The stall in sow housing

FOR FOURTH-GENERATION swine producer John Kellogg, sow gestation stalls are the latest in a line of housing systems designed to address sow welfare. It's a system he feels is superior, because it protects the sows from aggression and allows his staff to monitor the sows individually. But it's also a system Kellogg knows the public isn't entirely comfortable with. That's one of the reasons he makes a special effort to reach out to his neighbors and schools in his increasingly suburban community about 45 minutes southwest of Chicago.

"Some people do some things just because they have a passion for it," Kellogg said, "and one of mine is to help my urban neighbors understand what happens in agriculture these days."

As a conscientious swine producer using a controversial housing system, Kellogg was at the center of a spirited—and at times, heated—debate at the 2004 AVMA Animal Welfare Forum: "Sow Housing and Welfare," on Nov. 5, 2004, in Hoffman Estates, Ill. He joined four researchers and an ethicist in discussing the scientific, ethical, and practical issues surrounding the housing of pregnant sows with the more than 130 forum attendees. Summaries of these presentations will be published in JAVMA this spring.

The issue of whether it is humane to house pregnant sows in gestation stalls, which are too small for the animal to turn around in, has become one of the most contentious issues facing not only the profession, but also animal welfare scientists, producers, and consumers. Yet, speakers at the forum found some common ground.

Some shared principles emerged from the six presentations at the forum. These included the need to address public concerns about the welfare of pregnant sows, the importance of applying good stockmanship to improve welfare in all types of pregnant sow housing, the overarching economic considerations, and the need for more research on indicators of sow welfare.

One of the strongest messages coming out of the forum was that sows' mental and behavioral needs, not just their physical health and productivity, must be assessed when evaluating the appropriateness of various sow housing systems. Some attendees questioned whether scientifically valid ways of measuring sows' mental well-being currently exist. The researchers provided an overview of studies examining those and other facets of sow welfare.

As swine representative on the AVMA Animal Welfare Committee, Dr. Christa Irwin worked with the American Association of Swine Veterinarians Pig Welfare Committee to propose speakers and agenda for the forum. She also chaired the forum. Dr. Irwin, who represents the AVMA welfare committee on the Task Force on the Housing of Pregnant Sows, said the forum speakers were chosen for their areas of expertise, to give a thorough review of the issue.

"I have received very positive feedback from the forum," Dr. Irwin said. "Many were unaware that the issue is much more complex than typically projected. In addition, the topic was addressed comprehensively." She added, "I think (the forum) brought clarity to what the topics of discussion should be with regard to sow housing."  

The ethical challenge

As he sees it, animal welfare must be rethought in a dramatic way that puts the needs of animals first and balances them with the realities of the economic environment and producers, and the changing expectations of society.

Dr. Thompson outlined each of those elements. The requirements of pigs include not only straightforward ones such as freedom from disease, he said, but also behavioral dimensions.

"The behavior of animals including human beings is usually a fairly good indicator of how well they're faring in a particular situation," Dr. Thompson said, "and also, cognitive needs, which may be quite difficult to measure."

Dr. Irwin asked Dr. Thompson to elaborate on cognition, calling it "something that we can't necessarily get our arms around." Until 25 or 30 years ago, Dr. Thompson replied, psychologists ignored the qualitative experience of human behavior when designing scientific studies, for methodological reasons. If it is so difficult to study the cognitive aspects of human experience, he said it should be no surprise that it's difficult to study the cognitive aspects of a species with a very different cognitive apparatus and qualitative cognitive experiences. The fact that it is a hard thing to do should not be a reason for not trying to do it at all, he stressed. "One of the things you can do is just look at the basic neurological structure, and based on what you know happens in humans ... you can surmise that at least similar kinds of things may be happening in other animals."

Dr. Thompson described what the public wants—accountability, and not just enough to meet pigs' needs, but a well-being assurance program that can be trusted; honesty (eg, labels and terminology that are not misleading); and cost-effectiveness. Animal protection groups seek such things as stronger regulatory protection at the federal level, change in the legal status of animals, and ethical vegetarianism. Producers seek profitability, flexibility, and fairness.

In balancing those multiple factors, Dr. Thompson said that improving any one dimension may lead to decline in another, so the problem is not one of finding an ideal—though it may be useful to envision an ideal approach—but of finding a way to balance the trade-offs.

He said one approach would be to weigh the cost to the animals, determine what consumers are willing to pay, and arrive at an equation that represents a possible balance. Instead of thinking in terms of production systems or the economics of production systems, however, he suggested an alternative. "Frankly, I think it's going to be much more useful to think of this balance as a process of negotiation, one in which we have all of these things in play and we ... keep all of them on the table as we work through the process of thinking about welfare."

