105 sickened, 1 death linked to Salmonella Newport outbreak originating in beef from dairy cattle, 2016-17

Contaminated ground beef was the likely source of a protracted outbreak of 106 Salmonella Newport infections, 42 hospitalizations, and one death in 21 states during October 2016–July 2017. While no direct link was found, whole genome sequencing suggests dairy cows were the ultimate outbreak source.

Foodborne outbreak investigations could be enhanced by improvements in the traceability of cows from their originating farms or sale barns, through slaughter and processing establishments, to ground beef sold to consumers.

In January 2017, the U.S. Centers for Disease Control (CDC) identified a cluster of Salmonella enterica serotype Newport infections with isolates sharing an indistinguishable pulsed-field gel electrophoresis (PFGE) pattern, JJPX01.0010 (pattern 10), through PulseNet, the national molecular subtyping network for foodborne disease surveillance. This report summarizes the investigation by CDC, state and local health and agriculture departments, and the U.S. Department of Agriculture’s Food Safety and Inspection Service (USDA-FSIS) and discusses the possible role of dairy cows as a reservoir for strains of Salmonella that persistently cause human illness. This investigation combined epidemiologic and whole genome sequencing (WGS) data to link the outbreak to contaminated ground beef; dairy cows were hypothesized to be the ultimate source of Salmonella contamination.

A case was defined as infection with Salmonella Newport with PFGE pattern 10 closely related to the outbreak strain by WGS, with bacterial isolation during October 1, 2016, through July 31, 2017. A total of 106 cases were identified in 21 states (Figure 1). Most illnesses ([72%]) were reported from southwestern states, including Arizona (30), California (25), New Mexico (14), and Texas (seven). Illness onset dates ranged from October 4, 2016, through July 19, 2017. Patients ranged in age from <1–88 years (median = 44 years), and 53 (50%) were female. Among 88 (83%) patients with known outcomes, 42 (48%) were hospitalized, and one died.

Initial interviews identified consumption of ground beef as a common exposure among patients. A focused questionnaire was developed to collect detailed information on ground beef exposure and to obtain shopper card information and receipts. Among 65 interviewed patients, 52 (80%) reported eating ground beef at home in the week before illness began. This percentage was significantly higher than the 2006–2007 FoodNet Population Survey, in which 40% of healthy persons reported eating ground beef at home in the week before they were interviewed (p<0.001) (1). Among the 52 patients who ate ground beef at home, 31 (60%) reported that they bought it or maybe bought it from multiple locations of two national grocery chains, and 21 (40%) reported that they bought ground beef from locations of 15 other grocery chains. Specific ground beef information was available for 35 patients. Among these, 15 (43%) purchased ground beef as chubs (rolls) of varying sizes (range = 2–10 lbs), 18 purchased it on a tray wrapped in plastic, and two purchased preformed hamburger patties. Twenty-nine patients reported that they bought fresh ground beef, four bought frozen ground beef, and four did not recall whether it was fresh or frozen when purchased. When asked about ground beef preparation, 12 (36%) of 33 patients reported that they definitely or possibly undercooked it.

Traceback Investigation

USDA-FSIS conducted traceback on ground beef purchased within 3 months of illness onset for 11 patients who provided shopper card records or receipts. Approximately 20 ground beef suppliers belonging to at least 10 corporations were identified; 10 of the 11 records traced back to five company A slaughter/processing establishments, seven of 11 traced back to five company B slaughter/processing establishments, and four of 11 traced back to two company C slaughter/processing establishments.

Product and Animal Testing

Opened, leftover samples of ground beef from three patients’ homes were collected for testing. All were purchased from one of two national grocery chains that had been identified by a majority of patients. One sample, collected from ground beef removed from its original packaging, yielded the outbreak strain. The other two samples did not yield Salmonella.

The outbreak strain was also isolated from four New Mexico dairy cattle. One was collected from a spontaneously aborted fetus in July 2016, and one was isolated from feces from a young calf in November 2016. The third isolate was identified by searching the USDA Animal and Plant Health Inspection Service National Veterinary Services Laboratory (USDA-APHIS NVSL) database for Salmonella Newport isolates collected from cattle in Arizona, California, Texas, New Mexico, and Wisconsin during January 2016–March 2017. Eighteen Salmonella Newport isolates were identified, including 13 from Texas, three from New Mexico, and two from Wisconsin. The only Salmonella Newport pattern 10 isolate identified was from a fecal sample from a New Mexico dairy cow collected during November 2016. The fourth isolate was from a USDA-FSIS routine cattle fecal sample collected at a Texas slaughter establishment in December 2016; USDA-FSIS determined the sample was from a dairy cow and identified the New Mexico farm of origin. Because of confidentiality practices, officials were not able to identify the farm or farms of origin for the dairy cows associated with the other three samples or whether the four dairy cows were associated with a single farm. None of the 11 patients with information for traceback ate ground beef produced at the Texas slaughter establishment.

Whole genome high-quality single nucleotide polymorphism (SNP) analysis* showed that 106 clinical isolates were closely related to each other genetically, to the four dairy cattle isolates, and to the leftover ground beef isolate (range = 0–12 SNP differences), suggesting that the Salmonella bacteria found in patients, ground beef, and dairy cattle all shared a common source. Thirty-nine additional clinical isolates with PFGE pattern 10 were determined to not be closely related and were excluded from the outbreak. No antibiotic resistance was detected among three clinical isolates tested by CDC’s National Antimicrobial Resistance Monitoring Laboratory.

