The importance of on-farm biosecurity: Sero-prevalence and risk factors of bacterial and viral pathogens in smallholder pig systems in Uganda

Background The productivity of pigs in smallholder systems is affected by high disease burden, most of which might not be obvious, with their epidemiology and impact being poorly understood. This study estimated the seroprevalence and identified the risk factors of a range of bacterial and viral pathogens of potential economic and public health importance in domestic pigs in Uganda. A total of 522 clinically healthy pigs were randomly selected from 276 pig farms in Masaka (142) and Lira (134) districts of Uganda in 2015. Results Overall the highest animal prevalence was found for Streptococcus suis 73.0% (CI95: 67.0–78.3) in Lira and 68.2% (CI95: 62.7–73.4) in Masaka; followed by Porcine circovirus type 2 with 50.8% (CI95: 44.5–57.2) in Lira and 40.7% (CI95: 35.2–46.5) in Masaka and Actinobacillus pleuro-pneumoniae, 25.6% (CI95: 20.4–31.6) in Lira and 20.5% (CI95: 16.2–25.6) in Masaka. Mycoplasma hyopneumonia prevalence was 20.9% (CI95: 16.2–26.6) in Lira and 10.1% (CI95: 7.1–14.1) in Masaka, while Porcine parvovirus was 6.2% (CI95: 4.0–9.7) in Masaka and 3.4% (CI95: 1.7–6.6) in Lira. Less common pathogens were Influenza A, 8.5% (CI95: 5.6–12.8) in Lira and 2.0% (CI95: 0.9–4.5) in Masaka and Porcine Reproductive and Respiratory Syndrome Virus, 1.7% (CI95: 0.7–4.3) in Lira and 1.3% (CI95: 0.5–3.5) in Masaka. Even less common was Rotavirus A with 0.8% (CI95: 0.2–3.0) in Lira and 0.7% (CI95: 0.2–2.5) in Masaka; the same was for Aujeszky virus with 0.4% (CI95: 0.7–2.4) in Lira and 0.0% (CI95: 0.0-0.1) in Masaka. Co-infection with two pathogens was common and there was a significant association of M. hyo and PCV2 co-occurrence (p = 0.016). Multivariate analysis showed that for S. suis the use of disinfectant reduced odds of sero-positivitey (OR = 0.15; p = 0.017) and pigs less than 6 months were more likely to be infected than older pigs (OR = 3.35; p = 0.047). For M. hyo, crossbred pigs had higher odd of infection compared to local breeds (OR = 1.59; p < 0.001). Conclusions The studied pathogens have high prevalences in smallholder pig production systems and might be silent killers, thus affecting productivity and there is a possibility that some pathogens could spread to humans. Given the limited knowledge of veterinary workers and the poor diagnostic capacities and capabilities in these systems, the diseases are potentially usually under-diagnosed. These findings constitute baseline data to measure the impact of future interventions aiming to reduce disease burden in the pig production systems in Uganda.