Purpose: Rickettsiosis is a re-emergent disease that is not the subject of epidemiological surveillance in Colombia. Rickettsia cases are generally undiagnosed and several lethal outbreaks have been reported in Colombia during the last decade. The aim of this study is to analyze eco-epidemiological aspects of rickettsia infection in two of the towns with previous reported outbreaks. Methods & Materials: A cross-sectional study was conducted enrolling 597 people living in 246 houses from nine villages in two towns in Colombia from November 2015 to January 2016. Sociodemographic and housing conditions information (exposures) was collected by interview. Blood samples were collected to assess rickettsia infection in humans, horses and dogs (IFA, cut-off=1/128). In addition, ticks from humans and animals were evaluated for Rickettsia infection by qPCR targeting gltA and OmpB genes. All participants signed an informed consent. Data were analyzed by a multilevel logistic regression model using two levels (house and village) and seropositivity as the main outcome. Results: Overall prevalence of infection by Rickettsia was 25.6% (95%CI 22.2-29.3). Four zones with higher proportion of infection in both people and animals were identified. Spotted Fever Group Rickettsia was detected in Amblyomma ticks from humans and animals. The multivariate model showed the association between working outdoors and infection was confounded by age (years) and sex (male) (ORcrude= 2.25, 95%CI 1.47-3.46; ORadjust=1.06, 95%CI 0.61-1.85). Sex and age is a risk marker for rickettsia infection in this zone (ORmen=1.88, 95%CI 1.15-3.07) and OR35years vs 10years=1.59, 95%CI 1.19-2.10). Not farming into the forest is a protection factor (OR=0.49, 95%CI 95:0.29-0.82) and the proportion of seropositive animals is strongly associated with human infection (ORproportion52.9% vs proportion8.7%=3.58, IC95:1.69-3.73). The model revealed variability among houses but not among villages (RandomEffecthouses=0.65 ± Std. Err=0.25, 95%CI 0.30-1.39 and RandomEffectvillages=0.11±StdErr=0.29, 95%CI 0.00-21.46). Conclusion: Studies on rickettsia infection in areas where previous outbreaks occurred are useful as an early approach for surveillance of re-emergence diseases in Colombia. Domestic animals could have an important role as sentinels for detecting zones with Rickettsia circulation. Finally, prevention measures should be taken to avoid invasion of primary and secondary forests where increased transmission of emergent and re-emergent pathogens such as Rickettsia can occur.