True life: Someone cares about our research (even the tabloids)

Katrina Levine, extension associate and lead author of Evaluating Food Safety Risk Messages in Popular Cookbooks writes,

If you have been following the discourse epic battle between Ben Chapman and Gwenyth Paltrow, you may be wondering why a celebrity wants anything to do with us.  Me too.

It all started a few years ago with a conversation about recipes and cooking, and just how little was said in recipes about handling food safely. I mean, “cook until done”? What does that even mean?

So with Chapman’s support, I set off on a mission – to look through recipes in cookbooks (29 books and over 1700 recipes) for evidence of safe food handling guidance – giving a safe internal temperatures and ways to avoid cross-contamination.

It took about a year to collect the data (remember, 1700+ recipes…), and another couple to finish the article, Evaluating Food Safety Risk Messages in Popular Cookbooks, just published in the British Food Journal.

Then the press release went out, and something notable happened. Someone, somewhere, decided it was worth sharing. So it got shared – and talked about – a lot.

And there was an opportunity here: Talking with the media and posting about our study has never been about bashing celebrities, but about a chance to get our messages out there while we are being listened to.

I’ve done a few interviews and while the journalists may want to talk about Gwyneth, and who was the worst, I get to interject stuff into the pop culture conversation like, use a thermometer; follow safe endpoint temperatures; and, keep your hands and food surfaces clean and sanitized.

This is a researcher’s dream – to have your work noticed, discussed, and sometimes understood – by lots of people.

Putting in the work was worth it because what we did got noticed. And people are talking about it. Maybe they’ll be changing what they do because of it.

I am living the dream.

Our battle with Gwyneth: cookbook edition

The coverage of extension associate Katrina Levine’s research on cookbook food safety messages took an unexpected turn yesterday. Gwyneth Paltrow’s ‘people’ weighed in.

By ‘people’ I think it’s the folks who published her cookbooks.

It started with a string of emails from some folks in the UK who saw the NC State press release about the research. After analyzing 1700+ recipes from cookbooks on the New York Times best seller list we found that safe endpoint temperatures only appeared in just over 8% of the instructions.

Not great.

A few journalists want to know who are the biggest offenders are (quick answer: it’s pretty well everyone we looked at – but not all the time).

One of the books included in our study was Paltrow’s It’s All Good. In a flurry of questions, and without being able to find all the recipes online, I sent one of the enquiring minds a recipe from another book, My Father’s Daughter as an example of what we were looking at, with this note:

“Here’s one from chef Paltrow that does not have a safe endpoint temperature included (165F or 74C).

Heat oven to 400°. Mix butter, garlic salt, paprika, pepper and salt in a bowl. Rinse chicken inside and out; pat dry. Insert fingers between skin and breast to separate the two. Rub seasoned butter over chicken and under skin. Tuck wings underneath bird and tie together with a piece of twine. Tie legs together with another piece of twine. Place chicken on its side in a heavy roasting pan and roast 25 minutes. Turn onto its other side and sprinkle with several tbsp water; roast 25 minutes more. Turn chicken on its back; roast 10 minutes more. Turn on its breast; roast until skin is crispy and chicken is golden brown, 10 minutes more. Remove from pan and let rest, breast side down, 15 minutes, before carving (remove skin).”

The Paltrow folks responded, through the journalist with this:

“The recipe for “Roast Chicken, Rotisserie Style” was published in MY FATHER’S DAUGHTER in April 2011. While it did not have an endpoint temperature included, the directions called for the chicken to be roasted at 400F for 70 minutes, which is ample time to cook a 3-4 pound chicken.

IT’S ALL GOOD, which was published in April 2013, does include endpoint temperatures. “Super-Crispy Roast Chicken” in IT’S ALL GOOD is baked for 1-1/2 hours at 425 degrees and the recipe advises “The chicken thigh should register 165 degrees F on a digital thermometer at the very least (I usually let it get to 180 Degrees F just to be completely sure it’s cooked all the way through the bone).”

So we went back to the data – and yep, we noted that the Super-Crispy Roast Chicken had a safe end point temperature. What they omitted was that the first instruction in the recipe was to wash the chicken; one of the steps that can increase the risk of foodborne illnesses.

There were these other recipes from It’s All Good that don’t have the safe endpoint temperatures (and tell the reader to do non science-based things like touch it, look for clear juices or color to ensure doneness):

The row (I think that’s the correct colloquial British term) made the front page of the Daily Mail (above, exactly as shown).

As for this comment, ‘the directions called for the chicken to be roasted at 400F for 70 minutes, which is ample time to cook a 3-4 pound chicken.’

