Abstract
Data on husbandry practices, performance, disease and drug use were collected during a cross-sectional survey of 89 poultry meat farms in England and Wales to provide information on possible risk factors for the occurrence of fluoroquinolone (FQ) resistant bacteria. Faeces samples were used to classify farms as ‘affected’ or ‘not affected’ by FQ-resistant E. coli or Campylobacter spp. Risk factor analysis identified the use of FQ on the farms as having by far the strongest association, among the factors considered, with the occurrence of FQ-resistant bacteria. Resistant E. coli and/or Campylobacter spp. were found on 86% of the farms with a history of FQ use. However, a substantial proportion of farms with no history of FQ use also yielded FQ-resistant organisms, suggesting that resistant bacteria may transfer between farms. Further analysis suggested that for Campylobacter spp., onfarm hygiene, cleaning and disinfection between batches of birds and wildlife control were of most significance. By contrast, for E. coli biosecurity from external contamination was of particular importance, although the modelling indicated that other factors were likely to be involved. Detailed studies on a small number of sites showed that FQ-resistant E. coli can survive routine cleaning and disinfection. It appears difficult to avoid the occurrence of resistant bacteria when FQ are used on a farm, but the present findings provide evidence to support recommendations to reduce the substantial risk of the incidental acquisition of such resistance by farms where FQ are not used.