Discover more from The Rotten Apple
Issue #25 2022-02-14
Honey fraud allegations questioned, lipstick safety, volcanoes and 3D printed food
Eating lipstick (Happy Valentine’s Day!)
Wrong about honey fraud?
Volcanos + food = yuck
Quick primer on 3D printed food
Welcome! Thank you for subscribing to The Rotten Apple, where you can stay up to date with food integrity news in 5 minutes per week.
This week’s issue starts with a not-quite-food safety warning for those of us who wear lipstick. Counterfeit lipstick can contain heavy metals like lead, and recent test results are alarming.
Today’s big news is honey authenticity testing. I explore two brand new papers that are really important to future conversations about food fraud testing. As always, I’ve included links to the original sources, but if you don’t have time to wade through all the academic stuff, read on to get the key takeaways.
Also this week, food supply challenges after a volcano. And what exactly is 3D printed food? I know you - sort of - know, but I give you a juicy technical overview, without all the waffle.
By the way if you’re curious about how I find and curate the ideas for these newsletters, stay tuned. I’ll be hosting Ask Me Anything sessions later this year, where you can ask me questions about the newsletter, our food fraud Trello board, and anything else in a live, real-time format.
Food Safety (sort of)
Warning for Lipstick Wearers (Happy Valentines day)
Okay, so this isn’t about food. But it is fraud, and it is dangerous. Plus, I reckon more than half of The Rotten Apple subscribers wear lipstick on a regular basis, and unfortunately, we do eat lipstick…. Around 300 g per year according to one source (but probably much less*).
Lipstick needs strong, bright, colour-fast red pigments. Red pigments are tricky. Achieving a deep, strong and vibrant red or orange colour is a challenge for product formulators. Having said that, red pigments are only tricky if you don’t want to use pigments that contain heavy metals like lead, cadmium and barium. It’s easy to get great red colours if you use pigments made with heavy metals.
If you were a criminal and wanted to make cheap fake lipstick with strong attractive colours, what would you do? You’d reach for a lead-based colourant.
Last year, a consumer group in The Philippines found premium-brand lipsticks on sale for just 15 pesos (30 cents) each in local outlets. The lipsticks were counterfeits, fake versions of MAC cosmetics. They contained between 5 times and 1810 times more lead than is allowed**.
Yes, that’s right, super-dangerous levels of the heavy metal lead in cheap lipstick.
Think this is only happening in The Philippines? Don’t be so sure.
What to do:
Buy lipsticks made by well-known, trusted brands, from authorised stockists. Why? Because if a brand is valuable, the brand-owner has more incentive to properly manage their supply chains to avoid potential scandals with unsafe ingredients.
Avoid purchasing from discount outlets, market stalls and street corners, as the products sold there are more likely to be counterfeits than products from authorised retail stores.
* I have not included links to the dodgy source, but have included a reference to a legitimate source below.
** The ASEAN Cosmetic Directive limit for lead is 20 ppm.
In short: 🍏 Counterfeit lipstick can contain scary amounts of lead 🍏 Buy your lipstick with care 🍏
The Problem with Honey Analysis
Honey fraud reports strike fear into me. My goal as a commentator and information-sharer is to only report food fraud allegations that are plausible, genuine and scientifically supported. Accusations of honey fraud don’t always meet those criteria, but it’s difficult to know for sure.
In the last few years, there have been a couple of honey fraud allegations that were a bit questionable. Those haven’t made it into our Food Fraud Risk Information Database (Trello-hosted database), nor into this publication, because the publicly available information about the tests behind the allegations has been too opaque for me to form an opinion as to their legitimacy.
I’m not the only one who questions these results. Professor Chris Elliott, of the Institute for Global Food Security and – in my mind- the granddaddy and champion of modern food fraud prevention, has publicly called out honey fraud claims that were based on less-than-robust test protocols.
It’s not just honey. A similar set of concerns about herbal supplement testing using DNA barcoding has just surfaced. More on that next week.
Now we have got proper scientific evidence that all is not well in some honey fraud allegations. Great news!
Two new papers, by researchers from the (United Kingdom) Laboratory of the Government Chemist and the Institute for Global Food Security at The Queen’s University of Belfast discuss the problems with honey test reports. And propose a solution, hooray.
The first paper discusses and defines the problems with honey analytical reports. In the second paper the researchers discuss their analysis of test reports related to a honey fraud allegation published in the popular press in the United Kingdom (UK) in 2020.
The test report(s) alleged non-compliances and reported the presence of syrup markers and foreign sugars. However, after closely examining the test results, the researchers concluded that the data instead revealed a “nuanced and partially conflicting picture of the authenticity” of the alleged inauthentic honeys.
That is, the data didn’t fully support claims that the honey samples were adulterated.
The researchers recommend that test reports should not just describe the conclusions made by experts but should always include all the data on which authenticity findings are based. In the case of honey, for example, a report should state whether the results have been derived from NMR profiles or from molecular markers.