Maybe, show me the data. Lots of variables that can impact the final temperature – starting temperature of the chicken, thickness, oven heat calibration.

Isn’t it just easier to tell folks what the safe temperature is and tell them to stick it in?

Food and Wine points out exactly what we found. It’s not just Gwyneth.

But for once, let’s cut Paltrow some slack. Out of the whopping 29 best-selling cookbooks these experts analyzed, only nine percent of them included specific temperature information. She’s in good company. Meanwhile, only 89 — 89! — of the 1,497 recipes included in the study were deemed instructionally safe.

Honestly, none of this seems too egregious, and we almost wish Paltrow didn’t have to deal with the PR headache.

Oh well.

Cookbooks Give Readers (Mostly) Bad Advice On Food Safety

Cookbooks could be a much better source of food safety information than they are. So could online recipes (like those from Epicurious, Allrecipes.com and foodnetwork.com.

Katrina reminded me today that those are next.

The NC State University press release on our cookbook paper came out today.

For Immediate Release

March 27, 2017

A recent study finds that bestselling cookbooks offer readers little useful advice about reducing food-safety risks, and that much of the advice they do provide is inaccurate and not based on sound science.

“Cookbooks aren’t widely viewed as a primary source of food-safety information, but cookbook sales are strong and they’re intended to be instructional,” says Ben Chapman, senior author of a paper on the work and an associate professor of agricultural and human sciences at North Carolina State University.

“Cookbooks tell people how to cook, so we wanted to see if cookbooks were providing any food-safety information related to cooking meat, poultry, seafood or eggs, and whether they were telling people to cook in a way that could affect the risk of contracting foodborne illness,” Chapman says.

To that end, the researchers evaluated a total 1,497 recipes from 29 cookbooks that appeared on the New York Times best sellers list for food and diet books. All of the recipes included handling raw animal ingredients: meat, poultry, seafood or eggs.

Specifically, the researchers looked at three things:

Does the recipe tell readers to cook the dish to a specific internal temperature?
If it does include a temperature, is that temperature one that has been shown to be “safe”? For example, cooking chicken to 165°F.

Does the recipe perpetuate food-safety myths – such as saying to cook poultry until the juices “run clear” – that have been proven unreliable as ways of determining if the dish has reached a safe temperature?

The researchers found that only 123 recipes – 8 percent of those reviewed – mentioned cooking the dish to a specific temperature. And not all of the temperatures listed were high enough to reduce the risk of foodborne illness.

“In other words, very few recipes provided relevant food-safety information, and 34 of those 123 recipes gave readers information that wasn’t safe,” Chapman says. “Put another way, only 89 out of 1,497 recipes gave readers reliable information that they could use to reduce their risk of foodborne illness.”

In addition, 99.7 percent of recipes gave readers “subjective indicators” to determine when a dish was done cooking. And none of those indicators were reliable ways to tell if a dish was cooked to a safe temperature.

“The most common indicator was cooking time, which appeared in 44 percent of the recipes,” says Katrina Levine, lead author of the paper and an extension associate in NC State’s Department of Agricultural and Human Sciences. “And cooking time is particularly unreliable, because so many factors can affect how long it takes to cook something: the size of the dish being cooked, how cold it was before going into the oven, differences in cooking equipment, and so on.”

Other common indicators used in the cookbooks included references to the color or texture of the meat, as well as vague language such as “cook until done.”

“This is important because cooking meat, poultry, seafood and eggs to a safe internal temperature kills off pathogens that cause foodborne illness,” Levine says. “These temperatures were established based on extensive research, targeting the most likely pathogens found in each food.”

A list of safe cooking temperatures can be found here.

“Ideally, cookbooks can help us make food tasty and reduce our risk of getting sick, so we’d like to see recipes include good endpoint cooking temperatures,” Chapman says. “A similar study was done 25 years ago and found similar results – so nothing has changed in the past quarter century. But by talking about these new results, we’re hoping to encourage that change.”

The paper, “Evaluating food safety risk messages in popular cookbooks,” is published in British Food Journal. The paper was coauthored by Ashley Chaifetz, a former Ph.D. student in Chapman’s group at NC State who now works for the U.S. Department of Agriculture’s Food and Nutrition Service. The work was supported by USDA’s National Institute of Food and Agriculture under grant number 2012-68003-30155.

Note to Editors: The study abstract follows.