Because the USDA-FSIS traceback investigation did not converge on a common production lot of ground beef or a single slaughter/processing establishment, and no ground beef in the original packaging yielded the outbreak strain, a recall of specific product was not requested. A public warning was not issued to consumers because specific, actionable information was not available (e.g., a specific brand or type of ground beef). Officials in New Mexico visited the dairy farm that was the source of the cow at the Texas establishment and noted no concerns about conditions or practices. However, this visit occurred late in the investigation, and conditions at the time of the visit might not have represented those present immediately before and during the outbreak. No samples from the environment or cows were collected during this visit.

Epidemiologic and laboratory evidence indicated that contaminated ground beef was the likely source of this protracted outbreak of Salmonella Newport infections. A significantly higher percentage of patients than expected ate ground beef at home, and a patient’s leftover ground beef yielded the outbreak strain. Dairy cows colonized or infected with the outbreak strain before slaughter are hypothesized to be the ultimate outbreak source. Most U.S. ground beef is produced from beef cattle; however, 18% is produced from dairy cows (2). Dairy cows are sold for beef production through sale barns or directly to slaughter establishments as they age or if their milk production is insufficient (2). Previous studies have demonstrated long-term persistence of Salmonella Newport in dairy herds (3,4), and a 1987 Salmonella Newport outbreak was linked to contaminated ground beef from slaughtered dairy cows (5). In the current outbreak, as has been observed in previous outbreaks, ground beef purchases traced back to numerous lots and slaughter/processing establishments (6). One possible explanation is that dairy cows carrying a high Salmonella load that overwhelmed antimicrobial interventions could have gone to multiple slaughter/processing establishments (7), resulting in contamination of multiple brands and lots of ground beef. This might explain the reason for failure to identify a single, specific source of contaminated ground beef.

This investigation identified the outbreak strain only in samples from dairy cattle from New Mexico. All four isolates from dairy cattle samples were closely related genetically by WGS to isolates from patients, providing further evidence of a connection between dairy cattle in New Mexico and the outbreak. The disproportionate geographic distribution of cases in the U.S. Southwest, including New Mexico, also suggests a possible regional outbreak source. Although limited in scope, the query of the USDA-APHIS NVSL data identified the outbreak strain only from one New Mexico dairy cow (isolate 3), and the sample collection date was consistent with the timing of illnesses in this outbreak. The overall prevalence and geographic distribution of the outbreak strain in cattle is not known, and it is possible that cattle in states other than New Mexico might have been infected or colonized with the outbreak strain.

This was a complex and challenging investigation for several reasons. First, the PFGE pattern in the outbreak was not uncommon in PulseNet, making it difficult to distinguish outbreak cases from sporadic illnesses associated with the same Salmonella Newport pattern. WGS analysis provided more discriminatory power to refine the outbreak case definition and excluded 39 cases of illness from the outbreak. However, sequencing is not currently performed in real time for Salmonella, thereby slowing the process of determining which cases were likely outbreak-associated. In addition, a direct pathway linking outbreak cases to dairy cows infected with the outbreak strain of Salmonella Newport could not be established. This is because product traceback did not converge on a single contaminated lot of ground beef, and investigators were unable to ascertain a link between the beef slaughter/processing establishments identified during traceback and the farms with dairy cows that yielded the outbreak strain. Tracing back ground beef purchased by patients to slaughter/processing establishments requires documentation such as receipts or shopper card records, and only 10% of patients had this information available. For this outbreak, tracing back cows at slaughter/processing establishments to the farm from which they originated was problematic because cows were not systematically tracked from farm to slaughter/processing establishments.

Four points along the “farm to fork” continuum provide opportunities to prevent consumers from becoming ill from contaminated ground beef. First, farms can implement good management practices for cattle health, including vaccination, biosecurity (e.g., controlling movement of persons and animals on farms, keeping a closed herd [so that no animals on the farm are purchased, loaned to other farms, or have contact with other animals], planning introduction of new animals and quarantining them, and performing microbiologic testing of animals), and cleaning and disinfection measures to decrease Salmonella burden in animals and the environments in which they reside, reducing the likelihood that Salmonella will enter beef slaughter/processing establishments (8). Second, slaughter/processing establishments are required to maintain Hazard Analysis and Critical Control Points systems to reduce Salmonella contamination as well as slaughter and sanitary dressing procedures to prevent carcass contamination (9). Third, although Salmonella is not considered an adulterant in not-ready-to eat (NRTE) meat products, USDA-FSIS likely will consider the product to be adulterated when NRTE meat products are associated with an outbreak (9). Finally, consumers are advised to cook ground beef to 160°F (71°C) as measured by a food thermometer to destroy any bacteria that might be present. Consumers are also advised to wash hands, utensils, and surfaces often; separate and not cross-contaminate foods; and refrigerate foods promptly and properly.

This investigation emphasizes the utility of WGS during outbreak investigations and identifies the need for improvements in traceability from the consumer to the farm. It also highlights the importance of continued evaluation of farm practices to help reduce persistent Salmonella contamination on farms, contamination of ground beef, and ultimately human illness.

