Here is your permission to indulge in delicious Christmas treats this holiday season: a study shows鈥攏ay,聽proves!鈥攖hat yuletide desserts often use ingredients that have been scientifically demonstrated to lower your risk of disease and death. You will still die, as we all do eventually, but not quite as early if you devour a Christmas log.
Before you start baking, though, I need to point out that this study should be taken with a grain of salt, as it was written with tongue firmly planted in cheek.
A year ago, the聽British Medical Journal聽(BMJ) released聽聽as part of its Christmas edition, an annual tradition where the journal publishes聽sincere research about zany topics. In it, four authors took on the recipes behind famous Christmas-themed desserts featured on the reality television show聽The Great British Bake Off鈥�鈥渋n our opinion,鈥� the authors write, 鈥渢he greatest television baking competition of all time.鈥� They wanted to look up their ingredients in the scientific literature to see if, on the whole, they were associated with health benefits or harms, which is how they came to the frivolous (though technically scientific) conclusion that, overall, you聽can聽have your cake and eat it too.
Christmas desserts didn鈥檛 use to be so sweet. As the authors mention, early Christmas puddings encouraged by the Roman Catholic Church in medieval England were stews filled with prunes, carrots, nuts, eggs, beef, and mutton. Not exactly something I鈥檓 eager to buy from my local bakery. Over the centuries, the protein content went down and delicious sugar levels crept up.
If you watch聽The Great British Bake Off,聽you may have seen Jon鈥檚 pecan and maple buns with candied bacon or Rob鈥檚 apple and cinnamon baked Alaska tarts. Are they any good for your health? The authors of the聽叠惭闯听study mirrored the excess of the desserts you typically see on television by looking at the very top of an extravagant layer cake of evidence: not individual studies on the health benefits or harms of their ingredients, not meta-analyses of these studies, but umbrella reviews of meta-analyses. They thus performed, as the title of their study explains, an umbrella review of umbrella reviews of meta-analyses of observational studies. Still with me? The thing is that the health effects of, let鈥檚 say, butter have been studied so much, the evidence has been synthesized many times over. The authors here simply put the appraisals of the appraisals into a mixer and whipped them up into a cream of evidence.
Their paper is a lot of fun to read because it playfully teases you with the central problem it is commenting on. There鈥檚 an entire paragraph on Prue Leith鈥檚 chocolate yule log and why it鈥檚 wrong that it was described on the show as a Swiss roll subtly laced with Irish cream liqueur. 鈥淲e are not convinced,鈥� write the authors, 鈥渢hat this dessert adds to the 鈥榝estive spirit鈥� because it would not be appropriate to 鈥榮ubtly lace鈥� a dessert that you serve to your family and friends with alcohol that increases your risk of developing liver cancer, gastric cancer, colon cancer, upper aero-digestive tract cancer, gout, and atrial fibrillation.鈥� Of course, they parenthetically indicate the numerical relative risk increase after each disease to drive the point home with math.
When we look at聽The Great British Bake Off聽through this skewed scientific lens, we can come to all sorts of conclusions. Steven鈥檚 telephone cake has alcohol in it, which is bad, but also chocolate, which the weight of the evidence indicates is good for us, apparently. So, does the chocolate outweigh the alcohol? The authors looked at a total of 48 recipes for Christmas desserts featured on the show and judge that 鈥渢he health benefits of most ingredients in聽The Great British Bake Off聽Christmas desserts outweigh the harms.鈥� After all, many of the recipes called for fruits in one form or another, and fruits are good for us. 鈥淎ny recipe with fruit,鈥� the authors mention in passing, 鈥渆ven if it was only one berry, was weighted equally in terms of its protective effect in relation to the harmful effect of butter, even if it was four sticks!鈥� They could have taken quantities into consideration, but where would the festive spirit be?
What we have just witnessed is scientific satire. This is exactly the sort of soggy science that gets exaggerated by journalists to tell us that eggs are bad one day, good the next, and yet evil again a year later. So much of research into nutrition is based on food questionnaires, which rely on questions such as, 鈥淚n the last ten years, how often did you consume eggs weekly on average?鈥� People鈥檚 diets change over time, and most of us can鈥檛 remember what we had for breakfast, let alone how often we鈥檝e had eggs in the last decade. Plus, we may anticipate that the researchers think eggs are bad and adjust our answers to please them. Food questionnaires should be at the bottom of the scientific research food chain.
This umbrella review reminded me of an important paper that opened my eyes to some of the problems of scientific research: Jonathan Schoenfeld and John Ioannidis鈥櫬�, in which they picked food ingredients at random from a cookbook and looked them up in the scientific literature to see if they were known to increase cancer risk or decrease it. The bottom line was that聽别惫别谤测迟丑颈苍驳听common enough in a kitchen simultaneously caused cancer and protected you from it. Of the ingredients that had at least 10 studies behind them, only one was unilaterally shown to cause cancer and not protect against it: bacon. (Funnily enough, there was bacon in one of the Christmas desserts in the聽叠惭闯听paper, but the authors did not include it in their review 鈥渂ecause it is not a proper dessert ingredient and the first author is vegetarian.鈥�)
Back then, the flaw being highlighted was that scientists were making bold claims about food in their papers that were not supported by the data they had. The risks and benefits were exaggerated, and journalists were quick to further amplify them in the media, leading to mass confusion.
Here, we have a bit of festive satire that again demonstrates that food science can be processed in such a way that you can extrude any conclusion you want. Do you want to feel like you鈥檙e improving your health by eating an entire rhubarb cake? You can now point to this paper as justification!
Common sense needs to enter the recipe. We know that too much sugar is bad for us. We know that the mere presence of candied fruits in a cake does not make the cake healthy. We also know that food is not simply a way to health: it鈥檚 a social activity and a bringer of joy for many.
On the first page of the paper, the authors put together a three-fold answer to the question, 鈥淲hat does this study add?鈥� It鈥檚 a great reminder that scientific findings鈥攅ven from satire articles鈥攈ave to be聽interpreted, and food science is especially prone to a wide diversity of interpretations.
Here is what the authors wrote, with a caustic wit worthy of the British reality show:
- Social media: 鈥淵ou should eat Christmas desserts from聽The Great British Bake Off聽if you want to be healthier and live longer!鈥�
- Newspaper: 鈥淐an you have your cake, and eat it too? Study finds most Christmas dessert recipes from聽The Great British Bake Off聽might reduce the risk of death or disease.鈥�
- Real-life journal club: 鈥淭his umbrella of umbrella reviews does not consider the complexities of nutritional epidemiology (eg, overall diet and lifestyle) and health, and therefore does not contribute meaningfully to the literature.
Like a Christmas buffet, you get to pick which conclusion looks the most appetizing to you.
