Language selection

Government of Canada / Gouvernement du Canada

Search

Operational guideline: Animal welfare preventive control plans and self-audits for the slaughter of food animals

Table of contents

1.0 Purpose

The purpose of this document is to provide guidance to Canadian Food Inspection Agency (CFIA) Inspection Staff on Animal Welfare Preventive Control Plans (AWPCP) and self-audits for the slaughter of food animals.

This document is intended to be used in conjunction with other guidance documents as referenced in section 3.0 since it is limited to information not included in these documents.

The guidance outlined below should be used when verifying compliance with regulatory requirements related to AWPCP, which include self-audits.

2.0 Authorities

The inspection powers, control actions and enforcement actions authorized by the above legislation are identified and explained in the Food regulatory response guidelines.

3.0 Reference documents

Other reference documents:

4.0 Definitions

Unless specified below, definitions are located in either the:

Additional technical definitions are located in the Guidelines for animal welfare Preventive Control Plans and self-audits for the slaughter of food animals.

5.0 Acronyms

Acronyms are spelled out the first time they are used in this document and are consolidated in the food business line acronyms list.

Additional acronyms include:

6.0 Operational guideline

The legal framework for a Preventive Control Plan (PCP) for animal welfare is provistions 86 (1), 88, 89(1)(d) and (e) and 89(2) of the Safe Food for Canadians Regulations (SFCR).

The regulatory requirements for preparing, keeping, maintaining and implementing an animal welfare PCP (AWPCP) for operators who handles (including slaughter) food animals are explained in section 8.0 - Additional PCP content for animal welfare of Regulatory requirements: Preventive control plan (PCP).

An establishment's AWPCP replaces the former animal welfare written control program. Its purpose and elements are described in Guidelines for animal welfare Preventive Control Plans and self-audits for the slaughter of food animals.

The information presented in this document is complementary to information intended for the industry. CFIA must understand Industry requirements to be able to verify compliance to SFCR. Additionally, this document provides guidance intended specifically for inspectors to enable them to conduct compliance verification activities and to take enforcement action.

6.1 AWPCP - Additional information for inspectors

6.1.1 General information

The AWPCP may be based on existing models, but cannot be limited to vague and general statements. The AWPCP must contain detailed information specific to each activity and each type of food animal handled and slaughtered in the establishment covered by the licence.

The AWPCP must:

As a CFIA inspector, you do not have to know all the types of equipment, parameters, performance criteria that exist to ensure the welfare of food animals at the slaughterhouse. The operator must be able to provide you with evidence of the effectiveness of the measures to prevent or eliminate animal welfare risks and the performance criteria used as it is his responsibility to ensure that his AWPCP ensures compliance with SFCR.

For example, the AWPCP section on stunning in a slaughter establishment should include:

Refer to the following Guidelines for a detailed description of expected AWPCP specifications for live animal receiving, handling and slaughter:

6.1.2 CVS Task 5.1.12 – Animal welfare control program

Use CVS Task 5.1.12 to verify that the AWPCP is designed and implemented correctly, reassessed at least annually and is effective in controlling risks to animal welfare.

CVS Task 5.1.12 must be conducted as per the prescribe frequency in CVS, but should also be conducted in part every time there is an animal welfare concern identified while conducting any of the CVS section 5 tasks. It is not necessary to complete Task 5.1.12 in its entirety and review the entire AWPCP every time an IR-CAR is issued for section 5 Tasks. Review either the entire AWPCP or specific AWPCP elements linked to the IR-CAR finding(s), as deemed necessary, and evaluate the entire AWPCP as prescribed by CVS.

6.1.3 Repeat failures (patterns of failure)

If an animal welfare issue keeps reoccurring despite corrective actions and preventive measures, this can represent a pattern of failure, or it may indicate the preventive measures were ineffective. The AWPCP must include performance criteria and take measures to remain within these limits. An IR-CAR can be issued for repeat failures even when the operator identifies the failure and initiates corrective action, because repeat failures indicate an inability to consistently meet acceptable performance.

An operator may be tempted to remove preventive measures that were previously implemented once an issue appears to be resolved. If removal of the preventive measure causes the issue to reoccur, this signifies that the preventive measure either has to be permanent, or different.

