Yo-yo dieting: not so bad?
You know about “yo-yo dieting”? The weight lost in dieting is regained afterwards, sometimes with a bit extra. Scientists call it “weight cycling”, many overweight people call it a nightmare.
People who have done several weight-loss diets (serial dieters) have on average a higher weight than people who have dieted only once, or never. By itself, this observation seems to indicate that slimming diets make you fat… What? No, let’s look at it again. Suppose I tell you: “the eyesight of people who have changed their spectacles several times is less good than the eyesight of people who use the same glasses for years, of who use no glasses at all: therefore glasses are bad for your eyesight”. You will put me straight at once, because you know that it is the eyesight problems that cause people to change their spectacles, and not the other way around.
You see where this is going: might it be that some people have a tendency to put on weight, and that this causes them to undertake one or more diets? Their weight would then be the cause of the (serial) dieting, and not its consequence.
The same reasoning could apply to risks of cardiovascular disease and mortality. There are observational studies which suggest that these risks are higher for people who have done many diets. But then, when a doctor sees an overweight patient who has other issues linked to cardiovascular problems, e.g. hypertension, that doctor will tell that patient to lose weight. So… People who are at risk for cardiovascular disease undertake more diets than people who are not. You can find more on this in the paper by Mehta et al from the University of Alabama (1).
Now you might well object: “But still, those diets don’t do any good! If they did, we’d know about it!”
Would we, though? How would we know if diets work, and to what extent? How can we know what the weight of the serial dieters would have been if they had not undertaken any diet? Higher or lower? And their general health, would it be better or worse if they had never dieted? To answer such questions, one would need a large group of people to study; these people would have to be randomly attributed to two separate groups, and what they eat would have to be subjected to standard controls throughout their life: one group would diet regularly, the other would be forbidden to. This is not feasible, at least not in humans. But with mice, it can be done, and a team from the University of Alabama has done it (2). Obese mice were randomised to 1) remain obese or 2) follow a succession of weight-loss diets and ad libitum feeding (weight cycling or yo-yoing) or 3) lose weight and remain stable.
The mice in group 1 survived less well than either of the others. In other words, remaining obese was worse than yo-yoing. More surprisingly, survival was similar in groups 2 and 3! This seems to mean that maintaining a small loss of weight has the same effect as losing weight several times in succession, even if the weight cycles back up between diets.
So, is weight cycling not so bad after all? Well, we are not mice, so let’s not be hasty in transferring conclusions. But on the other hand, we should accept that it may be unfair to treat yo-yo dieting as damaging to one’s health. Perhaps we should see weight cycling as a process where the higher or lower weight is linked to each individual’s potential and lifestyle.
For the moment, we have no proof that weight cycling has any health benefits for humans. But there is no proof either that it does any harm… to bodily health, that is. For morale and self esteem, yo-yoing is very bad indeed, that is a fact and a very important one. Since regaining weight causes feelings of failure and guilt, if healthcare professionals could transmit a clear message that any loss of weight, even transitory, is not worse than obesity, it is very probably that the psychological health of their patients would benefit.
Let me end with a little reminder. First, it is not weight as such that has the largest impact on health and mortality, it is the quality of what one eats; and many weight-loss diets do not cover all fundamental nutritional needs (3). Second, changing one’s eating habits is anything but trivial, and may aggravate any existing eating disorder. I can only recommend consulting a qualified professional dietician on this question.
A few references:
1Mehta T. et al. Impact of weight cycling on risk of morbidity and mortality. Obesity Reviews 2014 ;15 :870-881
2Smith D.L et al. Weight cycling increases longevity compared with sustained obesity in mice. Obesity 2018 ;26 :1733-1739
3Rapport de l’ANSES (mis a jour 2016). Quels sont les risques liés aux pratiques alimentaires d’amaigrissement ?