Risk-based cost modelling of oral rabies vaccine interventions for raccoon rabies

Distribution of oral rabies vaccine (ORV) is an effective but costly strategy to control raccoon rabies. Because of high costs, ORV for raccoon rabies in the U.S. has been limited primarily to epizootic areas, leaving extensive raccoon rabies regions without any ORV intervention. Several cost scenarios for ORV application in raccoon rabies enzootic and epizootic regions were modelled in New York State to obtain estimated costs of ORV baits per scenario and potential savings compared with a uniform ORV baiting strategy. These cost scenarios modelled at the census tract, level the application of ORV baits at different densities according to levels of risk defined by the observed number of raccoon rabies cases per km2 and the expected number of cases per km2 estimated with a Poisson regression model. Bait purchase costs were lower using the modelled cost scenarios than a uniform baiting strategy, for both the NYS enzootic region and the Long Island epizootic zone. The proportion of savings for the NYS enzootic region was 29.57%, and the proportion of savings for the Long Island epizootic zone was 38.9%. Use of these cost scenarios to determine bait distribution by rabies risk level should be considered to maximize efficacy and reduce costs of ORV interventions.     

Challenges to controlling rabies in skunk populations using oral rabies vaccination: A review

Controlling rabies in skunk populations is an important public health concern in many parts of the United States due to the potential for skunk rabies outbreaks in urban centres and the possible role for skunks in raccoon rabies variant circulation. Oral rabies vaccination (ORV) programmes have supported wildlife rabies control efforts globally but using ORV to control rabies in skunk populations has proven more challenging than with other target species, like foxes, coyotes and raccoons. A review of published studies found that some ORV constructs are immunogenic in skunks and protect against virulent rabies virus challenges, especially when delivered by direct installation into the oral cavity. However, in field ORV programmes using currently available vaccine‐bait formats and distribution methods targeting other rabies reservoir species, skunks often fail to seroconvert. Field effectiveness of ORV in skunks appears to be limited by poor bait uptake or inadequate ingestion of vaccine rather than from poor vaccine efficacy. Observations of captive skunks revealed vaccine spillage when handling and biting into baits such that modification of bait formats might improve field effectiveness. In addition, a dose–response relationship between bait distribution density and post‐baiting seroconversion among skunks was observed across the limited number of field studies. Additional research is needed to identify opportunities to modify ORV baits and distribution strategies to improve the viability of ORV as a rabies control strategy in skunks.     

Control of terrestrial animal rabies in Anne Arundel County, Maryland, after oral vaccination of raccoons (1998-2007)

Objective-To evaluate the effectiveness of an oral rabies vaccination (ORV) project conducted from 1998 through 2007 in Anne Arundel County, Md, for the control of rabies in terrestrial animals. Design-Retrospective analysis of surveillance data (1997 through 2007). Animals-Free-ranging raccoons (Procyon lotor) and other terrestrial mammals. Procedures-Vaccinia-rabies glycoprotein recombinant virus oral rabies vaccine-bait units were distributed annually by aircraft and ground teams targeting free-ranging raccoons. Approximately 2 to 4 weeks following the vaccine-bait placement, raccoons were live trapped, sedated, processed, and then released. Serologic samples were tested for the presence of rabies virus-neutralizing antibodies (RVNAs). Bait acceptance was estimated by analysis of tetracycline biomarking of sampled teeth. Rabies incidence was determined by the passive identification of rabid terrestrial animals. Results-The incidence of rabies in terrestrial animals decreased 92% between 1997 (the year prior to the start of the ORV project) and 2007. The mean RVNA prevalence across all years was 33% among trapped raccoons in areas baited with a fish meal polymer bait type, whereas the mean bait acceptance was 30%. Adult raccoons had a seropositivity rate twice that of juvenile raccoons, whereas the bait acceptance rate between adults and juveniles did not differ significantly. For areas baited with a coated sachet bait, adults and juveniles had the same seroprevalence. Juveniles had better seroprevalence when the annual campaign started in September and October, compared with August. Conclusions and Clinical Relevance-The ORV project contributed to a significant decrease in annual incidence of terrestrial animal rabies in Anne Arundel County, Md, during the 10-year project period. For fish meal polymer baits, juvenile raccoons accessed bait at the same rate as adult raccoons but had a significantly lower prevalence of RVNAs. For coated sachet baits, seroprevalence was the same in both age groups. The time of year the bait distribution occurred and the bait type used may be partial explanations for the difference in RVNA seroprevalence between adults and juvenile raccoons.     

