The objectives of this study were: 1) determine the frequency with which uninfected and infected cows "switch" ELISA results when tested on multiple sequentially-obtained samples, 2) determine whether fecal culture or ELISA becomes positive first in infected cattle and 3) compare change in ELISA OD values over time for infected and uninfected cattle to determine if a rise in OD value (but below the "cutoff" ) could provide an early indication of infection.Approximately 3000 serum samples from 282 infected cows and 263 uninfected herdmates were analyzed using a commercially available ELISA. Cows were considered infected if any fecal culture result was positive, and was considered uninfected if all fecal culture results were negative. Cows had fecal and serum samples collected every 6 to 12 months. Number of samples per cow ranged from 2 to 15. Of the infected cows, 38% had at least one ELISA positive serum sample. Of these, 50% were fecal culture -positive at least one test prior to becoming ELISA-positive (median of 1.0 year earlier, range=0.5 to 3 years), 38% were positive for the first time on fecal culture and serum ELISA concurrently, and 12% were ELISA-positive at least one test prior to being detected by fecal culture (median 0.5 year earlier, range=0.5 to 2 years). Of the infected cows that remained in the herd after a positive ELISA result, 75% remained ELISA-positive on all subsequent ELISA tests.For cows that were seronegative, the difference between ELISA OD values on serial test dates was compared for infected and uninfected cows. There was no difference between groups, and detecting a rise in ELISA OD value below the cutoff was not useful to discriminate between infected and uninfected seronegative cows.