Introduction

There are two areas where welfare decision-making [in swine production] needs further analysis. The first is the current focus on the design and administration of welfare audits. This demands a community of caregivers and experts that agree on the approaches required to come to some commonality. This should be driven by the second and less discussed area of welfare decision-making. This is, namely, the day-to-day decisions that people make on the farms in taking care of animals. These decisions have often not been quantified or explained in sufficient detail. This has driven an assumption that inadequate care is given to the animals, and thus the purpose of welfare audits is not simply to measure the welfare of animals or the measures taken to ensure the welfare of animals, but to actually force substantial change on swine farms. We should question the reasons for welfare audits—and their place in animal care—before wholesale participation.

In many disciplines there is a saying that administrators should not be trusted if they actually want the job. Likewise, I think that anyone who actually welcomes [amimal] welfare audits should be questioned. Welfare audits are the result of a dysfunctional relationship between animal agriculture and the general public. They reflect the argument that farmlevel decisions and decision-makers cannot be trusted. If we accept this lack of trust, and invariably a lack of understanding, there is a place for audits, though audits may not be the best method to address the problem.

Welfare audits are problematic in many different aspects. We have problems in definition of the elements of welfare, measurement of those elements, and measurement of processes that lead to unsatisfactory welfare in animals. In application, it is fraught with personal bias, and it is conflicted by many different agendas.

We must come to the issue of welfare measurement with an understanding that our measures can only be partial, that all parties bring inherent biases, and that good communication methods are required to bring efficiencies to the process. In addition, all must admit that there are real needs for improvement in the welfare of farmed animals. It must also be recognized that all improvements are made under the restriction of limited resources. Particularly when we speak of welfare policy, where regulatory aspects are considered, we must view welfare considerations in terms of the allocation of limited resources. It really does become an economic question, though it is difficult for many parties to admit to this.

With all these points, it is absurd for any person to stand up and claim to have an answer to all questions concerning welfare audits. It takes a community to ask the correct questions and define the correct route to solving the considerable challenges in animal agriculture. What producers should know is whether the correct questions are being asked to come to a proper decision. Here are my 10 questions:

Do we respect each other?

One of the first challenges in coming to a consensus on measuring and evaluating welfare is in having a group that respects each other. Too often I have seen farm owners portrayed as profit-maximizing ogres, animal activists portrayed as anarchistic zealots, and veterinarians [portrayed] as unprofessional and pliable puppets of animal agriculture. Yet, the most frustrating aspect of some of these consensus models is self-aggrandizement by scientists. Science has been portrayed as allowing for an absolute truth, yet most scientists have a narrow view of welfare through a specific discipline.

Who are the experts in welfare?

The experts in welfare have to be defined as those who are involved in the care of animals and the day-to-day allocation of limited resources. Farmers, stockpersons, caregivers … are the experts we should [bring] to the discussion. Yet, too often, these are the people that are kept outside of the discussions. We need to personify the decisions and the decision-makers to allow a recognition that intentional and expert caregiving is the aim of endeavors on the farm. There needs to be an argument that empowerment of caregivers is a central requirement of welfare improvement.

Are audits the best way to improve welfare of animals?

We must agree that audits have limited capability in improving the welfare of farmed animals. Audits are designed to identify and punish bad apples. They are not designed to change the average. Any welfare improvement program thus needs to address concerns about ignorance and apathy. There are areas of genuine lack of information, and further research is needed. There is a need for further education of producers and for methods for inducing producers who are apathetic. Programs such as Swine Welfare Assurance ProgramSM (SWAP)1 are designed for self assessment and education and meet a real need outside of an audit program. Audits are almost always done where there is some level of distrust between parties involved. Demand for audits has come from animal activists and meat retailers. The reason for demanding audits differs between these two parties. Animal activists argue that the majority of farmers cannot be trusted and are unethical. Their purpose for audits is to illustrate systemic deficiencies and induce wholesale change. For meat retailers, there is a desire to avoid surprises and, in some cases, differentiate their product.

How do we increase the level of trust?

I don’t think that this question has been thoroughly addressed by our community. Certifying animal welfare must be more than simply examining animals, facilities, and processes. In my discussions with people concerned about animal care, the main question is whether there is intentional care. The criticisms of animal agriculture almost always use the words “corporate” or “industrial” as a descriptor of farms and are an attempt to portray a lack of intentional care. The real response to that distrust cannot simply be audits. The professionalism and care given by stockpersons must be given as much emphasis as audits.

So how do we measure welfare if we do audits?

