Comparative Study of Free-Roaming Domestic Dog Management and Roaming Behavior Across Four Countries: Chad, Guatemala, Indonesia, and Uganda

Dogs play a major role in public health because of potential transmission of zoonotic diseases, such as rabies. Dog roaming behavior has been studied worldwide, including countries in Asia, Latin America, and Oceania, while studies on dog roaming behavior are lacking in Africa. Many of those studies investigated potential drivers for roaming, which could be used to refine disease control measures. However, it appears that results are often contradictory between countries, which could be caused by differences in study design or the influence of context-specific factors. Comparative studies on dog roaming behavior are needed to better understand domestic dog roaming behavior and address these discrepancies. The aim of this study was to investigate dog demography, management, and roaming behavior across four countries: Chad, Guatemala, Indonesia, and Uganda. We equipped 773 dogs with georeferenced contact sensors (106 in Chad, 303 in Guatemala, 217 in Indonesia, and 149 in Uganda) and interviewed the owners to collect information about the dog [e.g., sex, age, body condition score (BCS)] and its management (e.g., role of the dog, origin of the dog, owner-mediated transportation, confinement, vaccination, and feeding practices). Dog home range was computed using the biased random bridge method, and the core and extended home range sizes were considered. Using an AIC-based approach to select variables, country-specific linear models were developed to identify potential predictors for roaming. We highlighted similarities and differences in term of demography, dog management, and roaming behavior between countries. The median of the core home range size was 0.30 ha (95% range: 0.17-0.92 ha) in Chad, 0.33 ha (0.17-1.1 ha) in Guatemala, 0.30 ha (0.20-0.61 ha) in Indonesia, and 0.25 ha (0.15-0.72 ha) in Uganda. The median of the extended home range size was 7.7 ha (95% range: 1.1-103 ha) in Chad, 5.7 ha (1.5-27.5 ha) in Guatemala, 5.6 ha (1.6-26.5 ha) in Indonesia, and 5.7 ha (1.3-19.1 ha) in Uganda. Factors having a significant impact on the home range size in some of the countries included being male dog (positively), being younger than one year (negatively), being older than 6 years (negatively), having a low or a high BCS (negatively), being a hunting dog (positively), being a shepherd dog (positively), and time when the dog was not supervised or restricted (positively). However, the same outcome could have an impact in a country and no impact in another. We suggest that dog roaming behavior is complex and is closely related to the owner's socioeconomic context and transportation habits and the local environment. Free-roaming domestic dogs are not completely under human control but, contrary to wildlife, they strongly depend upon humans. This particular dog-human bound has to be better understood to explain their behavior and deal with free-roaming domestic dogs related issues.