Why so many outbreaks in 2018? I’m not sure there were, but math is hard

Don and I recorded a podcast today and one of the potential show titles was math is hard.

I’m writing from experience, I had to take calculus in high school twice to get it. Explaining probability and risk is even tougher.

Some of the math of food safety illness burden comes down to this for me: There are billions of meals every year in the U.S. that don’t lead to foodborne illness.

That’s good.

There are millions of meals every year in the U.S. that do lead to foodborne illness.

That’s bad.

There was some buzz in the food and health media after a CNN article stated that CDC had investigated more multistate outbreaks in 2018, 21 in total, than any previous year.

That’s sorta true, but sorta not, depends on how it’s counted.

Rachel Rettner of Live Science talked to me about this last week and I shared the National Outbreak Reporting System (NORS) dashboard with her. This is a super cool tool that CDC has where they track all the reported outbreaks in the U.S. and the number isn’t 21. It’s probably closer to 4000. At least that’s around what has been reported into NORS annually since 2009 when reporting got better.

I’m throwing out the multistate part in my calculations, because the microbes don’t care about state lines or borders.

From the Live Science article:

Experts say that, although we heard a lot about foodborne disease in 2018, it doesn’t mean that we had any more outbreaks than usual. Indeed, it’s likely that the U.S. always has about the same number of outbreaks every year, said Benjamin Chapman, an associate professor and food safety specialist at North Carolina State University. But critically, health officials are getting better at detecting these outbreaks, Chapman said, leading to an increase in reported outbreaks in recent years.
“The science is getting better, and the public health resources are getting better, and we’re just getting better at finding things,” Chapman told Live Science

And although these outbreaks made headlines, there are hundreds more outbreaks that we don’t necessarily hear about that get investigated and reported every year. (An outbreak refers to an instance when two or more people get the same illness from the same contaminated food or ingredient, according to the CDC.)
Indeed, according to the CDC’s National Outbreak Reporting System, which summarizes data on U.S. reports of foodborne illness, there were about 4,000 foodborne illness outbreaks each year from 2012 to 2016, (the most recent years for which data is available). That’s up from only about 1,000 reported outbreaks in 2008.
That “looks like this big jump” in outbreaks, Chapman said. But the increase is really due to health officials getting better at “connecting the dots” to find more foodborne illness outbreaks, he said. In other words, the outbreaks were happening, but health officials just weren’t as good as detecting them.

Unfortunately, better detection of outbreaks means that the total number of reported outbreaks likely won’t be going down anytime soon.
“As we get better at reducing risk [of foodborne illness], we also get better at finding things we didn’t know were there,” Chapman said. “I don’t expect that we would have any less or any more outbreaks in 2019.”

My leafy green conundrum

Caesar salad isn’t even the best salad.

It’s the kind of salad you expect at a sports team banquet or during lunch at an all-day meeting.

It’s a safe menu choice.

Except for the past month when foodborne illness outbreak investigators have focused on Romaine lettuce as the culprit of an outbreak of E. coli O157 linked to over 170 illnesses in 32 states.

E. coli O157 historically was once only associated with ground beef making it’s first appearance as what was thought to be a rare strain in 1982 after an outbreak was traced to McDonald’s. In 1993 over 500 illnesses and 4 deaths were linked to Jack-in-the-Box restaurants , an event that brought foodborne illness to the national stage. Fast forward 35+ years and the devastating pathogen has caused illnesses after being consumed in cookie dough, hazelnuts, alfalfa sprouts, soy nut butter, chicken salad – and a whole bunch of fresh produce including leafy greens.

The very type of food we should eat more of betrays us at a higher rate compared to other foods: Fresh produce is believed to be the source of almost half of all foodborne illnesses in the U.S.

Because it is consumed raw, anything that fresh fruit and vegetables come in contact with from the field to the home can really only increase risk. Washing and rinsing can remove at most 99% of what’s there. Microbiologically speaking, because there may be tens of thousands of cells on a leaf of Romaine, that’s not a whole lot. Often produce-related outbreaks are linked to poop getting into the food somewhere — wildlife on the farm; water used for irrigation or rinsing; soil and/or manure; or, the people who harvest, pack, handle and prepare it.

The problem with this outbreak is that the world of food safety sleuths have yet to figure it out. This one is particularly hard because the supply chain is a mess and  investigators are trying to piece together what the farms and packing facilities looked like, food safety-wise, retroactively. Partners in figuring out outbreaks, the U.S. Centers for Disease Control and Prevention and the U.S. Food and Drug Administration have only been able to trace the problem to a specific geographic location – Romaine that was grown in the Yuma, AZ region.

