Comparison of challenge models for determining the colonization dose of Campylobacter jejuni in broiler chicks.

Coprophagous activity is normal among broiler chickens. The purpose of this study was to compare an individually housed chick model (where bird-to-bird coprophagia was prevented) to a group-housed chick model (where bird-to-bird coprophagia was allowed) for determining estimates of the number of Campylobacter jejuni RM1221 necessary to colonize 50% of broiler chicks inoculated (colonization dose 50% or CD(50)). Campylobacter jejuni RM1221 was orally administered in measured doses to newly hatched chicks. The chicks were housed either individually in cages designed to minimize coprophagous activity or in isolation units containing groups of birds where coprophagia was allowed. The birds were killed and analyzed for Campylobacter in the ceca on d 7 postinoculation. The CD(50) was calculated, and results from the 2 models were compared. Elimination of transmission of Campylobacter, through coprophagia or other means, led to a more clear determination of the estimated CD(50) of about 524 cfu of C. jejuni RM1221 as demonstrated in the individually housed chick model. Bayesian inference based on the beta-Poisson statistical modeling procedures were found to be superior to standard single-hit dose-response modeling for estimation of the CD(50). This study demonstrated that the individual bird challenge model is superior to the group challenge model for trials designed to determine colonization dose.