The Fast Diet: A Fast Route to Disordered Eating?
The false promise of intermittent fasting
Posted November 5, 2014
‘I’m looking forward to wearing my party dress and eating lots of Christmas food – I’ll be fasting when I can but I also know that if I don’t lose weight in December then there’s still January and, well, the rest of my life’.
This, from a book (Kate Harrison's modestly titled The 5:2 Diet Book: Feast for 5 Days a Week and Fast for 2 to Lose Weight, Boost Your Brain and Transform Your Health [2012, p. 228]) whose opening line is ‘I am on a diet. But this one is different. No, really’ (p. 16). Really?
Ultimately, 5:2, the Fast Diet, and all their intermittent-fasting relatives are the single thing they least want to be: just another diet. That is, just one more moneyspinning way of entrenching people’s disempowered dissatisfaction with their bodies by promising to cure it. The diet promises control over bodyweight -- meaning, of course, the power to reduce it: ‘Step One: How much do you want to lose, and how much can you afford to eat?’ (Harrison, 2012, p. 14). It also promises better health and greater longevity. It backs up its claims with a smattering of science (in this case centred on the reduction of insulin-like growth factor, IGF-1, through intermittent fasting). Crucially, from the perspective of disordered eating in general and anorexia in particular, it involves calorie-counting and extended periods of hunger, because it requires the dieter to drastically reduce his or her food intake (down to 500 kcal) two days a week, even though an overall calorie intake isn’t prescribed for the week as a whole.
The basic idea is that intermittent partial fasting reduces levels of circulating IGF-1 and increases insulin sensitivity without lowering resting metabolic rate, and that this in turn reduces the risk of cancer and other age-related diseases, as well as promoting weight loss. Plenty of other people have criticised Michael Mosley’s brainchild, the Fast Diet (set out in a book co-authored with Mimi Spencer called The FastDiet: Lose Weight, Stay Healthy, and Live Longer with the Simple Secret of Intermittent Fasting [2013/2015]), for dressing up personal anecdote in scientific rhetoric and overstating the strength of the available scientific evidence on the biochemistry of the whole thing. It's also been criticised for fostering ambiguity as to whether the apparently distinctive benefits of ‘fasting’ require you not to eat at all for some amount of time, or just not to eat very much during that time. I won’t rehearse all those arguments here, but I do want to give a brief outline of what seem to me the main problems with the biological science, and then explore some of the related psychological problems raised by a diet strategy based on intermittent partial fasting, focusing in particular on how approaches like these may act as gateway diets to more disordered eating. I’ll focus on the variations on the 5:2 diet, since they’re the best known of the fasting diets around at the moment, but similar considerations probably apply to others in the same vein, like James Johnson’s Day Up Day Down alternate-day fasting protocol (complete with trademarked super-supplements).
First, the science behind it all. Even if you stick it out, stay on the wagon for more than a few months, and don’t develop an eating disorder along the way, is the Fast Diet really likely to help you lose more weight than the age-old staple of commercial diets everywhere: calorie restriction by any other name? And how likely are its other putative benefits to materialise? To hear Mosley talk, you’d think IGF-1 was single-handedly responsible for stopping us all being immortal superhumans – and, of course, that intermittent (partial) fasting is the best ever way to reduce it. Unfortunately, neither of these things is true.
IGF-1 does play a role in the ageing process, especially when present to excess, but the flipside of that is that it also – in interaction with growth hormone – has valuable anabolic functions in cellular growth and repair, while of course innumerable other variables feed into the processes of us ageing and falling ill. And the main weakness of the whole set-up is the current shakiness of the evidence that intermittent fasting reduces IGF-1 more than continuous calorie-restriction or restriction of carbohydrate. I was surprised to learn that, in line with Mosley's theory, resting metabolic rate isn’t reduced – indeed, may even be slightly increased as part of the body’s stress response – during fasts of up to 60 or maybe even 72 hours (Owen et al. 1979, Zauner et al. 2000, Heilbronn et al. 2005), since I’d always believed the old advice about kickstarting your metabolism by eating breakfast. Cortisol levels seem also not to be increased by short-term fasting, though its secretion rhythm may be altered (Ben Salem et al. 2003). But these pieces of supporting evidence don’t on their own, of course, make the whole theory viable. A review of the evidence on alternate-day fasting (ADF) and chronic disease prevention from human and rodent trials concludes that ‘ADF may effectively modulate several risk factors, thereby preventing chronic disease, and that ADF may modulate disease risk to an extent similar to that of CR [calorie restriction]’ (Varady and Hellerstein 2007). In other words, if you think intermittent fasting suits you better than continual calorie restriction, go ahead, but don’t expect wildly different results. (See also this CMAJ news item and Harvie et al.'s 2011 randomised trial involving young overweight women, which found that ‘IER [intermittent energy restriction] is as effective as CER [continuous energy restriction] with regard to weight loss, insulin sensitivity and other health biomarkers’ [my italics]. A 2016 systematic review (Davis et al. 2016), finding eight eligible studies, also found no differences in weight loss (and proportion of fat loss) between standard energy restriction and intermittent energy restriction, where overall energy restriction was similar.)
The state of the current evidence is such that even the most vocal advocates of intermittent fasting are forced to conclude that although various health markers associated with longevity and anti-ageing are improved in both humans and other animals in fasting, ‘fasting may not work by some magical pathway separate from caloric restriction. It may, but it hasn’t been established.' Any biochemical improvements brought about may well be explained by reference simply to loss of bodyweight, which is sought here by intermittent calorie restriction, but is achievable by all sorts of other means, with equivalent end effects.
And as always when it comes to nutritional science, things are more complicated than they seem at first glance. The distinctions between bound and circulating IGF-1, for instance, and the difference between IGF-1 produced by the liver (the standard route) and by the muscles (Gatti et al. 2012), seem to warrant more investigation, as does the relationship between IGF-1 and its seven binding proteins in their contributions to coordinating cell growth, including in cancer. And protein consumption seems to be another key variable when it comes to modulating IGF-1 levels – more protein, more IGF-1 (Bartke 2008) – but then so is carbohydrate intake. In a study using rodents, IGF-1's binding protein 3, which is negatively associated with tumour growth, was increased (and the IGF-1 : IGFBP-3 ratio lowered) by a no-carb ketogenic diet, which also reduced circulating insulin levels and made the mice lose weight (Freedland et al. 2008). Meanwhile, a ‘Western diet’ (with roughly equal parts fat and carbohydrate) resulted in the highest serum insulin and IGF-1 levels (and no weight change), with a low-carb diet somewhere in between.
There’s some evidence that the SIRT-1 gene, which is involved in cell maintenance and repair during times of minimal food consumption, and is often referred to as ‘the skinny gene’, is activated during intermittent partial fasting, and also when insulin levels are lowered, which occurs when consuming a diet low in carbohydrate (Draznin et al. 2012). If this is true, then intermittent fasting may just be a biochemically indirect, as well as a psychosocially problematic, route to the same beneficial place: lower insulin. If sugar is the real problem, there are more obviously sensible responses than eating almost nothing two days a week. I suggested in a previous post that lower carb (in combination with higher protein and fat) consumption could be a helpful way of regulating hunger post-recovery, and though of course there are potential problems with any kind of systematic restriction after recovery from an eating disorder, they seem far greater in intermittent fasting, even of the partial kind, than in lower-carb diets.
This is all complicated research in progress, and on all sides more work is needed, in particular the development of experimental methods to make possible more and better-quality studies involving human participants. In the end, there are always tradeoffs to be made: between performance and longevity – ‘Which do you prefer, having better muscle and cognitive performance or living longer?’ – and between control over bodyweight/bodyfat and a flexible, relaxed, and socially compatible diet. The Fast Diet’s general ethos, though, is to insist that it’s possible to have both: enhanced performance and a longer life, reduced calorie intake and unfettered feasting. This is obviously a large part of its appeal, especially among men who may not have dieted before, and may not be comfortable with the gendered sociocultural implications of ‘ordinary dieting’. But not only are the diet’s promises based on extreme cherrypicking from often equivocal scientific findings, they're also founded on an assumption – or rather, a covertly insinuated expectation – that the physical, psychological, and social effects of partial intermittent fasting should be easily dealt with: that someone eating ‘what they want’ five days a week, and 500 calories the other two days, can really be said to be ‘having it all’.
The original BBC Horizon programme that kicked this all off, ‘Eat, Fast and Live Longer’, should have made it painfully clear to anyone that fasting does dangerous things to people. Mosley began his new life with a 3½-day fast, consuming only water, black tea, and one 50-calorie Cup-a-Soup a day, and watching the intimately charted story of his responses to fasting was like watching someone moving in fastforward through the early love–hate phase of a seduction by anorexia.
First he goes through the process of coming to know all the ins and outs of hunger, like how it pulls you down – ‘It’s really quite discouraging going into work when you’re hungry’ – and how to ride its waves – ‘I’ve learned that hunger does not build and build but comes in waves that pass’ – and how its very deepest troughs are when you give in to it: the ‘empty spaces’ feel worst when he starts eating. Then there’s the training in how to feel guilty about eating, even when you’re not awake: ‘I dreamt that I ate a sandwich and then I felt fantastically guilty.’ There’s the initiation into what it means to be socially excluded by a refusal to eat: he goes out to dinner with his film crew, and sits there staring at the food they’re eating, eating nothing, his dinner waiting for him in his hotel room in the form of a kettle to boil for his mug of soup-water. There’s the bafflingly blinkered certainty of the fasting person – there’s nothing else, he says, that doesn’t hurt you and does this much good – and the social selfishness of that person, too: he never once thinks to ask (at least not to camera) how it makes anyone else feel, to have him sitting there refusing to partake, or having to go outside because his stomach’s rumbling so much. Ultimately what we’re left with is a caricature of the unanchored anorexic, full of false breeziness: ‘It’s actually twenty to two, and I don’t feel remotely hungry, but it is lunchtime, so if nothing else I think I’m going to go and prowl around.’ The fact that Daft Punk’s ‘Harder, Better, Faster, Stronger’ is the soundtrack to this scene would be hilarious if it weren’t so agonisingly ironic (how about their ‘More than ever / Hour after / Our work is never over’ versus his ‘there’s the rest of my life to lose weight’?).
The negative effects of fasting on sleep, alertness, cognitive-motor performance, and mood are well documented (Roky et al. 2000, 2004). Some online commentators who have tried 5:2 have expanded on these, adding in other candidates like irritability, obsession with food, lethargy and a constant desire just to retreat to bed. This isn’t to say that drastic calorie reduction is going to be equally difficult for everyone, but Kate Harrison’s book pays no more than lip service to the problems it entails. In ‘Step Two: Your First Fast’ it quotes 5:2 dieters’ top tips for coping with hunger, and these combine all the standard tropes of disordered eating with the jokey pretence that ‘hey, it’s all fine’. Just as in Mosley’s documentary, this is what makes them all the more alarming:
‘Have a plan for the week and measure out what 500 calories looks like, this will stop you obsessing about food ALL day long. Thinking of what you will reward yourself with is also good. I plan to have a nice Thai meal with all those “naughty” carbs.’ (p. 134) (Classic ED traits: and conversion of nutritional qualities to numerical form, leading to preoccupation with food due to undereating, leading to inflexible meal planning and disproportionate use of food as reward, plus attribution of moral qualities to specific food groups and their consumption.)
‘Try to leave breakfast as late as possible. When you eat in the morning it makes you feel like you want to eat more. I prefer to leave my meals as late as I can.’ (p. 135) (Deferring a restricted pleasure in order to control hunger.)
‘Drink lots of boiled water or herbal teas on your fasting day. Every time you feel peckish, have a herbal tea and you’ll find the flavour makes you feel as if you’ve eaten something.’ (p. 135) (Attempting to sate hunger using calorie-free bulk, typically liquids.)
Mosley and the dieters quoted in Harrison’s book all testify to the same thing: to succeed in this diet, you have to draw up very explicit battle lines between your mind and your body, between hunger you feel on fast days and your natural instinct to act on it on one side and the 5:2 doctrine that forbids eating on the other. This dynamic is at the heart of all restrictive eating disorders, and while of course it’s central to all calorie-restricting diets too, it becomes especially acute in fasting diets, where it’s the timing itself – the periods of extremely low energy intake and hence heightened hunger – that’s supposedly the source of their supposed benefits.