Dr. Thompson said it is important to think about the groups that should be at the negotiating table to ensure that requirements are met.  

A producer perspective

"In my lifetime, I've had experience with four different types of gestation systems," Kellogg said.

Although the family farm has never used an open-pasture facility, Kellogg is familiar with it from working at his uncle's farm while a youth. At Yorkville, Kellogg has had considerable experience with open-front buildings with group pens. It was difficult to keep the pigs warm and dry there, so they moved to enclosed buildings with group pens to better deal with the elements.

"Then we did experience more aggression there than we had expected, and correctly, we (moved to) gestation stalls," Kellogg said. He said that the evolution to each new system was the industry's attempt and his own to always find something better.

Kellogg said that different producers can manage different systems well, but he favors gestation stalls because of the individual care they make possible, and the northern Illinois climate. Each stall has a feed box that can be preset for the exact amount of feed an animal is to receive each day, and a constant supply of clean water. The farm employees do daily checks for health, feed intake, and behavior. They monitor body condition. They conduct daily environmental checks for ventilation and heating and cooling, and daily facility checks for pen maintenance.

A card is attached to each crate for record keeping and observations.

Asked whether confinement leads to lameness in sows housed in stalls or development of sores, the producer said genetics enables him to select for animals that do not have projecting shoulder bones, for example, which are prone to developing calluses and sores. There is much less aggression, he said, and there are far fewer injuries and cases of lameness than when they used mechanical equipment to clean the group pens and sometimes injured sow feet and limbs. Kellogg said fewer than three percent of his sows die of disease or trauma in gestation stalls.

Parasites and sunburn were mentioned as other shortcomings of outdoor systems.

"Twenty years ago, we really thought we were doing a good job when we weaned 9.3 pigs per litter and were weaning 19 pigs per sow per year," Kellogg said. In comparison, over the past 12 months they weaned an average 10.9 pigs per litter and 26.2 pigs per sow per year.

"We've changed genetics and we've changed everything that any good producer would to try and make things better," Kellogg said, "but if we were still dealing with the elements outside and many of the challenges we went through with groups, I am convinced we would not be that productive, and to me, productivity really is an indication that the sow is comfortable and being well-taken-care-of."

What about those who feel that the gestation stall is a confining housing system that suppresses too many of the natural behaviors of swine? Dr. Peter Theran, vice president of animal science for the Massachusetts SPCA, asked. Kellogg said he does not see signs the sows are frustrated with being in a stall. "As you get used to being around pigs, you can recognize when they're content and when they're not. I don't know how you determine whether it's a happy pig or not, but I can tell you when she's comfortable, and that's my goal is to keep them comfortable."

As an assessment tool for animal welfare, the Swine Welfare Assurance Program has benefited his animals and been a "tremendous guideline" to help the Kelloggs set up training programs for their employees. SWAP measures parameters associated with welfare across all types of housing systems, rather than focusing on a particular system.

Economics central to debate

Several attendees questioned some of Dr. Appleby's assertions about the economic feasibility of improving sow welfare, given competition and the public demand for cheaper food.

"I'm not criticizing, in general, farmers, who are in an extremely tight economic squeeze," Dr. Appleby said. "But I do believe there could be significant improvements in farm animal welfare with a very small impact on food prices. I don't know how to get there; I would like more economists to help us discuss how we can get there."

Stanley Curtis, PhD, an attendee and professor emeritus in the Department of Animal Sciences at the University of Illinois, criticized Dr. Appleby's presentation, saying it might give attendees a misimpression, being based on theoretical economics and on "perceived animal welfare."

"If those assumptions turn out to be wrong, the whole model goes down the drain," Dr. Curtis said. "Even though it's interesting to see what we can do with theoretical models, we certainly aren't at the place where we ought to be giving advice to producers or advice to governmental regulators or lawmakers about how these things ought to be, because the assumptions might be wrong."

Dr. Appleby responded, "We understand quite a lot of aspects of the reality of sow welfare in relation to housing, sufficient that we can offer advice to those who need that advice. The idea that we cannot offer any advice until we have definitive answers to all these questions is unrealistic."

Gene Bauston, the co-founder of the animal rights organization Farm Sanctuary, asked whether sow welfare could truly be addressed, given economic constraints. "But in terms of what a pig really wants, is that possible to achieve, or are the economic factors always going to mitigate against true welfare?" Bauston asked.

"The science needs to be interpreted in an ethical context, and in industry, the science and ethics will have to be interpreted in an economic context," Dr. Appleby said. "What I was arguing against is simplistic, short-term, economic jumping-to-conclusions that we are limited by what we can do today. This system, this economic squeeze, has led us to that point."

