In study 1 we evaluated 49 swine workers and 79 non-exposed controls for antibodies to swine influenza viruses. Multivariate modeling revealed that workers who seldom used gloves (OR=30.3) or who smoked (OR=18.7) were at highest risk of having evidence of previous H1N1 swine virus. These findings have value in pandemic influenza planning.

In study 2 evaluate the within-herd rate of transmission of porcine reproductive and respiratory syndrome virus (PRRSV) and risk factors affecting the rate of seroconversion in growing pigs post-weaning. Twenty six barns were selected to participate and 20 pigs per barn (for a total of 520 pigs) were housed together in a pen, tagged and tested for PRRSV antibodies for approximately 24 weeks, sampling every two weeks from the time of enrollment in the nursery or wean-to-finish building until marketed. At the start of the project, these pigs were also tested for PRSSV circulation using PCR. A total of 101 (22%) pigs were identified to be viremic for PRRSV by PCR at enrollment. The cumulative mean incidence density rate for PRRSV seroconversion for all pigs was 2.57 per 100 pig days. This means that on average, it took pigs 39.0 days (95% CI, 35.3 to 43.2) from enrollment to seroconvert to PRRSV as detected by ELISA. On average it would take 3.6 bleeding periods (between 7 and 8 weeks) from when the first pig in a group was identified to have seroconverted until the last pig in the group seroconverted. Cox proportional hazard multivariate modeling identified enrollment PRRSV PCR positive result (HR=34.0, 95%CI 8.06 to 143.44) and number of PCR positive pigs in the cohort at enrollment (HR=1.66, 95%CI 1.23 to 2.23) as significant risk factors for decreased time-to-seroconversion. PRRSV is not highly contagious. The smallest pigs in the group are not more likely to be PCR positive or seroconvert to PRRSV sooner than the largest pigs in the group. PRRSV ELISA S/P values at weaning are not a predictor of viremia or time-to-seroconversion. Serologic PRRSV homogeneity is dynamic and difficult to achieve even in a small group of pigs with constant exposure to each other.

In study 3 we evaluate if oral fluids collected by barn personnel could be used as a method of surveillance based on PCR testing. Approximately 12,150 pigs in 10 wean-to-finish barns on 10 different farms were monitored for the circulation of porcine circovirus type 2 (PCV2), porcine reproductive and respiratory syndrome virus (PRRSV), influenza A virus (IAV), and Torque teno virus genogroups 1 (TTV1) and 2 (TTV2) using oral fluid specimens. Oral fluid samples were collected from 6 pens at each site starting at the time of pig placement (~3 weeks of age) and continuing thereafter at 2-week intervals for a period of 18 weeks. Data were analyzed both on a pen basis as well as a barn basis. Overall, 508 (85%) samples were PCR positive for PCV2, 73 (12%) for PRRSV, 46 (8%) for IAV, 483 (81%) for TTV2, and 155 (26%) TTV1 during the study period. The estimated arithmetic means of the quantitative PCR-positive oral fluids for PCV2, PRRSV, and IAV were 1 x 104.62, 1 x 104.97, and 1 x 105.49 per ml. With a single exception, all barns were positive for PCV2 and TTV2 at every sampling point in the study. Pathogen circulation varied among barns, particularly for IAV and PRRSV. Chi squared analysis of the cumulative distribution of pen level agent combinations between all 10 barns indicated they were statistically different from each other. The most commonly observed patterns were PCV2+TTV2 (239, 40%), PCV2+TTV1+TTV2 (88, 15%), and PCV2 alone (66, 11%). Although weekly mortality was the only herd health parameter available for evaluation, the Cox proportional hazard survival analyses showed that the issue of herd health is complex and will require thoughtful experimental design and analysis. This "proof-of-concept" project showed that a variety of pathogens circulate both intermittently and continuously in pig populations and demonstrated that barn health is highly variable, even among barns in the same production system. Oral fluid sampling and testing may ultimately meet the objective of being a more convenient and cost effective means for pig barn surveillance.

The last chapter of this dissertation focuses on biological risk management (BRM). The term BRM has been specifically selected rather than simply biosecurity. Biosecurity implies a yes or no approach to things whereas BRM takes the approach that it is all a matter of managing risk as most risks cannot be totally eliminated. For many thesis and dissertations, the literature review serves as a way to be "introduced" into a topic. It lays the foundation for the studies to follow. In this case, my literature review work will serve as a guide on how to work to prevent disease transmission (animal-to-animal as well as animal-to-human and human-to-animal) in the swine industry. It is a comprehensive document that tries to move away from specific pathogens and focus on disease transmission (focus of this dissertation) from the perspective of routes. This is important because once again, many times we are focused on only one pathogen and ignore many others. This is especially true when looking at a new or re-emerging disease as well as a foreign animal disease (FAD). Focusing on routes of transmission is the core to all BRM practices.