Protracted outbreak of Salmonella Newport infections linked to ground beef: Possible role of Dairy Cows-21 states, 2016-2017

CDC

https://www.cdc.gov/mmwr/volumes/67/wr/mm6715a2.htm

Kis Robertson Hale, Food Safety and Inspection Service, U.S. Department of Agriculture; territorial, state, city, and county health departments and laboratories; Danya Alvarez, John Crandall, Hillary Berman-Watson, California Department of Public Health Microbial Diseases Laboratory.

 

Going public, Salmonella-in-French-cheese-style: Morbier and Mont d’Or cheese behind 10 deaths in France, 2015-16

In a country where reporting foodborne illness is deemed unpatriotic an investigation by France Inter radio revealed that at least 10 people died in the Franche-Comté region in the east of France linked to two cheeses made from unpasteurized milk  in late 2015 and early 2016.

The investigation produced a document which showed that in January 2016 national health authorities had discovered an unusually high number of salmonella contaminations in France that was centred on Franche-Comté.

Five cheese making companies in the region, between them making 60 different brands, were later identified as being at the source of the contaminations that began in November 2015 and continued until April the following year.

In a way that is truly French in its description, those who died in the outbreak were old people who were physically weak or who suffered from another illness.

Jean-Yves Mano, the president of the CLCV consumer association, said he was surprised that a product recall had not been ordered of products that might have been infected with salmonella.

“We do not understand why a general alert was not issued by state officials, or at least information given on what precautions to take,” he told France Inter.

The state food agency, the Direction générale de l’alimentation (DGAL), said there were two reasons why a recall was not ordered.

The first was that it would have allegedly been very difficult to identify which exact brand of the cheeses were contaminated because there were a total of 60 that were produced in the cheese-making firms where the outbreak originated.

The second was that by the time the authorities found out where the outbreak had come from, the contaminated cheeses had already been consumed and the new batches in the cheesemakers’ premises were not infected.

“It is perhaps due to these two factors that this contamination was not in the media, even though all the data was public nothing was hidden,” said Fany Molin of the DGAL food agency.

That’s French-bureau-speak.

Go public: Further illnesses may be prevented; others learn; citizens may not come with torches demanding change; and it’s the right thing to do.

Going public: Early disclosure of food risks for the benefit of public health

Mar.17

NEHA, Volume 79.7, Pages 8-14

Benjamin Chapman, Maria Sol Erdozaim, Douglas Powell

http://www.neha.org/node/58904

Often during an outbreak of foodborne illness, there are health officials who have data indicating that there is a risk prior to notifying the public. During the lag period between the first public health signal and some release of public information, there are decision makers who are weighing evidence with the impacts of going public. Multiple agencies and analysts have lamented that there is not a common playbook or decision tree for how public health agencies determine what information to release and when. Regularly, health authorities suggest that how and when public information is released is evaluated on a case-by-case basis without sharing the steps and criteria used to make decisions. Information provision on its own is not enough. Risk communication, to be effective and grounded in behavior theory, should provide control measure options for risk management decisions. There is no indication in the literature that consumers benefit from paternalistic protection decisions to guard against information overload. A review of the risk communication literature related to outbreaks, as well as case studies of actual incidents, are explored and a blueprint for health authorities to follow is provided.

Going public (not): One dead 33 sick from Salmonella at Colorado restaurant in November

The Tri-County Health Department said one person has died from salmonella poisoning related to eating at a restaurant in Aurora (not the Aurora, Ill. of Wayne’s World). The outbreak put three other people in the hospital.

The health department said 33 people were sickened by eating at La California restaurant on Peoria Street in November.

The health department said lab tests show the family combination meal might have led to the poisoning. The meal includes ingredients such as meat, beans and cilantro.

But investigators could not pinpoint what started the outbreak.

The health department said the outbreak affected people who ate at La California from Nov. 4-26. Almost all of the people infected experienced diarrhea and abdominal cramping.

La California is at 1685 Peoria St.

The health department’s report said 13 of the 33 cases are confirmed, and 20 of the cases are probable for salmonella.

The illnesses involved 32 restaurant patrons and one employee.

Twenty-five cases had exposures at the restaurant with their meals within a five-day period from Nov. 10-14.

“It was significantly associated with the illness. But we couldn’t ID a single item in the family combo that was associated with the illness. … Everybody ate everything in the family combo,” Tri County Health spokeswoman Jen Chase told CBS Denver.

Going public: US FDA version

The purpose of this guidance is to assist and provide recommendations to industry and FDA staff regarding the use, content, and circumstances for issuance of public warnings and public notifications for firm-initiated or FDA-requested recalls under 21 CFR Part 7, Subpart C – Recalls (Including Product Corrections) – Guidance on Policy, Procedures, and Industry Responsibilities.

The guidance also discusses what information should be included in a public warning, as well as the parties responsible for issuing it.

It represents FDA’s current thinking on public warning and notification of recalls under 21 CFR Part 7. 1 This guidance has been prepared by the Office of Strategic Planning and Operational Policy (OSPOP), in the Office of Regulatory Affairs (ORA), in cooperation with the Center for Biologics Evaluation and Research (CBER), the Center for Drug Evaluation and Research (CDER), the Center for Devices and Radiological Health (CDRH), the Center for Veterinary Medicine (CVM), the Center for Tobacco Products (CTP), and the Center for Food Safety and Applied Nutrition (CFSAN) at the U.S. Food and Drug Administration.

