Anti-Diet: Reclaim Your Time, Money, Well-Being, and Happiness Through Intuitive Eating

Overall score

Scientific accuracy

Reference accuracy

Healthfulness

How hard would it be to apply the book's advice? Very easy

Book published in 2019

Published by Little, Brown Sparks

First Edition, Hardcover

Review posted January 3, 2023

Primary reviewer: Morgan Pfiffner

Peer reviewer: Stephan Guyenet

Table of contents

1. Introduction
2. Scientific accuracy
3. Reference accuracy
4. Healthfulness
5. Most unusual claim
6. Other
7. Conclusion
8. Updates

Claim 1

Dieting does not lead to long term weight loss

Supporting quote(s) and page number(s)

Page 85: “…most diets appear to work in the short term but don’t lead to long-term weight loss”

Page 84-85: “A robust body of evidence shows that intentional weight-loss efforts don’t work; with a failure rate that many researchers agree is north of 95 percent”

Page 89: “And a large-scale 2015 study of more than 278,000 people found that within five years, the proportion of people who’ve regained all their lost weight (or more) is somewhere between 95 and 98 percent”

Page 92: “The only thing diets are really good at is causing people to gain weight over time”

Page 158: “If you think your current body size is a problem, dieting gives your a 66 percent change to make that ‘problem’ a bigger problem”

Criterion 1.1. How well is the claim supported by current evidence?

3 out of 4

This claim received a score of 3, indicating that it is moderately supported by current evidence. AD describes dieting — defined as ways of eating that restrict calorie intake — as unsustainable and ultimately unable to produce long term weight loss except in rare cases involving extreme effort. AD agrees that weight loss can occur initially, but claims that over the long term various factors (including changes in hunger, cravings, and energy expenditure) lead to complete regain of the weight lost, even stating on page 47 “intentional weight loss efforts have been shown to cause long-term weight gain for up to two-thirds of the people who embark on them”. AD suggests that at least 95% of diets fail, meaning that dieters gain nearly all lost weight back within around 5 years.

Diet trials do consistently indicate that people placed on diets, regardless of the diet, typically experience weight loss within the first 6 to 12 months, followed by weight regain over the next several years. This is indeed a consistent finding in the scientific literature, although the ineffectiveness of dieting seems to be overstated in AD. One meta-analysis of diet trials found that in the five years following initial weight loss, an average of around 80% of the weight lost is regained. This finding, while disappointing, is a more optimistic picture than the one painted in AD.

Additionally, the same meta-analysis noted that dieters engaged in high amounts of exercise had maintained around double the amount of weight loss compared to the dieters getting low amounts of exercise. This suggests dieting may be more successful when combined with regular exercise (this does not mean exercise leads to long-term weight loss — it usually has little impact on its own — but rather that it can support weight loss maintenance). As for the claim that 95% of diets fail, a number of long-term clinical trials suggests this is by no means a hard-and-fast rule:

– In the Look AHEAD trial, 5,145 people with overweight or obesity and type 2 diabetes were randomly assigned to either an intervention involving diet, exercise, and psychological support or a control group receiving general diabetes support and education. After 8 years, 50.3% of people in the intervention group lost at least 5% of their initial body weight and 26.9% of people lost at least 10% of their initial body weight.

– The Lifestyle Heart Trial was intended to study the effect of diet and lifestyle changes on cardiovascular disease. Participants were randomly assigned to an intervention involving a low fat diet based primarily on whole plant foods (along with stress management, psychosocial support, and aerobic exercise) or a usual diet control group. After 5 years, participants in the intervention group had achieved a weight loss of 12.8 pounds (after initially losing 24 pounds after 1 year) compared with little change in weight in the control group.

– In one noncontrolled trial, 112 people with obesity followed a weight loss program, which centered around following a very-low-calorie, liquid diet for 12–14 weeks, followed by a weight maintenance program, which involved advice to eat a low fat, high fiber diet and skills to help participants with dieting. After 7 years, 25% of program participants had maintained a weight loss of at least 10% of their initial body weight (around 11kg/24 lbs).

So while it is well-established that, for the majority of people, most of the weight initially lost via dieting is later regained, there are clearly people for whom certain diets (often combined with exercise and psychosocial support) can lead to significant amounts of long-term weight loss.