6.2 Self-audits – additional information for inspectors

6.2.1 General information

The self-audit portion of CVS Task 5.1.12 asks CFIA to verify that industry is doing regular self-audits; to observe a self-audit, and to review self-audit records. CFIA is not required to perform self-audits, nor to correlate with industry.

Self-audits are tools for industry to evaluate their performance at any given moment in time using criteria that are measurable and that can be measured in a consistent manner each time and by different people. A self-audit is not the same thing as monitoring procedures in the AWPCP, and it does not replace monitoring. It simply gives a snapshot of the slaughter process to gauge performance at that moment in time.

The objective of self-audits is to ensure the Objective performance criteria for humane slaughter are met for mammalian and avian food animal. Additional criteria may be audited or higher performance standards may be used, but performance standards must not be inferior to those indicated in Objective performance criteria for humane slaughter. The establishment may also perform self-audits of performance criteria associated with other activities in addition to humane slaughter, such as transportation.

Self-audits should be conducted on all species being slaughtered, not just cattle, pigs, sheep, horses, chickens and turkeys. The criteria are adaptable to all species and all slaughter methods. The operator can adapt an audit for a minor species by using criteria for another similar species and a similar slaughter method as described in Objective performance criteria for humane slaughter. For example, use cattle criteria for bison stunned with firearms, or turkey criteria for geese or ducks stunned with an electrical water bath. Quail self-audits is similar to poultry self-audits and should use the same criteria as poultry. Rabbit self-audits should use the same criteria as poultry for electric stunning but in case of mechanical stunning (captive bolt), should use the same performance standard as for mammalian. Refer to section 6.2.4.1.

Self-audits can be performed by a trained employee or by an external third-party hired for that purpose. The auditor must not be the same person as the one doing the task that is being audited or the one responsible for taking corrective action (e.g. re-stunning animals).

The person conducting a self-audit will not normally step in to initiate corrective action or draw attention to an ineffective stun or other deficiency while the audit is taking place, with the exception of an observed deliberate act of cruelty which always requires immediate intervention. The auditor's role is not that of the monitor or of the employee tasked with responding to any deviation. Instead, the auditor's role is to silently count the number of animals/events without drawing attention to deficiencies in order to avoid influencing the audit outcome.

6.2.2 Audit frequency and sample size

Audit frequency and sample size is determined by the operator and is specified in the AWPCP. Factors to consider when determining audit frequency and sample size are explained in Guidelines for animal welfare Preventive Control Plans and self-audits for the slaughter of food animals and accompanying footnotes. In very small plants, for example those slaughtering 5 animals per day, a running tally should be kept so that meaningful data can be extrapolated.

The recommended sample size of 100 cattle, pigs, sheep/goats, 500 chickens or 100 turkeys, as per Objective performance criteria for humane slaughter is not mandatory. The sample size must be discussed with the veterinarian with supervisory authority and should be appropriate to the slaughter volume. Sample size for self-audits does not have to be the same as for monitoring. During monitoring or random checks of slaughter effectiveness, smaller or larger sample sizes may be used depending on the situation or the issue and its extent, and only the particular area of concern needs to be evaluated.

6.2.3 Audit techniques

There are several different animal welfare auditing services, each evaluating different phases of animal production, transport, handling or slaughter with different criteria being measured. Each audit organization uses its own audit tool to conduct professional third-party audits, and these audit tools can be adapted for self-audits. Some examples include Professional Animal Auditor Certification Organization (PAACO), Farm Animal Care Training and Auditing (FACTA), Validus, Silliker (Merieux); industry-specific audits from the North American Meat Institute (NAMI) for cattle, pigs, sheep, National Chicken Council (NCC) for broilers, National Turkey Federation (NTF) for turkeys; and newer models tailored to specific needs, e.g. the Canadian cattle feedlot industry, many of which follow the PAACO model.

The Objective performance criteria for humane slaughter are based on NAMI, NCC and NTF performance criteria.

Different audit tools may tell the auditor to measure some criteria in certain locations using a given set of parameters. For example, one audit tool may specify that vocalization is to be counted for all provoked squeals in 100 pigs in the restrainer, and only when the posterior part of the pig's hind end is past the restrainer entrance, while another audit tool may include pigs in the chute leading to the restrainer, and only for electric prod use.

Some poultry establishments may choose to add criteria for broken wings as per NCC and NTF audit tools. The NCC audit tool counts the number of chickens with at least one broken wing, while the NTF tool counts the number of broken wings and not the number of turkeys with at least one broken wing.