Having a good sense of the supply chain for food, so outbreaks and incidents can be solved, is something that the industry has struggled with for decades. Even with the increased use of electronic records and the promise of blockchain, the data that gets recorded and shared relate to location and how food safety is managed from production to distribution to the grocery store still relies on people to input it.

Better traceability is often held up as a magic bullet but can’t really stop outbreaks from happening alone. Being able to trace a product is wholly reactive. While it is part of a good food safety culture even a good traceability program doesn’t wipe raw poop off of foods.

What keeps food safe is vigilance by the food industry, learning from past outbreaks and focusing on carrying out best practices daily. Lots of food companies talk about food safety. Implementing it daily is much harder. It takes a system throughout the entire company from the front-line staff all the way to the CEO that values food safety. Everyone needs to understand why food safety matters, what their role is and care about the folks who eat their products.

Where I grew up, there was a small tailgate farmers market Saturday mornings in the parking lot adjacent to the grocery store. I never really wondered whether the food sold there was safe. I didn’t think a whole lot about food safety and regulation until years later. I figured that if someone could sell it, they must know what they are doing, and I didn’t have to worry about it.

Food safety is all about trust, and I had lots of it.

I still do.

But over 75 outbreaks linked to leafy greens since 1995 is eroding that trust.

 

Chipotle is getting sued again

Another shareholder who didn’t believe the press releases. According to NOLA.com, 

Chipotle Mexican Grill Inc. is facing another class-action lawsuit by a disgruntled shareholder over its alleged inability to keep its restaurants clean, a new hurdle for a company still trying to regain consumer and investor confidence after a food scare two years ago. The latest complaint follows Tuesday’s revelation (July 18) that one of the burrito chain’s Virginia stores had temporarily closed because of a suspected norovirus outbreak, and Wednesday’s news media reports about Texas customers complaining of rodents dropping from the ceiling.

In March, Chipotle won dismissal of a similar lawsuit brought by investors over stock price drops after several food-borne illness outbreaks in 2015 were traced to its restaurants. A judge concluded that the case was “long on text but it is short on adequately-pleaded claims.” Those plaintiffs are now trying to revive the case in Manhattan federal court.

In the case filed Thursday, shareholder Elizabeth Kelley said the Colorado-based company made misleading statements to bolster confidence that it had resolved the health and safety troubles from 2015, when it was forced to close all of its U.S. stores after hundreds of consumers got sick. Chipotle spokesman Chris Arnold didn’t immediately respond to a request for comment on the lawsuit.

 

Two trichina outbreaks in Alaska; it was the walrus

There’s not a lot of trichina in the U.S. food supply anymore. It used to be a much more important pathogen. In the 1940s, when the US Public Health Service started tracking the illness, there was around 400 cases a year. Now there’s about 20.

A couple of the more notable incidents were reported in MMWR last week – two outbreaks in Alaska linked to raw walrus.

During July 2016–May 2017, the Alaska Division of Public Health (ADPH) investigated two outbreaks of trichinellosis in the Norton Sound region associated with consumption of raw or undercooked walrus (Odobenus rosmarus) meat; five cases were identified in each of the two outbreaks. These were the first multiple-case outbreaks of walrus-associated trichinellosis in Alaska since 1992.

The walrus consumed during the implicated meal in the second outbreak had been harvested and butchered by patients F and I during the previous 1–3 months, and the meat had been stored frozen in unlabeled bags in their respective household chest freezers. The meat was prepared by patient H, who reported that she boiled it for approximately 1 hour, after which the exterior was fully cooked, but the interior remained undercooked or raw, which was the desired result; interviewed persons reported that many community members prefer the taste and texture of undercooked or raw walrus meat to that of fully cooked meat.

These outbreaks also highlight the importance of culturally sensitive public health messaging. In areas where wild game species are harvested for subsistence, traditional methods of collecting, handling, preparing, storing, and consuming meat often have great cultural significance; however, some of these methods can be inconsistent with public health best practices. Rather than promoting or proscribing specific methods, public health messages that focus on communicating risks and explaining the manner and magnitude of risk reduction that can be achieved using different approaches (e.g., alternative methods of preparing meat for consumption) enable members of the target population to make informed decisions that integrate their traditional practices with their awareness and tolerance of risks.

Today in norovirus news

Winter vomiting virus, noro, Norwalk, whatever you want to call it, is all the rage right now. There are so many outbreaks it’s hardly new and novel.

Today, two schools in Denver shut due to an outbreak. And one in Florida.10849902_719581291471357_3442145704847569295_n1-300x3001-300x300-1

A hotel in Ireland also closed for to control the virus through cleaning and disinfection after residents, staff and guests came down with gastrointestinal illness symptoms. Sounds like a memorable wedding.