This draft guidance, when finalized, will represent the current thinking of the Food and Drug Administration (FDA or Agency) on this topic. It does not establish any rights for any person and is not binding on FDA or the public. You can use an alternative approach if it satisfies the requirements of the applicable statutes and regulations. To discuss an alternative approach, contact the FDA staff responsible for this guidance as listed on the title page.

This guidance applies to voluntary recalls of products subject to FDA’s jurisdiction, including any food, drug, and device intended for human or animal use, any cosmetic and biologic intended for human use, any tobacco product intended for human use, and any item subject to a quarantine regulation under part 21 Part 1240. However, it does not apply to radiation emitting electronics which are governed by 21 CFR Part 1003 and 1004.

In general, FDA’s guidance documents do not establish legally enforceable responsibilities. Instead, guidances describe the Agency’s current thinking on a topic and should be viewed only as recommendations, unless specific regulatory or statutory requirements are cited. The use of the word should in Agency guidances means that something is suggested or recommended, but not required.

Public warning and notification recalls under 21 CFR part 7, subpart C guidance for industry and FDA staff, 17 January 2018

FDA

https://www.fda.gov/downloads/Safety/Recalls/IndustryGuidance/UCM592851.pdf

Going public: Early disclosure of food risks for the benefit of public health

Mar.17

NEHA, Volume 79.7, Pages 8-14

Benjamin Chapman, Maria Sol Erdozaim, Douglas Powell

http://www.neha.org/node/58904

Often during an outbreak of foodborne illness, there are health officials who have data indicating that there is a risk prior to notifying the public. During the lag period between the first public health signal and some release of public information, there are decision makers who are weighing evidence with the impacts of going public. Multiple agencies and analysts have lamented that there is not a common playbook or decision tree for how public health agencies determine what information to release and when. Regularly, health authorities suggest that how and when public information is released is evaluated on a case-by-case basis without sharing the steps and criteria used to make decisions. Information provision on its own is not enough. Risk communication, to be effective and grounded in behavior theory, should provide control measure options for risk management decisions. There is no indication in the literature that consumers benefit from paternalistic protection decisions to guard against information overload. A review of the risk communication literature related to outbreaks, as well as case studies of actual incidents, are explored and a blueprint for health authorities to follow is provided.

 

Going public: FDA leaves tainted foods on shelves too long, report finds

The U.S. Food and Drug Administration is not moving quickly enough to ensure that contaminated food is removed from store shelves, despite being given the necessary authority, federal investigators have concluded.

The inspector general of the Department of Health and Human Services examined 30 of 1,557 food recalls between 2012 and 2015. The investigators found that the F.D.A. did not always evaluate foodborne hazards in a timely manner or ensure that companies initiated recalls promptly, leaving consumers at risk.

Food companies took an average of 57 days to recall items after the F.D.A. was apprised of the potential hazards. One recall did not begin for 303 days, the investigators said.

“Each and every day is important, because every day the product remains on the shelf, consumers are potentially at risk for serious illness or death,’’ said George Nedder, an assistant regional inspector general at Health and Human Services and lead author of the new report.

The F.D.A. has jurisdiction over most of the food supply in the United States, including virtually all processed food. (The Department of Agriculture oversees meat and poultry, and handles recalls of those products.)

About 48 million Americans get sick from food-borne diseases each year, according to the agency. Of those, 128,000 are hospitalized and 3,000 die.

Food recalls are announced frequently, but the vast majority are voluntary. Sometimes a company comes forward to report contamination, adulteration or mislabeling to the F.D.A. Sometimes the agency approaches a manufacturer after learning of a hazard from a state agency or other source.

In the past few months, the F.D.A. has announced recalls of contaminated apple slices (with listeria), biscotti (undeclared nuts) and adobo sauce (salmonella), among other problems.

The report noted numerous failings, among them “deficiencies in F.D.A.’s oversight of recall initiation, monitoring of recalls,” and in collecting and tracking recall data. Investigators also found that the F.D.A. did always not evaluate health hazards in a timely manner.

Many of the problems detailed in the report were the subject of an unusual “early alert memorandum” from the inspector general’s office in June 2016, which warned that “consumers remained at risk of illness or death for several weeks after F.D.A. was aware of a potentially hazardous food in the supply chain.’’

The new report noted that it took the agency 165 days to start a recall of nut butters that may have been tainted with salmonella; 151 days to recall hazelnuts that also suspected of salmonella contamination; 82 days to recall frozen spinach suspected of high levels of cadmium; and 27 days to start a recall of cooked duck eggs carrying the bacteria that cause botulism.

F.D.A. Commissioner Scott Gottlieb said that the agency had taken to heart the inspector general’s earlier warning and had already started to address recall enforcement problems that have persisted for years.

Dr. Gottlieb said that most food recalls occur within four days of a problem being reported, an assertion that Mr. Nedder disputed.

“Delay is deeply concerning to me as well,” Gottlieb, a physician, said in an interview. “I think it’s working a lot better now.”

FDA will also release guidance in the first half of 2018 on what information it can release about where recalled food and diet supplements were sold. A coalition of safety advocates urged Gottlieb in August to release the names of the stores that sell recalled food items, which FDA has claimed is confidential business information. The Agriculture Department releases it, as do other agencies including the Consumer Product Safety Commission.