“Evaluating food safety risk messages in popular cookbooks”

Authors: Katrina Levine and Benjamin Chapman, North Carolina State University; Ashley Chaifetz, U.S. Department of Agriculture

Published: March 17, British Food Journal

DOI: 10.1108/BFJ-02-2017-0066

Abstract:

Purpose: Medeiros et al. (2001) estimate 3.5 million cases of foodborne illness in the U.S. annually are associated with inadequate cooking of animal foods or cross-contamination from these foods. Past research shows home food handling practices can be risk factors for foodborne illness. The purpose of this study was to evaluate the communication of food safety guidance, specifically safe endpoint temperatures and cross-contamination risk reduction practices, in popular cookbook recipes.

Design/methodology/approach: Recipes containing raw animal ingredients in 29 popular cookbooks were evaluated through content analysis for messages related to safe endpoint temperature recommendations and reducing cross-contamination risks.

Findings: Of 1,749 recipes meeting study criteria of cooking raw animal ingredients, 1,497 contained a raw animal that could effectively be measured with a digital thermometer. Only 123 (8.2%) of these recipes included an endpoint temperature, of which 89 (72.3%) gave a correct temperature. Neutral and positive food safety behavior messages were provided in just 7.2% (n=126) and 5.1% (n=90) of recipes, respectively. When endpoint temperatures were not included, authors often provided subjective and risky recommendations.

Research limitations/implications: Further research is needed on the effect of these results on consumer behavior and to develop interventions for writing recipes with better food safety guidance.

Practical implications: Including correct food safety guidance in cookbooks may increase the potential of reducing the risk of foodborne illness.

Originality/value: Popular cookbooks are an underutilized avenue for communicating safe food handling practices and currently cookbook authors are risk amplifiers.

Cookbooks aren’t the greatest source of food safety info

Every year for Christmas Dani and I buy a couple of cookbooks for each other. Maybe it’s from a chef at a restaurant we’ve been to. Or someone we’ve seen on Top Chef. Yeah, thousands of recipes are available on the Internets but the books become something we keep going back to.

According to the LA Times, we’re not alone.

Ten Speed will have published at least 30 books in the food and spirits category by the end of 2016; Phaidon will have released 20 titles. Culinary dynamo Ina Garten’s latest “Cooking for Jeffrey: A Barefoot Contessa Cookbook” is currently holding the No. 8 spot in all categories on Amazon.

Print isn’t totally dead.

Since folks still go to these books for ideas, instructions and information on what and how to cook, extension associate Katrina Levine and I (with the stats help of Ashley Chaifetz) decided to go look at the type of food safety info is in the books. Katrina looked at the New York Times best seller lists for food and diet from Sept 2013-Jan 2014 and evaluated the food safety messages in over 1700 recipes. Some of the stuff she found was that although there are a lot of recipes that could benefit from instructions about safe endpoint temperatures only about 8 percent mentioned a specific temperature. And just under three quarters of those recipes even had the science-based temperature.

Lots of other insights, the paper was published as an early cite late last week and will be in an upcoming special issue of the British Food Journal. The abstract is below:

Evaluating food safety risk messages in popular cookbooks

British Food Journal, Volume 119, Issue 5 2017

Katrina Levine , Ashley Chaifetz , Benjamin Chapman

Abstract

Purpose

Medeiros et al. (2001) estimate 3.5 million cases of foodborne illness in the U.S. annually are associated with inadequate cooking of animal foods or cross-contamination from these foods. Past research shows home food handling practices can be risk factors for foodborne illness. The purpose of this study was to evaluate the communication of food safety guidance, specifically safe endpoint temperatures and cross-contamination risk reduction practices, in popular cookbook recipes.

Design/methodology/approach

Recipes containing raw animal ingredients in 29 popular cookbooks were evaluated through content analysis for messages related to safe endpoint temperature recommendations and reducing cross-contamination risks.

Findings

Of 1,749 recipes meeting study criteria of cooking raw animal ingredients, 1,497 contained a raw animal that could effectively be measured with a digital thermometer. Only 123 (8.2%) of these recipes included an endpoint temperature, of which 89 (72.3%) gave a correct temperature. Neutral and positive food safety behavior messages were provided in just 7.2% (n=126) and 5.1% (n=90) of recipes, respectively. When endpoint temperatures were not included, authors often provided subjective and risky recommendations.

Research limitations/implications

Further research is needed on the effect of these results on consumer behavior and to develop interventions for writing recipes with better food safety guidance.

Practical implications

Including correct food safety guidance in cookbooks may increase the potential of reducing the risk of foodborne illness.

Originality/value

Popular cookbooks are an underutilized avenue for communicating safe food handling practices and currently cookbook authors are risk amplifiers.

Would you like some eggs with your cocktail?