There is considerable controversy and no straightforward answer [to this question]. The measures can be divided into five areas: the pigs, historic performance records, the caregivers, contingencies, and production processes. There are absolutely stunning differences in the estimates of relative importance of components. These differences should lead to real questioning of the utility of welfare audits. Prioritizing measures is a very important step, as it should reflect the priorities of the community. In addition, it has to be recognized that there are limited resources available for welfare audits and that they will be a biased toward simpler measures. Simpler measures are repeatable between auditors and can be performed in a short period.

In the discussions I have seen three major biases. The first is to rely on experimental studies to critique processes such as castration and [the use of] gestation stalls. We then are not auditing the welfare of animals but the application of the results of experimental studies. There are numerous potential failures in scientific studies. There are differences in genotype environment and management that limit the representativeness of studies. We are also limited by the breadth of issues studied. For instance, if pain is a concern, is castration the most painful condition for pig? It can be argued that lameness should be much more of a concern than castration, and yet lameness has had little study. This second bias is toward culpability.

I have seen too many arguments of whether disease is a welfare concern. Many critics are much more interested in controlling the direct interaction of pigs and people. Thus, again, there is more interest in castration than lameness. Likewise, contingencies such as alarm systems are often underemphasized. The third bias is against production records. Admittedly, animal productivity is not linearly correlated to animal welfare. Yet deviations of productivity, particularly in mortality and morbidity, are excellent measures of potential failures. It is interesting that human welfare measures often emphasize the mortality and morbidity records of different communities. Second to that, basic health procedures such as vaccination and prenatal care are also emphasized. We see little of these discussions of animal welfare.

Can the audit results be served cafeteria style?

The obvious answer is no. The National Council of Chain Restaurants (NCCR) and the Food Marketing Institute (FMI) argue that the results of audits can be given as a database to its members, as they do in their Animal Welfare Audit Program2. This will allow buyers to define differing welfare priorities, as required. This avoids the whole controversy of prioritizing aspects of the audit. It also balkanizes our efforts to improve the welfare of pigs. Furthermore, it is difficult to imagine its application, as it requires good segregation of product within the processing plant. This is especially difficult to achieve for ground meat products. It also results in an unstable relationship between retailers and producers.

Who should administer the audits?

There are really two choices in this area. The first is to create audit that can be administered by almost anyone. The aim is to have simple measures, usually biased towards process verification, and a simple pass/fail system. Thus the audit is limited by the skills of the auditor. The opposite is a professional and educated auditor that can recognize more complex conditions. The Professional Animal Auditor Certification Organization Inc. (PAACO)3 is currently organizing a professional body of auditors to address the latter requirement. It is made up of animal scientists and veterinarians in the United States and involves training and certifying auditors.

Who should design the audits?

In theory it should be a body reflecting all constituencies. In reality, so far, it has been single constituencies. We need to have buy-in for the forms to allow utility and also stability. As long as there is a power-play between constituencies and their audit forms, the producer is at risk. In addition, audit forms must be designed with good statistical methods to evaluate sample size and the repeatability of welfare measures. Finally, audits must be pre-tested to ensure they can be done efficiently.

So what about gestation stalls?

No discussion on animal welfare can be complete without a discussion on gestation stalls. The justifiable use of gestation stalls cannot be comprehended by many members of society. Conversely, animal welfare audits cannot simply focus on facility and other process measures to reflect the care of animals, and stalls cannot be a central measure of an animal welfare audit. There is evidently a need for further education of the public on the requirements and alternatives for gestation sow housing. Given that, decisions on animal care are often based on aesthetic evaluation, and this is an aspect that will be difficult to address in the future.

Where will it end?

Many producers express a real unease, arguing that we are on a slippery slope toward loss of control and overregulation. They are worried about academicians, administrators, government officials, and bureaucrats embracing assurance programs that have real no endpoint. That concern is justified, yet assurance programs do have a real place as animal agriculture has lost a strong link with the general society. The building up of relationships, so that pork producers can be trusted, should be the long-term goal, with audits being, at best, just part of the answer.

Conclusions

Audit creation is a complex and demanding exercise. It can only reflect the heart of the decision-making on welfare that occurs on the farm. If done incorrectly, it will open a Pandora’s Box for pig farmers. The simple answer to that question on what you should know is that you need to be at the table. A poorly designed and administered welfare audit can do more damage than good, both for pig farmers and pigs. We need to make sure that welfare audits empower the people on the farm to use their resources in the best way possible to care for the pigs entrusted to them.

References

National Pork Board. 2005. Swine Welfare Assurance Program, www.porkboard.org/SWAPHome/
SES, Inc. 2004. Animal Welfare Audit Program, www.awaudit.org
JAVMA News. 2004. Animal auditor organization formed www.avma.org/onlnews/javma/nov04/041115i.asp

July 2006