The economic implications of sylvatic rabies eradication in Italy

After more than 10 years of absence, sylvatic rabies re‐appeared in Italy in 2008. To prevent disease spread, three oral rabies vaccination (ORV) campaigns targeting red foxes were performed through manual distribution of vaccine baits between January and September 2009. As these campaigns proved unsuccessful, at the end of December 2009, baits started being distributed using helicopters, allowing uniform coverage of larger areas in a shorter time period. From winter 2009 to autumn 2016, a total of 15 ORV campaigns (four emergency, four regular and seven preventive ORV) were implemented through aerial distribution of baits. In this study, we assessed the costs of the aerial ORV campaigns, which were aimed at eradicating the disease and reobtaining the rabies‐free status. Cumulative costs per km² were estimated at €59.45 during emergency campaigns and ranged between €51.94 and €65.67 in the regular vaccinations. The main portion of costs for ORV programmes were related to baits supply and distribution: €49.24 (82.83%) in emergency campaigns and from €40.33 to € 43.35 in regular ORVs (71.97% and 66.02%, respectively). At the end of each ORV campaign, the efficacy of vaccination activities was estimated by assessing the proportion of foxes testing positive for tetracycline biomarker in jawbone, indicating bait intake. Results revealed that the proportion of foxes that ingested baits varied between 70.97% and 95.51%. Statistical analysis indicated that reducing the density of dropped baits could potentially lead to a cost‐saving of 22.81%, still maintaining a satisfactory level of bait intake by the fox population.     

A Retrospective Economic Analysis of the Ontario Red Fox Oral Rabies Vaccination Programme

Ontario initiated a red fox (Vulpes vulpes ) oral rabies vaccination (ORV) programme in 1989. This study utilized a benefit‐cost analysis to determine if this ORV programme was economically worthwhile. Between 1979 and 1989, prior to ORV baiting, the average annual human post‐exposure treatments, positive red fox rabies diagnostic tests and indemnity payments for livestock lost to rabies were 2248, 1861 and $246 809, respectively. After baiting, from 1990 to 2000, a 35%, 66% and 41% decrease in post‐exposure treatments, animal rabies tests and indemnity payments was observed, respectively. These reductions were viewed as benefits of the ORV programme, whereas total costs were those associated with ORV baiting. Multiple techniques were used to estimate four different benefit streams and the total estimated benefits ranged from $35 486 316 to $98 413 217. The annual mean ORV programme cost was $6 447 720, with total programme costs of $77 372 637. The average benefit‐cost ratios over the analysis period were .49, 1.06, 1.27 and 1.36, indicating overall programme efficiency in three of the four conservative scenarios.     

Safety and immunogenicity of a vaccine bait containing ERA strain of attenuated rabies virus.

Ninety percent of foxes fed commercial ERA vaccine in a specially designed bait developed rabies serum neutralizing antibodies. The vaccine bait did not cause clinical signs of rabies when consumed by foxes, raccoons, skunks, dogs, cats, cattle and monkeys. When presented, in the laboratory, to wild rodents of the species Microtus, Mus musculus and Peromyscus, the vaccine baits caused vaccine-induced rabies only in Mus musculus. Laboratory mice of the CD-1 and CLL strain were susceptible to vaccine-induced rabies; however, studies showed that transmission of virus to other animals did not occur. These studies suggest that the vaccine bait described could be useful in a rabies control program in areas where foxes and wild dogs are the principal vectors.     