Criterion 1.2. Are the references cited in the book to support the claim convincing?

2 out of 4

The references received a score of 2, indicating that they are weakly convincing. A number of studies are cited to support the claim that dieting usually does not lead to much long term weight loss. However, some of the studies cited have important limitations. For example, AD cites a UK study on free-living people, claiming it shows that 95 to 98 percent of people regain all their lost weight within 5 years. However, this study was based on health records and did not actually assess whether participants were dieting, so it probably shouldn’t be used to make claims about the success rate of dieting.

At several points, AD implies or directly claims dieting leads to long term weight gain. On page 91 AD writes of diets: “they tend to result in weight gain over time”. Several studies are referenced to support this idea, but there are important caveats with these studies that are not well addressed. AD cites observational evidence in which people who frequently dieted were more likely to gain weight, but one issue with such data is that the people more likely to diet may be more prone to gaining weight. The references only show that people who diet also frequently gain weight later, not that dieting causes this weight gain. And when looking at randomized controlled trials, which are the most rigorous way of answering this question, dieting does not seem to result in long-term weight gain.

Criterion 1.3. How well does the strength of the claim line up with the strength of the evidence?

2 out of 4

The strength of the claim received a score of 2, indicating that it is moderately overstated. Existing evidence does support that diets don’t produce significant amounts of sustained weight loss in most people, but the available literature does not appear in line with claims by AD that dieting is ineffective for 95 to 98% of people (when looking at weight loss after 5 years). Furthermore, AD claims dieting leads to long term weight-gain for most people, which is not supported by randomized controlled trials.

Ultimately, a more nuanced interpretation of the literature is that most diets don’t cause as much weight loss as people hope for, and the effectiveness of a diet depends on various factors, including diet adherence, physical activity, positive social and psychological support, the food environment, and having healthy sleep habits.

Overall (average) score for claim 1

2.3 out of 4

Claim 2

Obesity does not directly harm health

Supporting quote(s) and page number(s)

Page 47: “Instead, the ‘obesity epidemic’ is really a moral panic that has a lot more to do with diet culture’s skewed beliefs about weight than with any actual threat to public health”.

Page 56: “She [Marion Nestle] builds her argument against the food industry on the assumption that larger body size equals poor health, and that overeating of eating particular types of food is the cause of larger body size—suppositions that are based on diet-culture beliefs rather than on scientific fact”.

Page 144: “…if all studies were to control for weight cycling, any excess risk from so-called ‘overweight’ or ‘obesity” might disappear”.

Page 158: “By now I hope you understand that most chronic diseases blamed on weight can most likely be explained by other phenomena, such as weight stigma and weight cycling”.

Criterion 1.1. How well is the claim supported by current evidence?

0 out of 4

This claim received a score of 0, indicating the evidence for the claim is opposed by current evidence. When it comes to the relationship between excess body fat and disease outcomes there is a large body of evidence noting a connection between the two. Observational evidence shows that people with obesity are at a higher risk of developing type 2 diabetes, cardiovascular disease, certain types of cancer, among other health conditions, compared to people of “normal” body weight.

AD does not dispute that such associations exist, but rather claims the link is not causal. AD suggests that confounding variables not considered in observational studies explain the association between obesity and poor health outcomes. However, much of the evidence suggesting excess body fat increases the risk of certain diseases is based not only on observational evidence but intervention studies as well.

Clinical trials have found that losing weight (achieved via diet, exercise, and medications, often in combination) improves risk factors for heart disease and type 2 diabetes, including systolic blood pressure, LDL, and blood glucose levels, with greater weight loss tending to cause greater improvements in these risk factors. Additionally, a 2017 meta-analysis of randomized trials found that weight loss resulting from a variety of strategies (often dietary changes and exercise) among people with obesity led to an 18% reduction in all-cause mortality risk. Still, this data is limited by the fact that some of the interventions may have had benefits independent of their effect on weight loss.

The beneficial effect of reducing excess body fat is also seen with bariatric surgery. AD dismisses the seemingly protective effect of weight loss achieved via these surgeries, claiming these findings are the result of study limitations: “In other words, the control group may have had a higher risk of death than the bariatric surgery group even before surgery, rendering the comparisons essentially meaningless” (Page 157). To support this, AD cites an observational study on older men which failed to find a reduction in mortality following bariatric surgery when the results were adjusted for a number of differences between the surgery group and a non-surgery group.

