Evaluation of selected antibiotic residue screening tests for milk from individual cows and examination of factors that affect the probability of false-positive outcomes

Total composite milk samples from 131 cows in one herd were analyzed. Eight beta-lactam residue screening tests were evaluated for performance using milk from individual cows and factors that affect the rate of false-positive outcomes were determined. Cows were not treated with an antibiotic for at least 30 d prior to sampling. Tests evaluated were Delvotest(R) P, Charm Cowside, Charm Farm, Penzyme(R), Valio T101, LacTek(TM), CITE Probe(R), and Charm Bacillus stearothermophilus disk assay. Cows averaged 155 d of lactation. Milk production at the time of sampling ranged from 3.6 to 26.3 kg per milking per cow. The somatic cell count of milk averaged 243 x 10(3)/ml and ranged from 8.5 x 10(3)/ml to 3437 x 10(3)/ml. Total viable bacteria counts averaged 197.8 x 10(3)/ml. Total coliform counts ranged from 0 to 205/ml. Selectivity rates (rate of truly negative samples that were found to be negative by the assay) were greater than 90% for all tests except the CITE Probe(R) test. Use of logistic regression showed that an increase in colony-forming units was associated with a decrease in the probability of a false-positive outcome for the CITE Probe(R) test. Milk production, coliform counts, and parity each affected the probabilities of positive outcomes for different tests. Except for one test, selectivity rates of the beta-lactam residue screening tests for milk from individual cows was greater than 0.9.