As a CFIA inspector, you do not need to know specific details about what is being evaluated other than the main objective criteria in Objective performance criteria for humane slaughter. If you have concerns about what additional criteria are being audited , how/where they are evaluated , how individual criteria results are tabulated (pass/fail, or points), and how the overall pass/fail score is determined, consult the AWPCP to see if includes details that will clarify the concern, or discuss with the operator.

The audit procedure should be the same for all self-audits conducted in the establishment for a given species and slaughter method. However the audit procedure may be different in another establishment even if the same species is slaughtered using the same method and this may be because the operator wants to focus on particular criteria (e.g. slips), or because different auditors perform the audit in the second establishment (e.g. a PAACO-certified 3rd party auditor instead of an establishment employee).

6.2.4 Specific audit items

6.2.4.1 Mammalian food animals

Generally, the acceptable performance standard for any mechanical stunning (captive bolt, firearms) is ≥ 96% of the animals are rendered insensible by the first stun, while for electrical or gas stunning, it is ≥ 98% of the animals are rendered insensible following the first stun application. Refer to section 6.2.6.

For pigs, formerly the only performance criteria evaluated for stunning effectiveness were stunning electrode placement, hot wanding and overloading of gondolas. Stunning effectiveness now includes evaluation of the percentage of animals rendered insensible following electrical or gas stunning (objective of ≥ 98% rendered insensible for each method). If an operator chooses to stun pigs with a captive bolt stunner, acceptable performance would be ≥ 96% rendered insensible by the first stun (IG under development).

For sheep and goats, acceptable performance is ≥ 98% rendered insensible by an electrical stun or ≥ 96% rendered insensible by the first captive bolt stun.

During self-audits for animals stunned with captive-bolt equipment, a failed stun is scored if the stunner causes an injury or leaves a mark on the animal. A stun is considered missed and is not counted if the stunner does not touch the animal, or the outer housing of the equipment touches the animal but the bolt does not exteriorize and so does not leave a mark or penetrate the head.

Security knocks may be used following mechanical stunning methods. The effectiveness of the first stun is always evaluated and the auditor must have time to assess the first shot before the second one is delivered. If the first stun is effective, the second stun is considered a security knock, but if the first stun is not effective, this is counted as a failed stun and the second stun is considered a re-stun.

For detailed descriptions of self-audit methodology for cattle, pigs and sheep, download the latest Recommended Animal Handling Guidelines and Audit Guide from North American Meat Institute. Note that the operator does not necessarily need to follow the NAMI self-audit style, but this guide will help you understand how self-audits work.

6.2.4.2 Avian food animals

Pre-stun shocks are not included in the Objective performance criteria for humane slaughter. An operator may choose to audit pre-stun shocks, and if so, this separate performance criterion would be evaluated at the entrance to the water-bath stunner. However, unless specified in the AWPCP for self-audits, pre-stun shock occurrence is not documented and does not count as a failure of the stunning effectiveness criterion. On the other hand, pre-stun shocks are an animal welfare issue and the operator should address this in his AWPCP (avoidable suffering).

Auditors performing self-audits according to NCC and NTF standards normally evaluate the stunning effectiveness criterion immediately upon stunner exit. Although birds may show signs of return to sensibility beyond this point because of a weak stun or electro-immobilization, birds showing signs of returning to sensibility after the automatic knife are not counted as an ineffective stun under the "Stunner" criterion. The reason for this is that poultry audit standards were developed in the United States of America (USA), where poultry are exempted from the USA Humane Methods of Slaughter Act and therefore do not need to be rendered insensible before slaughter. In Canada, poultry must be rendered unconscious prior to slaughter in a manner that prevents it from regaining consciousness before death. Therefore, the operator should address this in his AWPCP (avoidable suffering).

The criterion "Cut/neck slitter (automatic)" evaluates whether the equipment cuts blood vessels or not, and under NCC/NTF auditing standards, birds are counted as being either cut or uncut. An operator may choose to tighten these standards and add a secondary criterion for a good versus a poor cut, but that will be at the discretion of the licence holder.

Under NCC/NTF auditing standards, the presence of an uncut bird entering the scalder and witnessed by the auditor is an automatic audit failure. Deliberate acts of cruelty will also automatically result in audit failure.