However, it doesn’t appear that all non-randomized studies can be explained by inadequate adjustment. The SOS study, for example, found that bariatric surgery led to significant weight loss over a follow-up of up to 20 years. After adjustment for a large number of baseline characteristics, those receiving bariatric surgery were at a lower risk of total mortality, cardiovascular mortality, and type 2 diabetes. Additionally, there are randomized controlled trials of bariatric surgery that are not susceptible to the critique raised by AD. A 2019 meta-analysis of randomized controlled trials found that the likelihood of study-defined diabetes remission was 10 times higher in those receiving bariatric surgery versus general medical management after 2 years of follow-up, strongly suggesting that excess body fat is an important cause of type 2 diabetes. Randomized controlled trials also show that bariatric surgery is effective for long-term weight loss.

In addition, AD tends to focus on less effective forms of bariatric surgery, like gastric banding. Yet studies that mostly represent best-practice surgeries like Roux-en-Y and gastric sleeve surgeries suggest that they cause large long-term weight loss, and very large long-term reductions in the risk of heart attacks, stroke, type 2 diabetes, some cancers, and the overall risk of dying. This suggests that there is probably a strong long-term relationship between weight and health.

A significant contribution of body fat to type 2 diabetes risk is also seen in clinical trials involving dietary interventions. A notable example of this comes from the DiRECT study. In this randomized controlled trial, people with type 2 diabetes were assigned to adopt a very low calorie diet for 3–5 months, followed by a program to help maintain weight loss. After 1 year, participants in the intervention group had achieved an average weight loss of 10 kg (22 lbs), compared with 1 kg (2.2 lbs) of weight loss in a control group. Diabetes remission (based on a commonly proposed definition) was achieved by 46% of participants in the intervention group, compared with only 4% of participants in the control group. Furthermore, diabetes remission was directly related to weight loss, illustrated by the fact that no diabetes remission occurred in participants who didn’t lose weight and 86% of participants achieved remission who lost at least 15 kg/33 lbs). Evidence from the DiRECT study and several other clinical trials indicate that weight loss achieved via caloric restriction (and in some cases along with physical activity) can result in remission of type 2 diabetes and, furthermore, this remission is associated with the amount of weight lost.

Some clinical trials have also found that diet and exercise interventions resulting in weight loss reduce the risk of developing type 2 diabetes. The Diabetes Prevention Trial examined the effect of an intervention designed to produce weight loss using diet (via restriction of calories and fat) and exercise (with the goal of 150 minutes weekly). After 10 years of follow-up, the intervention reduced the risk of developing type 2 diabetes by 34% compared to a control group. Interestingly, this benefit was observed even though much of the initial weight loss (around 7 kgs/15 lbs by the end of the first year) was gradually regained. Two other clinical trials, one conducted in Japan and one conducted in Finland, also found that weight-loss inducing diet and exercise interventions reduced the risk of type 2 diabetes by 67% and 53%, respectively. However, these studies involved complex dietary changes, sometimes with increasing physical activity, making it hard to isolate the effect of weight loss per se.

Ultimately, a constellation of observational and interventional studies support a causal link between elevated body fat and a higher risk of various negative health outcomes.

Criterion 1.2. Are the references cited in the book to support the claim convincing?

1 out of 4

The references received a score of 1, indicating that they do not provide convincing support for the claim. AD references a number of studies supporting the harms of factors potentially related to losing weight. This includes complications arising from bariatric surgery, an increased risk of alcohol-use disorder following gastric bypass surgery, and a higher risk of dying in people with a history of weight cycling. However, none of these directly support the claim made in AD that elevated levels of body weight and body fat do not directly increase the risk of negative health outcomes.

Criterion 1.3. How well does the strength of the claim line up with the strength of the evidence?

0 out of 4

The strength of the claim received a score of 0, meaning it is greatly overstated. The idea that there is no direct relationship between body fat and health is refuted by a large body of evidence, as previously covered.

