Housing of farrowing and lactating sows in farrowing crates can cause a number of welfare problems. Recently, alternative farrowing pens have been developed which allow the sow to turn around and move freely. In this experiment a conventional farrowing crate (C; 5.0 m2; the sow is crated permanently), the Trapez pen (T; 6.7 m2; with a crate opened two days after farrowing) and a modified FAT2 pen (F; 6.7 m2; loose-housing system with nest and dunging area), were compared. The behaviour of sows was recorded from 6 hours ante partum to 24 hours post partum and analysed for duration of birth, number of posture changes, activities indicating nest building and piglet crushing. Sows were inspected for skin lesions at days 5 and 23, post farrowing. There was a tendency for longer birth duration and restlessness during farrowing when sows were crated. Sows in F and T crushed more piglets than sows in C but only few sows were concerned. In C more sows had severe injuries on the udder and on the limbs than in F and T. It can be concluded that the health and behaviour of farrowing and lactating sows are negatively affected when being housed in conventional pens with crates and slatted flooring. Further development is required to minimise piglet crushing in loose-housing systems.

Piglet crushing remains a major problem in pig production. Reduced crushing might be obtained through genetic selection on sow behavioural traits. The aim of the study was to assess the relationship between behavioural responses at 6 months of age, around farrowing, and sows’ reproductive performance including crushing levels. At 6 months of age, behavioural responses of 75 nulliparous sows were observed both during behavioural tests to human presence and to the presence of a novel object in their home pen, and their responses when placed in a weighing device. At first farrowing, nervousness of the sows was observed when placed in the farrowing crate 1 week before and the day of farrowing, as well as their fear responses when approached by a human from behind or at the front of the farrowing crate. At 6 months of age, escape from a human tended to be correlated with the reactivity in the weighing device (rs = 0.21, P = 0.09). Around first farrowing, the withdrawal reaction when a human approached at the front was correlated with the fear response when approached from behind and the nervousness of the sow in the crate (rs = 0.29, P < 0.05; rs = 0.37, P < 0.01). The fear response when approached from behind was correlated with nervousness in the crate and around farrowing (rs = 0.70, P < 0.001; rs = 0.25, P < 0.05), and nervousness in the crate was significantly correlated with the nervousness around farrowing (rs = 0.34, P < 0.01). The escape from a human at 6 months was correlated with withdrawal when approached from the front before farrowing (rs = 0.38, P < 0.01) and with nervousness of the sow in the crate (rs = 0.24, P < 0.05). The number of piglets crushed at first farrowing was correlated with the latency to approach a novel object at 6 months and nervousness around farrowing (rs = −0.27, P < 0.05; rs = 0.28, P < 0.05), and tended to be correlated with the escape behaviour from human at 6 months and withdrawal away from human presence before farrowing (rs = 0.21, P = 0.09; rs = 0.22, P = 0.08). These results suggest that behavioural responses to humans and during management practices of nulliparous sows at 6 months of age are, to some extent, related with their behaviour around farrowing and crushing levels of piglets at farrowing.