Spread of porcine circovirus associated disease (PCVAD) in Ontario (Canada) swine herds: Part II. Matched case-control study

BACKGROUND The emergence of porcine circovirus associated disease (PCVAD) was associated with high mortality in swine populations worldwide. Studies performed in different regions identified spatial, temporal, and spatio-temporal trends as factors contributing to patterns of the disease spread. Patterns consistent with spatial trend and spatio-temporal clustering were already identified in this dataset. On the basis of these results, we have further investigated the nature of local spread in this report. The primary objective of this study was to evaluate risk factors for incidence cases of reported PCVAD. RESULTS A time-matched case-control study was used as a study design approach, and conditional logistic regression as the analytical method. The main exposure of interest was local spread, which was defined as an unidentified mechanism of PCVAD spread between premises located within 3 kilometers of the Euclidean distance. Various modifications of variables indicative of local spread were also evaluated. The dataset contained 278 swine herds from Ontario originally sampled either from diagnostic laboratory submissions or directly from the target population. A PCVAD case was defined on the basis of the producer's recall. Existence of apparent local spread over the entire study period was confirmed (OR = 2.26, 95% CI: 1.06, 4.83), and was further identified to be time-varying in nature - herds experiencing outbreaks in the later part of the epidemic were more likely than control herds to be exposed to neighboring herds experiencing recent PCVAD outbreaks. More importantly, the pattern of local spread was driven by concurrent occurrence of PCVAD on premises under the same ownership (OREXACTwithin ownership = 25.6, 95% CI: 3.4, +inf; OREXACToutside ownership = 1.3, 95% CI: 0.45, 3.3). Other significant factors included PRRSv status of a herd (OREXACT = 1.9, 95% CI: 1.0, 3.9), after adjusting for geographical location by including the binary effect of the easting coordinate (Easting > 600 km = 1; OREXACT = 1.8, 95% CI: 0.5, 5.6). CONCLUSIONS These results preclude any conclusion regarding the existence of a mechanism of local spread through airborne transmission or indirectly through contaminated fomites or vectors, as simultaneous emergence of PCVAD could also be a result of concurrent change in contributing factors due to other mechanisms within ownerships.