Overall (average) score for claim 2

0.3 out of 4

Claim 3

Weight stigma and weight cycling explain why obesity is linked to health problems

Supporting quote(s) and page number(s)

Page 52: “Instead, we shouldn’t be framing ‘obesity’ as a cause of poor health outcomes at all—both because the idea of an ‘obesity epidemic’ was largely fabricated by people with a vested interest in the weight loss industry and because weight stigma and the pressure to diet may explain any correlations we see between larger body size and poor health outcomes”.

Page 144: “…if all studies were to control for weight cycling, any excess risk from so-called ‘overweight’ or ‘obesity” might disappear”.

Page 158: “By now I hope you understand that most chronic diseases blamed on weight can most likely be explained by other phenomena, such as weight stigma and weight cycling”.

Criterion 1.1. How well is the claim supported by current evidence?

2 out of 4

This claim received a score of 2, indicating that it is weakly supported by current evidence.

AD claims that weight cycling, in which there are large fluctuations in body weight over time (as opposed to remaining at a relatively consistent body weight), is harmful for health. AD claims large amounts of weight cycling, attributable to frequent dieting, therefore may explain why people with obesity are at a higher risk of negative health outcomes. AD writes that unsuccessful dieting attempts “often lead to weight cycling—the yo-yo of weight loss and regain that’s all too familiar to most dieters—which, it turns out, is also bad for your health” (page 144).

There is indeed a large body of evidence suggesting weight cycling is associated with negative outcomes. A 2019 meta-analysis of 23 cohort studies, for example, found that individuals with the largest amount of weight fluctuation were at a higher risk of cardiovascular disease and all-cause mortality. However, most studies in the meta-analysis did not determine if the fluctuations in body weight were the result of intentional actions (e.g. from frequent dieting) or were unintentional (e.g., due to bouts of illness). In fact, when looking at the few studies that actually reported on weight cycling due to intentional weight loss there was no increase in risk. Thus, the implication in AD that weight cycling as a result of dieting leads to disease is not supported by this evidence.

AD suggests that weight stigma contributes to (and may, along with weight cycling, entirely explain) the negative health outcomes in people of higher body weights. Broadly speaking, weight stigma has been defined as “discriminatory acts and ideologies targeted towards individuals because of their weight and size”, with this weight stigma potentially coming from various sources (friends, colleagues, medical professionals).

In theory, weight stigma could increase the risk of certain negative health outcomes known to be more likely among people with obesity. For example, weight stigma seems to increase stress and depression. This could increase the risk of cardiovascular disease, since both psychological stress and depression have been linked to a higher risk of some cardiovascular disease outcomes.

However, in some studies, obesity and poor psychological health seem to have additive, rather than overlapping, effects on health outcomes. In one cross-sectional study on men, having depression was associated with more than double the odds of having diabetes, having obesity was associated with more than triple the odds of having diabetes, and having a combination of depression and obesity was associated with more than seven times the odds of having diabetes. Similar additive associations were also seen with high blood pressure and high cholesterol. Another cross-sectional study found that having depression and obesity were additive in their association with risk of ischaemic heart disease. Of course, there are other potentially harmful impacts of weight stigma, so these studies may not perfectly reflect the full impact of weight stigma.

There is limited evidence looking at whether weight stigma itself is related to negative health outcomes like diabetes and cardiovascular disease. In the NESARC study, perceived weight stigma was associated with a higher risk of having a heart attack, even after adjusting for BMI. This tentatively suggests weight stigma does indeed contribute to negative health outcomes, but the evidence is not strong and, more importantly, it by no means establishes that weight stigma fully explains the relationship between obesity and negative health outcomes.

Based on the available evidence, it seems reasonable to suspect weight stigma can contribute to the health risks associated with obesity, but it is not established that weight stigma is the main driver of such risks. In addition, the evidence does not appear to support the idea that weight cycling due to dieting causes or contributes to the health risks linked to obesity.

Criterion 1.2. Are the references cited in the book to support the claim convincing?

1 out of 4

The references received a score of 1, meaning they do not provide convincing evidence for the claim. At several points AD cites research accurately, but then arrives at conclusions about this research in ways that are fairly speculative. For example, the book cites studies that report an association between weight cycling and health problems, but then a logical leap is made that this shows losing weight on purpose via dieting causes these health problems. This doesn’t necessarily follow because the weight cycling measured in these studies isn’t necessarily due to dieting; some of it could be unintentional weight changes due to health conditions or cigarette smoking. In support of this, when these studies only consider weight cycling due to intentional weight loss, this association disappears (although to be fair, the evidence on that is limited to three studies). Also relevant, although not quite as directly, a meta-analysis of diet trials reports that intentional weight loss via dieting lowers all-cause mortality.

Not much evidence is provided to support the claims that weight stigma explains any association between body weight and the negative health outcomes, with AD even indicating that research on the topic is lacking because conducting such a study would be “a tall order in our society” (page 158).

Criterion 1.3. How well does the strength of the claim line up with the strength of the evidence?

1 out of 4

The strength of the claim received a score of 1, indicating the claim is substantially overstated. While weight stigma and weight cycling are associated with negative health outcomes, there is little evidence available to indicate these fully explain why there is a link between elevated levels of body fat and poor health. And for weight cycling specifically, other factors (like illness and lifestyle) seem likely to explain its association with worse health, as opposed to a directly harmful effect of weight cycling itself.

Overall (average) score for claim 3

1.3 out of 4

Overall (average) score for scientific accuracy

1.3 out of 4

Reference 1

Reference

Chapter 8, reference 8.  Polivy and Herman. “Restrained eating and food cues: Recent Findings and Conclusions.” Curr Obes Rep 6(1):79-85. 2017.

Associated quote(s) and page number(s)

Page 213-214: “A 2017 review of the scientific research on how chronic dieting affects food consumption found that not only do dieters eat more in response to food-related advertisements or large portion sizes, they also eat more of both high-fat foods and foods labeled as healthy or low-calorie, even if those labels are false. Dieters, in other words, are highly susceptible to both food-industry and diet-industry marketing”

Criterion 2.1. Does the reference support the claim?

3 out of 4

This reference received a score of 3, indicating that it offers moderate support for the claim. The reference is to a review article covering research comparing “restrained” and “unrestrained” eaters. The review notes that, in general, restrained eaters seem to be more responsive to various food-related cues. More specifically, restrained eaters often eat more of a food when it is advertised or when it is perceived as healthy and tend to eat less when given “diet cues” (e.g., exposure to healthy foods). Still, it’s difficult to infer whether (or to what extent) this phenomenon is the result of “chronic dieting” (as inferred in AD), since research has found that being a restrained eater is not an exact proxy for a history of dieting attempts (“chronic dieting”), even though there is a correlation between the two.

Reference 2

Reference

Chapter 1, reference 19.  Rees. “Industrialization and Urbanization in the United States, 1880–1929.” Oxford Research Encyclopedia of American History.

Associated quote(s) and page number(s)

Page 24: “From 1850 to 1900 the percentage of Americans living in cities more than doubled, and by 1920 more Americans would live in cities than in the countryside.”

Criterion 2.1. Does the reference support the claim?

4 out of 4

This reference received a score of 4, indicating that it offers strong support for the claim.

Reference 3

Reference

Chapter 7, reference 2.  Leslie Jamison. “I Used to Insist I Didn’t Get Angry. Not Anymore”, New York Times Magazine, January 17, 2018.

Associated quote(s) and page number(s)

Page 187–188: “It has been easier to shunt female sadness and female anger into the ‘watertight compartments’ of opposing archetypes, rather than acknowledging the way they run together in the cargo hold of every female psyche.”

Criterion 2.1. Does the reference support the claim?

4 out of 4

This reference received a score of 4. This claim here is not as straightforward to score, given that it is a poetic description of a perceived social phenomenon. Nevertheless, we felt the reference — an article in The New York Times Magazine about female anger — is sufficiently in line with the statement.

Reference 4

Reference

Chapter 4, reference 14.  Farrel et al. Accounting for the Cost of US Health Care: A New Look at Why Americans Spend More, December 2008,

Associated quote(s) and page number(s)

Page 132: “In reality, though, the costs attributed to “obesity” are grossly inflated: excess spending on health care in the United States has nothing to do with body size.”

Criterion 2.1. Does the reference support the claim?

