The Animal Health Institute (AHI) filed a data quality petition January 22, 2003 to the Food and Drug Administration (FDA) challenging a risk assessment of the human health impacts from fluoroquinolone resistant bacteria (Camplylobacter). Background Fluoroquinolones are a class of antibiotics used on humans, pets, and more recently poultry. The Centers for Disease Control (CDC) opposed the use of fluoroquinolone in poultry, but the Food and Drug Administration (FDA) approved two drugs -- one developed by Abbott Laboratories (approved in 1995) and another by the Bayer Corporation (approved in 1996). CDC believes that use of fluoroquinolones in poultry creates antibiotic-resistant food borne bacteria, making it harder to treat people infected from eating undercooked poultry with illnesses such as campylobacteriosis. In 2000, FDA's Center for Veterinary Medicine (CVM) proposed withdrawing approval for the Bayer fluoroquinolone, enrofloxacin, for use in poultry. Bayer filed a request for a hearing, which took place earlier this year, and the proposed withdrawal is awaiting a judge's decision. Abbott Laboratories voluntarily requested withdrawal of its drug application in 2001 for its fluoroquinolone, sarafloxacin hydrochloride. AHI filed a related data quality petition with CDC December 6, 2002 that challenged statements the agency made on human illness related to fluoroquinolone resistant bacteria, the percentage of resistant bacteria in humans, and the number of annual cases of campylobacteriosis in the U.S. AHI appears to be using the Data Quality Act to disqualify information on fluoroquinolones in the event the court supports the withdrawal proposal. If AHI's data quality challenge undermines the evidence supporting CVM's proposal, then the organization might overturn the withdrawal through a separate avenue. Request for Correction The petition challenges the Campylobacter risk assessment referred to as the "Vose Rick Assessment", which became available on CVM's website in January 2001. AHI asserts that because the document has been widely relied upon, referenced, and disseminated by the CVM, CDC and others it is subject to the FDA's data quality guidelines. The petitioner contends the risk assessment contains several flaws and does not meet the data quality standards of utility, objectivity and integrity because:

• It does not use the best available science and supporting studies including peer reviewed studies.
• It uses data collected by unacceptable methods.
• It does not ensure that the information is comprehensive.
• It does not explain why additional studies were not used in developing the risk estimate produced from the assessment.

AHI also claims that because the risk assessment is used in regulatory actions such as the proposed drug withdrawal, it qualifies as influential information and should have met the higher data quality standards for that category of information. Additionally, the group believes the information should meet the standards of risk assessment, which mandate the assessment specify the population affected, expected risk for that population, the upper- and lower-bound risk estimate, data gaps and uncertainties, and additional studies not used for the risk assessment and why they were not used. In order to correct the information, AHI recommends that the FDA:

• Appoint a qualified and independent risk assessment expert to evaluate the Vose Risk Assessment model and the issues the petition has raised.
• Stop disseminating the risk assessment if the expert review confirms that the assessment does not comply with FDA data quality guidelines and have a qualified and independent expert group fully review the assessment if the review disproves the complaints.
• Stop CVM from disseminating the underlying data if it is found out of compliance with FDA guidelines.
• Prompt CVM to issue a corrective statement on its website, in other points of dissemination, and on archival documents that reference the information explaining it is erroneous.

Agency Response The March 20, 2003 FDA response informed AHI that the agency will issue a decision on the data quality challenge within 60 days of the judge's final decision in the hearing. The agency explains that this is consistent with OMB's guidelines stating that the data quality correction process should not disrupt agency processes. Additionally, the agency stated that hearing decision will be binding and the official response to the petition should not disagree with this decision. The FDA also contends that processing the request would waste resources because AHI's data quality complaints are being handled through the hearing. AHI Appeal Although FDA had not issued a final response on the data quality challenge, AHI filed an appeal April 16, 2003. AHI claims that the interim response constituted a denial and the agency is avoiding its requirements under the Data Quality Act. It notes several reasons for the reconsideration request.

1. AHI believes the improper person responded to the request, as the guidelines state that the supervisor of the employee that made the decision in question should review the challenge.
2. CVM's deferral of its decision until after the conclusion of the hearing violates the Administrative Procedure Act.
3. The organization asserts that the data quality aspects of the risk assessment are not being considered in the court hearing and cites CVM's statement in the hearing that the judge disregard the FDA guidelines.
4. AHI contends that a review of its request will not disrupt the hearing or waste agency resources.
5. CVM's choice to wait until after the hearing eliminates AHI's only opportunity to correct the information before the hearing is over, and allows the information to be used to support withdrawal of the drug approval.
6. AHI restates its belief that the risk assessment does not meet the OMB, HHS, or FDA's data quality guidelines.

Appeal Response The CVM denied the appeal on September 16, 2003, stating that the issues AHI raised in the data quality challenge were currently being addressed through the administrative hearing on proposal to withdraw enrofloxacin. As outlined in FDA's data quality guidelines, this is one of the existing processes that the agency can use to address information quality concerns. Such provisions enable agencies to prevent the data quality guidelines from becoming duplicative mechanisms that waste resources. This final decision leaves AHI with no further administrative options under the data quality guidelines. For more information on the court hearing, see the FDA website.