State models cited as ways to improve outbreak responses

Robert Roos

Jul 31, 2008 (CIDRAP News) Experts and industry leaders speaking at congressional hearings this week on the nationwide Salmonella outbreak said federal agencies should take cues from state programs if they want to improve the traceability of fresh produce and the success of foodborne disease outbreak investigations.

At a House subcommittee hearing yesterday, a Minnesota expert said investigations of multistate foodborne disease outbreaks are hindered by a lack of standardized techniques and approaches from state to state. The expert, Michael T. Osterholm, PhD, MPH, proposed that other states adopt a set of best practices like those used by the Minnesota Department of Health (MDH), which played a major role in linking the Salmonella cases to jalapeno peppers early in July.

Osterholm, who is director of the University of Minnesota Center for Infectious Disease Research and Policy, publisher of CIDRAP News, also proposed the establishment of regional surveillance teams or a national surveillance team patterned after teams used by the MDH.

At a separate House subcommittee hearing today, tomato industry leaders from California and Florida said that programs in those states make it possible to quickly trace fresh tomatoes back from the retail level through the distribution chain to the grower.

The Food and Drug Administration (FDA) had a difficult time tracing the sources of tomatoes, which were the prime suspect in the huge outbreak for several weeks, until suspicion fell on jalapeno and Serrano peppers in early July. The FDA maintains that the main problem was that many businesses which handle tomatoes use paper instead of electronic records. In the early weeks of the outbreak, the agency published an often-revised list of growing areas that were considered safe, while warning consumers to avoid certain types of raw tomatoes from other areas.

The Salmonella outbreak included 1,319 cases in 43 states, Washington, DC, and Canada as of yesterday. The FDA began issuing advisories about raw jalapeno and Serrano peppers on Jul 9, and the warning about tomatoes, which was based on statistical associations in the absence of any findings of tomatoes contaminated with the outbreak strain, was canceled on Jul 17.

The FDA announced Jul 21 that a jalapeno pepper contaminated with the outbreak strain had been found at a tomato distributor in McAllen, Tex. Then on Jul 28, Colorado officials reported finding a jalapeno tainted with the outbreak strain in the home of a person who had the illness. Yesterday the FDA said that jalapeno was traced to a farm in Mexico and contaminated irrigation water and a tainted Serrano pepper had been found at another Mexican farm.

Lack of standardization
In written testimony presented to the House Agriculture Subcommittee on Horticulture and Organic Agriculture yesterday, Osterholm said epidemiologic investigations are carried out by many different jurisdictions, with no general agreement on best practices. His statement was co-written by Craig Hedberg, PhD, a foodborne disease expert with the University of Minnesota School of Public Health, and John Besser, PhD, clinical laboratory manager at the MDH.

"There are great differences in the ability of states to collect and analyze the basic information needed to resolve outbreaks, which places intrinsic limitations on the ability of CDC [Centers for Disease Control and Prevention] to investigate multistate outbreaks," Osterholm stated. "This in turn limits the ability of FDA or USDA [US Department of Agriculture] to pinpoint the sources of contamination and to break the chain of transmission."

Contributing to the lack of standardization is the fact that foodborne disease investigations are handled at different levels in different states, Osterholm reported. He said a survey reported last year showed that gastrointestinal disease surveillance was conducted by local agencies in about half of the states, was centralized in a state office in about a quarter of the states, and was handled by regional state offices in another 20%.

He explained that the role of CDC in multistate foodborne outbreak responses is to aggregate surveillance data on a national level and provide consultation and coordination; the agency does not have the authority to independently investigate an outbreak in a state, though it can respond to a state request.

In many foodborne disease outbreaks, the food vehicle is never found, Osterholm said. The prime reason many outbreak investigations fail is the long time lag between when people get sick and when the outbreak is recognized, he said. It can take 3 to 4 weeks for investigators to learn from DNA fingerprinting that they have a cluster of cases caused by the same strain of pathogen. When patients are interviewed, they have to try to recall what and where they ate as long as 5 to 6 weeks earlier.