This set of problems may be particularly acute for women. Stefani Ruper’s excellent review of the evidence on intermittent fasting with an emphasis on the issue of sex differences draws on studies with rats that suggest a heightened sensitivity to starvation in females as opposed to males, in particular in terms of the adrenal stress response, which is harmful to reproductive systems and sleep cycles. One of these studies (Martin et al. 2007) suggests that this kind of physiological response in women may be a clue to the higher prevalence of anorexia in women than men: elevated hunger and heightened cognitive and motor activity in starvation, along with the accompanying reproductive shutdown, may increase female survival in periods of famine (see also Hoyenga and Hoyenga 1982), but may also represent an evolutionary basis for women’s greater vulnerability to anorexia thanks to its distinctive cognitive-physiological accompaniments. This means fasting may be more likely to act as a dangerously energising trigger for eating disorders in women than in men.
Along with the other differential effects of fasting on cholesterol, triglycerides, and glucose tolerance between men and women, as well as between overweight and normal-weight dieters, these considerations mean that women should perhaps be particularly wary of embarking on this kind of diet. Several of the commenters to Ruper’s post remarked on the addictive nature of the ‘hunger high’, which is raised by another online experimenter with fasting, Hugh Fearnley-Whittingstall: ‘there’s no evidence it triggers anorexia, but fasting can be addictive’, he writes. Of course the research hasn’t (yet) been done, but the risks seem clear enough. And for those suffering or recovering from eating disorders, being confronted on a daily basis with this kind of material can’t be easy. Though this post might just seem to add to the barrage, I hope an informed perspective on both intermittent fasting's questionable factual foundations and its obviously unhealthy superstructure might help forewarn, forearm, and generally protect.
Maybe in the end we’re reduced to the sad irony that the happiest possible outcome of adopting this diet seems to be a fairly high probability of failure to stay on it. One of the studies mentioned above (Harvie et al. 2011) found that among participants following six months of an intermittent-fasting diet (all of whom were overweight or obese at the start of the trial) only 58% planned to continue it after the end of the experiment, as opposed to 85% for the continual energy-restriction diet. There may be various reasons for this discrepancy. Davis and colleagues suggest in their 2016 review that the difficulty with long-term compliance for intermittent energy restriction [IER] 'may be attributed to the effect of the diet on quality of life, with a greater level of adverse effects experienced in the IER group, such as headache, lack of energy and problems fitting the diet into their daily routine (the diet suggested in this study was based on milk plus one piece of fruit and some vegetables, which may have been tedious)' (p. 298). The simple familiarity advantage of conventional dieting structures may perhaps also play a role. In any case, this finding goes against all the claims made about the sustainability of 5:2, and recalls Mosley’s accidentally insightful comment that ‘the biggest problem with prolonged fasting is me; despite knowing all the wonderful benefits, I just can’t bring myself to do it’. Sometimes the body is its own best protector.
The primary reason why the whole idea has been so popular is the same reason why it’s all such a bad idea. The Fast Diet appeals to people’s sense of drama: a defiant two days a week of being bloody hungry seems preferable to plodding along day after day with low-level hunger and restriction. And it dresses itself up as carte blanche to indulge most of the time, to give you what you want (fat loss) while letting you have what you really want (chocolate and almond croissants, in Mosley’s case). In this respect it’s similar to the initial hype around the Atkins diet (endless fatty fry-ups, yay!), but with an even more disingenuously disguised structure of opposing extremes, and much less science behind it.
Proponents of the 5:2 method don’t think it should stop at food: Oliver Burkeman wonders whether 5:2ing your whole life, from your drinking habits to your spending or even just generally worrying about stuff, could make you happier and more balanced, proposing the basic argument that moderation is better than abstinence. But the thing is, this isn’t moderation, this is yo-yoing between extremes of self-denial and self-indulgence. And in this key respect it’s terrifyingly like anorexia: like the 23 hours of gnawing hunger building up to a single faultlessly orchestrated feast, which is all that gives life pleasure and meaning. Harrison suggests that this diet has helped her rediscover what appetite and food mean to her, and that it ‘reminds me how lucky we are to be able to choose what and when we eat’. Let’s pause and think, hard, before we exercise our privilege to make dietary choices by making one that looks very much like slipping into the realm where preference gives way to compulsion, and an eating disorder begins.
I thank Graham Barrett for the original inspiration to write this post, and a blog reader for a more recent prod in the right direction.