 

Seek and ye shall find: Shiga-toxin producing E. coli sources of most USDA outbreaks

The U.S. Department of Agriculture’s Food Safety and Inspection Service (FSIS) works closely with federal, state, and local public health partners to investigate foodborne illness outbreaks associated with its regulated products.

regulators-younggunmountupTo provide insight into outbreaks associated with meat and poultry, outbreaks reported to FSIS during fiscal years 2007 through 2012 were evaluated.

Outbreaks were classified according to the strength of evidence linking them to an FSIS-regulated product and by their epidemiological, etiological, and vehicle characteristics. Differences in outbreak characteristics between the period 2007 through 2009 and the period 2010 through 2012 were assessed using a chi-square test or Mann-Whitney U test.

Of the 163 reported outbreaks eligible for analysis, 89 (55%) were identified as possibly linked to FSIS-regulated products and 74 (45%) were definitively linked to FSIS-regulated products. Overall, these outbreaks were associated with 4,132 illnesses, 772 hospitalizations, and 19 deaths.

Shiga toxin–producing Escherichia coli was associated with the greatest proportion of reported outbreaks (55%), followed by Salmonella enterica (34%) and Listeria monocytogenes (7%). Meat and poultry products commercially sold as raw were linked to 125 (77%) outbreaks, and of these, 105 (80%) involved beef. Over the study period, the number of reported outbreaks definitively linked to FSIS-regulated products (P = 0.03) declined, while the proportion of culture-confirmed cases (P = 0.0001) increased.

Our findings provide insight into the characteristics of outbreaks associated with meat and poultry products.

Foodborne outbreaks reported to the U.S. Food Safety and Inspection Service, fiscal years 2007 through 2012

Journal of Food Protection, Number 3, March 2016, Pages 442-447, DOI: http://dx.doi.org/10.4315/0362-028X.JFP-15-376

Robertson, A. Green, L. Allen, T. Ihry, P. White, W. S. Chen, A. Douris, and J. Levine

http://www.ingentaconnect.com/content/iafp/jfp/2016/00000079/00000003/art00011

UPI gets into the Buzzfeed list business: Notable E. coli outbreaks in U.S. fast food restaurants

Most of the U.S. mainstream food safety news coming through Google Alerts over the last week has been recycled and Chipotle-related.

Much of the focus has been on the business with the same questions are being asked by many journos: Will the fast casual Mexican restaurant rebound? What will happen to their stocks? When can we eat there again? Sorta lost in the media are the stories of the folks who went to grab a lunch and ended up ill. The folks that couldn’t go to work, missed life events and may have a long recovery. The affected have been digested down to a list of numbers.Jimmy-Johns-Gourmet-sandwiches

UPI gets into the food safety list business and revisits stigma-creating events over the past 30+ years, here are some highlights:

McDonald’s (1982)
Nearly 50 people in Oregon and Michigan fell ill after eating burgers at McDonald’s. The confirmed outbreak was the first time E. coli O157:H7 was linked to food poisoning, but wouldn’t be the last time ground beef would be recalled for outbreaks of the dangerous pathogen, including Topps Meat Co. recalling nearly 22 million pounds in 2007 and Con Agra Foods pulling nearly 20 million pounds of ground beef in 2002.

Jack-in-the-Box (1993)
The 1993 Jack-in-the-Box outbreak occurred when more than 500 people became infected after eating undercooked beef patties associated with 73 restaurants in Washington, Idaho, California, and Nevada. Four children died and hundreds of customers were left with permanent injuries, including kidney damage, resulting in numerous lawsuits.

Kentucky Fried Chicken (1999)
In July of 1999, public health officials confirmed four Cincinnati-area Kentucky Fried Chickens were to blame for an outbreak of E. coli that led to 18 illnesses and at least 11 hospitalizations. Investigators identified poorly prepared coleslaw as the source of the contamination.

Sizzler (2000)
Two Sizzler restaurants in Wisconsin were responsible for 64 confirmed cases of E. coli and dozens of hospitalizations. Four patients developed hemolytic uremic syndrome, or HUS, a serious illness that can result in kidney failure. One child died. Officials linked the contamination back to watermelon, which was cross-contaminated with raw meat products. Eight years later, the family of the 3-year-old girl who died from exposure at a Sizzler restaurant reached a $13.5 million settlement with the company’s meat supplier.

Taco Bell (multiple)
In December, 2006, 71 illnesses linked to Taco Bell were reported to the CDC from five states: New Jersey, New York, Pennsylvania, Delaware, and South Carolina. Investigations indicated shredded lettuce was the likely source of the outbreak. Two years later, all Taco Bell restaurants in Philadelphia were temporarily closed and green onions removed from all 5,800 of its U.S. restaurants after tests indicated they were to blame for an E. coli outbreak that sickened at least five dozen people in New Jersey, New York and Pennsylvania.