Katrina Levine, extension associate at NC State University writes:

The holidays are a time when we come together to celebrate and mingle over food and drinks, especially eggnog and other egg-containing drinks. Restaurants, bars, and your bartending neighbor often feature foamy, frothy egg cocktails this time of year, but are they all that they’re cracked up to be?

Back in the summer, it was a dark and stormy night when I entered the dimly-lit Raleigh speakeasy. A jazz band was playing a chaotic Charlie Parker-inspired song and the bar was packed with people sipping some exotic-looking cocktails. faceI approached the bar to order and out of the corner of my eye I saw the bartender whip out an egg and crack it directly into someone’s drink (me, right, not quite exactly as shown).

Raw egg-containing cocktails are an emerging trend in the local bar scene. Fizzes, flips, sours, and even eggnog often contain raw whole eggs, whites, or yolks. While they may be tasty, the risk of acquiring Salmonella from one of them is uncertain.

egg-breakConsuming raw or undercooked eggs has been linked to many foodborne illness outbreaks of salmonellosis. A recent Michigan outbreak sickening 32 and one in North Carolina in 2012 making 29 people ill were both likely caused by eating raw egg-based sauces. A major outbreak of salmonellosis in 2010 infected almost 2,000 people in 11 states (CDC, 2010). A table of raw egg-related outbreaks in can be found here. Hens that are infected with Salmonella Enteritidis, the most common strain in eggs, can pass on the bacteria directly to the eggs forming in their ovaries (Gantois et al., 2009). This is the mostly likely route of contamination (Gantois et al., 2009). There is not enough research to show which part of the egg has the greatest amount of contamination from this route, although the shell, membranes, white (albumen), and yolk can all become contaminated (Gantois et al., 2009).

Eggs can also become contaminated by penetration through the shell when exposed to things like feces contaminated with Salmonella (Gantois et al., 2009).  Because a shell is porous, the bacteria can still find ways to get through the shell and into the egg. The shell of the egg is more likely to become contaminated than the inside of the egg, and egg yolks are less likely than whites to contain Salmonella because the antimicrobial properties of the white can reduce or eliminate the bacteria (Gantois et al., 2009). But if the shell has Salmonella and it comes in contact with the egg, such as when you crack it, you could contaminate the egg inside (i.e., cross-contamination). Washing eggs won’t help you – it actually makes contamination more likely (USDA FSIS, 2011).

Salmonella Enteritidis is found in an average of 1 in every 20,000 eggs (Ebel & Schlosser, 2000). Contamination rates are also influenced by percentage of infected hens in a flock and time between infection and producing eggs (Braden 2006).

The U.S. produces about 80 billion eggs for consumers annually, roughly 30% of which are pasteurized  (USDA NASS, 2014; USDA FSIS 2013). If 1 in 20,000 have Salmonella Entertidis, that equals about 2.8 million contaminated eggs, some of which may end up at bars, restaurants, or home kitchens.

Ok, so say you’re lucky #20,000. What are the chances you’ll get sick? It’s unclear whether the risk of illness is impacted by adding Salmonella-containing egg into an alcoholic drink. The literature suggests that there are at least a couple of factors that may affect survival and destruction of the pathogen: pH and alcohol content.

Consuming alcohol along with a pathogen may have a protective effect. Epidemiologists showed in a 2002 outbreak investigation report that there was a protective effect among people who drank more than 40 grams of alcohol (that’s about 2 pints of beer or 2-3 glasses of wine) (Bellido-Blasco et al., 2002). The infection rate of those who had more than 40 grams was 54%, compared with 78% and 95% for those who had less than 40 grams and no alcohol, respectively (Bellido-Blasco et al., 2002).drink

In 2008, researchers at Rockefeller University investigated whether spiking eggnog would kill Salmonella, but their results were inconclusive (and weren’t published in a peer-reviewed journal). Even though the drink was made with 14% alcohol, it was initially packed with bacteria, no Salmonella was recovered after sitting in the fridge about 3 weeks (Rockefeller University, 2008).

Yet another study on wine examined the antimicrobial effects of unadulterated wine and the role of some of its constituents – total phenols, ethanol, and pH – on antimicrobial activity (Boban et al, 2010).  They tested intact wine, wine with the phenols removed, wine with the alcohol removed, ethanol, a low pH, and ethanol and a low pH combined, against Salmonella Enteritis and E. coli (Boban et al., 2010). Even though the pH and alcohol content were similar among all samples, intact wine had the highest amount of antimicrobial activity, while non-intact versions had lower antimicrobial activity (Boban et al., 2010). Ethanol alone and low pH alone had negligible antibacterial activity, but when combined had a greater activity (Boban et al., 2010). While the alcohol content and pH seem to have some sort of effect on the antimicrobial effects of wine, the effect can’t be attributed to the alcohol, pH, or any of the other components specifically (Boban et al., 2010).

