Relation between reproduction performance and indicators of feed intake, fear and social stress in commercial herds with group-housed non-lactating sows

Group-housing of non-lactating sows is becoming increasingly widespread in commercial sow herds in European countries as a result of changed legislation. Group-housing may lead to individual variation in feed intake, stress and fear, which may impair the reproduction performance. However, whether the individual variation in feed intake and the level of stress and fear under commercial conditions is severe enough to impair the reproduction performance is not known. In a detailed farm study including 14 herds with different layouts the relations between various indicators of feed intake, stress and fear and reproduction performance were studied based on 553 focal sows. Twelve percent of all mated sows were re-mated and average litter size was 14.8 born piglets per litter. Chance of pregnancy (P < 0.05) and litter size (P = 0.08) correlated positive with back fat gain from weaning to 3 weeks after mating. Sows eating in less than 20% of all observations at feeding had significant higher risk of returning to oestrus compared to sows eating more frequently. Number of skin lesions correlated positively (P = 0.07) with interval from weaning to first mating. No relations between reproduction performance and lying behaviour, aggressions after mixing or the outcome of three fear tests were found. The results indicate that the unintended individual variation in feed intake in sows group-housed in commercial herds may be large enough to affect the reproduction performance. This calls for management initiatives to reduce unequal feed intake in group-housed sows.