Safety and immunogenicity of ERA strain of rabies virus propagated in a BHK-21 cell line.

The ERA strain of rabies virus was propagated in a baby hamster kidney cell line (BHK-21/C13). The viral titer was 10(1.8) tissue culture infective doses (TCID) higher than that of commercial ERA vaccine. The ERA/BHK-21 vaccine in baits retained titers of 10(6.3) to 10(6.4), TCID when subjected to daily temperature fluctuations from 9 degrees C to 24 degrees C for 21 days. This titer, according to a dose response in laboratory foxes, was still capable of immunizing up to 100% of foxes consuming a bait. The ERA/BHK-21 vaccine, when presented in baits, produced antibodies in 80 to 100% of dogs consuming more than one bait. Duration of immunity in foxes, from feeding the ERA strain rabies virus in baits, as determined by resistance to challenge with virulent virus, was at least 48 months. The vaccine strain retained some pathogenicity for nontarget species. In tests carried out on foxes, raccoons, dogs, cats and cattle, the vaccine did not cause vaccine-induced rabies. One of 14 skunks which consumed four baits developed vaccine-induced rabies, but virus could not be isolated from the salivary glands of this animal. The vaccine, when presented in baits, caused vaccine-induced rabies in 37% of laboratory mice, 3.4% of Microtus and 2.6% of Peromyscus species. Rabies virus could not be isolated from the salivary glands of rodents with vaccine-induced rabies. It was concluded that ERA virus propagated in BHK-21/C13 cells and incorporated in an acceptable bait produced a high titer, stable, immunogenic and safe vaccine for foxes.     

Evaluation of 6 ONRAB Ultralite bait flavor matrices delivered to coyotes (Canis latrans): Implications for oral rabies vaccination

The spread of rabies in terrestrial wildlife throughout the United States is primarily controlled through oral rabies vaccination. Relatively low bait acceptance and seroconversion rates by some target species have prompted investigation into an alternative to the RABORAL V-RG bait currently used. In Canada, ONRAB Ultralite baits are used to vaccinate raccoons (Procyon lotor) and striped skunks (Mephitis mephitis). Comparative studies between RABORAL V-RG and ONRAB found higher seroconversion rates among raccoons that ingested ONRAB, suggesting that it may be a suitable alternative. However, ONRAB has not been evaluated in many rabies reservoir species, including coyotes (Canis latrans). Vaccination of coyotes is a critical element in preventing reemergence of canine strain of rabies in the United States. We evaluated flavor preference of ONRAB Ultralite oral rabies vaccine baits by coyotes. Preferences among bait types differed (Friedman χ² = 13.28; df = 5; P = 0.02). Of the 6 bait flavors evaluated, cheese ranked the highest, followed by fish, chicken, sugar-vanilla, egg, and bacon flavors. Pairwise trials among the top 3 flavors (cheese, fish, and chicken) showed no difference (Friedman χ² = 3.00; df = 2; P = 0.22). Our research suggests that among the bait flavors we evaluated, cheese, fish, or chicken-flavored baits may be an appropriate flavor for delivery of ONRAB Ultralite baits to coyotes.     

The spatial and temporal disappearance of different oral rabies vaccine baits

Bait disappearance can give valuable information for the assessment of oral vaccination campaigns of foxes against rabies. In this study, the spatial and temporal disappearance of three different vaccine baits under almost identical conditions was investigated. In the study area, 350 baits were placed at previously marked positions during two different periods; late autumn and early spring. The distribution of baits was in accordance with the method as recommended by the European Union; a density of 20 baits per km2 along flight lines 500m apart. Bait disappearance was checked 1, 3, 5 and 7 days after distribution. At least 80% of the baits had disappeared within one week after distribution. No difference in bait disappearance was observed between the two selected periods. However, a significant higher bait disappearance was observed in forested areas when compared to open agricultural areas. Furthermore, the differences in bait disappearance between the three type of baits tested were relatively small and not significant.     