FDA Commissioner Scott Gottlieb, MD said in a statement (edited below) that, “One of our most important jobs at the U.S. Food and Drug Administration is ensuring the safety of the U.S. food supply. When we learn about a food in the marketplace that may be unsafe, we must act quickly to keep people from getting sick or being harmed. If foodborne illness has already occurred, we also must act quickly to keep more people from becoming ill. The re-issued, final version of the report by the Office of the Inspector General (OIG), which examined our food recall practices over the time period from Oct. 1, 2012 to May 4, 2015, raised some significant concerns for me. While the FDA has addressed many of the findings after the draft version was first released in 2016, we still have more work to do. I take these obligations very seriously. Making sure the FDA has effective recall practices in place, and that we take immediate action to address unsafe foods, are high priorities of mine. Our recall authorities – and how we deploy them – are a cornerstone of our vital, consumer protection mission. …

“Among other steps, the agency will issue guidance on recall communications in the first half of 2018. As one example of the new steps we’re considering, the FDA is examining in what situations it can help consumers get information about the stores and food service locations that may have sold or distributed a potentially unsafe, recalled food, and what company may have supplied the product. If we’re able to disclose this information, consumers would have an easier time knowing if they might have, or have been, exposed to a recalled product that could cause potential risks if it were consumed.”

Here’s some suggestions:

Going public: Early disclosure of food risks for the benefit of public health

Mar.17

NEHA, Volume 79.7, Pages 8-14

Benjamin Chapman, Maria Sol Erdozaim, Douglas Powell

http://www.neha.org/node/58904

Often during an outbreak of foodborne illness, there are health officials who have data indicating that there is a risk prior to notifying the public. During the lag period between the first public health signal and some release of public information, there are decision makers who are weighing evidence with the impacts of going public. Multiple agencies and analysts have lamented that there is not a common playbook or decision tree for how public health agencies determine what information to release and when. Regularly, health authorities suggest that how and when public information is released is evaluated on a case-by-case basis without sharing the steps and criteria used to make decisions. Information provision on its own is not enough. Risk communication, to be effective and grounded in behavior theory, should provide control measure options for risk management decisions. There is no indication in the literature that consumers benefit from paternalistic protection decisions to guard against information overload. A review of the risk communication literature related to outbreaks, as well as case studies of actual incidents, are explored and a blueprint for health authorities to follow is provided.

Australia still sucks at going public: Salmonella outbreak sickened 100 in March, shows up in annual report yesterday

In early March, 2017, more than 100 people reported becoming sick after dining at Ricardo’s Cafe in Canberra.

The outbreak was briefly noted in media accounts – only because so many people took to facebook – and then disappeared from public discussion.

Until yesterday.

According to ACT Health’s annual report there was a “disturbing” salmonella outbreak linked to Ricardo’s Cafe earlier this year.

Daniella White of The Age reports Ricardo’s was listed in the ACT Health report. 

None of the food outlets have faced prosecution, but ACT Health said it would provide reports on all three cases to the ACT Director of Public Prosecutions.

The biggest outbreak was associated with where 100 people reported suffering from gastroenteritis with 75 confirmed cases of Salmonella.

Of the confirmed cases 19 people were hospitalised, ACT Health said.

A second outbreak of Salmonella was also investigated about the same time with four confirmed salmonella infections from people who reported eating at Central Cafe in Gungahlin between January 30 and February 2.

Both cafes were found to have flaws in their food handling processes and procedures and forced to temporarily close.

Ricardo’s Cafe’s owner previously stated salmonella had been found on a dish cloth and tea towel in the cafe.

ACT Health said both cafes have been inspected since the outbreaks and found to be compliant.

A third Salmonella outbreak investigation was conducted in February and March, with 11 cases where people ate at the same restaurant over a five week period.

Several inspections of the premises did not identify any issues and the source of that outbreak remains unknown, ACT Health said.

It declined to identify the food outlet given it was found to be compliant and could not be held responsible for the outbreak.

An ACT Health spokesman said any foodborne outbreak was taken seriously by the Health Protection Service.

Associate professor Martyn Kirk, from ANU’s College of Health and Medicine said the food service industry had a responsibility to make sure it handled food safely and avoided high risk foods, such as raw eggs and improperly cured foods.

Going public: Early disclosure of food risks for the benefit of public health

Mar.17

NEHA, Volume 79.7, Pages 8-14

Benjamin Chapman, Maria Sol Erdozaim, Douglas Powell

http://www.neha.org/node/58904

Often during an outbreak of foodborne illness, there are health officials who have data indicating that there is a risk prior to notifying the public. During the lag period between the first public health signal and some release of public information, there are decision makers who are weighing evidence with the impacts of going public. Multiple agencies and analysts have lamented that there is not a common playbook or decision tree for how public health agencies determine what information to release and when. Regularly, health authorities suggest that how and when public information is released is evaluated on a case-by-case basis without sharing the steps and criteria used to make decisions. Information provision on its own is not enough. Risk communication, to be effective and grounded in behavior theory, should provide control measure options for risk management decisions. There is no indication in the literature that consumers benefit from paternalistic protection decisions to guard against information overload. A review of the risk communication literature related to outbreaks, as well as case studies of actual incidents, are explored and a blueprint for health authorities to follow is provided.

Going public: Embrace it; supermarket x’ and the ethics of food risk disclosure

I figured out about 45 years ago that going public was the best way to handle conflict.

I was throwing stones at the grill of a car in an abandoned lot in Brantford, Ontario (that’s in Canada), trying out cigarettes for the first time, and we got caught.

When I was 18-years-old, and went to jail, I had nothing to hide.

