REVIEWS: Compilation of the Scientific Literature Comparing Housing Systems for Gestating Sows and Gilts Using Measures of Physiology, Behavior, Performance, and Health1

The objective of this review was two-fold. First, a series of meta-analyses (analyses of treatment effects across studies) were performed on available data from scientific literature to determine whether sow behavior, performance, or physiology differed for sows in group pens or individual stalls. Second, research publications in areas of performance and health, physiology, and behavior of pregnant gilts and sows in studies that directly compared gestation sow housing systems were summarized. Common systems were stalls, tethers, and various types of group housing systems. Results of meta-analyses showed that the average levels of productivity, oral-nasal-facial behaviors (ONF), and blood cortisol were statistically similar for sows in group pens and stalls. For the review, in some studies, circulating cortisol concentrations were greater among gestating females kept in tethers compared with other systems; however, overall cortisol was not altered by housing system. Immune parameters were largely not influenced by housing system. Housing system did not alter heart rate. Gestation housing system may influence sow behavior including stereotypic ONF, postural locomotory, feeding behaviors, or social behaviors. Overall, total ONF behaviors were comparable between gestation sow housing systems. However, tethered and stalled sows exhibited more stereotypic ONF compared with sows in group or outdoor systems. Compared with group housing, individually confining sows during gestation resulted in postural and movement restrictions. Stall size and design can impact postural adjustments and inter-stall aggression of individually housed sows. Inconsistent performance and health results were found among sow housing studies. Sows in stalls consistently had equal or greater reproductive performance compared with sows in other systems. Farrowing rate for sows in individual stalls was equal to or superior to sows in other systems. Farrowing rate was clearly superior among sows in stalls compared with group systems, where dynamic social groups were employed. However, tethered sows may have reduced litter size and increased piglet birth weight. Sows in group housing systems, particularly electronic sow feeder (ESF) systems, had injury scores greater than sows in either stalls or tethers. Gestation housing system (individual vs group) may impact sow welfare in the farrowing area (using stalls or pens). In conclusion, although individual studies found significant housing system effects, subjected to the overall evidence from adequately designed studies meta-analyses revealed that gestation stalls (non-tethered) or well-managed pens generally (but not in all cases) produced similar states of welfare for pregnant gilts or sows in terms of physiology, behavior, performance, and health.