Background
Antimicrobial use (AMU) in livestock is a debated topic, mainly due to the risk of associated development of antimicrobial resistance. There is focus on reducing AMU in the Danish pig production, which accounts for the largest proportion of AMU for animals in Denmark. Due to special restriction on AMU in organic pig production, the AMU in organic pig production is lower than in conventional pig production. There is concern that reduced AMU could jeopardize animal health and welfare, if it reflects insufficient treatment of sick animals, which might be reflected by the prevalence and types of lesions found at meat inspection. However, little is known about the associations between AMU and meat inspection findings in pigs from organic farms. Furthermore, excess amount of antimicrobial product after a treatment cannot be re-prescribed in organic pig herds. The initial prescription is recorded in the national database VetStat, but the unused amount is not deducted leading to uncertainty when reporting AMU. The objectives of this study were to 1) describe AMU patterns based on prescription data for organic pig production and compare with those of the conventional pig production for year 2016, 2) study the associations between herd-level AMU prescription data and meat inspection data for organic pig herds and 3) validate herd-level AMU prescription data in VetStat against treatment records collected on-farm in organic Danish pig herds.

Results and conclusions
Gastrointestinal indications account for the largest proportion of AMU in both organic pig herds (65 and 54% of treatment doses for weaners and finishers, respectively) and conventional pig herds (80 and 68% of doses for weaners and finishers, respectively). A larger proportion is prescribed for respiratory indications in organic than conventional weaners and arthropathic indications in finishers. No associations between AMU and meat inspection data were found. This needs further investigation as the prevalence of lesions at slaughter was slightly (non-significantly) higher in herds with no registered AMU than with AMU prescriptions. Only 8 out of 31 herds had recorded their AMU sufficiently detailed to compare, and using VetStat as a proxy for AMU led to 9–88% overestimation of the actual use in 7 out of these 8 herds and 120% underestimation in one herd.