Letter to the Hon. Phyllis Fong, Inspector General of the US Department of Agriculture - Delauro, Merkley Call on USDA Inspector General to Expand Swine Slaughter Inspection Rule Audit

Letter

Dear Inspector General Fong:

We write to request that your office expand its ongoing audit into the U.S. Department of Agriculture's (USDA) proposed Modernization of Swine Slaughter Inspection Rule (Rule) to include data quality and methodological concerns that call into question the validity of the Rule's estimated impacts on food safety and animal welfare.

On June 21, 2019, at the request of sixteen Members of Congress, USDA's Office of Inspector General (OIG) confirmed that it had opened an inspection into the quality of worker safety data used to develop the Rule, as well as whether the Department followed applicable requirements for public transparency.

Earlier this summer, the U.S. House of Representatives passed the Fiscal Year 2020 Agriculture Appropriations bill, which included language that would prevent USDA from finalizing, issuing, or implementing the Rule until OIG has presented findings on all data methods used in crafting the new swine slaughter inspection system and the Department has resolved any issues identified. The Rule has significant implications for worker protection, food safety, and animal welfare, clearly highlighting the need for this additional oversight.

According to the Centers for Disease Control and Prevention (CDC), consumption of pork causes over half a million cases of foodborne illness, and over ten percent of Salmonella infections, each year in the United States. USDA itself has concluded that higher line speeds result in more pathogen contamination "in the absence of compensating measures." Similarly, the agency has indicated that having plants "fully staffed" with inspectors improves food safety, and that microbiological performance standards have resulted in better control of pathogens like Salmonella. Yet FSIS has failed to explain why the proposed rule--despite increasing line speeds, reducing the number of inspectors, and eliminating microbiological testing standards--would not hurt food safety. Moreover, the agency has failed to comply with requests for data and other information cited in its analyses of the proposed changes.

In support of its claim that "the new system is unlikely to result" in more foodborne illness, FSIS has relied primarily on a risk assessment, entitled "Assessment of the Potential Change in Human Risk of Salmonella Illnesses Associated with Modernizing Inspection of Market Hog Slaughter Establishments." Under Office of Management and Budget Guidelines, an external peer review of this risk assessment should have been conducted before FSIS "announce[d] its regulatory options." However, FSIS instead waited until after it had published its proposed rule to conduct an external peer review, and then published the results of that peer review after the close of the comment period on the proposed rule.

The peer review strongly suggests that the risk assessment is invalid, with three of the five reviewers raising fundamental objections. The first of these reviewers concluded that "the regression model assumption that there is a relationship between the HACCP-Based Inspection Models Project (HIMP) inspection activities and post-chill prevalence of Salmonella contamination has not been established," and that "this makes the risk assessment model invalid." The second concluded that the risk assessment's modeling was "completely inadequate to describe the hog slaughter data." The third reviewer proclaimed to be "not convinced" that the risk assessment takes "a statistically legitimate approach." In response, FSIS has issued a revised version of its risk assessment, but has made no changes to the proposed rule.

The flawed evidentiary basis for moving forward with the proposed rule is particularly concerning given the absence of any pathogen-specific performance standard for pork. Such a standard would provide an important guardrail, and a direct measure of the proposed changes' impact on food safety. However, while a standard for Salmonella in whole hog carcasses was put in place in 1996, it has not been updated, and the agency decided to abandon testing for compliance with the standard in 2010. Since then, FSIS has conducted "exploratory" sampling of pork parts that has revealed high levels of Salmonella contamination, as much as 22 percent in some products. Nevertheless, the agency has not developed a new performance standard.

To better understand whether FSIS accurately analyzed food safety data and appropriately shared that data with the public, we ask that you broaden your ongoing investigation of the rulemaking process to determine whether FSIS:

Complied with Office of Management and Budget requirements for peer review with respect to the Market Hog Risk Assessment prior to announcing the proposed rule;
Allowed for informed public comment on the proposed rule by making information from the peer review accessible during the comment period on the rule itself;
Responded appropriately to criticism from the peer reviewers, including by making adjustments to the proposed rule where warranted;
Came to a reasonable determination in proceeding with the rulemaking without first adopting performance standards for pathogen testing; and
Failed to respond in a timely manner to Freedom of Information Act requests from stakeholders for the data the agency used to support the proposed rule.
Additionally, we are deeply concerned that USDA failed to adequately assess and consider the impacts of the proposed rule on animal welfare. Most notably, the proposed rule (1) requires slaughter facility employees, rather than FSIS inspectors, to sort pigs "before the animals are presented for ante-mortem inspection" and (2) eliminates slaughter line speed caps. Increased line speeds will result in an increased volume of animals being slaughtered. FSIS cannot credibly argue that decreasing inspector oversight of the sorting of pigs while increasing the volume of animals slaughtered will not undermine animal welfare. Slaughter employees face a much greater threat of retaliation and intimidation by management than do FSIS inspectors; employee pre-sorting of pigs could result in the failure of slaughter employees to properly sort and euthanize animals unfit for slaughter. Even worse, the proposed rule fails to even consider the impact it would have on non-ambulatory disabled pigs--a class of animals especially at risk of inhumane handling during sorting. The agency had previously insisted that its inspectors needed to see every non-ambulatory pig, in part because they may be the bellwether of an outbreak that could cause the culling of massive numbers of animals. The proposed rule ignores this issue, despite the threat to animal welfare and public health.

Further, USDA is responsible for enforcing the Humane Methods of Slaughter Act (HMSA), which requires that all farmed animals be rendered fully unconscious before slaughter (with exceptions for animals ritually killed for religious purposes). A 2015 undercover investigation at a HIMP plant in Minnesota (Quality Pork Processors) revealed that plant workers who were facing the pressure of faster slaughter line speeds improperly stunned pigs or did not stun them at all. The undercover investigator observed pigs regaining consciousness after their throats had been cut open and continuing to move down the slaughter line--without the line ever being stopped. The plant's Animal Welfare Supervisor acknowledged that pigs sometimes regain sensibility after stunning.

USDA's evaluation of HIMP relies heavily on a 2014 report analyzing data from the operation of HIMP in "market hog establishments." As USDA admits, the 2014 hog HIMP report did not address compliance with the HMSA, nor did it even mention animal welfare. USDA claims it analyzed HMSA compliance in constructing the proposed rule, but its review was limited to two sets of data from January 2013 through September 2015: a comparison of Humane Activities Tracking System (HATS) task hours performed at HIMP plants vs. non-HIMP plants, and a comparison of Noncompliance Record (NR) rates between HIMP plants and non-HIMP plants.

Based on these data, USDA concluded that HIMP establishments exhibit higher compliance rates with humane handling regulations than do non-HIMP establishments, and that increased offline inspection may improve compliance with the HMSA. We are concerned that these conclusions may run counter to the full evidence that USDA should have considered when determining the animal welfare impacts of the proposed rule and that USDA may be drawing the wrong conclusions from the data. Allowing faster line speeds and assigning pre-sorting duties to plant employees could yield fewer NR rates, though not because HIMP plants are in higher compliance, with animals handled more humanely, but because HIMP provides less-thorough oversight and makes it more difficult for inspectors to observe violations.

Taken together, the available evidence indicates that, if finalized and implemented, USDA's rule may jeopardize not only worker safety but also public health and animal welfare. To address these concerns, we strongly encourage OIG to conduct a thorough review of the food safety and animal safety data collection and analysis. Thank you for your consideration of this important request. We look forward to your response.

Source