Press journalists should ensure they have proof beyond reasonable doubt, as for a criminal prosecution, before publishing allegations of food law contraventions. There is a reporting standard used in criminal forensic test reporting and the researchers propose a similar standard be followed when reporting food fraud in the media.
The reporting standard they recommend is called Evaluative Reporting and includes a method for evaluating evidence called likelihood ratio (LR). In criminal forensic law, the LR method compares the likelihood of observing evidence of a crime for two competing propositions: prosecution or defence. Other researchers have also suggested LR be used when interpreting wine and olive oil authenticity tests.
In honey, an LR approach to reporting a single chemical test result might look like this:
Honey is expected to contain a certain level of the enzyme diastase and there is a ‘legal’ limit for diastase in Europe. If a honey sample contains less than the limit it might be reported as inauthentic. However, if the LR framework is used, any report would also include alternative reasons for the sample to contain less diastase than the limit, perhaps due to being stored at high temperatures. In the case of diastase, historical data shows that there is a 12% likelihood of genuine honey sampled from retail premises containing less diastase than the ‘legal’ limit. The analytical report should mention this probability and also include the interpretation that the diastase content is not necessarily evidence for fraud.
I’ve simplified (sorry, not sorry!). The true LR approach is more complicated: there are mathematical formulas and defined numerical bands for categorising the strength of evidence. There are also important questions about how to combine LR numerical results from multiple tests. For example...
- If a honey sample ‘fails’ a diastase test but passes a colour additive test, how much weight do you put on each result?
- If you find evidence of caramel colour in a sample, but don’t have a good dataset for the likelihood of naturally-occurring caramels (say from beekeepers’ smoke), how can you assign a probability that the caramel is there by chance?
So many questions!
The takeaway here is that analytical test reports should either:
(1) Present a set of data without including any interpretation or judgement about the samples tested; or
(2) Include a full set of data and expert interpretations. In this case the report must include enough information that knowledgeable readers can understand the individual test results, including quantitative results, limits of detection and the uncertainty of each result. They should also contain an indication of the strength of the conclusions, based on the likelihood ratio (LR) method. Conclusions should be phrased like this: “This sample has compositional or matrix parameters that are untypical (that is, outside those represented in a reference dataset) because … [give a reason, e.g. possible adulteration] or arising by chance”.
In short: 🍏 Food authenticity testing is hard 🍏 It’s easy for results to be misinterpreted 🍏 Incomplete test reports are unhelpful (duh) 🍏 Allegations of inauthenticity should not be published unless test reports have shown evidence beyond reasonable doubt, accompanied by alternative explanations for results that appear to indicate fraud 🍏
Prof Elliott refutes the NMR method: https://www.newfoodmagazine.com/article/127872/honey-analysis/
Newspaper report of the alleged fraud in 2020:
Volcanic Ash and Food: Don’t Eat Kidneys (?)
I’m feeling super thankful not to be in Tonga right now, where their food supply and everything else has been badly affected by the January volcano eruption and tsunami.
Crops, livestock and fisheries have been affected by ash fall, saltwater inundation and acid rain, with an estimated 60,000 people affected. Another unfortunate side effect: Tonga was Covid-free until military aid arrived in the island state.
The FAO has shared its advice about food safety and volcanic ash fall. It includes obvious instructions, like “water may be polluted”, and less obvious advice such as “avoid consuming kidney meat”.
What Exactly is ‘3D Printed Food’?
Like a scene from the Jetsons, we can now eat food that is printed on demand in our own homes. That’s right, you can purchase 3D food printers for domestic use.
Printing food = using machines to deposit layers of edible material into pre-designed shapes. The machines can cook as they go, or they can deposit the foodstuff unchanged.
There are technical challenges. To be successful, the food needs to be free-flowing during printing but self-supporting after printing. Food properties, especially rheological properties, are important. It’s a food engineer’s wet dream!
How do you make 3D printed foods?
1. You can extrude liquids and powders, through a die hole, without or without temperature and pressure. You can cook during the extrusion process, or cook the printed food afterwards, or just melt and set/freeze foods like chocolate.
2. You can jet food liquids or powders onto surfaces, like an inkjet office printer. Example: jet food dyes onto cakes or cookies to make decorative images.
3. You can jet a powder plus a binder at the same time, so the printed powder sticks together into the desired shape.
4. You can use laser sintering. The laser selectively heats and fuses certain parts of a layer of powder. After that another layer is put down. Sweets/candies have been made in this manner.
5. You can have more than one print head printing at the same time. Print a “steak”, cover it with gravy.
What can you make with 3D printing?
Chocolate called Mona Lisa chocolate, is made by famous Swiss company Barry Callebaut using 3D printing tech.
Plant based meats can be structured into shapes using 3D printing. Redefine Meat and NovaMeat are two companies doing this.