Bartke, A. (2008). Growth hormone and aging: A challenging controversy. Clinical Interventions in Aging, 3(4), 659. Open-access full text here.
Ben, L. S., Bchir, S., Bouguerra, R., & Ben, C. S. (2003). Cortisol rhythm during the month of Ramadan. Eastern Mediterranean Health Journal, 9(5-6), 1093-1098. Paywall-protected journal record here.
Davis, C. S., Clarke, R. E., Coulter, S. N., Rounsefell, K. N., Walker, R. E., Rauch, C. E., ... & Ryan, L. (2016). Intermittent energy restriction and weight loss: A systematic review. European Journal of Clinical Nutrition, 70(3), 292. Paywall-protected journal record here.
Draznin, B., Wang, C., Adochio, R., Leitner, J. W., & Cornier, M. A. (2012). Effect of dietary macronutrient composition on AMPK and SIRT1 expression and activity in human skeletal muscle. Hormone and Metabolic Research, 44(09), 650-655. Paywall-protected journal record here.
Freedland, S. J., Mavropoulos, J., Wang, A., Darshan, M., Demark‐Wahnefried, W., Aronson, W. J., ... & Pizzo, S. V. (2008). Carbohydrate restriction, prostate cancer growth, and the insulin‐like growth factor axis. The Prostate, 68(1), 11-19. Direct PDF download here.
Gatti, R., De Palo, E. F., Antonelli, G., & Spinella, P. (2012). IGF-I/IGFBP system: Metabolism outline and physical exercise. Journal of Endocrinological Investigation, 35(7), 699-707. Direct PDF download here.
Harvie, M. N., Pegington, M., Mattson, M. P., Frystyk, J., Dillon, B., Evans, G., ... & Son, T. G. (2011). The effects of intermittent or continuous energy restriction on weight loss and metabolic disease risk markers: A randomized trial in young overweight women. International Journal of Obesity, 35(5), 714. Open-access full text here.
Heilbronn, L. K., Smith, S. R., Martin, C. K., Anton, S. D., & Ravussin, E. (2005). Alternate-day fasting in nonobese subjects: Effects on body weight, body composition, and energy metabolism. The American Journal of Clinical Nutrition, 81(1), 69-73. Open-access full text here.
Hoyenga, K. B., & Hoyenga, K. T. (1982). Gender and energy balance: sex differences in adaptations for feast and famine. Physiology & Behavior, 28(3), 545-563. Paywall-protected journal record here.
Martin, B., Pearson, M., Kebejian, L., Golden, E., Keselman, A., Bender, M., ... & Becker, K. G. (2007). Sex-dependent metabolic, neuroendocrine, and cognitive responses to dietary energy restriction and excess. Endocrinology, 148(9), 4318-4333. Open-access full text here.
Owen, O. E., Reichard, G. A., Patel, M. S., & Boden, G. (1979). Energy metabolism in feasting and fasting. In D. M. Klachko, R. R. Anderson, & M. Heimberg (Eds), Hormones and energy metabolism (pp. 169-188). Boston, MA: Springer. Paywall-protected journal record here.
Roky, R., Iraki, L., HajKhlifa, R., Ghazal, N. L., & Hakkou, F. (2000). Daytime alertness, mood, psychomotor performances, and oral temperature during Ramadan intermittent fasting. Annals of Nutrition and Metabolism, 44(3), 101-107. Direct PDF download here.
Roky, R., Houti, I., Moussamih, S., Qotbi, S., & Aadil, N. (2004). Physiological and chronobiological changes during Ramadan intermittent fasting. Annals of Nutrition and Metabolism, 48(4), 296-303. Direct PDF download here.
Varady, K. A., & Hellerstein, M. K. (2007). Alternate-day fasting and chronic disease prevention: a review of human and animal trials. The American Journal of Clinical Nutrition, 86(1), 7-13. Open-access full text here.
Zauner, C., Schneeweiss, B., Kranz, A., Madl, C., Ratheiser, K., Kramer, L., ... & Lenz, K. (2000). Resting energy expenditure in short-term starvation is increased as a result of an increase in serum norepinephrine. The American Journal of Clinical Nutrition, 71(6), 1511-1515. Open-access full text here.