Managing Grouped Sows

By Janeen L. Salak-Johnson and Stanley E. Curtis, Department of Animal Sciences, University of Illinois at Urbana-Champaign.
calendar icon 7 December 2007
clock icon 14 minute read


A negative public perception of the use of gestation stalls (crates) is a continual problem for the pork industry. Researchers and producers must continue to seek a solution this problem which, first and foremost, engenders and enhances the well-being of the sow. However, those solutions should not create a negative economic impact for the American pork industry. We must scientifically develop, validate, and implement alternative sow-keeping systems and sow-management strategies and tactics that are both practical and economically feasible but that effectively sustain sow well-being and preserve or increase the efficiency of producing pork. A well-defined, science-based assessment of the welfare of the pregnant sow is essential to the future of the entire livestock industry.

Unfortunately, there is a lack of data to establish welfare-friendly guidelines on how to effectively manage sows in groups during gestation. It is not as simple as removing individual stalls and forming group pens. Despite the notion held by some that keeping sows in groups is a relatively welfare-friendly practice because sows in groups can turn around and interact socially, keeping sows in groups can lead to its own welfare problems. These problems importantly include increased aggression at mixing and feeding, increased injuries and lesions, increased variability in body-condition score, among others. All of these factors can be influenced by feeding method, social status, floor space per animal, group size, genetics, and management procedures.

Despite the choices of group-keeping systems that are being tried, many of these have not been shown to improve the dry sow’s state of being. Thus, some of the many factors that should be considered when designing, implementing, and managing sows in groups during gestation are: group size, floor-space allowance, group composition (static vs. dynamic), diet type, method of feed delivery, genetics, sow temperament, and so forth. Most importantly, existing group-keeping systems differ in terms of feeding, group management, and floor type, but group keeping systems are more complex than these few factors.


Farm-animal welfare is a major public issue for US agriculture today. Finding and using scientifically sound approaches to assess the state of being of farm animals is crucial to the long-term sustainability of American agriculture. State and federal lawmakers are continually bombarded by humane activists who offer proposed legislation on animal welfare. Unfortunately, there is a dearth of scientific information as to how to defend several contemporary farm-animal husbandry practices.

Today’s typical sow-keeping system—the individual 0.61m wide × 2.12 m long stall (or crate)—is perhaps the most controversial welfare concern now facing the US pork industry. Following the European trend, which are based on public intuition as much as scientific evidence, the use of individual gestation stalls has been banned via public referenda in Florida and Arizona. New bills and referenda are set to be introduced and held in several other states, including California. Also, there have been recent announcements of intent to abandon sow stalls by Smithfield Foods, Maple Leaf Farms, McDonalds, and Burger King, among others. The national mass media have given much space and time to the goodness and rightness of abandoning the individual dry-sow stall in favor of keeping pregnant sows is groups. As a consequence, US pork producers face a dilemma: They now must decide how they will keep their sows in the future.

The term “sow keeping” refers to the accommodating and caring for breeding, gestating, and lactating sows. But the sow-keeping issue refers for the most part to “should pregnant sows be kept individually or in groups?” As for keeping dry sows in groups a variety of system alternatives have been proposed and are being tried. But neither a set of acceptable group systems nor an ideal group system has been identified. So far, results of industry and experiment-station research from around the world indicate that alternative keeping systems do not necessarily result in improving sow state of being. In fact, sows in individual stalls as compared to group pens have similar values across measurement perspectives (McGlone et al., 2004). No one system has been identified as being better than others based on current notions of sow welfare (Barnett et al., 2001; McGlone et al., 2004; Rhodes et al., 2005).

Group Sow-keeping Systems are not Jigsaw Puzzles

Group keeping systems are extremely complex, and difficult with which to come to grips, simply because there are so many factors that must be considered and integrated. Animal accommodations are not jigsaw puzzles (Curtis, 1995). Jigsaw puzzles once were a whole picture that was, in its original form, jig-sawed into integral pieces. The pieces fit together perfectly because at one time they were together. Not so with the contemporary group sow-keeping systems we are in the process of contriving, sometimes in Rube Goldberg fashion ( Rube_Goldberg_machine). The pieces, we are finding, as expected, do not always fit together perfectly or even acceptably, simply because they were not designed in the first place to fit together in the present arrangements. In the case of group-housing systems, the pieces include, among others:

  • indoors or outdoors, drylot or pasture, and insulated, mechanically ventilated frame structure or hoop structure;
  • floor type—slots, solid, or bedding;
  • group size and floor-space allowance (will vary with group size);
  • group management-static or dynamic (will vary with group size and floor-space allowance);
  • group feeding systems-drop, trickle, or electronic or individual feeding stalls;
  • sow-handling-labor requirement; herder safety while handling and managing manure;
  • manure-handling system—suspension, scrape semi-solids to auger or belt, or Bobcat®; labor requirement;
  • capital costs, maintenance costs, and operating costs;
  • behavioral management, disease management, and disease-preventive and treatment management; and
  • sow state of being, sow productivity, and sow productive lifetime.