6.2.5 CFIA actions in response to self-audit failures

The self-audit is a tool for industry to measure its performance at any given moment. CFIA only needs to verify that an operator is conducting regular self-audits while performing the annual CVS task 5.1.12. Otherwise, CFIA's role is to verify compliance with regulations, and the tool for CFIA to do so is section 5 CVS tasks.

Do not be overly concerned with self-audit details. Your main concerns should be:

If you have any concerns with regulatory compliance, conduct the appropriate CVS task. For the other concerns listed above, conduct a portion of CVS task 5.1.12 to see what the AWPCP says about self-audit procedures, documentation, rating and follow-up procedures, and discuss any concerns with the operator.

Some self-audit performance criteria are regulatory requirements while some are not. Regulatory requirements include effective stunning (SFCR 141), bleed rail insensibility (SFCR 142 and 143), access to water in holding pens (SFCR 71 and 136) and absence of deliberate acts of cruelty (SFCR 128 or 129 (1)). Performance criteria which are not regulatory requirements are slips/falls, vocalization and prod use, but they may be a regulatory requirement if slipping, falling and vocalization are the result of the facilities (floors-SFCR 57 (b) (iv) and 58 (3)), poor handling or equipment causing avoidable suffering or injury (SFCR 135), or if prod use is contrary to the requirements of SFCR 129(2).

CFIA action during observed self-audits will not depend on the final result of a self -audit but will depend on the actions of establishment employees or of the operator when regulations are contravened (e.g. when an animal fails to be rendered insensible or returns to sensibility on the bleed rail) with the exception of an observed deliberate act of cruelty which always requires immediate intervention such as compliance and/or enforcement actions. Please refer to a list of deliberate acts of cruelty in the Humane care and handling of food animals (internal access only - RDIMS 11001057).

6.2.5.1 Circumstances whereby an IR-CAR is issued following self-audit observations

Perform the corresponding CVS task and issue an IR-CAR if a regulatory non-compliance is not identified or effective corrective action is not taken by establishment employees during self-audit of regulatory criteria. Note that the auditor is not the one taking corrective action.

Examples include:

Perform the corresponding CVS task and issue an IR-CAR if a regulatory non-compliance is not identified or effective corrective action is not taken by establishment employees during self-audit of non-regulatory criteria, where handling, use of equipment or the facility contravenes SFCR requirements.

Examples include:

6.2.5.2 Circumstances whereby CVS Task 5.1.12 should be conducted

Perform at least a portion of CVS task 5.1.12 whenever there are concerns or questions about the self-audit. Refer to the Guidelines for animal welfare Preventive Control Plans and self-audits for the slaughter of food animals. Clarify questions and discuss concerns with the operator as needed.

Examples include:

6.2.5.3 Circumstances whereby a corresponding CVS task should be conducted

Perform the corresponding CVS task if you have concerns when:

Examples include:

6.2.5.4 Circumstances whereby no CFIA action is required

You do not need to take any action when:

Examples include:

6.2.6 Zero tolerance versus "tolerance" levels

During self-audits, the term "zero tolerance" means that the only way each of the 3 criteria listed below can pass is if they have a perfect score.

The 3 zero tolerance criteria are:

Most self-audit tools (NAMI, NCC, NTF) assign an automatic audit failure for these 3 criteria.

Other criteria have a "tolerance", e.g. 96% for captive bolt stunning or 98% for electric or gas stunning, in recognition of the fact that humans and equipment do not function perfectly at all times. The more automated a procedure is (electrical or gas stunning versus captive bolt stunning), the more effective it is expected to be.

A less than perfect procedure can be tolerated, but animal suffering must never be tolerated. If the performance standard is ≥ 98%, the equipment itself does not require corrective action unless performance falls below 98%, but corrective action is mandatory for each and every animal affected by poor equipment performance. Therefore if performance is within acceptable criteria, the equipment does not require adjustment, but each affected animal requires immediate re-stunning or immediate killing as the case may be. If performance is below acceptable criteria, corrective action must be taken for both equipment and affected animals.

6.3 Assigning compliance level to CVS Task 5.1.12

Assess deficiencies identified while completing CVS task 5.1.12 and assign a compliance level based on the extent, severity and impact of these deficiencies on the integrity of the AWPCP.

Assign an unacceptable compliance level and issue an IR-CAR when:

Assign an acceptable compliance level and document items requiring correction when:

Date modified: