A triumphant headline recently popped into my inbox – ‘How Intermittent Fasting Could Transform Adolescent Obesity.’ Yes folks, the controversial ‘Fast Track To Health’ Trial results have been published in JAMA. I’ve written at length about this previously, but here’s a quick recap:
The Fast Track Trial Controversy
The Fast Track trial was a year long semi-starvation experiment conducted on higher weight Australian teens aged 13 to 17. For the first month, the kids endured a very low calorie diet (VLCD), where meals were replaced with Optifast shakes. This provided just 800 calories a day – less than a quarter of daily requirements. For the next 11 months, teens were separated into either a Fasting group, where for 3 days of the week they starved, ingesting just 600-700 calories, or a Continuous Dieting group, where calorie intake was restricted to between 1400 and 1700 calories a day.
The ‘Fast Track’ plan to inflict prolonged extreme diets on growing teenagers was shocking to me, and to many of my colleagues. There’s an enormous body of research demonstrating that diets produce small weight reductions in the short term followed by weight regain, and that prolonged restriction causes metabolic slowing.
Not only is it ineffective, dieting in adolescence – particularly extreme dieting – is strongly linked to eating disorders. Appalled, I submitted a complaint, co-signed by 28 other health professionals, to the ethics committee who’d approved the trial, calling for it to be stopped.
Upon receiving our complaint (and several others), recruitment for the trial was paused, but ultimately the committee (which was headed up by one of the Fast Track researchers) doubled down, claiming that the risks were justified by its anticipated benefits. While acknowledging the risk of eating disorders, they stated that they would adapt their ‘risk monitoring’ plan in order to better to manage them.
As an eating disorder therapist I know that there’s no such thing as a ‘safe’ way to starve teenagers. Frustrated and worried, in February 2019 I started a petition calling for it to be stopped, which quickly attracted more than 20000 signatures. People all over the world were horrified that in this day and age, teenagers could be exposed to such a pointless and risky experiment. Multiple eating disorder organisations released public statements calling for it to be abandoned. Elissa Myers, then President of The Academy for Eating Disorders, accused the team of playing ‘a very dangerous game’.
Media coverage exploded, but the Fast Track team doubled down, and it went ahead. All we could do was try to warn potential participants about the risks, so with the help of a wonderful group of advocates, we set up a website – fasttracktrial.com.au – which presented the scientific evidence related to the inefficacy and harms, both physical and psychological, of extreme dieting in adolescence. It also included lived experience stories from survivors of child and adolescent dieting. Our website didn’t go down well with the Fast Track team, who unleashed their lawyers, demanding that it be taken down and the domain name given to them. After a stressful legal battle, I was allowed to keep the website up (yay to free speech!), which is still there today.
For decades there’s been tension between researchers who have built glittery careers peddling weight loss and those – including eating disorder researchers and clinicians, fat activists, fat scholars, and those with lived experience of being harmed by weight loss regimes – who continue to raise the alarm over its inefficacy and harm. But incredibly, weight loss research continues to obtain public funding and those who push back are seen as ‘fringe’ or somehow ignoring human health.
The unprecedented Fast Track trial protests came as a huge surprise to the researchers, who were – for the first time – held to account. While complaints were being reviewed, trial recruitment was paused. Representatives from eating disorder organisations met with the team, and worked with them, making changes to consent forms to add information about efficacy and potential harms. Staff received eating disorder awareness training. The Fast Trackers increased the number and frequency of eating disorder screens during the trial, and hired a clinical psychologist to oversee the kids’ mental health. They also added a (non-compulsory) 2 year follow up to better monitor longer term impacts. An ‘independent data safety monitoring committee’ was appointed to oversee risk. The identity of this committee was concealed from the public.
All of this is important information. And yet it’s not mentioned in the published paper. A supplementary table is included, listing some of these changes, but with no context. Erasing the controversy upholds an illusion of acceptability. If a scientific experiment stirs unprecedented public concern, multiple complaints, and trial design changes, this should be fully disclosed and openly discussed. Such glaring omissions raise significant ethical questions about the transparency of the Fast Track team.
What Was The Point of The Fast Track?
The clinical trial register, which was written before publication, states that they wanted to ‘find out if intermittent energy restriction results in significant weight loss (via BMI z-score) after 52 weeks compared to a standard care reduced calorie intervention.’
This was a pointless question, as we already knew from adult research that fasting fares no better than everyday dieting, except that more people tend to drop out because it’s pretty awful to semi starve. At the time, fasting was all the rage, and the Fast Trackers insisted this was ‘important research question to answer’, but in the post-Ozempic world, when even the weight loss industry admit that diets don’t work, fasting seems so….old fashioned.
The Fast Track Teens
The teenagers were recruited from The Children’s Hospital Westmead in Sydney and Monash Children’s Hospital in Melbourne. Teens with an equivalent adult BMI between 30 and 45 were eligible to take part (those with BMI’s over 45 were deemed to have too many comorbidities, an interesting admission given that the trial was allegedly a fast-track to health). Teens were also required to have what the researchers termed an ‘obesity-related condition’ – prediabetes, insulin resistance (IR), acanthosis nigricans, hypertension, low HDL cholesterol (HDL-C) level, high triglycerides, elevated liver enzymes, or polycystic ovary syndrome*.
In order to run an experiment with enough statistical power to figure out if fasting was any different to regular dieting, they needed to recruit 186 kids – but failed to meet the target. Only 141 were persuaded to take part.
The average age was 14.8, half were male and half female. 2 kids were Indigenous, and 17 were born overseas. The team’s historical fondness for inflicting extreme weight loss experiments on children from Non English Speaking Backgrounds (NESB) continued into the Fast Track trial: almost half of the teens were from migrant parents, and 40% were from Non English Speaking Backgrounds (NESB). Around the same percentage were from socially disadvantaged backgrounds.
Stacking The Deck
At the beginning of the experiment the 71 Fasting (F) teens group were quite different to the 70 Continuous Diet (CD) group. CD teens were on average heavier, with a baseline BMI of 35.95 vs 34.83, with higher rates of insulin resistance (86.6% vs 76.5%).
To put this into perspective, at the start of the trial, as a group the CD’s weighed 377kg more than the F’s – roughly equivalent to the weight of 1500 Optifast shakes.
In scientific experiments it’s critical that the 2 groups being compared are very similar at the start, because any differences found in the end might be due to the pre-existing group differences (i.e., if the CD’s were larger & had more health issues to begin with, this could advantage the F group). When the 2 groups are significantly different from each other it’s called a ‘failure of randomisation,’ which reduces validity. In the real world we call it ‘stacking the deck’.
Jumping Off The Fast Track Train
31% of the teens dropped out before it ended, and surprise surprise – dropouts were almost DOUBLE in the F (40%) than in the CD group (23%). 7 kids in the Fasting group reported that the fasting diet was why they dropped out, compared to 1 in the CD group.
This came as a great surprise to the Fast Track researchers, because in their 6 month Fast Track pilot trial, the 21 teens who didn’t drop out filled in a survey rating the extreme diets as ‘easy’ and ‘pleasant’ to follow, and 12 ticked ‘yes’ in response to whether they thought they could stick to it for 1 year. I’ve written in detail about the unrelenting optimism of the pilot.
The Fast Trackers’ dedication to alternative facts is nothing short of breathtaking. In spite of the teens explicitly stating that the fasting diet was the reason they dropped out, they speculated that it was actually due to ‘the longer study duration and decreased contact with the clinicians’, or COVID lockdowns – which makes no sense at all, given the CD’s experienced the same conditions.
High dropouts meant that they only ended up with data for 97 teenagers, well short of their initial target of 186. This means that the entire study was statistically underpowered, falling below the threshold to detect differences between the two diets.
Fast Track to…Tiny Temporary Weight Change
The Fast Trackers chose BMI z score (BMIz) at 1 year as their primary outcome, a strange choice given that they’ve previously acknowledged that the BMI95 is a more accurate way to detect changes in higher weight kids. A quick (and hopefully painless) statistics lesson here – Z-scores are a standardised way of presenting how far a value sits from the average. At baseline, the teens’ BMIz score was 2.55, meaning that their weight was 2.55 standard deviations above the population average. By the end of the experiment, this reduced by 0.28. Shaving 3/10ths of a standard deviation off a BMIz score will not ‘transform adolescent obesity.’ In the words of the great Ragen Chastain, this degree of weight loss could be achieved with a loofah and a haircut.
Because the Fast Track didn’t include a control group – a group of kids whose diets weren’t tinkered with – there’s no way of detecting if the (small) weight changes may have happened anyway, during the course of natural growth. The kids did grow on average by 2cm during the 52 weeks.
Both groups lost an equally tiny amount of weight, so the very important research question – is intermittent fasting better than everyday dieting for teens? – was a firm no. The Fast Track was an expensive failure: $1.2 million dollars of public research money was spent on this guff.
Scrambling for meaning, the researchers bragged that the teens didn’t show signs of weight regain, and that: ‘…these findings contrast with adult data, which document weight regain between 6 and 12 months’. But this isn’t true – the Fast Track BMI data** collected at baseline, week 4, week 16, and week 52 clearly show a pattern of weight regain:
And of course this weight regain trajectory doesn’t count the high number of teens who’d dropped out (who are likely to have regained).
We can see from the pink shaded area in the graph that the VLCD in the first 4 weeks (where all kids endured 800/calories a day on weight loss shakes) is doing most of the weight loss heavy lifting. Hospital based weight loss clinics rely heavily on VLCD’s, which produce short term weight changes and teach people exactly nothing about having a healthy relationship with food.
The Fast Trackers’ JAMA article dazzles readers with complicated statistics (like the % of kids who reduced BMI95 by 5%), but no amount of spin can hide the fact that these kids endured a year of extreme restriction for very little weight loss, and were busily regaining it by the time the experiment ended.
Fast Track to….Health?
Five secondary health markers were reported in the paper: cholesterol, pre-diabetes, liver function, blood pressure, and insulin resistance. There were no improvements in cholesterol in either group, and no improvement in pre-diabetes. At one year there was a small improvement in impaired liver function tests (from 27.7% down to 17%) for the combined group of 90 kids who finished. Of the 90 completers, 47 started with high blood pressure, and 22 improved, but of 43 kids who started with normal blood pressure, 10 developed high blood pressure. This is concerning, but not discussed.
For insulin resistance, at 16 weeks both groups showed reductions (@ 20% reduction for the F’s and 35% for CD’s), but at 1 year only the CD’s showed IR improvements, and these were rebounding (@ 25% reduction).
This is wild – the only group difference found in the whole Fast Track trial showed CD was superior to Fasting for IR. And even in the CD group, we see a similar pattern to the weight regain trajectory, because IR was creeping back up.
Two important health outcomes listed in the Fast Track clinical trial registry were absent from the published paper: inflammatory markers and measures of resting energy expenditure (which measure damage inflicted by dieting). Also absent was a description of whether or not the kids actually stuck to the diets, and how much they found them to be ‘acceptable’. Perhaps the team are busily preparing another optimistic publication?
In summary, the health benefits obtained from this year long semi-starvation experiment were, to put it mildly, not exactly robust: small improvements in liver function and some blood pressure improvements counterbalanced by those who developed high blood pressure. There was a temporary suppression in insulin resistance, which was climbing back up at 1 year. The fasting protocol itself was no ‘healthier’ and in fact worse than everyday dieting for insulin resistance.
Fast Track to Health. Are you kidding? Are these lacklustre results really ‘worth the risk’?
Fast Track to…Harm
Dieting is a high risk behaviour, especially in adolescence. ‘Adverse events’ occurred in almost half (47.5%) of the teens. The most common were viral illnesses including COVID, ‘acute illness or injury unrelated to intervention’ , and gastrointestinal disturbance. One teen in the CD group developed gallstones and had an operation to remove their gallbladder. This is a common consequence of extreme dieting.
One teen in the Fasting group developed an eating disorder – Atypical Anorexia Nervosa (AAN), and 2 were withdrawn due to mental health concerns.
It’s incredibly hard to read about an adolescent developing AAN in this study. This was my concern from the first moment I heard about the trial, a fear shared by a myriad of health professionals, survivors of childhood and teen dieting, and eating disorder organisations all over the world. Anorexia is the most deadly of all the eating disorders***, and dieting in adolescence is a well established risk factor for developing them. We all knew this, and tried to stop this trial, but concerns were overridden because of the researchers’ misplaced conviction that the ‘benefits’ would outweigh the risks. Even 1 detected case of AAN is a risk not worth taking. I’ll say it again: there’s just no safe way to starve teenagers.
Fun (Concealed) Fact About Goal Weight
The Fast Trackers report that 5 F’s and 4 CD’s ‘met their goal weight’. But 2 of these kids – 1 from each group ‘did not attend the week 52 appointment’. This is obviously a red flag, but it’s not discussed. As it turns out, the ‘hit goal weight’ F group teen didn’t return for the 52 week appointment because they developed AAN and had been removed from the experiment. Goal weight was a WARNING SIGN, not an achievement. This is relevant information which should be clearly communicated – not left to an exasperated whistle blower.
Conclusions
The Fast Track trial was an expensive, underpowered failure which harmed several participants and created one severe eating disorder (that we know of). Risky experiments like this should be stopped. But the reality- denying Fast Track team claim that:
‘these findings suggest that intermittent and continuous energy restriction delivered as part of an intensive behavioral weight management program may both be beneficial options (emphasis added)’.
On what planet does it makes sense to recommend adding an intervention with no evidence of superiority, which caused way more people to drop out, AND caused at least one eating disorder ? Isn’t science supposed to be about adding benefits and minimising risks? The only ones who are benefitting here are the researchers, who are truly masterful storytellers.
I’ll be back to dissect the Fast Trackers’ second publication, which reported the results of ‘screening and monitoring’ for depression and eating disorders, in which more fantastic storytelling will be on display. If you’d like to be the first to read this, you can get early access by subscribing to my newsletter.
Stay strong diet culture dropouts!
*These ‘conditions’ occur in people of all sizes.
**Researchers have recommended using BMI change as a more powerful and interpretable way of measuring longitudinal weight change in adolescents than BMIz scores.
***Eating disorders in larger bodied people are just as serious and life threatening as eating disorders in low weight people.
If you are struggling with issues relating to food, your body, exercise, or with an eating disorder, I can help. I have offices at Flourish Kirribilli, in Sydney, and am also available for online appointments. If you’re a health professional I also provide clinical supervision, and my books are currently open.