As empirical observations and experimental results accumulate, it is clear that each piece has advantages and each piece has disadvantages in terms of productivity, profitability, and, yes, animal welfare. All of the current group-system alternatives allow sows, some freedom of movement, opportunity for social interaction, and individual choice among (available) microenvironments. But there are greater or lesser welfare problems associated with each of these Rube Goldberg group-keeping systems, too, including, stress due to aggression early on and social tension for the duration, variable body-condition score, and injury.

Moreover, clearly, more highly skilled and more attentive stockmanship is required to successfully manage sows in groups. And, even more importantly, when a group-keeping system fails to work well, some sows inevitably experience a poor state of being. The sow-welfare problems in a group system can be horrendous (McGlone, 2006; Curtis, 2007).

Keeping Sows Individually versus in Groups

Based on scientific evidence and empirical observations there are positive and negative aspects of all systems of sow accommodation that have been tried. The sow is a creature that is difficult to keep. As for reproduction, few studies have shown reproductive impairment in pregnant sows kept in stalls as against sows in groups. Although some studies have shown greater reproductive performance for sows in groups than in stalls (Bates et al., 2003; Lammers et al., 2007), some findings have been that reproductive performance is similar for sows either in groups or in stalls (Langendijk et al., 200; Harris et al., 2006; Salak-Johnson et al., 2007). Whereas others have found greater reproductive performance for sows that gestated in stalls as compared with those that gestated in groups (Barbari, 2000; Bates et al., 2003; Karlen et al., 2006).

How might the differences in results and conclusions have arisen?

Phase of Reproduction

It recently has been documented that welfare challenges to the sow change over time. Sows in groups during early gestation have increased incidence of scratches, higher estrus return, and higher cortisol concentration, whereas, sows in stalls have increased lameness during late gestation (Karlen et al., 2006). These findings are similar to those reported by Salak-Johnson et al. (2007), wherein lesion scores for group-housed sows were high at group-formation (mixing) time, plateau during mid-gestation, but increased again during late-gestation. Even the European Union regulations already recognize this as they provide for sows to be kept in individual accommodation during breeding and the first few weeks of gestation, when reproductive functions are most sensitive to deleterious effects of social strife and other stresses. This, of course, further complicates the rational design of the ideal sow-keeping system. Thus, we probably should accommodate sows differently during different stages of reproduction.


Stereotypic behavior patterns (e. g., repetitive, invariant, apparently functionless and purposeless bar- and trough-biting, sham-chewing, drinker-pressing, floor-nosing, head-weaving) often are used as a measure of sow welfare. Differences in stereotypic behaviors between sows in stalls and those in groups have been found (Vieuille-Thomas et al., 1995). Specifically, sows in groups spend less time interacting with bars and troughs (Karlen et al., 2006). Others, however, have reported no differences in stereotypic snout behaviors between sows kept in indoor stalls and those kept in outdoor groups (Dailey and Mc- Glone, 1997).

Although sow behavior in this regard may differ among housing systems, often it seems some component such as direction of bars or other design feature that is responsible for the behavior displayed by the sow (Dailey and McGlone, 1997; McGlone et al., 2004). In other words, sow repetitive behavior patterns seem to be partly driven by external stimuli and opportunities, and therefore can be designed out of a system. Only if it has been established that a behavior is internally driven and clearly associated with reduced welfare need we have ethical concerns about its display.

Also, conclusions that these behavior patterns are a bad sign in terms of sow welfare may have been premature and erroneous. Pregnant gilts either on pasture or on drylot or in stalls showed similar amounts of snout behaviors (Daily and McGlone, 1997). These behaviors did not start with the gestation stall, and they apparently are at least partly triggered internally. So, if anything, they ought to be accommodated and encouraged not neutralized and discouraged. Most stereotypic snout behaviors in sows come around the time of once-daily feeding of a restricted ration. So there also seems to be an external component in the motivation for such snout behaviors. So it now seems that the sow’s motivation to display repetitive snout behaviors is mixed—internal and external—and therefore it is complex and will not be easy to come to understand. Finally, there is good evidence that restricting a small ration of highconcentrate diet to a once-daily feeding causes hyperacidic indigestion in the sow. Those repetitive snout behaviors result in increasing saliva secretion, the buffering capacity of which neutralizes the heartburn. Thus, the bottom line is that the sow’s so-called snout stereotypies might not be functionless or purposeless after all.

Plusses and Drawbacks Around

There are both advantages and disadvantages associated with dry sows being kept either individually or in groups during gestation. No keeping system in either category has been identified, developed, or optimized such that it ultimately and invariably sustains the well-being of the pregnant sow.

Possible explanations are that

  • most studies have not taken a holistic approach by assessing multiple measures of state of being,
  • most studies have compared so-called “systems” but have not considered all of the features of those respective systems that can influence how a sow responds to and interacts with her environment, or
  • both of the above.

It is crucial that producers realize that there is more than the several approaches for each of the several design features, and thus many permutations of design features. Numerous feature combinations are possible, and not all of them work.

Management of Sows in Groups

How should we manage sows in groups? This is a million dollar question for which we do not yet have a definitive answer.