Personalised nutrition foods can be made with 3D technology. In personalised nutrition, foods are specifically tailored to meet an individual’s texture and nutritional needs. UK company Nourished makes high nutrient snacks using 3D printing. Foods with modified textures that meet the needs of the high proportion of elderly people with swallowing difficulties can also be made with 3D printing.
Coming soon… 4D printing (what the?)
3D printing = three-dimensional printing. Next up: 4D printing. The fourth dimension is time. In 4D printing, the food is designed so that its structure changes after it has been printed. An example: pasta that is printed to be flat when dried – and hence easy to pack – can be designed to form into three-dimensional shapes when cooked. This has been achieved by researchers at MIT who used layers of cellulose and gelatine that respond differently to moist cooking.
Not so fast
If you’ve got nerdy friends (like me), who are into 3D printing plastic items, you will know it’s super difficult to achieve consistent, repeatable results when printing plastic. It also takes a long time to print a single item. If one little thing goes wrong in the depositing of one layer of the design, the finished result is ruined, and you have to start over. And that’s with synthetic, completely uniform plastic printing filament. One can only imagine what it would be like to try to use printing stock made from natural agricultural products with all their inherent variability.
Food companies, it seems, are up for the challenge. Food engineers, I salute you.
Food Fraud Incidents and Horizon Scanning
Food fraud incidents added to Food Fraud Risk Information Database in the past week
Edible cannabidiol (CBD) hemp oil has been recalled because it was made at an unregistered facility and contains an ingredient (concentrated CBD) that is not approved as a "food" in the jurisdiction - Ireland
Thousands of bottles of contraband and/or adulterated liquor were seized by authorities - Columbia
Five men were arrested for supplying alcohol in Afghanistan in two separate incidents. Alcohol is illegal in that country - Afghanistan
Large quantities of smuggled food and drink were seized by authorities in December. They are alleged to have been imported without proper documentation - Bolivia
Foreign liquor, being smuggled in vegetable trucks, was found by authorities after a tip-off. In a separate operation, liquor that was being smuggled at sea was intercepted by the navy - Pakistan
Four people were detained in relation to deaths caused by beverages containing methanol - Russia
A food business owner has been accused of smuggling hundreds of thousands of dollars worth of food into the country after disgruntled ex-employees informed authorities - Zimbabwe
Illegally imported pork - which presents risks of African Swine Fever - was destroyed by authorities - Italy
Authorities have arrested suspects in two separate cases of frozen meat smuggling by sea - Hong Kong
Twelve thousand packs of coffee beans from three different brands were seized by authorities for a variety of infringements including containing bark, sticks and corn - Spain
A consumer group tested seafood and meat balls and found some lobster balls contained no seafood DNA and some beef balls contained pork or chicken - Hong Kong
A man was arrested for stealing 30 tonnes of rice that was meant for public distribution - India
Oats from Eastern Europe that were conventionally grown were found mislabelled as 'organic' by authorities in Italy
A pasta company has been accused of country of origin labelling fraud over its use of imported wheat. The company claims it uses a small amount of wheat from a country not listed on the label and that this is acceptable - Italy
A market operator has been accused of country of origin fraud in relation to dried fish, legumes and spices which were claimed to be from Africa and Japan but which were instead from countries deemed 'at risk', meaning they require official clearances before being sold - Italy
Fish samples obtained from supermarkets and restaurants in Singapore were found to be mislabelled with respect to species identification at a rate of 26% (n = unknown). Samples from restaurants were not mislabelled, with all mislabelled samples purchased from supermarkets. Black Cod, Cod, Seabass, Dory and Bocourti were the products that were most often mislabelled, being replaced with Sablefish, Patagonian toothfish and Iridescent shark – Singapore
Food fraud horizon scanning (non-incident updates to the Food Fraud Risk Information Database in the past week)
Update added to some previous supplement entries: DNA bar-coding methods for herbal supplement authentication are being called into question because DNA can degrade during processing and because such methods are usually not quantitative. One person associated with a North American authenticity test provider has been accused of unethical practices, falsification of data and plagiarism. [Next week’s issue of The Rotten Apple will have more details about this] 02/02/2022
Seeds (for cultivation)
Seeds of soybeans, sunflowers, beans, corn, sorghum, sesame, birdseed, millet and fodder, which were smuggled into the country illegally, have been destroyed by authorities. The smuggling is done to avoid taxes and is said to harm the local seed production industry – Bolivia. 16/01/2022
The Indian government has created standard procedures for food businesses to ensure that any used cooking oil they dispose of goes only to authorised producers of non-food products like biodiesel and does not end up back in the human food chain - India 11/02/2022
Here’s what you missed last year:
When the world’s leading food fraud guy warns us about a new food fraud risk, we’d better sit up and pay attention. Professor Chris Elliott of Queens University Belfast has published a warning about a ‘new’, potentially deadly fraud type. Sadly, it’s happened before, but no one was warning us about it back in 2007… This time, we have no excuses. Read about it in Issue #15
Thanks for reading, see you next week.