Forum speaker Dr. John Deen, an associate professor of swine production systems and director of the Swine Center at the University of Minnesota, said the provision of animal welfare is best viewed as a classic economic problem. "I've been told this is a bad argument—that it gets into the idea of prostituting yourself, that there are moral absolutes and you work your way from that," he said. "I would argue the opposite—that all welfare policy reduces itself to a question of allocation of resources."

Limited resources must be allocated across unlimited needs, he said. To avoid overemphasizing any one need, all concerns must be addressed at the same time and same level. Visual measures, for example, are overemphasized, he said. "Part of the reason why stalls have become the poster child of animal agriculture, to a great extent, is because pictorially and on a visceral basis, stalls don't look as good; they look unnatural."

Given the economic model, Dr. Deen contends that the experts in swine welfare are the caregivers—the producers, stockpersons, and herdspersons who are involved in day-to-day allocation of the various needs. "The rest of us—as scientists, as veterinarians, as nutritionists—allow better information to be available to those decision makers," he added. Dr. Thompson agreed that the people who spend time with pigs are the most qualified to make the decisions.

The science of sow welfare

"We can do things that are good for us and not good for them," Dr. Appleby said.

Determining what is good for sows is also a challenge, because stakeholders have different approaches to evaluating animal welfare. Dr. Appleby and other speakers referred to three approaches that assess (1) physical attributes (eg, physiology, production), (2) feelings (beyond behavior to cognition, for example), and (3) the natural environment (whether the pigs live in the natural conditions where they started).

Producers and veterinarians often focus on the physical needs of the animal, while the public focuses on the mental needs of the animal and whether the animal is able to express its natural (innate) behaviors, Dr. Appleby said.

Ed Pajor, PhD, an animal science researcher at Purdue University who serves on several industry animal welfare committees, emphasized the importance of addressing the public's concerns about animal welfare in his presentation.

"Focusing on the function of the body is not sufficient to address the concerns of the public," he said.

"I challenge you, as veterinarians, to think outside one circle (the physical)," Dr. Pajor said.

One of the seminal papers on animal welfare, the Brambell Committee Report, was published in 1965 by some of the top animal welfare experts in the United Kingdom. It established the five freedoms for animals—the freedom to stand up, lie down, turn around, groom, and stretch limbs—as a way of meeting animal's welfare needs. It was updated in 1993 by the UK Farm Animal Welfare Council to include the following:

These five freedoms serve as the cornerstone for animal welfare standards in many countries, Dr. Pajor said.

The lack of freedom of movement for animals on large-scale farms is one of the key welfare concerns that have garnered the public's attention, said Harold W. Gonyou, PhD, a researcher at the Prairie Swine Center in Saskatoon, Canada. It is an issue that has been raised with veal calf crates, sow gestation stalls, and battery cages for hens, he said.

Dr. Gonyou explained the scientific rationale for allowing animals freedom of movement, saying it gives the animal control over its environment, reduces frustration, and allows it to maintain physical condition. In fact, studies have shown decreased muscle mass and decreased bone strength in sows kept in stalls, compared with sows housed in groups, Dr. Gonyou said.

"I personally feel there's merit to the sows having freedom of movement," he said during the question-and-answer session.

When asked by a forum attendee whether turnaround stalls, which are roughly the same size as gestation stalls but have a swinging sides that allow the sow to turn around, adequately provide sows with freedom of movement, Dr. Gonyou said turnaround stalls and wider gestation stalls are both compromises, but don't offer the highest level of welfare.

Dr. E. Wayne Johnson, University of Illinois, asked Dr. Gonyou, "What kind of natural behavior is good for animals?" Dr. Johnson used his sons' fighting with each other as an example of natural behavior that's not necessarily good for them.

"I think we can try to take things to the extreme," Dr. Gonyou responded. "But that's not what we're asking for. We're asking for a sow to have the ability to go off in a corner and defecate or go off in a corner and urinate. We're asking for a sow to have the ability to get out from under a drippy water line, to go over and lie in the sunshine on cool days, to go over and lie in a cool area on a hot day. Those are some of the things that come about by (allowing sows) freedom of movement."

Dr. Pajor said animals have certain natural behaviors that they are highly motivated to perform and if they are unable to execute these behaviors, they may resort to stereotypic behaviors. In sows, nest building and foraging are two such natural behaviors.

"A lot of stereotypic behavior is related to the inability to forage for food," he said.

Higher rates of stereotypic behavior have been documented in sows kept in stalls than in other housing systems, Dr. Pajor said.

Another welfare consideration is freedom from aggression. It is addressed in a section of the Brambell report that is not often cited, but states that "confinement may well confer advantages, notably shelter from the weather, predators, and bullying," according to Dr. Gonyou.