According to the AVMA website, sow keeping systems should attempt to

  • minimize aggression and competition among sows;
  • protect sows from detrimental effects associated with environmental extremes, particularly temperature’
  • reduce exposure to hazards that result in injury, pain, or disease;
  • provide every animal with daily access to appropriate feed and water;
  • facilitate observation of individual sow feed intake, respiratory rate, urination and defecation, and
  • allow sows to express most normal patterns of behavior.

Additional management strategies that should be considered are:

  • once a small group (4 or 6) of sows has been established, no more sows should be added to the group;
  • mix sows only if they are being kept in larger groups (20+);
  • if sows are mixed, mix in a new pen not a hometerritory pen; and
  • if an individual feeding system is not being used, sort and establish sow groups according to sow eating speed to reduce fighting and undernourishment (Temple Grandin, http://www.grandin. com/welfare/tips.sow.housing.html).

Group Housing Systems: Factors to Consider

Keeping sows in groups during gestation is not as simple as removing individual stalls, forming group pens. Despite the perception that keeping sows in groups is a relatively welfare-friendly practice as compared to using individual stalls simply because sows in groups can turn-around and interact socially, keeping sows in groups can lead to its own welfare problems. These mainly include increased aggression at mixing and feeding, increased injuries and lesions, increased variability in body-condition score, and so forth. All of these factors can be influenced by feeding method, social status, floor space per animal, group size, genetics, and management procedures. Thus, some of the many factors that should be considered when designing and implementing group keeping systems are group size, floor space allowance, group composition (static vs. dynamic), diet type and method of feed delivery, genetics, sow temperament, and so forth (Levis, 2007). Most importantly, group-keeping systems differ in terms of feeding, group management and floor type.

Feeding Systems and Management

What are our feeding options? Feeding systems can be first categorized as individual or group systems. Feeding grouped sows individually can be accomplished by using individual feeding stalls of some sort or electronic sow feeders (ESF).

Individual stalls. Individual feeding stalls will minimize aggression at feeding time, but some aggression still occurs. Also, it is not possible to feed individual rations because sows enter stalls somewhat at random at feeding time.

Electronic sow feeders. With ESF, each sow wears a transponder and eats by entering a feeding station, and each sow has her own daily feed allotment. Each feeding station can accommodate ~40 sows. A sow can consume her entire daily ration in one meal during a single daily visit or in several smaller meals spaced throughout the day.

Aggressive physical acts can occur while sows are waiting to enter the feeding station (Jensen et al., 2000; Anil et al., 2006). Vulva-biting is a welfare concern associated with ESF, but attempts have been made to reduce vulva biting. Appropriate feeder design and placement have reduced the incidence and severity of vulva biting (Levis, 2007).

Group-feeding options. There are three general approaches to group feeding: trough feeding, single-drop feeding, and floor feeding. These feeding systems have been used for small groups of ~6 as well groups of ≥100 sows.

Floor feeding, where the feed is simply dumped on the solid floor, engenders the most feeding-time aggression because sows compete for feed. Dominant sows overeat and become fat, submissive sows become thin. However, Salak-Johnson et al. (2007) found that floor feeding sows in groups of 5 at floor10 space allowances of 2.3 and 3.3 m2 had greater body weight and backfat depth than did sows in groups of 5 at 1.4 m2 per sow or individual stalls.

Single-drop feeding, in which a single drop of dry feed across a wide area is made once a day, also has been tried. This system leads to increased aggression because some sows eat faster than others and will consume their own allotment and then go looking for more, stealing the feed of others. Threats, attacks, and fights were evident in a study by Jansen et al. (2007) that involved 32 drop-feeders depositing feed into 3 troughs across the width of a pen for 50 sows.

A modification of single drop feeding is the trickle-feeding system wherein feed is continuously delivered at a slower rate over a period of time. There seems to be less aggression expressed in this system. Both the drop-feed and trickle–feed systems operate well with a wide range of group sizes, the most common group size being 5 to 6 sows.

Hulbert and McGlone (2006) reported that neither pen type (individual or group) and nor feeding system (drop or trickle) was associated with physiological stress responses in sows. Feeding system had no affect on social interaction or overall activity, although gilts in group pens equipped with drop feeding did show more snout behavior at 1200 h than did those in trickle-feeding pens. Over a 24-h period, however, snout behavior did not differ with feeding system. Neither did time spent feeding differ between drop- or trickle-fed sows. Treatment differences in behavior were due to available floor space not feeding system.

Despite the general perception that sow aggression is reduced with trickle feeding, it still is not perfect in this regard. Aggressive behavior and poor and variable body-condition score are still apparent in group systems with trickle feeding (personal observation, November, 2006).

Static versus Dynamic Group Management

The question remains: How can we minimize aggression and competition among group-kept sows? The most serious and injurious aggression and competition among group-kept sows occurs upon introduction of new sows to a group (mixing) and at feeding-time. Upon mixing, sows will fight with a high level of aggression, part of their natural attempt to form a new dominance hierarchy. The aim of sow management must be to enable the formation of the social order with as little stress and physical injury as possible.

November 2007
© 2000 - 2023 - Global Ag Media. All Rights Reserved | No part of this site may be reproduced without permission.