Recent announcements by a number of production companies in the USA and Canada regarding group sow housing have many producers wondering whether this signals the beginning of the end for sow stalls.  The answer to this question is, to a large degree, a matter of conjecture bearing in mind that there seems to be no big push by Canadian food retailers to go down this path or indeed any significant political pressure on sow stalls.  However, lobby groups are actively pursuing the sow stall agenda and, at some point in the future, sentiment on the issue will change, maybe quite quickly.  It’s therefore a good idea for producers to start making themselves aware of the practical considerations relating to group housing and even considering the unthinkable – what would you do if sow stalls were to be phased out over a relatively short time period?  And perhaps producers and others in the industry should be looking at what has happened in Europe over the last 20 years to learn from experience there.

In the UK, the short time-scale (eight years) that the industry had to convert to group housing was a major cause of both practical and financial problems.  For example, a significant number of producers converted sow stall houses to group pens to reduce the cost of meeting the legislation and, in many cases, this resulted in an unsuitable solution for the pigs, the operators and the unit’s productivity.  If the North American industry moves in an orderly way towards group housing over a relatively long period of time, then many errors can be avoided. In fact, assuming a loose housing system for sows is introduced as part of a farm’s normal building replacement program, there is no reason why it should not be as productive and cost-effective as a stall system, while bringing benefits from a consumer perception viewpoint.

Performance not an issue

Many comparisons of group housing and sows stalls have been carried out over the last 25 years and the majority show little differences in terms of the major breeding herd parameters such as litter size, farrowing rate, litters/sow/year and pigs/sow/year.  If anything, group systems show a number of advantages over stalls such as a shorter weaning to oestrus interval, lower stillbirth rate and better sow longevity. Work carried out at Britain’s National Agricultural Centre, where one of the country’s first electronic sow feeding (ESF) systems was built in 1986, has compared performance in this ESF system to that for sows housed in groups of six (sow yards) and individual sow stalls (Table 1). 

Results from group housing compared to stalls do vary somewhat, depending on the type of group system, whether sows in groups are on slatted or bedded floors and with the timing of mixing into groups relative to the day of weaning. However, in Europe, after 20 years of comparison, it is widely accepted that there are no significant performance advantages of either individual or group housing.  In practice, any differences noted by producers tend to be due to variations in the design and construction of group systems and the way they are managed.

Table 1:  Performance of three sow housing systems

                                                              E.S.F. yard      Sow yards                   Sow stalls

No. of sows recorded                                    482                             559                              331

Av. wean to first service (days)                       6.5                              5.7                               5.7

Returns to first service (%)                            11.8                            14.3                             12.1

Farrowing rate (%)                                        83.8                            80.9                             83.7

Av. nos. born alive/litter                                 11.2                            10.8                             11.0

Av. nos. born dead/litter                                 0.6                              0.7                               0.8

Av. nos. mummified/litter                                0.3                              0.2                               0.3

Pre-weaning mortality (%)                             12.4                            10.4                             11.3

Av. nos. reared/litter                                       9.4                              9.9                               9.4       

From: NAC Pig Unit, UK, 1991

Re-engineering facilities may compromise efficiency

For an existing production system, one of the most important issues is to decide how facilities can be re-engineered in order to convert to group housing.  However, direct conversion from stalls to groups is fraught with practical difficulties.  The first and most important aspect is that, typically, a sow stall building has an average of about 1.86 – 2.14m² (20 – 23ft²) per sow, including access areas, whereas group housing systems require in the range 2.3 – 2.6 m² (25 – 28 ft²) per sow of total space.  That means a reduction in sow numbers or a compromise on space for the sow, either of which will have negative economic consequences. Some systems in the USA have opted for a lower than optimum space allowance and will reap the consequences in lost performance and higher sow mortality and morbidity.  The second disadvantage of conversion is that the layout of most sow stall buildings does not lend itself well to group penning, particularly due to the positions of the slatted areas.  The alternative is to go ahead anyway, in which case pen cleanliness is likely to be a problem, or to replace all the floors, which is very expensive. It is often more practical and cost-effective in the long term to build a new sow barn than to carry out a conversion.

During the 1990’s, I helped many clients in the UK go through the process of deciding how to re-engineer their units.  In most cases we opted to utilize the sow stall barn for additional farrowing space, nursery pens or finishing rooms, depending on the requirements determined by pig flow through the new system.  The main objective here was to look for opportunities to increase efficiency and output at lowest cost because the investment in new group housing would bring very little financial benefit on its own.  For example, utilizing a sow stall barn for additional finishing space in order to increase carcass weight was always a very cost-effective option. Of course, utilizing a sow stall barn for nursery or finishing is only possible on a farrow-to-finish site and in multi-site systems sow stall barns either have to be converted, probably with some additional space added, or replaced.  Building additional space will require the appropriate development permits and, in some situations, could be a limiting factor.  However, in the UK the authorities were quite understanding, bearing in mind that the change to a group system was a legal requirement.

Flooring type influences choice of system

One of the key decisions about group housing is whether to use bedding or not. In Europe a majority of systems have straw-bedded floors, with either a slatted or solid dunging area, although un-bedded systems are also used.  The floor type and whether bedding is used also influences choice of group system and certain aspects of management.  However, the use of bedding may not be feasible in many North American systems because they currently handle manure as a liquid and are not likely to want to operate two manure disposal systems.  Despite this, I would urge people to consider the advantages and disadvantages of using some (not necessarily large amounts) of bedding, compared with slatted floors.   One aspect of performance that is significantly improved where solid floors and bedding are used is sow mortality and culling rates.  Table 2 shows that average sow death losses in Britain, where the majority of gestating sows are housed on straw, are about half of those in the USA and Canada, where most sows are kept in slatted stalls. Culling rates are also lower.

Table 2:  Sow death rates for 2005

                                       Britain            Canada              USA

Average                              4.7                   8.1                   8.9

Top 10%                           N/A                   5.5                   4.8

Worst 10%                         6.6                 13.2                 13.2

   Source:  PigChamp Benchmarking Database/MLC Pig Yearbook 

Systems that combine the use of a bedded lying area with a slatted dunging area are sometimes used in Europe.  The slatted area may be raised above the level of the bedded area to retain straw and prevent it being dragged onto the slats by the sows.  Alternatively, systems with a solid floored lying area and slatted dunging area may use sufficient bedding to provide some rooting material and gut-fill for sows, without creating the need for a solid manure handling system.  Such compromises are worth considering.

Despite the benefits of bedding, I suspect that slatted floors will be most widely used in North America.  If that is the case, close attention must be paid to the quality of slats used in group housing systems, to minimize the amount of injury to feet and legs. While they are more expensive to manufacture, slats with rounded, moulded edges will result in less injury than those with the standard ground-off edge and will therefore be most cost-effective in the long term.  Slat width and gap is also important and sows are more comfortable on a wide slat that allows them to easily stand with their whole foot on the solid part of the slat, rather than having one half of a hoof down the edge of the gap. Slat widths of 125mm (5”) with an 19-20mm (3/4”+) gap are ideal.

Time of mixing affects group management

Sow groups should be formed at weaning, immediately after service or at about 28-30 days into gestation.  The big advantage of keeping sows in stalls for the first part of gestation is that it makes management of breeding, checking for returns and scanning so much easier.  Research in several countries suggests that, overall, there is no performance advantages either way, although some trials showed a slight advantage where sows are housed in stalls for 28 days.  Table 3 shows the results of Danish trials on two farms where sows were grouped either at weaning or on day 28 after service and there were no statistical differences in performance.

Table 3: Results from two herds with sows in large groups and ESF

                                                                     Herd 1                                       Herd 2

Time of entry                                   After           4 weeks                     After           4 weeks

                                                         service      after service               service      after service

No. of litters                                         281                 299                          361                 309

Total born* per litter                            12.0                12.0                         12.8                12.7

Farrowing rate (%)                                87                   90                            86                   83

* Liveborn + stillborn

From: National Committee for Pig Production, Annual Report, 1998, Denmark

Another very important reason for delaying grouping until around 28 days is that it maximizes space utilization.  Where sows are mixed either at weaning or after breeding, a decision has to be made about what to do with sows that return or are found non-pregnant.  If they are left in the original group, this makes management much more difficult, but if they are removed and mixed with a contemporary group, the pen is then under-utilized.  Not only that but re-mixing of sows at any stage is undesirable.  After the time that sows have been scanned, very few dropouts should occur and groups can remain stable.  Large, dynamic groups make more efficient use of space than systems with smaller groups because they are more flexible and involve regular mixing of sows anyway.

Good pig skills are key to success

One thing that is very clear from my 25-year involvement with group housing systems is that a higher level of ability is required in the managers and technicians operating the system if they are to realize excellent results.  There is no doubt that it is harder to identify individual animals, recognize sick sows and spot abnormalities than it is in stall systems.  Operators must have an excellent knowledge of pig behaviour in order to be effective.  Unless you are confident that staff have the abilities required, group housing should be avoided.

Hospital pens essential

In any group system it is inevitable that a few sows will be sick, injured or become disadvantaged for various reasons.  In some cases those sows may be bullied by others in the group. In a stall system, such animals cannot be bullied by other sows and can be given individual treatment, whereas in a group system, they must be taken out of the group.  Consequently a number of hospital pens are required, where sows can be housed individually or in small groups.  These should preferably have solid floors and straw bedding because a significant proportion of the sows removed will have foot and leg problems.

Learn from existing systems

There is an enormous amount of experience and information about the design, construction and operation of group housing systems, especially from Europe.  Of course not all of it is applicable to North American conditions but there is no point either re-inventing the wheel or making the same mistakes that were made in Europe 20 years ago, although there is no doubt this will happen to a degree.  Producers thinking about group housing should make sure they gather as much information as possible because there are so many aspects to consider. 

Photo captions:

UK-ESF-1 – Group housing systems in the UK have resulted in equally good performance to sow stalls

Slatted ESF-1 – In slatted systems the slats should be wide enough to fully support the sows’  feet and have rounded edges