Preference among 7 bait flavors delivered to domestic dogs in Arizona: Implications for oral rabies vaccination on the Navajo Nation

Less than 20% of the domestic dogs on tribal lands in the United States are vaccinated against rabies. One method to increase vaccination rates may be the distribution of oral rabies vaccines (ORVs). ONRAB^® (Artemis Technologies, Inc., Ontario, Canada) is the primary ORV used in Canada to vaccinate striped skunks and raccoons. To investigate the potential use of ONRAB^® ORV baits to vaccinate feral domestic dogs against rabies on tribal lands and beyond, we performed a flavor preference study. A total of 7 bait flavors (bacon, cheese, dog food, hazelnut, sugar-vanilla, peanut butter, and sardine) were offered in pairs to 13 domestic dogs. Each dog was offered all possible combinations of bait pairs over a period of 10 days, with each bait offered 6 times. The proportion of times each bait was consumed first by individual dogs was calculated and comparisons among dogs were conducted using the MIXED procedure in SAS (SAS Institute, Cary, NC). Pairwise comparisons between baits were performed using “contrast” statements with sugar-vanilla flavor as the default for comparison. Type 3 tests of fixed effects showed a significant treatment effect (F_6,72 = 9.74, P < 0.0001). Sugar-vanilla was selected first during 14% of the offerings and exhibited the least preference among all bait types (F_1,72 = 22.46, P < 0.0001). Dog food was selected first 56% of the time, and more frequently than all other bait types (F_1,72 = 13.09, P = 0.0005).     

Persistence of elevated rabies prevention costs following post-epizootic declines in rates of rabies among raccoons (Procyon lotor)

Determining the benefits to cost relationships among different approaches to rabies control and prevention has been hindered by the inherent temporal variability in the dynamics of disease among wildlife reservoir hosts and a tangible and objective measure of the cost of rabies prevention. A major and unavoidable component of rabies prevention programs involves diagnostic testing of animals and the subsequent initiation of appropriate public health responses. The unit cost per negative and positive diagnostic test outcome can be reasonably estimated. This metric when linked to methodologies subdividing the epizootic process into distinct temporal stages provided the requisite detail to estimate benefits derived from rabies control strategies. Oral rabies vaccine (ORV), for prevention of the raccoon-associated variant of rabies, has been distributed in Ohio and adjoining states in an effort to develop an immune barrier to the westward spread of epizootic raccoon rabies. The costs of ORV delivery have been quantified. Herein, the cost structures required to assess the benefits accrued by prevention were developed. A regression model was developed effectively predicting (r² = 0.70) the total number of rabies diagnostic tests performed by 53 counties in five northeastern (NE) states from 1992 to 2001. Five temporal stages sufficed to capture the range of variability in the raccoon rabies epizootic process. Unit costs, dollars per diagnostic test outcome, were calculated for negative and positive results from published reports. Ohio counties were matched to NE counties based on similar socioeconomic characters. A “pseudo-epizootic” of raccoon rabies was introduced into Ohio and the costs savings from ORV were derived as the excess costs imposed by epizootic spread throughout the state. At 46 km/year (range modeled, 30–60 km/year), the pseudo epizootic spread, and reached the enzootic stage, in all Ohio counties by year 13 (range modeled, 11–17 years). Cumulative excess costs for Ohio ranged between $11 and $21 million; counties of low socioeconomic status experienced the greatest relative excess costs. The costs for rabies prevention activities reached apices during the epizootic stage of raccoon rabies (2.7–10.8 times baseline) an unforeseen finding indicated elevated costs persisted (1.7–7.2 times baseline) into the enzootic stage.     

