Academic journal article Bulletin of the World Health Organization

Urban Epizootic of Rabies in Mexico: Epidemiology and Impact of Animal Bite Injuries

Academic journal article Bulletin of the World Health Organization

Urban Epizootic of Rabies in Mexico: Epidemiology and Impact of Animal Bite Injuries

Article excerpt


In 1989, a total of 2776 human rabies deaths and 1 041 031 rabies post-exposure prophylaxis treatments were reported to WHO.(a) In developing countries, more than 200 persons undergo post-exposure prophylaxis for every human rabies case [1], often at substantial cost and sometimes with serious side-effects.

Control of dog rabies is crucial, since almost all human rabies result from bites by rabid dogs.(a) In several developing countries, dog rabies seems to be recurring in areas were it had once been controlled, possibly because of increasing population density and mobility of people and dogs,(b) and the failure to maintain rabies control programmers. In most countries, dog rabies is an urban problem (2).(c) Although the epidemiology of enzootic rabies in dogs has been amply described [3-11] reports of epizootics in dogs are relatively rare [12-14].

Animal bite injuries cause considerable morbidity and occasional mortality throughout the world. In areas where rabies is enzootic, such injuries also place an additional burden on health resources because of the frequent need to administer rabies post-exposure prophylaxis. We are unaware of any studies that have described the impact of animal bite injuries during a rabies epizootic. Accordingly, we report on the epidemiology of a major urban epizootic of rabies in Hermosillo, Mexico, and describe the impact of animal bite injuries during this outbreak.

Materials and methods

Hermosillo, the state capital of Sonora, is located in north-western Mexico, approximately 275 km south of the U.S. border. In 1987 the city's estimated population was 431 000, and is rapidly growing. At the time of the epizootic, Hermosillo was divided into seven administrative health sectors. No cases of animal rabies were reported in Hermosillo from 1980 to 1984. In 1985 and 1986, respectively, 1 (2%) of 58 and 2 (4%) of 52 animals tested were positive for rabies. In the second half of 1987, however, 121 (74%) of 164 animals tested were positive for rabies, and the number of cases increased despite a mass vaccination programme.


We initiated our investigation in January 1988. Records of all laboratory-confirmed cases of animal rabies (by direct fluorescence antibody (FA) examination of brain tissue) from 1 January 1985 to 28 February 1989 were obtained from the Hermosillo Centro de Salud. We collected information on characteristics of the animal and owner (if any) and the number of persons exposed by the animal.

Laboratory diagnosis of animal rabies

Samples of brain tissue from 70 suspect rabid dogs and cats examined by the Hermosillo Centro de Salud were also tested blind by the Centers for Disease Control (CDC) and the Arizona Department of Health Services. A series of monoclonal antibodies against rabies virus was used to compare three virus isolates from rabid dogs in Hermosillo with previous isolates from Mexico [15].

Household survey

A house-to house survey of 1104 households was conducted in February 1988 to evaluate the owned dog population, rabies vaccination coverage, and other possible risk factors for rabies in dogs. The sampling was based on household clusters of the city developed by health officials for a previous childhood immunization survey. Each of the 550 clusters consisted of approximately 160 socioeconomically homogeneous households. We randomly selected 69 household clusters and, starting from a random point, surveyed 16 consecutive houses (approximately 10% of those in each cluster) for dog ownership, vaccination status, and household characteristics. Because many households were reluctant to disclose their income, we used the mean household size and average monthly expenditure as surrogate measures of socioeconomic status (SES). The expenditure ranges we selected were too wide, resulting in a median monthly household expenditure that was similar in most sectors; as a measure of the median household expenditure in each sector we therefore used the proportion of households that reported expenditures below the overall median level. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.