Skip to main content

Verified by Psychology Today


Anorexia and the Diet Delusion: Healthy Eating Post-Recovery

Can eating more fat help regulate satiety and energy after weight restoration?

[**Update March 2019: Following a few recent ambivalent/critical reactions to this post, I've made a number of additions to help 1) clarify what I originally meant and 2) reflect changes in my thinking since. The most important point I want to convey that was probably not clear enough in the original post is that what I think may be helpful post-recovery is to eat more fat and protein in order to keep satiety and energy levels more stable. I.e., I'm not recommending eating fewer carbs in order to lose body fat.

Please also see the comments section below for questions, comments, and replies on topics like 'fearing' hunger and how hunger changes for different people at different stages in recovery, and generally more elaborations on my views and valuable perspectives from others.

Above all, please note: this post is directed at people who are physically fully recovered from anorexia: for whom weight restoration to a point where bodyweight maintains itself stably without restriction has been completed. If you're not at that point yet, read the post if you want, but please remember that it isn't relevant to you now, and that a lot will have changed for you before it is. For now, your most important task is to eat whatever you think will help you keep leaving anorexia further behind.**]

For some time now, I’ve felt as though I have been eating healthily. When I have this thought, I do so of course in the context of my history of anorexia. By this I don’t mean, however, that the statement is merely relativised: I eat as healthily as can be expected for someone who used to have anorexia. Rather, a history of clinically disordered eating perhaps entails qualitatively different parameters for its opposite, healthy eating, than apply to the wider population. However, there are many similarities too; it’s by no means the case, I think, that sufferers from eating disorders are destined always to eat differently from others. It would be hard to maintain this whilst also believing (and presenting evidence for the fact that) full recovery from eating disorders is possible. Clinical surveys vary in their assessment of the long-term prognosis for anorexia—and indeed defining recovery is problematic in itself—but a review of relevant studies ten years ago suggested that more than 50 percent of sufferers fail to recover fully. Nonetheless, that means that nearly 50 percent (46.9 percent) do recover fully. We might argue about definitions (I criticise the use of implausibly low BMI thresholds as markers of recovery here); but I suppose people know when they are fully recovered, if not exactly when they make the transition—certainly I’m quite sure.

Perhaps in fact there’s only really one different parameter when it comes to healthy eating specifically for someone who used to have anorexia, and this can be summed up as: everything in moderation. I’ll try to expand on this as I go along. (I apologise in advance for the length of this post; it just ended up needing this much space.)

A word first about why I’m writing this post. I was about to say that in some ways it goes a little beyond the remit of the blog, since it’ll include elements of nutritional science and will focus on life after physical recovery is complete and psychological recovery is well on the way to being, not on the illness or the weight-restoration phase of recovery. But then I realised that this isn’t atypical of my posts more generally, and that later phases of life after recovery have a crucial place in the discussion of eating disorders.

In any case, one recent prompt to write about this was a message from a reader who is trying to resist comparing her food intake with that of various female members of her family, who practise various diet-related habits in the interests of eating ‘healthily’. This of course raises the question as to what assumptions her family are making about what eating healthily means. It also ties in with anorexia’s origins, which often lie in teenage dieting (i.e. eating ‘more healthily’ in order to lose weight), often through emulation of older women, often relatives, with the same aims. In what follows, I’ll explore what healthy eating may or may not mean for the general population, and how these broad considerations yield a particular conclusion as regards anorexia and hunger.

Everyone in recovery from anorexia has to come to terms with the fact that the world of ‘normal eating’ which (s)he is at last regaining entry to is full of people who eat unhealthily, whether primarily in physical or mental terms, whether to extremes or barely perceptibly. Some people are addicted to junk food, others are addicted to ‘healthy eating’ regimes (sometimes referred to as orthorexia); some people order their lives around their diets and measure their self-esteem with their waist sizes; a great many people in the industrialised world live with low-level uninterrogated longings to be thinner and therefore, they believe, happier, and live more or less permanently with the calorie-restricted diets they believe will achieve this for them – because they assume everyone else does too, and that this is normal and as good as things can be.