All incoming inmates had to strip naked in front of some deliberately inattentive guard, as part of admission.

There was nothing left to hide, and there isn’t now.

I know lots of people who hide behind their vanity; I know few whom are truly public, warts and all.

Whether it’s some academic institution or a local hockey club, bring it on; your rumors only ferment your own failures.

Julia Glotz of The Grocer in the UK writes that in the hepatitis E outbreak, supermarket X is Tesco.

Julia is referring to explosive claims that, as initially reported by the Sunday Times, hepatitis E is sickening Brits at an alarming rate.

But that didn’t stop Public Health England from sticking its fingers in its ears and loudly singing ‘la la la’ for the best part of a week. It didn’t want to name ‘supermarket x’ because the findings of its hepatitis E study do ‘not infer blame on the supermarket’.

Oh okay. Fair enough. I see the strategy here. So instead of some uncomfortable articles about a possible link between hep E infections and retail pork products, we’re going for a week of near-hysterical coverage shouting Silence On Banger Virus, and insinuating an agency tasked with protecting public health is refusing to drop a big supermarket in it. What must these brutes at Tesco be doing to poor PHE to force it into silence? Round of applause, everyone.

Anyway, the kitty is now firmly out of the bag, so it’s time to reflect on some lessons from the whole saga. After all, hepatitis E is far from the only food scandal in town right now. And, in PHE’s defence, this is really tricky territory. Risk communication is notoriously hard, and informing the public about risks related to food can be especially thorny. Food regulators and agencies across the globe struggle with it.

Doug Powell, a former professor of food safety who runs the wonderfully named BarfBlog (yes, it’s about people getting sick from dodgy food), has an entire section on his website called ‘Going public’. In it, he chronicles example after example of public disclosure foul-ups from food regulators around the world: vague press releases rushed out after hours; vital information made available too late for the sake of double, triple and quadruple checking; and bare facts and stats dumped on the public without context to help them make sense of risk.

Powell is also the co-author of a fascinating article published in the Journal of Environmental Health earlier this year, which examines how regulators approach public comms during food scares. It concludes there is widespread confusion – and hugely varying standards – largely because there is little concrete guidance on food risk disclosure. Too many decisions are made ad hoc; what little guidance there is is couched in flabby, unhelpful language like ‘timely release of information’, without defining what ‘timely’ means.

So does that mean PHE should have named Tesco right from the start? Not necessarily. Its study did not establish that Tesco products were the direct cause of hepatitis infections; it was a statistical analysis only. There are valid concerns about the potential for premature information to mislead consumers and do significant harm (though it’s worth pointing out Powell’s report found no evidence to support a paternalistic approach where members of the public are protected from ‘too much’ information). ‘Naming and shaming’ of individual companies could potentially distract consumers from other, more important information about how they can protect themselves. Might a shopper who doesn’t buy ham and sausages at Tesco, for example, tune out and not pay attention to cooking advice to minimise the risk of hepatitis E infections?

It’s a fair question. So by all means don’t name names in your scientific paper. But accept that a paper like it will very likely get media pick-up – and once it’s out in the public domain, the game is up. You have to come clean. “To support consumer decision making, available information must support what consumers want and need to know, which might or might not be what information authorities want to relay,” Powell writes in his JEH article. And in this case, consumers wanted to know who ‘supermarket x’ was.

“While PHE may have reasons for not going public – whether legislative or bureaucratic – whoever is involved will eventually be found out,” Powell added when I caught up with him about the case this week. “Trust is earned, not given, and consumers have a right to know – especially if it’s a public health issue like hepatitis E in pork.”

 It’s not all down to the regulators, though. Industry, too, has a part to play in this. If there’s any risk to public health, information disclosure must be led by the relevant agencies, says Chris Elliott of the Institute for Global Food Security. But when there is no (or a very remote chance) of a food safety problem, “then I think supermarket x should take the lead in the comms,” he adds. “But this should be agreed with the regulator, so no contradiction or confusion should result.”

It’s undoubtedly a difficult balance to get right. In Powell’s article, the killer line comes from epidemiologist Paul Mead: “Food safety recalls are always either too early or too late. If you’re right, it’s always too late. If you’re wrong, it’s always too early.”

So yes, it can be a poisoned chalice. But regulators can’t dodge it – and they have to start dealing with it more assuredly than they often do at the moment. 

Going public: Early disclosure of food risks for the benefit of public health

Mar.17

NEHA, Volume 79.7, Pages 8-14

Benjamin Chapman, Maria Sol Erdozaim, Douglas Powell

http://www.neha.org/node/58904

Often during an outbreak of foodborne illness, there are health officials who have data indicating that there is a risk prior to notifying the public. During the lag period between the first public health signal and some release of public information, there are decision makers who are weighing evidence with the impacts of going public. Multiple agencies and analysts have lamented that there is not a common playbook or decision tree for how public health agencies determine what information to release and when. Regularly, health authorities suggest that how and when public information is released is evaluated on a case-by-case basis without sharing the steps and criteria used to make decisions. Information provision on its own is not enough. Risk communication, to be effective and grounded in behavior theory, should provide control measure options for risk management decisions. There is no indication in the literature that consumers benefit from paternalistic protection decisions to guard against information overload. A review of the risk communication literature related to outbreaks, as well as case studies of actual incidents, are explored and a blueprint for health authorities to follow is provided.