Acidity is also a factor. Salmonella won’t grow below a pH of about 3.6-4.0 (Lanciotti et al., 2001; Perales & Garcia, 1990), so the overall concentration and combination of ingredients in the drink would have to be at 4.0 or below to kill the bacteria.

Different forms of alcohol may be more or less acidic (Lazar, 2011). Grape-based forms, like sherries and vermouths, are more acidic, while distilled spirits, like gin and vodka, generally have a neutral pH, because the distillation process removes the acidity. Other additions to the alcohol can impact pH as well. Citrus ingredients like lemon and lime juices are by far the most acidic with a pH of about 2-3 (Lazar, 2011). Several experiments have mixed an acid, like lemon juice, vinegar, or citric acid, with mayonnaise made with raw eggs to see if it reduces Salmonella. Results show, though, that adding acid alone does not seem to reduce the bacteria to undetectable levels unless left to sit for at least an hour at 25°C (about 77°F) (Zhu, Li, and Chen, 2012). Warmer temperatures actually seemed to help kill more bacteria because cell membranes are more permeable, allowing more acid to enter bacterial cells (Zhu, Li, and Chen, 2012). Since most raw egg cocktails are kept cold and consumed quickly, a splash of lime juice probably wouldn’t be enough to lower the pH of the whole drink.

The bottom line is that the risk of getting a raw egg cocktail contaminated with Salmonella is low based on the egg contamination rates, but if you win the contaminated egg lottery, your grand prize could be a visit to the hospital. You’ll have to decide if you are willing to take the risk.

Opt for drinks that have already been pasteurized or that are made with a pasteurized egg product or pasteurized shell eggs – it’s the easiest way to reduce the risk of Salmonella.

References

Bellido-Blasco, J. B., Arnedo-Pena, A., Cordero-Cutillas, E., Canós-Cabedo, M., Herrero-Carot, C., & Safont-Adsuara, L. (2002). The protective effect of alcoholic beverages on the occurrence of a Salmonella food-borne outbreak. Epidemiology, 13(2), 228-230.

Boban, N., Tonkic, M., Budimir, D., Modun, D., Sutlovic, D., Punda‐Polic, V., & Boban, M. (2010). Antimicrobial effects of wine: separating the role of polyphenols, pH, ethanol, and other wine components. Journal of food science, 75(5), M322-M326.

Braden, C. R. (2006). Salmonella enterica serotype Enteritidis and eggs: a national epidemic in the United States. Clinical Infectious Diseases, 43(4), 512-517.

Chickens and Eggs 2013 Summary. USDA NASS. February 2014.

Ebel, E., & Schlosser, W. (2000). Estimating the annual fraction of eggs contaminated with Salmonella enteritidis in the United States. International journal of food microbiology, 61(1), 51-62.

Egg Products and Food Safety. USDA FSIS. August 2013.

Gantois, I., Ducatelle, R., Pasmans, F., Haesebrouck, F., Gast, R., Humphrey, T. J., & Van Immerseel, F. (2009). Mechanisms of egg contamination by Salmonella Enteritidis. FEMS microbiology reviews, 33(4), 718-738.

Lanciotti, R., Sinigaglia, M., Gardini, F., Vannini, L., & Guerzoni, M. E. (2001). Growth/no growth interfaces of Bacillus cereus, Staphylococcus aureus and Salmonella enteritidis in model systems based on water activity, pH, temperature and ethanol concentration. Food Microbiology, 18(6), 659-668.

Lazar, Michael. “The Electric Cocktail Acid Test.” Stirred, Not Shaken blog. December 28, 2011.

Multistate Outbreak of Human Salmonella Enteritidis Infections Associated with Shell Eggs (Final Update). CDC. December 2010.

Perales, I., & Garcia, M. I. (1990). The influence of pH and temperature on the behaviour of Salmonella enteritidis phage type 4 in home‐made mayonnaise. Letters in applied microbiology, 10(1), 19-22.

Rockefeller University. “Rockefeller microbiologist tests safety of spiked eggnog.” December 19, 2008. http://newswire.rockefeller.edu/2008/12/19/rockefeller-microbiologist-tests-safety-of-spiked-eggnog/

Shell Eggs from Farm to Table. USDA FSIS. April 2011.

Zhu, J., Li, J., & Chen, J. (2012). Survival of Salmonella in Home-Style Mayonnaise and Acid Solutions as Affected by Acidulant Type and Preservatives. Journal of Food Protection, 75(3), 465-471.