This is all a shock to the person who has invested enormous effort in reaching a healthy weight again, and who has been assured by her therapist or any number of self-help books and blogs and articles that a healthy body weight is the way to a healthy mind. It is, of course, the only way (for someone with anorexia), but it also isn’t a guarantee—it is necessary but not sufficient. A great deal of hard work is necessary to translate a healthy body into a fully healthy mind. But as I’ve described, all the effort of recovery can ultimately mean that the formerly sick person has more and better psychological and behavioural resources to combat the innumerable dietary delusions out there than people with no history of eating disorders have. The key is to keep these in practice as recovery recedes further into the past—in the sense of automatically helpful reflexes against encroaching immovable food rules, say, or against missed meals.

This well-trained healthy attitude is constantly confronted, though, with its opposite in other people—even, or especially, people who claim they are living the epitome of healthily. Perhaps the primary problem here is the pervasive equation of healthiness with thinness, or with the loss of weight. (And note that this is usually talked of in terms of undifferentiated loss of ‘weight’, rather than as targeted fat loss. Even if the latter is what people mean, the methods they use to achieve it often result primarily in fluid and muscle loss.) There are many reasons why thin doesn’t necessarily equal healthy, especially the importance of the specific location of fat deposits, as well as other factors like hormonal balance and cardiac fitness and muscular strength and mobility. And of course, anorexia is an excellent example of what happens when we take this received wisdom to its illogical extreme—the problem being that no one ever tells you where thin stops being good and starts being deadly.

The diet industry, in all its manifestations from diet books to diet foods to dieters’ fitness programmes to diets that aren't diets but lifestyles, feeds and feeds off this indiscriminate desire to be 'just a bit slimmer'—and it also feeds off the fact that the vast majority of the diets it peddles, and their many lifestyle accoutrements, don’t work, so people keep coming back for more. I’ve done quite a bit of reading on the subject of nutrition over the years, and the most persuasive writing I’ve come across on why conventional calorie-restricted diets don’t work—persuasive because, in the main, engaging carefully and thoroughly with the scientific literature—is by the science journalist Gary Taubes, notably his 2010 book Good Calories, Bad Calories. (This book was published as The Diet Delusion in the UK and Australia, and I’ll refer to it by this title, because I prefer it for its richer associations: the pun on ‘diet’ prefigures the book’s debunking of the myth that ‘diets’ in the conventional sense are our only weight-loss option, and hints at the delusions we may be subject to as regards diet more generally.) Taubes covers a great deal of ground here (there’s a digested version for those without time for the full account, called Why We Get Fat and What to Do About It), but his conclusions are set out concisely in the epilogue (p. 454—also reproduced here).

Because much of the developed world seems to be obsessed with weight/fat loss, Taubes's work has been interpreted mostly in terms of its relevance to that side of things. But what's more relevant to our purposes here is that on his account, calories aren’t key to weight loss and gain or fluctuations in hunger and satiety levels, carbohydrates are:

Insulin is the primary regulator of fat storage. When insulin levels are elevated – either chronically or after a meal – we accumulate fat in our fat tissue. When insulin levels fall, we release fat from our fat tissue and use it for fuel. […] By driving fat accumulation, carbohydrates also increase hunger and decrease the amount of energy we expend in metabolism and physical activity.

These conclusions challenge the received wisdom that consuming excess calories is what makes us fat; that dietary fat is what makes us fat; and that insufficient exercise is what makes us fat. The standard way of thinking about body fat and food is that we become fat because we eat too much. Taubes argues that, on the contrary, we eat too much because of hormonal and other biological mechanisms that drive us to become fat (depositing and not burning fat). Other examples of these mechanisms include the glycerol phosphate molecule produced from glucose, which plays an important role in binding fatty acids together into triglycerides: ‘a product of carbohydrate metabolism—i.e., burning glucose for fuel—is an essential component in the regulation of fat metabolism: storing fat in the fat tissue’ (p. 388). Again, what matters most for my argument here is that the proposed mechanisms of fat loss versus gain entail greater fluctuations in hunger when there's proportionally less fat and more carbohydrate in the diet.

