Objective. The specific objective of this thesis was to address the U.S. pork industry's need for evaluation of alternative ante-mortem diagnostic samples for PRRSV. To accomplish this objective, three trials were untaken with the following goals: (1) Develop a sampling protocol for the blood swab method. (2) Analyze the diagnostic accuracies of the blood swab, the capillary tube and the jugular sampling methods on two standard PRRS diagnostic tests: quantitative real-time reverse transcriptase polymerase chain reaction (qRT-PCR) and ELISA in known positive and negative animals. (3) Analyze the diagnostic accuracy of the blood swab method for ELISA testing in commercial finishing swine.;Trial one. Trial 1 showed that under laboratory conditions, the Fisherbrand swabs absorbed a significantly higher volume of blood than the Puritan swabs. Trial 2 showed that when the 20g x 1/2&inches; needle was used, significantly more blood was absorbed than when the 25g x 5/8&inches; needle was used regardless of the swab type. Trial 3 confirmed that when the 20g x 1/2&inches; needle was used, the swab absorbed significantly more blood. Trial two. For qRT-PCR testing, the sensitivity and specificity for all sampling methods ranged from 93%--100% for weeks 1--3 PI. Results of ELISA testing depended on cut-off selection. Optimization of ELISA S/P cut-off points for swab sample data was substantially lower (S/P ratio of 0.08 +/- 0.05) than the industry standard (S/P ratio of 0.4). Trial three. The sensitivity and specificity of the swab sampling method was 22.3% (95% probability intervals = 16.0%, 29.2%) and 94.3% (80.1%, 99.8%) respectively when an S/P ratio cut-off of 0.4 was used. The sensitivity and specificity were 94% (89.8%, 97.2%) and 93.5% (77.8%, 99.8%) respectively when an S/P ratio cut-off of 0.08 was used.;Implications. (1) Fisherbrand swabs absorb a significantly greater blood volume (167mul) then Puritan swabs (142mul) under ideal sampling conditions. (2) Under field conditions, the Fisherbrand swabs absorbed numerically less blood (118mul) then under laboratory conditions (167mul). (3) A Fisherbrand swab and a 20g x 1/2&feet; needle combination would be the best diagnostic sample for sows and finisher pigs when collection time is less than 15 seconds. (4) The capillary tube method suffers from neither inadequate volume nor differences in sample type collected (as compared to the blood swab method). This study indicated that the capillary sampling method can be used with ELISA and real-time qRT-PCR testing with diagnostic accuracy equal to the jugular sampling method. The total expense for the capillary sampling method was {dollar}1.44/sample ({dollar}0.94/tube in product and {dollar}0.50/sample in labor cost (calculated for 3 minutes at {dollar}10/hr)). In comparison, the total cost of the jugular sample was {dollar}0.91/sample. (5) Early diagnosis (weeks 1--3 PI) of PRRSV infected nursery pigs using real-time qRT-PCR under study conditions can be equivalently accomplished using the capillary, swab, or jugular sampling methods. (6) No change in cut-off values for qRT-PCR data dichotomization is necessary for data obtained via any of the sampling methods. (7) The diagnostic accuracy of PRRSV ELISA was poor for the swab sampling method when an S/P ration cut-off of 0.4 was utilized (sensitivity ranged from 20%--55.6% over weeks 2--7 PI). (8) When optimal cut-off values are employed, as determined by AUC analysis, all sampling methods are capable of achieving very high diagnostic accuracy on PRRSV ELISA testing. These cut-off values may not be clinically useful. (9) Under commercial (field) conditions, the sensitivity of the swab sampling method was low (22.3%) for ELISA results dichotomized at an S/P ratio of 0.4. (10) The sensitivity of the swab method improved when a lower S/P cut-off was used (94%) indicating this method may have application in routine ELISA diagnostic monitoring programs. (11) In comparison to the jugular sampling method, the sensitivity and specificity of the swab method is lower when used in commercial settings; this will result in more false negative and false positive test results. (12) Under the assumptions of the partial budget, the jugular sampling method would cost a 1000-head sow operation {dollar}0.15/pig produced while the swab sampling method would cost {dollar}0.36/pig produced. (Abstract shortened by UMI.)