31 sickened by E. coli O55 in Dorset: 3 years later, health-types’ report remains a secret

In Dec. 2014, an outbreak of E. coli O55 was identified in Dorset, U.K. with at least 31 sickened. Public Health England (PHE) and local environmental health officials investigated and found nothing, other than cats were also being affected.

Tara Russell of Bournemouth Echo reports again this week that a review into the outbreak in Dorset was carried out, health chiefs have insisted – but the report is not available to the public.

Public Health England (PHE) says the public can only request to see the report detailing exactly what happened when 31 people contracted the O55 strain between July 2014 and November 2015 through a Freedom of Information request.

Families including some whose children have been left with lifelong health complications say they did not know the review existed and have branded it ‘disappointing and disgusting’ they have been kept in the dark.

The Daily Echo has lodged an official FOI request on behalf of the affected families and will receive a response in July.

Nurse Jessica Archer, who today suffers crippling head pains, fatigue and depression while her nephew Isaac Mortlock (right) endures severe seizures, must be peg fed every night and will need a kidney transplant as a result of the outbreak, said: “Without the Daily Echo we wouldn’t even know this report even existed and we are very interested to see it and we have the right to know. The families affected have so many unanswered questions and have to live with the effects of this outbreak forever but yet again we feel Public Health England are trying to sweep it under the carpet and hope that it will just go away.

“It is disappointing and disgusting this report has not already been made public let alone having to wait and wait still. We feel there have been a series of failures and this is the latest.”

The news comes after Jessica last month called for PHE to be held to account telling how her and her five-year-old nephew’s Isaac Mortlock’s lives have changed irreversibly, and accused the organisation of ‘a cover up.’

In response, PHE told the Daily Echo it carries out ‘routine outbreak reviews once investigations have ended’, adding it is ‘a learning organisation and reflects on outbreaks to identify lessons learnt and to continually improve our response.’

However at the time, the organisation refused to tell the Daily Echo exactly which lessons were learned.

It was only following a further request from this newspaper, PHE said a report was compiled however it has not been available to the public.

A spokesman said: “This report was not intended for external publication – it’s not standard procedure to publish outbreak reports externally due to patient confidentiality – however if interested parties would like to request a copy they can do this via our Freedom of Information portal.”

That’s bullshit.

Outbreak investigations are routinely published while ensuring patient confidentiality.

Families say it is the latest in a string of ‘failures’ by Public Health England.

A spokesman from PHE added: “As with all outbreaks, PHE Health Protection Team ensured throughout their investigation that those affected were kept informed of any information that was uncovered at that time.”

That’s also bullshit.

And why UK health types feature prominently in our paper on when to go public for the benefit of public health.

Three years seems a bit long.

Going public: Early disclosure of food risks for the benefit of public health

Mar.17

NEHA, Volume 79.7, Pages 8-14

Benjamin Chapman, Maria Sol Erdozaim, Douglas Powell

http://www.neha.org/node/58904

Often during an outbreak of foodborne illness, there are health officials who have data indicating that there is a risk prior to notifying the public. During the lag period between the first public health signal and some release of public information, there are decision makers who are weighing evidence with the impacts of going public.

Multiple agencies and analysts have lamented that there is not a common playbook or decision tree for how public health agencies determine what information to release and when. Regularly, health authorities suggest that how and when public information is released is evaluated on a case-by-case basis without sharing the steps and criteria used to make decisions.

Information provision on its own is not enough. Risk communication, to be effective and grounded in behavior theory, should provide control measure options for risk management decisions. There is no indication in the literature that consumers benefit from paternalistic protection decisions to guard against information overload. A review of the risk communication literature related to outbreaks, as well as case studies of actual incidents, are explored and a blueprint for health authorities to follow is provided.

Going public: Why FDA doesn’t share retail outlet information (but should) edition

Earlier this month Doug and I had a paper published in the Journal of Environmental Health about the need for public health folks (especially at the local and state levels) who are dealing with an outbreak to have a plan on when to go public. The plan should include what info the release; how they release it; and, what triggers release.

There are a bunch of great folks in these agencies who are often understaffed, overworked and dealing with political pressures – but often don’t look to the risk communication world for tips on this stuff.

Caitlin Dewey at the Washington Post wrote about why FDA doesn’t practice sharing outlets/retailers where recalled products are sold, even those linked to illnesses.

The FDA does not specify, however, which stores, centers or schools — because that would violate its interpretation of an obscure trade secret rule.

This interpretation differs from that of other agencies in the federal food safety system, an overlapping and often illogical network of regulatory fiefdoms. The system, which is responsible for keeping food free of bacteria and other pathogens, frequently has to weigh the very real interests of private food companies against potential risks to the public. In the case of releasing retailer lists during major outbreaks, the FDA has historically sided with business, ruling that such lists constitute “confidential commercial information” and thus should not be available for public consumption.

Critics say that the agency’s unwillingness to share this information poses a clear danger to public health, particularly in cases like the current E. coli outbreak, where parents may not know if their child consumed the recalled product. According to the Centers for Disease Control and Prevention, 14 of the 16 people who have fallen ill were children.

It could also prove relevant in incidents like last year’s multistate hepatitis A outbreak, which was traced to frozen strawberries imported from Egypt and sold at several Tropical Smoothie Cafes. The FDA did not specifically reveal which locations, however — a measure that some experts say would have gotten the news to ill consumers faster. That’s important in the case of an illness like hepatitis A, which can be treated with a vaccine for a limited period after exposure.