Oral vaccination against rabies and the behavioural ecology of the red fox (Vulpes vulpes)

As a result of oral vaccination of foxes (Vulpes vulpes) against rabies, this virus disease has almost been completely eradicated from West- and Central Europe. In most countries, vaccine baits were distributed twice a year: during spring (March to May) and autumn (September to October). This strategy has shown to be able to control and eventually eradicate rabies. However, it remains to be clarified if this is the most cost-effective strategy. Astonishingly, the behavioural ecology of the target species, the red fox, did receive only limited attention selecting the periods when baits should be distributed. Considering the behavioural ecology and rabies epidemiology of foxes, territory owners seem to play a key role in the spread and maintenance of rabies. Thus, oral vaccination campaigns should be targeted primarily at these animals. It is suggested that the optimal timing for bait distribution in Europe is late autumn (November) or early winter (December), depending on the prevailing climatic conditions. Additional campaigns, when financially feasible, can be implemented in order to maintain a high vaccination coverage during the remaining year. Furthermore, different baiting strategies can be selected in case of re-infection or persistent residual foci.     

An epidemic of sylvatic rabies in Finland--descriptive epidemiology and results of oral vaccination.

When rabies reappeared in Finland in April 1988, the country had been rabies free since 1959. Soon a picture of sylvatic rabies become evident, its main vector and victim being the raccoon dog (Nyctereutes procyonoides). Between 8 April 1988 and 16 February 1989, 66 virologically verified cases were recorded (48 raccoon dogs, 12 red foxes, 2 badgers, 2 cats, 1 dog and 1 dairy bull) in an area estimated at 1700 km2 in south-eastern Finland. The greatest distance between recorded cases was 67 km. A positive reaction with monoclonal antibody p-41 indicated that the virus was an arctic-type strain. A field trial on oral immunization of small predators was initiated in September 1988 using Tübingen fox baits according to the Bavarian model of bait distribution. Each bait contained 5*10(7) TCID50/ml modified live rabies virus (SAD-B19). The 6 months' surveillance indicate a seroconversion rate of 72% (N = 126) in the raccoon dog population, 67% (N = 56) in the red foxes and 13% (N = 16) in the badgers, when titers greater than or equal to 1.0 IU/ml are considered seropositive. In the whole follow-up period, no statistically significant difference could be detected between the raccoon dogs and red foxes in the rate of seroconversion or in the uptake of tetracycline from the baits. Notably high antibody levels were recorded in both raccoon dogs and red foxes within 4-5 months after vaccination. Of the seropositive animals, the proportion of animals with titers 3.0 IU/ml or greater was higher in raccoon dogs (73%) than in red foxes (51%) (x2 = 5.29, p less than 0.05). The trial shows that raccoon dogs can be immunized against rabies in the field with vaccine baits originally developed for controlling sylvatic rabies in foxes.     

Antibody response to Raboral VR-G® oral rabies vaccine in captive and free-ranging black-backed jackals (Canis mesomelas)

Rabies is a zoonotic disease that remains endemic in large parts of southern Africa because of its persistence in wildlife and domestic dog vectors. The black-backed jackals (Canis mesomelas) is primarily the wildlife vector responsible for rabies outbreaks in northern parts of South Africa. Two trials were carried out to investigate antibody responses to the oral rabies vaccine Raboral V-RG® in black-backed jackals under captive and free-ranging conditions. In captive jackals 10/12 (83%; 95% confidence interval [CI]: 52% – 98%), seroconverted after single oral vaccination. Nine captive jackals had protective antibody titres (> 0.5 IU/mL) at 4 weeks (median: 2.1 IU/mL; inter quartile range [IQR]: 0.6–5.7) and 10 jackals had at 12 weeks (median: 3.5 IU/mL; IQR: 1.5–8.3) and three maintained antibody titres for up to 48 weeks (median: 3.4 IU/mL; IQR: 2.0–6.3). Four sites were baited with Raboral V-RG® vaccine for wild jackals, using fishmeal polymer and chicken heads. Baits were distributed by hand or from vehicle at three sites in north-eastern South Africa, with an average baiting density of 4.4 baits/km² and at one site in central South Africa, at 0.12 baits/km². This resulted in protective antibody titres in 3/11 jackals (27%; 95% Cl: 6–61) trapped between 3 and 12 months after baiting in north-eastern South Africa, compared with 4/7 jackals (57%; 95% Cl: 18–90) trapped after 3–18 months in central South Africa. This study shows the potential utility of oral rabies vaccination for the control of wildlife-associated rabies in north-eastern and central South Africa, but extensive studies with wider distribution of bait are needed to assess its potential impact on rabies control in wild jackals.     