[**Update March 2019: Taubes's conclusions and hypotheses are by no means mainstream, and like almost everything in the world of nutritional science, they are hotly contested. Empirical studies have been conducted finding support for the carbohydrate—insulin hypothesis and its many sub-hypotheses, and others have been conducted finding evidence against them. The experiments and the debates continue. A cogent set of criticisms of Taubes by neuroscientist Stephan Guyenet is available here, and you can find other exchanges between the two here and here. The science keeps moving, and my sense is that the carbohydrate—insulin hypothesis, in whatever refined and qualified form, will remain an important contributor to our slowly improving grasp on how food affects both appetite and body composition. But don't take Taubes as fact any more than you would any other set of scientific claims or interpretations of evidence. He certainly did everyone a favour in questioning the fat-is-bad dogma. Beyond that contribution, we'll have to wait and see how the science pans out.**]

To me, the insulin hypothesis is relevant to someone recovered from anorexia in one key area: hunger. Taubes presents evidence to suggest that carbohydrates in the diet create hunger by promoting fat storage (making fatty acids unavailable as fuel), and that undereating—that is, eating in order to create an energy deficit, in conventional diets by reducing fat and therefore increasing carbs—causes hunger, and is therefore ineffective as a dieting strategy in the long term. Sooner or later, the dieter will get sick of being hungry, and will abandon the diet. A slowed metabolism and likely muscle loss will then make it all the easier to regain the lost fat.

What about anorexia, then? Why doesn’t this logic apply to people with anorexia who very successfully lose weight and keep it off? Taubes, unfortunately, isn’t good on anorexia. He discusses it only once at any length in The Diet Delusion. He recaps the hypothesis that hunger and satiety are responses to the insulin-driven cycles of fat storage and fat mobilisation. And then he goes on:

One of the most radical implications of this hypothesis is that even such an intractable condition as anorexia nervosa—which, like obesity, is now universally considered a behavioral and psychological disorder—may be caused fundamentally by a physiological defect of fat metabolism and insulin. The behavior of undereating may be a compensatory response to a physiological condition, just as the behavior of overeating can. Any hormonal abnormality that makes it difficult to store calories as fat—the fat cells, for example, becoming prematurely or abnormally resistant to insulin—could conceivably induce a compensatory inhibition of eating behavior and/or an increase in energy expended. What appears to be purely a behavioural phenomenon, the anorexia itself (and perhaps even bulimia nervosa), would be the compensatory response to a physiological problem, the inability to store calories after a meal in the energy buffer of the fat tissue. Correctly identifying cause and effect in these conditions would be difficult, if not impossible, without the understanding that there is an alternative hypothesis to explain the observations. (p. 440)

These suggestions (and of course he does only present them as tentative suggestions) ignore—or reveal ignorance of—several key facts about anorexia.

First, for example, people recover from anorexia, and it is unlikely that recovery (which usually occurs through a combination of psychological therapy and increased calorific intake, in a wide variety of proportions of carbohydrate to fat and protein) always just happens to coincide with the end of the fat cells’ resistance to insulin or similar.

Secondly, as I’ve discussed before, the boundaries between anorexia and other eating disorders, notably binge-eating disorder and bulimia, are so fluid, and the clinical trajectory from one to another so common, that many practitioners find their categorisation as separate conditions unhelpful. It isn’t clear how Taubes’s hypothesis would account for the fact that the transition from, for example, anorexia to binge eating occurs more often than not.

Thirdly, mere enforced undereating (as in the Minnesota Starvation Study) effectively induces a condition remarkably similar to anorexia, while refeeding can then cure it.

Fourthly, the implication here is that people with anorexia don’t feel hunger—which is one of the myths I think most urgently needs demolishing, since for the majority it's clearly not true: denying hunger is not at all the same as not feeling it in the first place.