Adding to the difficulty, Osterholm wrote, is that many public health agencies do not use a standardized questionnaire or collect detailed source information about food items when they interview case-patients. "Systematically collecting detailed exposure information during early interviews with cases is a critical need to improve the effectiveness of our surveillance and outbreak investigation efforts," he said.

Osterholm said a group called CIFORthe Council to Improve Foodborne Outbreak Responsehas developed guidelines that could help to standardize the response to outbreaks. The practices, many of which have been used successfully in Minnesota, include interviewing all patients when their cases are first reported, using a standardized form to collect detailed exposure information when recall is the greatest, and then to interview patients again after possible new sources are suggested during the investigation. "We believe these should be adopted as best practices, and that where resources limit the adoption of these practices, we must find a way to build the infrastructure of our public health system to make it possible," he stated.

A further key to successful outbreak investigation in Minnesota, Osterholm said, has been the use of a group of eight to ten public health students, known as Team Diarrhea, to interview patients. Since interviewing patients quickly is crucial, "we believe a series of regional Team Ds or a national Team D would go a long way to providing precisely the real-time support for outbreak investigations at the state and local levels that is so sorely needed," he said.

Osterholm's testimony dovetailed with comments at today's hearing by Kirk Smith, head of the MDH's foodborne disease unit. He summarized how the MDH investigated a cluster of Salmonella Saintpaul cases that surfaced in Minnesota in late June, leading to the identification of jalapeno peppers as the food vehicle.

The first S Saintpaul isolates were identified on Jun 23, and by Jun 30 several patients reported they had eaten at the same restaurant. On Jul 3 the MDH investigators were able to tell the CDC that the restaurant investigation pointed to jalapenos. In addition, the Minnesota Department of Agriculture was able to trace the jalapenos to a farm in Mexico.

One reason for the successful investigation was that foodborne disease probes in Minnesota are centralized at the state level, Smith said. That makes it possible to confirm and type Salmonella isolates, usually within 2 to 3 days, and to interview patients quickly, he explained.

Does Bioterrorism Act need updating?
Today's hearing, held by the House Energy and Commerce Subcommittee on Oversight and Investigations, focused on lessons learned from the Salmonella outbreak and response.

In an opening statement, Rep. Bart Stupak, D-Mich., the subcommittee chair, said the group would consider whether the Bioterrorism Act of 2002, which was designed in part to improve the traceability of food products, needs to be amended for that purpose.

The panel heard from tomato industry and state officials that Florida and California have programs that require traceability for tomatoes and work well.

Ed Beckman, president of California Tomato Farmers, based in Fresno, said his group recently conducted a trace-back demonstration for staff members of the subcommittee. He said it took only 35 minutes to trace a box of tomatoes to the grower.

"What we did was link the 'one step up and one step back' requirements of this (Bioterrorism) act at each level of the supply chain," he said.

Concerning FDA claims that it was difficult to trace tomatoes, Beckman said, "We can't help but ask specifically, where was the problem?"

The reply from David Acheson, MD, the FDA's associate commissioner for foods, was that many of the businesses that handle tomatoes had only paper records, which took time to sift through.

Acheson was asked whether the Bioterrorism Act "worked" in the case of this outbreak. "The Bioterrorism Act worked as written," he replied. "We rarely ran into a situation where people were not keeping records. It was many of the small producers, the small restaurants . . . they do not have electronic records; the vast majority of information we got was paper."

Parker Boothe, president of a tomato company in Manteca, California, and other industry representatives asserted that the FDA should set national safety and traceability requirements for tomatoes. He called the cost of the traceability system minimal, saying, "Any size firm, large or small, can do this."

Acheson told the panel that the FDA, in proposing its Food Protection Plan last fall, asked for 10 specific legislative authorities. Of those, "probably the one that's most important is the one that requires preventive controls [in food production and processing]. That's absolutely critical across the board," he said.