Stability of attenuated live virus rabies vaccine in baits targeted to wild foxes under operational conditions

The viability of an attenuated live virus rabies vaccine in a bait targeted to red foxes was examined under various operational conditions in a series of experiments in Ontario. The virus was relatively stable over a 28-day period in the field, losing a mean 0.5, s = 0.2 log10 of virus titer. The micro-environment into which the bait was placed (open cultivated field, grassy meadow, wooded grove, sun or shade) did not make an appreciable difference in the viability of the virus. There was no significant difference (P < or = 0.05) between mean ambient temperatures and the temperature of fluids in blister packs of baits placed in sun or shade. Sixty-three percent of foxes fed baits exposed to sun and shade conditions for 21 days (titer 10(6.2) tissue culture infective doses per 1 mL) developed rabies virus-neutralizing antibodies. Storage of vaccine baits at -30 degrees C prior to bait distribution was important in maintaining virus viability.     

Vaccination of feral pigs (Sus scrofa) using iophenoxic acid as a simulated vaccine

OBJECTIVES: To develop an encapsulation method for delivery of vaccines to feral pigs, and quantify the effect of iophenoxic acid on captive feral pig blood iodine concentrations to assist in investigation of factors affecting vaccine uptake. DESIGN AND METHODS: Feral pigs were administered iophenoxic acid by oral gavage, and consumption was assessed for different encapsulation methods in baits. Blood iodine concentrations were monitored for eight days after consumption. The relationship between dose rate, time since dosing and blood iodine concentration was assessed for gavaged and baited captive feral pigs. Wild feral pigs were baited with PIGOUT baits containing 20 mg of encapsulated iophenoxic acid to simulate a vaccination program. Using knowledge from the pen studies, bait uptake and factors affecting bait uptake were investigated. RESULTS: Bait-delivered iophenoxic acid led to variable and inconsistent changes in blood iodine concentrations, in contrast to pigs receiving iophenoxic acid by gavage. This precluded accurate assessment of the quantity consumed, but still allowed a conservative determination of bait uptake. Iophenoxic acid in smaller capsules was consumed readily. Increasing baiting intensity appeared to increase bait uptake by wild feral pigs, and pigs of varying sexes, ages and weights appeared equally likely to consume baits. CONCLUSIONS: Encapsulated liquids can be delivered to feral pigs within baits, should the need to vaccinate feral pigs for fertility or disease management arise. High baiting intensities may be required.     

SURVIS: a fully-automated aerial baiting system for the distribution of vaccine baits for wildlife

Large-scale oral vaccination of wildlife against rabies using aerial bait distribution has been successfully used to control terrestrial wildlife rabies in Europe and North America. A technical milestone to large-scale oral rabies vaccination campaigns in Europe was the development of fully-automated, computer-supported and cost-efficient technology for aerial distribution of baits like the SURVIS -system. Each bait released is recorded by the control unit through a sensor, with the exact location, time and date of release and subsequently the collected data can be evaluated, e.g. in GIS programmes. Thus, bait delivery systems like SURVIS are an important management tool for flight services and the responsible authorities for the optimization and evaluation of oral vaccination campaigns of wildlife against rabies or the control of other relevant wildlife diseases targeted by oral baits.     