Doug and I argue that public health agencies (like FDA) should be in the business of sharing the info they have, the info they don’t have and all the uncertainties. This includes distribution data. There are lots of ways that folks get food safety and recall information. Sometimes it’s directly from their retailer of choice; or maybe it comes from a local media source. Or someone shared something on Facebook.

Bill Hallman and colleagues at Rutgers conducted a survey of consumers and their self-reported behaviors following 2008’s Salmonella saintpaul in tomatoes, er, peppers outbreak and found that lots of people (81%) say they they share recall info when they see it. 38% believe that the food they purchase is less likely to be recalled than their neighbors. And less than 60% report checking their fridges and pantries for the food.

Releasing retailer/distribution information might increase the chance individuals will say to themselves ‘I’ve bought some soynut butter recently, and I got it at that grocery store’ and they go check.

But I could just be optimistic.

A food safety Facebook friend posted a couple of days ago about a Listeria monocytogenes-linked recall. It was so important to him that he posted the info twice once on Feb 17 and again March 11, ‘I want to again stress that you should check any production codes immediately and if you have any of these products, either throw them away or return them to your grocery store. I just checked my cheese stash and had the pepperjack slices that are included.’

But, like Hallman and colleagues found, while he shared the info, he wasn’t motivated to actually go to his fridge to look for it the first time he posted. Maybe the distribution information would have triggered a behavioral response.

Going public: have a plan

‘Food safety recalls are always either too early or too late. If you’re right, it’s always too late. If you’re wrong, it’s always too early.’
That’s what Paul Mead was quoted as saying in response to when to go public with outbreak information over a decade ago. 
During foodborne illness outbreaks and incidents information is evolving – what people know, and when the share it can impact public health, and buyer decisions. Go public too late and stuff remains on the market. Go public too early risks making a wrong decision
Doug, Sol Erdozian and I wrote a paper a while back that finally got published this week in the Journal of Environmental Health where we look at the going public situation. There’s no magic answer; just have a plan and a set of criteria to look at when making the decision of what to share when. Talk about uncertainty. And don’t make it up on the fly.
Here’s NC State’s press release about our paper:

There’s an ongoing debate among public health officials about how quickly they should notify the public about foodborne illness outbreaks, and how much information should be shared. Is it better to tell people as early as possible, or could that create panic that is counterproductive?

Food safety researchers are now calling on public health agencies to develop clear guidelines on when to inform the public about foodborne illness outbreaks – something which is often handled on an ad hoc basis at the local, state and federal levels.

To learn more about how health agencies are currently addressing these questions, researchers evaluated 11 case studies of large outbreaks, dating back to 1996.

Not only is there no clear consensus on how to respond, they discovered, but there is no system in place to help officials decide when to tell the public about a foodborne illness outbreak.

“We found that pressure from social media, or from companies, has sometimes influenced when health officials release information, which is problematic,” says Ben Chapman, lead author of a paper on the work and an associate professor of agricultural and human sciences at North Carolina State University.

“Officials need to have clearly defined processes for determining when information should be made public, and those processes don’t appear to exist right now,” Chapman says.

Instead, researchers found that public health agencies – from the federal Centers for Disease Control and Prevention to state and local agencies – make decisions about sharing information on a case-by-case basis.

“There are advantages and risks to both sharing and withholding information,” Chapman says.

Sharing information early in an outbreak can allow consumers to make informed decisions about their food choices that limit risk. But there can also be a lot of uncertainty about that information.

“For example, officials may be investigating a particular restaurant or type of food, but the investigation could ultimately find that the culprit was actually a different source altogether,” Chapman says.

By the same token, withholding information until there is less uncertainty may increase public health risks because the source of the illness may remain accessible to unwitting consumers.

“The best case is to share what you know, and what you don’t know, in an open and transparent way,” Chapman says. “Talking about uncertainty may be uncomfortable for officials, but they need to have a plan for how to do so.”

The paper, “Going Public: Early Disclosure of Food Risks for the Benefit of Public Health,” is published in the Journal of Environmental Health. The paper was co-authored by Maria Sol Erdozaim, a former undergraduate at Kansas State University, and Douglas Powell of Powell Food Safety.

-shipman-

Note to Editors: The study abstract follows.

“Going Public: Early Disclosure of Food Risks for the Benefit of Public Health”

Authors: Benjamin Chapman, North Carolina State University; Maria Sol Erdozaim, Kansas State University; Douglas Powell, Powell Food Safety

Published: March 2017, Journal of Environmental Health

Abstract: Often during an outbreak of foodborne illness, there are health officials who have data indicating that there is a risk prior to notifying the public. During the lag period between the first public health signal and some release of public information, there are decision makers who are weighing evidence with the impacts of going public. Multiple agencies and analysts have lamented that there is not a common playbook or decision tree for how public health agencies determine what information to release and when. Regularly, health authorities suggest that how and when public information is released is evaluated on a case-by-case basis without sharing the steps and criteria used to make decisions. Information provision on its own is not enough. Risk communication, to be effective and grounded in behavior theory, should provide control measure options for risk management decisions. There is no indication in the literature that consumers benefit from paternalistic protection decisions to guard against information overload. A review of the risk communication literature related to outbreaks, as well as case studies of actual incidents, are explored and a blueprint for health authorities to follow is provided.