, the rabies virus most likely was transmitted to the deer through contact with wildlife, as reported in the survey. However, deer-to-deer transmission could not be excluded at farms where multiple deer were affected (i.e., farms A and D). At both locations, adult does died, followed by buck fawns, suggesting that infection could have occurred during close maternal activities involving saliva transmission. However, not all of the rabid fawns belonged to the rabid does, and some of the rabid fawn’s mothers remained healthy, even though doe can display maternal behavior toward multiple fawns. Regardless, measures to decrease contact between captive deer and rabies reservoir species should be implemented. Such measures might include trapping and removing such species, eliminating brush, groundhog burrows, or other potential sources of dens or cover for terrestrial carnivores; avoiding planting crops or storing food near deer pens; and using elevated, closed feeders (such as tube feeders) placed away from pen fences.