Jimmy John’s (multiple)
In 2013, the Colorado Department of Public Health and Environment identified nine cases of E.coli O157:H7 in the Denver area linked to the consumption of Jimmy John’s sandwiches containing cucumbers imported from Mexico. This wasn’t the first time Jimmy John’s, based in Champaign, Ill., has been linked to an outbreak. Since 2008, the sandwich chain has been cited for serving contaminated sprouts at least five times.

FiveThirtyEight analyzes Chipotle’s norovirus outbreak

I’ve been a fan of Nate Silver’s fivethirtyeight.com for a few years. It’s one of a handful of sites that I read daily. Today the numbers nerds tackled the Chipotle norovirus outbreak comparing it to other fast food outbreaks using CDC’s databases. Turns out this outbreak is unique for its size and location – at a fast food setting.Screen Shot 2015-12-10 at 10.17.38 PM

Just under 3 percent of the food-related norovirus outbreaks that the Centers for Disease Control and Prevention tracked from 2009 to 20141sickened 80 or more people. Only 1.3 percent of outbreaks sickened 140 or more people. Half of all outbreaks sickened 11 or fewer people.

It’s not just the size of the outbreak that’s unusual — it’s where it happened. Although norovirus is the most common cause of food-related illnesses nationwide (about 50 percent of outbreaks), it’s not the most common cause of illnesses at fast-food outlets like Chipotle. Since 2009, when the CDC began tracking where outbreaks occurred, fast-food restaurants have been much more likely to give their patrons salmonella than norovirus (28 percent of illnesses at fast-food chains were linked to norovirus, while 43 percent were linked to salmonella). Sit-down restaurants have the opposite risk profile (45 percent of illnesses linked to norovirus, 25 percent to salmonella).

And here’s some more context from CDC on where norovirus outbreaks happen.

Screen Shot 2015-12-10 at 10.13.16 PM

 

 

Ciders and juices can cause illnesses; here’s a big list

Every fall since 2010 there’s been at least one juice or cider related outbreak in North America. Some beverages were pasteurized, many weren’t. CiderPic1Right now there are two outbreaks linked to unpasteurized ciders in California and Illinois.

Here’s a list of all the ones we’ve been able to find (going back to 1924) – 84 outbreaks leading to over 3500 illnesses.

Click here to download the entire list.

Screen Shot 2015-11-05 at 3.18.55 PM

 

Surveillance for waterborne disease outbreaks associated with drinking water – United States, 2011-2012

The U.S. Centers for Disease Control reports in Morbidity and Mortality Weekly that advances in water management and sanitation have substantially reduced waterborne disease in the United States, although outbreaks continue to occur (1). Public health agencies in the U.S. states and territories* report information on waterborne disease outbreaks to the CDC Waterborne Disease and Outbreak Surveillance System (http://www.cdc.gov/healthywater/surveillance/index.html).

water.wellFor 2011–2012, 32 drinking water–associated outbreaks were reported, accounting for at least 431 cases of illness, 102 hospitalizations, and 14 deaths. Legionella was responsible for 66% of outbreaks and 26% of illnesses, and viruses and non-Legionella bacteria together accounted for 16% of outbreaks and 53% of illnesses. The two most commonly identified deficiencies† leading to drinking water–associated outbreaks were Legionella in building plumbing§ systems (66%) and untreated groundwater (13%). Continued vigilance by public health, regulatory, and industry professionals to identify and correct deficiencies associated with building plumbing systems and groundwater systems could prevent most reported outbreaks and illnesses associated with drinking water systems.

This report provides information on drinking water–associated¶ waterborne disease outbreaks in which the first illness occurred in 2011 or 2012** (http://www.cdc.gov/healthywater/surveillance/drinking-surveillance-reports.html), and summarizes outbreaks reported to the Waterborne Disease and Outbreak Surveillance System through the electronic National Outbreak Reporting System (http://www.cdc.gov/nors/about.html) as of October 30, 2014. For an event to be defined as a waterborne disease outbreak, two or more persons must be linked epidemiologically by time, location of water exposure, and case illness characteristics; and the epidemiologic evidence must implicate water as the probable source of illness. Data submitted for each outbreak include 1) the number of cases, hospitalizations, and deaths; 2) the etiologic agent (confirmed or suspected); 3) the implicated water system; 4) contributing factors in the outbreak; and 5) the setting of exposure.

Karlyn D. Beer, PhD1,2; Julia W. Gargano, PhD2; Virginia A. Roberts, MSPH2; Vincent R. Hill, PhD2; Laurel E. Garrison, MPH3; Preeta K. Kutty, MD3; Elizabeth D. Hilborn, DVM4; Timothy J. Wade, PhD4; Kathleen E. Fullerton, MPH2; Jonathan S. Yoder, MPH, MSW2