Fifth, this argument implies that people with anorexia eat and eat and nonetheless fail to put on weight, which is a bizarre misunderstanding of a condition that is well known to consist in eating too little. It is true that in some cases, as with me, a stable diet is established that seems relatively generous in calorific terms—certainly more so than many conventional weight-loss diets—but not to the extent that a genetic insulin resistance need be posited to account for the absence of weight gain. A calorie-restricted diet that is followed rigorously and permanently does result in continuing weight loss to a point beyond which the body can no longer function; this is why people can starve to death. (Conversely, a relatively high-calorie diet that is followed rigorously and permanently from an underweight starting point does result in continuing weight gain to a point where reduction in appetite and increase in metabolic rate together generate bodyweight stability. This is why, as I've described, weight gain in recovery from anorexia can proceed, at least in the early stages, with relatively predictable ratios of calories consumed to kilos regained.)

There are two crucial facts about eating disorders that Taubes neglects, which could in fact end up supporting his overall argument about hunger:

1) People who have anorexia grow addicted to the sensation of hunger itself, and 2) anorexia often culminates in binge eating or another eating disorder.

Hunger addiction in anorexia may well have hormonal elements (as in the endorphin-fuelled ‘hunger high’), and these contribute to buttressing its central position in the value system that is built up around it (hunger = self-denial, strength, power, purity, specialness, etc.), and thence to strengthening the behavioural web of habits that preserve it. This addiction is why anorexic ‘diets’ succeed where others (whose aim is significant and sustained weight loss) fail. The reasons why this addiction develops in some people and not others need to be better understood, but the fact is that some people with anorexia go on undereating year after year, sometimes decade after decade, living with hunger of an intensity and permanence that others are unable or unwilling to accept. This links to the second point, which is that in reality, in the majority of cases (where neither recovery nor death occurs first), anorexia mutates into another eating disorder: eating consistently too little gives way to alternating between too little and too much, and/or vomiting to reduce calorie absorption. This brings anorexia back into the realm of the ordinary, flawed diet, and supports Taubes’s claims as to why such diets ultimately fail. The only difference is how long anorexia often lasts compared to a typical calorie-restricted diet.

For those who do not make this transition to another form of eating disorder, recovering from anorexia involves learning to cope without hunger in the sense that other people have to learn to cope with it when they embark on conventional semi-starvation diets. Some people never manage this, and never get fully better. Others do manage it, retraining our appetites as our body weight increases, relearning how to listen to hunger, and creating lives in which we don’t feel hungry for extended periods, and in which hunger is a signal to eat rather than a signal to be ignored. I'm simply no longer willing to put up with the tedious pain of prolonged hunger, and I don’t need to any more. And this is as it should be. This kind of relationship with hunger, I think, is a key element of what it means to eat healthily.

Thanks to the general continuing demonisation of fat (particularly saturated fat), many people keep eating much more carbohydrate than fat post-recovery, which may be one factor that helps make this kind of balance hard to find. And the greater trouble in the post-anorexic context is that a small slip—a slight movement towards relapse in a time of stress, say, when the old coping mechanisms of dietary restriction and resulting weight loss fall imperceptibly back into place—can easily be magnified if, as is likely, the dietary restriction takes the form it takes in most of the diets we’re familiar with. The primary way we’ve probably all grown up being taught that weight loss happens is by restricting fats (because they give you the most energy per gram) and upping the proportion of (‘good’, unrefined) carbs. But the trouble with this route for the person not long recovered from anorexia is that hunger increases because insulin secretion increases. Then the old hunger addiction may be reawakened, and a full-blown relapse is more likely to occur.

This is why I think it worth writing about The Diet Delusion in this blog: it offers a different perspective on bodyweight maintenance and hunger. The common low-fat, (fairly low-protein,) calorie-restricted diet is unlikely to be a healthy way for anyone to eat for long periods, and it is certainly the way in which failure to recover fully from anorexia is most often manifested, as people ostensibly recovered insist on maintaining their body weight at a pre-determined level through dietary restriction. Because of the way most of us were brought up to think about healthy eating, this kind of relapse into restriction usually involves getting rid of a lot of the fat in the diet. It then usually results in muscle depletion, a slowing of the metabolic rate, and constant hunger and consequent preoccupation with food.

