Development and validation of a protocol to identify and recruit participants into a large scale study on liver fluke in cattle

BACKGROUND: Liver fluke infection caused by the parasite Fasciola hepatica is a major cause of production losses to the cattle industry in the UK. To investigate farm-level risk factors for fluke infection, a randomised method to recruit an appropriate number of herds from a defined geographical area into the study was required. The approach and hurdles that were encountered in designing and implementing this study are described. The county of Shropshire, England, was selected for the study because of the variation between farms in exposure to fluke infection observed in an earlier study. RESULTS: From a sampling list of 569 holdings in Shropshire randomly drawn from the RADAR cattle population dataset, 396 (69.6%) holdings were successfully contacted by telephone and asked if they would be interested in taking part in the study. Of 296 farmers who agreed to receive information packs by post, 195 (65.9%) agreed to take part in the study. Over the period October 2014 – April 2015 visits were made to 100 dairy and 95 non-dairy herds. During the farm visits 40 faecal samples +/− bulk-tank milk samples were collected and a questionnaire administered. Composite faecal samples were analysed for the presence of F. hepatica eggs by sedimentation and bulk tank milk samples were tested with an antibody ELISA for F. hepatica. Forty-five (49%) of non-dairy herds were positive for liver fluke infection as determined by the finding of one or more fluke eggs, while 36 (36%) dairy herds had fluke positive faecal samples and 41 (41%) dairy herds were positive for F. hepatica antibody. Eighty-seven (45.8%) farmers said that they monitored their cattle for liver fluke infection and 118 (62.1%) reported that they used flukicide drugs in their cattle. CONCLUSIONS: Using a protocol of contacting farmers directly by telephone and subsequently sending information by post, 79% of the target sample size was successfully recruited into the study. A dataset of farm-specific information on possible risk factors for liver fluke infection and corresponding liver-fluke infection status was generated for the development of statistical models to identify risk factors for liver fluke infection at the farm-level.