The use of tetracycline in anthelmintic baits to assess baiting rate and drug efficacy against Echinococcus multilocularis in foxes

Anthelmintic (praziquantel) baiting of wild red foxes against Echinococcus multilocularis infection was studied in a highly epizootic suburban area of Otaru, Hokkaido (the northern island of Japan) during the summer and autumn in the years 1999-2004. Acceptance of baits containing the biomarker tetracycline (TC) was evaluated. The prevalence of E. multilocularis infection in foxes before baiting (1999-2000) was 58% (88/153), whereas in the fourth year of bait distribution year (2004), it decreased to 11% (5/45). Analysis of TC marking in the teeth of foxes showed that 39% (77/195) of those captured after baiting were estimated to have consumed baits in the year of capture. Importantly, more juvenile (56%, 49/87) than adult foxes (26%, 28/108) were marked, indicating efficient baiting of juveniles, which tended to have a higher worm burden of E. multilocularis. Of 77 marked foxes, E. multilocularis and Alaria alata (monitored as the second indicator species of deworming) were not detected in 70 (90%) and 76 (99%) foxes, respectively. The results suggest effective deworming by bait consumption. However, it was also demonstrated that 9% of the marked foxes were infected or re-infected after bait consumption, suggesting high infection pressure and the importance of frequent baiting.     

Stability of vaccinia-vectored recombinant oral rabies vaccine under field conditions: a 3-year study

Rabies is an incurable zoonotic disease caused by rabies virus, a member of the rhabdovirus family. It is transmitted through the bite of an infected animal. Control methods, including oral rabies vaccination (ORV) programs, have led to a reduction in the spread and prevalence of the disease in wildlife. This study evaluated the stability of RABORAL, a recombinant vaccinia virus vaccine that is used in oral rabies vaccination programs. The vaccine was studied in various field microenvironments in order to describe its viability and facilitate effective baiting strategies. Field microenvironments influenced the stability of this vaccine in this study. This study emphasizes the importance of understanding how vaccines perform under varying field conditions in order to plan effective baiting strategies.     

Rabies oral vaccination of foxes during the summer with the VRG vaccine bait

The vaccination of foxes by distributing vaccine baits in the environment was initiated in France in 1986. Two campaigns per year were carried out: one in the spring and one in the autumn. After the spring campaigns, only 22-52% of fox cubs consumed vaccine baits compared to 75% of the adults and 70-80% of the adults or fox cubs after autumn campaigns. In order to reduce the period of time during which fox cubs do not have access to baits and are not immunised, a vaccination campaign was organised during the summer of 1992 over a contaminated area of 25,748 km2 where vaccines had never previously been given. Vaccine bait stability was assessed during the same summer in the field and their appetence tested on captive foxes. The efficacy of the campaign was evaluated by the relative decrease in rabies incidence and the rate of bait uptake by foxes compared to those from neighbouring areas vaccinated for the first time with the same vaccine during the spring or autumn. Summer vaccination significantly increased (P < 0.01) bait uptake by fox cubs (71%) compared with spring vaccination (39%), but no significant difference was observed for adult foxes. Moreover, the decrease in rabies incidence, measured during the 6-month period following the campaigns was less pronounced after summer vaccination (49% decrease) than when the first vaccination was carried out during the spring or autumn (79 and 72% decrease, respectively). Three campaigns led to an apparent elimination of rabies when the first campaign was performed in the spring or autumn, but only to a 76% decrease in rabies incidence density index when the first campaign was performed during the summer. The high thermostability of the Raboral VRG bait permits its use during the summer for an emergency campaign. For routine vaccination plans, however, the classical calendar of spring and autumn vaccination campaigns should continue to be preferred.