Dog demography and population estimates for rabies control in Bali, Indonesia

Purpose: The battle against rabies, a fatal zoonotic disease, is still ongoing on the island of Bali, Indonesia. Several rounds of mass vaccination have been conducted to control the disease in dogs. Characterizing the dog population is important for effective control measures. The objective of this study is to (1) characterize the demography of owned and free-roaming dogs and (2) estimate the abundance and explore potential predictors of dog populations in villages in Bali. Methods: The study was conducted in one year (March 2011 through March 2012) on two dog subpopulations, owned and free-roaming, in 37 villages in Bali. Data on owned dogs were acquired with a door to door (DTD) survey, while a photographic mark recapture (PMR) survey was used to collect information on free-roaming dogs. The sampling unit is a “banjar”, a community unit which forms a village, and the design was two-stage sampling. In the first stage villages were stratified by urbanization and selected using stratified random sampling. At the second stage all banjars within a village were subjected to a DTD survey whereas only 4 banjars per village were randomly selected for a PMR survey. Dog variables of interest were sex, age group, vaccination status and confinement status for owned dogs. In addition, data on human population and presence of a market, bus terminal, temple, school, beach, rice paddies, plantation, and forest was collected for each banjar. Results: A total of 17,376 owned dogs and 1972 free-roaming dogs were surveyed in the study. Of the owned dogs, 70 % were male, 84% were adults, 66% were allowed to roam freely and 83.6% were vaccinated against rabies. Meanwhile in free-roaming dogs, 76.8% were male, 96.6% were adults, and only 30.9% had vaccination collars. Conclusions: Further analysis is being conducted to test predictor variables of banjar dog populations while correcting for detection error in the PMR survey. DTD survey data is being analyzed using R and PMR data with Program MARK - a statistical software designed to analyze data derived from marked individuals