"This is a very serious concern," he said.

The lack of environmental enrichment in gestations stalls is another concern of the public and the animal welfare and rights communities. The European Union requires environmental enrichment, most often in the form of straw, for sows. In the United States, Dr. Gonyou noted, producers have resisted providing environmental enrichment because straw is not available in all parts of the country, and it is not compatible with liquid manure handling systems. Dr. Gonyou said the scientific literature shows, however, that there are welfare advantages to providing sows with enrichment. For instance, straw provides thermoregulatory benefits for the sows, it serves as a dietary supplement, and it gives them something to do, he said.

Another welfare issue is whether the sows can comfortably and safely lie down in the stall. The Food Marketing Institute and the National Council of Chain Restaurants indicate they prefer alternatives to housing sows in stalls, but if stalls are used, the sow should be able to lie down without her udder protruding from the stall, her head touching the feeder, or her hindquarters touching the back of the stall.

"We're not talking about the sow moving here; we're talking about the sow lying in comfort, " Dr. Gonyou said.

He said these are a few of the many variables that affect sow welfare, and scientists don't all agree which are the most important.

"If you were basing your decision solely on the freedom of movement, there's one system that gets dropped," Dr. Gonyou said. "If you are basing everything on controlling aggression, there is one system that wins. So, which is most important? We have different views of what's important to the sow. Until we can find a way to determine how the sow herself perceives these, we stumble about. As we do that, we have to use our professional opinion, judging the data we have, to come up with what is best."

Dr. Gonyou also emphasized that the choice is not simply between stalls and group housing. There are many variations of group housing, with different strategies for feeding, grouping, and monitoring the animals. Key to all these variations is a high level of stockmanship.

In certain variations of group housing, studies have found that the sows are as productive as, or more productive than sows kept in stalls, Dr. Pajor said. Also, researchers have found good strategies for reducing aggression by tweaking feeding, enrichment, and grouping strategies, Dr. Gonyou said.

According to Dr. Deen, veterinarians recognize that absolutely meeting all five freedoms is impossible. "My interpretation and what I'm driving toward is an incremental approach rather than an absolute approach."

More research comparing sow-housing methods is needed, all four researchers agreed.

"I don't think the science gives us a clear answer about sow housing," Dr. Pajor said. "Science can tell us what is, not what we ought to do."

Dr. Deen noted there are constructs of knowledge besides scientific. "Do I, as a veterinarian, simply go through and quote references to my clientele, or do I tell stories about successes and failures of care giving in other farms, and experiences that we see in various areas? And it really is a combination of both—we need the stories; we also need the science."

Forum attendee Dr. Holly Cheever, a member of the board of the Association of Veterinarians for Animal Rights, urged the AVMA to apply a commonsense, evolutionary perspective to evaluating the welfare of sows in various housing systems.

"In terms of common sense, any system that diverges the least from supplying any species' natural physical, psychological, and behavioral needs is most likely to be the least stressful," Dr. Cheever said. "Those needs evolved over millennia and are pretty much hardwired into the brain of the animal. We should try to take those very strongly into consideration as we propose our management systems."

Drs. Gonyou and Pajor pointed out that there are limitations to applying an ecologic approach to animal welfare. For instance, there have been tremendous selective pressures on domestic pigs for the past 5,000 years. Also, there are economic factors such as a lack of space for raising pigs and the need to feed multitudes of people that must be considered.

"It's just very difficult to accommodate that ecological approach in a large-scale operation," Dr. Gonyou said. He added, however, that research on pigs in a more natural setting might provide information that could be applied to large-scale production.

Dr. Appleby disagreed about the extent of genetic change in domestic pigs. "Yes, there have been changes, and they are genetic changes, but they are not very fundamental," Dr. Appleby said. "No behavior patterns have disappeared in domestic sows compared to wild sows. As such, the natural, biologic-ecologic approach is a valid one."

Ultimately, the public may decide what is acceptable, and that decision may be value-based rather than science-based, according to several speakers and audience members. Dr. Curtis, for example, asked Dr. Thompson whether science will prevail, if and when it becomes possible to build a strong consensus among scientists about what constitutes animal well-being based on scientific experimentation rather than empirical observations.

"In other words, does science trump human opinion?" Dr. Curtis asked. Dr. Thompson replied it is important to work with people who have identified themselves as genuinely interested in the well-being of animals so they understand that the differences between species do create very different sets of needs. "I think there actually is a basis to start to build a scientifically informed understanding of what animals' needs are and a conception of animal welfare," he added. "Having said all of that, I have spent much more of my life working on genetic engineering than I have on animal welfare, and I am keenly aware that science does not trump public opinion when it comes to genetic engineering and risks from genetic engineering."