False attribution 
18 January 2018

The heralding of the new year in the Western world brings with it a swathe of resolutions. Given the ever-increasing proportion of people who are either overweight or obese, a popular one is to lose weight. Knowing this, book publishers ensure they have their latest diet books on shelf before the new year begins, hoping to capitalise on resolutions when they still burn brightly in their owner’s minds.

While there are exceptions, the vast majority of each year’s diet books focus on the journey of a particular individual who has in some way “seen the light”, identified the dietary villain(s) that everyone else (or at least most) has been ignoring (due to every possible reason: from their own naivety; to corrupt businesses obfuscating the evidence) and has successfully lost a truck load of weight and has never felt better. The diet book is their testament to the truth and all others are wrong. Sadly, it sounds awfully familiar, doesn’t it?

Needless to say that the testimonial that underpins most diet books, otherwise known as the case-study or case-report, is not rated very highly in scientific terms – in fact it’s pretty much on the bottom of the scientific evidence pile, just above personal opinion, animal and test tube studies.

The reason why is simple – case-studies provide a detailed knowledge of an individual’s experience, but they are not carefully controlled scientific experiments and therefore the results cannot, and should not, be generalised to the broader population. Only the results of larger studies conducted in groups of people chosen to be representative of the general population can.

Due to their inherent nature, case-studies are very prone to false attribution. Think of false attribution as a case of mistaken identity – it’s when you misidentify the cause of an event like losing weight. For example, most of the case studies that appear in diet books identify a specific nutrient or ingredient like (current favourites) carbohydrates/sugars/fructose/gluten/dairy foods, etc … that they have avoided – religiously – and simply advise fellow sufferers to do the same – avoid all of the foods and drinks that contain the problematic nutrient/ingredient(s) at all costs (there’s typically no room for moderation – you can’t dance with the devil). In doing so, they inadvertently reduce their kilojoule intake – which is the real reason why they have lost the weight. It’s (kilojoule reduction) the common denominator and the reason why the actual dietary villain itself doesn’t really matter as long as avoiding it restricts your food and drink consumption sufficiently. All roads lead to Rome so to speak.

However, when the results of the case-study are put to the test using the best scientific method – the randomised controlled trial – they are rarely replicated. The reason why is again quite simple. In randomised controlled trials people are randomly selected to follow a control diet (typically a healthy diet as defined by the most recent dietary guidelines, for ethical reasons) and an intervention diet – say a low fructose diet to use a common dietary villain, where the individuals are instructed to reduce their fructose consumption to as low a level as possible. Other than the fructose content of the diet, all other known factors (known scientifically as confounders) are controlled for, or kept equal, like the total kilojoules consumed, the amount of protein, fat, carbohydrate, dietary fibre, vitamins, minerals, alcohol and the amount of exercise performed. Randomised controlled trials have in fact been performed in this case (fructose), and when all of the results are gathered together in a systematic review and meta-analysis – the highest level of evidence – they are quite clear: fructose is not uniquely fattening or detrimental to health. In fact moderate amounts (less than 50 g per day) of added fructose may have some health benefits.

So, despite the compelling story exquisitely told in the latest best-selling diet book, testimonials/case-studies are prone to false attribution and are rarely supported by higher levels of scientific evidence. They are best left on the shelf. Sadly, the dietary guidelines have never been best sellers, but they are available online for free …