Pereira is the main city of the Coffee-Triangle region, which includes three departments (first administrative territory level) and 53 municipalities (second administrative territory level). Pereira is the capital of Risaralda (967,780 habitants in 2020), a department surrounded by six other western departments (Antioquia, Caldas, Tolima, Quindio, Valle del Cauca and Choco) [23]. Pereira’s landscape embraces both urban and rural areas. The first consisting of 20 communities (the city) and the second by 12 corregiments (sub-municipalities) (both tertiary administrative territory level) (Fig. 1). As per the National Statistics Department (DANE, www.dane.gov.co) the total population for 2020 reached 477,027 inhabitants. The metropolitan area includes Dosquebradas and La Virginia municipalities (Fig. 1), with a reported 709,338 inhabitants for 2020. The communities of Rio Otun, Centro, San Joaquin, Del Café, Boston, El Oso, Consota and Cuba, are the most populated areas of the Pereira municipality, making up for 51% of its population (Fig. 2). The municipality of Pereira extends over an area of 702 Km2 (4°48′51″N 75°41′40″W). The climate is tropical with an annual median temperature of 18.8°C (median minimum of 15.8°C, median maximum of 26.3°C).
Based on the municipality dog's census, we calculated a minimum sample of 98 dogs to be assessed. Finally, 100 were included distributed in four urban communes (out of 20) and four rural corregiments (out of 12) of the municipality Pereira, and its neighboring municipality of Dosquebradas (Fig. 1). From those owners who voluntarily accepted to participate with their dogs, only one dog was included, regardless of where co-living was in the same house.
Blood samples from each dog was individually collected from the radial vein into a sterile vacuum tube (Vacutainer, Becton, Dickinson and Company Franklin Lakes, NJ, USA) without anticoagulant at morning time. In the evening, samples were centrifuged at 1300–1800 × g for 20 min, followed by serum separation from the clot.
To determine selected-vector-borne pathogens exposure, a rapid enzyme-linked immunosorbent assay (ELISA) kit (SNAP® 4Dx® Plus Test Kit, IDEXX Laboratories, Inc, Westbrook, ME, USA) was used following the manufacturer's instructions [8]. This qualitative test allowed us to simultaneously detect the presence of circulating antibodies (IgG and IgM) against immunodominant proteins of E. canis/ewingii (p30 and p30-1, sensitivity of 96.2%), A. phagocytophilum/platys (p44/MSP2, sensitivity of 99.1%), B. burgdorferi s.l. (C6, the sensitivity of 98.8%), and D. immitis antigens (antigens principally produced by adult females) based on specific antibodies (sensitivity of 99.2%). The SNAP® 4Dx® Plus Test Kit showed a specificity of ~ 100% for the above mentioned microorganisms [8, 24, 25]. 4Dx® Plus Test Kit has been validated in dogs [8, 24, 25].
Besides the general demographic, clinical and laboratory surveys, a questionnaire for social variables related to living conditions and households was also performed, including aspects such as: house vulnerability (defined according to five possible types of houses, from those with luxury and appropriate sanitary conditions to those without luxury and inappropriate sanitary conditions, last category considered as a vulnerable house); house location (in rural or urban areas); environmental elements close proximity to the house (e.g. small lakes, small rivers or wetlands); materials employed in the building of the walls (e.g. blocks, bricks, cement, wood, cardboard-tin) and the floors (e.g. cement, wood, soil); access to tap water; need of water collection and its keeping at appropriate receptacles; and disposal of sewage water and waste disposal (e.g. by an urban service) [26], among other social aspects such as the socioeconomic level of the house which is determined by the National Statistics Department (DANE).
Statistical analysis was performed using Stata 14®IC (Stata Corp., College Station, Texas, USA). Chi-square tests were used to compare proportions of positivity related to categorical dependent variables and establish statistical significance. Association between seropositivity and independent variables such as social house conditions, dog living conditions, and socioeconomic level was evaluated. For all the independent variables, chi-square (χ2) and Fisher tests were used to assess associations and significance. In those significant values (p < 0.05), also the odds ratio with their 95% confidence interval (95%CI) was calculated.
The location of the houses hosting the dogs was georeferenced with their correspondent coordinates by the free mobile cell-GPS software application "Herramientas de GPS" v.3.1.0.5, developed by Virtual Maze® (www.virtualmaze.com). Seroprevalence was also presented by geographical information systems (GIS)-based maps. The georeferenced places were incorporated in the software Google Earth Pro®, and the layers of communes, corregiments and municipalities, provided by the Open Data Pereira Geographical Information System (https://mapas-pereira.opendata.arcgis.com/), in .kml files, were included to locate the coordinate’s points in the corresponding shapes. Microsoft Access® software was used to design the spatial database to import incidence rates by corregiments and communas at Pereira municipality to the GIS software. The Client GIS software open source was Kosmo Desktop 3.0 RC1® (SAIG S.L., Madrid, Spain). For access to geographic data required and sharing results with institutions, support was provided by the spatial data infrastructure for the department by the Regional Information System of the Coffee-Triangle ecoregion (SIR) as standardized and reported before [27–29]. The shapefiles of corregiments and communas (.shp) were linked to a database through spatial joined operation to produce digital maps of the seroprevalence for the study area. Data used for the GIS-based maps were derived from the geographical origin of the serosurvey at the different geographic levels [29].