Taubes is writing for and about a Western population for whom semi-starvation diets are dangerously ineffective; people recovered from anorexia are a sub-section of that population for whom such diets can be dangerously effective. But an understanding of the biology is equally important to both groups, I think. In the latter case, I wonder whether understanding hunger and its causes better, rather than allowing us to lose or rigidly maintain bodyweight more effectively, might let us know that an inclination to lose fat can be acted upon without inducing hunger, and without triggering all those old destructive associations. This applies whether the inclination is what we might call either ‘unhealthy’ or ‘healthy’: prompted by some unrelated life event that increases stress or reduces self-esteem, or prompted by a sensible, measured desire to make oneself healthier in relation to, say, the heart or one’s joints.

I appreciate that this weapon could be a double-edged sword. Learning about and adopting any nutritional perspective that presents one food group as problematic may be dangerous for someone who has had anorexia. Once the insulin hypothesis is entertained, the demonisation of dietary fat is undone, but sugars become the enemy of (physical) health. And for someone recovered from anorexia, the specific associations between bodyweight and psychology mean that the physical benefits of a certain dietary strategy have to be weighed against the possible psychological costs of consistently pursuing any ‘abnormal’ strategy.

As I mentioned at the beginning, comparison and competition with others in the food context—primarily trying to make sure one eats less than everyone else—is a very prevalent characteristic of anorexia, and coping with other people’s diets can be difficult during illness, recovery, and to a lesser extent even afterwards. But perhaps a different perspective on the whole concept of dieting—an understanding, for example, that a reduction in ‘calories in’ (or energy intake) will always result in a reduction in ‘calories out’ (energy expenditure), through mechanisms like a change in metabolic rate; or that eating less fat in practice usually means eating more carbohydrate—can help us change the parameters of the comparisons and stop fixating on sheer quantity, or on probably misguided notions of healthiness. This is not to advocate that once you're recovered you adopt a holier-than-thou attitude to other people’s dietary decisions, nor that you open up a gulf between you and those around you; anorexia itself does both these things beautifully. But it does perhaps mean being willing to challenge inherited notions of what healthy eating means and why hunger takes hold, and to experiment with them in our lives if it seems like the benefits might outweigh the risks.

As I write this post I keep hesitating as to the wisdom of bringing nutritional science often reduced to a 'carbs-are-bad' soundbite into a blog about anorexia at all, and I am aware that for some recovered people ignorance may be bliss (or mental health) when it comes to the ongoing debates on nutrition: given that there is no single right answer, and given that immersing oneself in evidence and counter-evidence can easily increase one’s obsessiveness to no very salient purpose, it may be better just to eat what one likes, or what people around one eat, and go no further. I certainly did this myself for some time after recovery, and was happy that way. But I guess that, like me, lots of people in my situation don’t remain happy that way. It’s also simply unrealistic to maintain that no one recovered from anorexia will ever want to actively modulate his/her own body shape or weight, and it seems to me that being informed is better than being uninformed.

[**Update March 2019: I'm very much in two minds about that last sentence. Body weight and shape should perhaps be treated differently here. On body weight: I know I've acted on 'harmless' reasons for wanting to modulate my weight since recovery, specifically to 'make weight' for powerlifting competitions. But in the end I concluded that that was unhelpful, and that competing without worrying about ending up in the most competitive weight class was much more sensible—for anyone, and especially for someone who's learned as much as I have about what I want my bodily existence to be and not be. When it comes to body shape, I haven't ever tried to actively modulate it since I got better, but I do like how it's changed as I've got stronger and healthier over the seven or so years since I've been fully recovered: I like the fact that I look strong, because I am: that my lats and my quads, for example, are pretty big; that I don't look like the frail creature I used to be any more, because I'm not. I also like the fact that my inner thighs have plenty of anti-thigh-gap fat; that I don't look like the frail creature many people want to be, because I want nothing of the kind. As I say, this isn't active modulation, exactly, but it is active appreciation of what has changed because of things (sporting activity, diet) I've cultivated with awareness of their physical effects on size and shape. It's complicated. I read the original sentence now and it sounds like a bit of a pseudo-recovered sentence. But I think we all know when we're doing pseudo-recovered kinds of modulation and when we're not. So hold on to that instinctive knowledge, whatever you do.]

And ultimately, the philosophy of ‘everything in moderation’—so counter to the essence of anorexia—must be key here. In the early post-recovery days, you have to guard against fanaticism especially in food-related realms. (Guarding against fanaticism wouldn’t be a bad philosophy for everyone else, either, given how often it's the starting point for an eating disorder or the way subclinical disordered eating manifests.) The possible gains in cardiac health or hormonal balance of, say, never eating pudding again are outweighed, even more comprehensively for you than for anyone else, by the damage to mental health and wellbeing and the possibility of descent into disorder when any single rule about food becomes uncontravenable.

So, I like living with an informed perspective on these issues, but I’m careful not to live solely according to the nutritional theory. Taubes argues persuasively, in The Diet Delusion and elsewhere, that human beings (and all animals) aren’t ‘thermodynamic black boxes’ subject to the laws of energy in versus energy out. But we also need to remember that the ‘evolved complex systems of hormones and enzymes and proteins’ that constitute our bodies also bring with them psychological causes and effects, and these may well mean having a dessert in a given instance is healthier for the organism as a whole than not having it. Again, the balance is here slightly differently weighted, I think, for those who have had an eating disorder in the past than for those who haven’t. Thus I generally eat a diet high in fat and protein and a moderate amount of carbs (my enthusiasm for strength training means I eat more carbs as a convenient fuel than I perhaps otherwise would), but there are plenty of exceptions to this general rule, and I cherish the exceptions (the pasta, the chocolate, the alcohol, etc.) as contributing to rather than diminishing the healthiness of the way I eat, seen holistically.

[**Update March 2019: This is still a broadly accurate description of my diet, though my carb intake is higher now than when I first wrote this piece, partly because I now (for ethical reasons) eat much less meat than I did from mid-recovery through to a few years after recovery. I still find that eating plenty of fat and protein keeps my appetite and energy levels stable in a way a high-carb low-fat diet doesn't. I still love plenty of kinds of carbohydrate, and eat them as much as I want to. My strength training is much less goal-orientated now, more a refreshing interlude in my day (max. three times a week!) than something where progression and competition are central, and my hunger has doubtless changed accordingly too, in terms of both how much I eat and the kinds of thing I eat.]

Seen holistically, the many possible functions of food ought also, I think, to be balanced against one another. Food is, primarily, a fuel for everyday activities; it can also be a fuel for extraordinary ones. Food can be a means of moulding one’s body to one’s physical ideals, or a way to let one's body reflect one's broader political or existential ideals. Food is a socially cohesive force and an identity-shaping one. It promotes physical health and ill health. It alters mood; and it can be a source of simple pleasures, and a mediator of complex ones. Perhaps never letting any one of these functions or roles eclipse all the others is a decent recipe for healthy eating for those who once had anorexia.

In the experience of anorexia, food is freighted, far more than for most other people, with a great number of value judgements, and its selection, preparation, and consumption are distorted into spectacularly odd forms, from minuscule weighing and measuring and sub-dividing to elaborate accompanying rituals to simple deferral, all supporting the valuation of food as danger or threat to be resisted, or, conversely, and simultaneously, ultimate delight to be worshipped. After recovery, food can be released from these shackles of immovable routine, and can be the fuel for happier things. But I think that for anyone who has ever suffered from anorexia, it’s important to bear in mind that of the many things it can be, food is none of them exclusively, or even predominantly, in a really healthy body and mind.

More from Emily T. Troscianko
More from Psychology Today