2 out of 4

This reference received a score of 2, indicating that it offers weak support for the claim. This reference is a report from the McKinsey Global Institute examining the reasons for the much higher spending on health care in the United States compared to other countries. A variety of complex reasons are given to explain why the United States spends far more on healthcare than similar countries, including higher drug costs and greater spending on health administration. In line with the claim in AD, obesity does not seem to play an overt role in creating this health care spending disparity (as in, per capita spending in the US is similar to in many other countries). But it is still the case that a large portion of healthcare spending in the United States does seem to be related to obesity.

Reference 5

Reference

Chapter 3, reference 38.  Mari et al. “Adherence with a Low-FODMAP Diet in Irritable Bowel Syndrome,” Eur J Gastroenterol Hepatol 2019 Feb;31(2):178-182

Associated quote(s) and page number(s)

Page 110: “As a 2019 study put it, “Strict adherence [to] a low-FODMAP diet should raise the suspicion of a possible underlying eating disorder.”

Criterion 2.1. Does the reference support the claim?

3 out of 4

This reference received a score of 3, indicating that it offers moderate support for the claim. This study was conducted on 233 people with IBS, assigning them to follow a low-FODMAP diet for 6 weeks. Investigators determined how many participants had adhered to the diet and how many of those participants were at risk of Eating Disorder (ED) behavior. At the end of the study, it was found that participants who self-reported adhering to the diet were almost twice as likely to fit the criteria for ED behavior.

However, given that most people don’t have an eating disorder, this criteria is likely to be fairly unreliable. In the cited study, for example, most (about 67%) of the participants who adhered to the low-FODMAP diet did not actually have ED behavior. So it is perhaps debatable whether strictly adhering to a low-FODMAP diet is sufficient to raise suspicions of an eating disorder.

Reference 6

Reference

Chapter 9, reference 19.  Sabin et al. “Implicit and explicit anti-fat bias among a large sample of medical doctors by BMI, race/ethnicity and gender,” PLoS One 2012; 7(11) 

Associated quote(s) and page number(s)

Page 240: “Research has found that both implicit and explicit anti-fat stigma is just as prevalent among medical doctors as it is in the general public.”

Criterion 2.1. Does the reference support the claim?

4 out of 4

This reference received a score of 4, indicating it offers strong support of the claim. The study used an online test to examine the level of both implicit and explicit weight bias among 359,261 people from the general population as well among 2,284 medical doctors. The study found that implicit weight bias scores were similar between the general public and medical doctors and explicit weight bias scores tended to be higher among medical doctors compared to the general population.

Reference 7

Reference

Chapter 3, reference 30. Biesiekierski et al. “No effects of gluten in patients with self-reported non-celiac gluten sensitivity after dietary reduction of fermentable, poorly absorbed, short-chain carbohydrates,” Gastroenterology Aug; 145(2):320-8.e1-3 2013

Associated quote(s) and page number(s)

Page 107: “The best-designed of these studies, published in 2013, showed that when it comes to gluten, there’s a strong “nocebo effect” — a phenomenon where merely thinking that something (such as gluten) is making you sick causes actual symptoms…In the 2013 study, people weren’t told whether they were getting gluten or not. But many of them thought they were getting gluten and expected to feel bad when eating it, so they did—even those who had in fact been given gluten-free food.”

Criterion 2.1. Does the reference support the claim?

3 out of 4

This reference received a score of 3, indicating that it offers moderate support for the claim. This section discusses a study purportedly showing that people with non-celiac gluten sensitivity (NCGS) experience negative health effects in response to eating gluten because they expect to experience negative health effects (and not because eating gluten actually causes negative health effects).

In brief, the study included participants with NCGS, giving them capsules containing either gluten or a placebo (whey protein), with participants unaware of whether they would be ingesting gluten or not. In the days they were taking the assigned capsules, participants experienced worsening gastrointestinal symptoms, regardless of whether they were ingesting gluten or the placebo.

So while the study provides compelling evidence that eating gluten does not have negative effects (e.g., on gastrointestinal symptoms) among people with NGCS, it is not as clear whether the results can be attributed to a nocebo effect. To more sufficiently demonstrate a nocebo effect of gluten, a trial would also give participants an intervention in which they were not given gluten and knew they were not given gluten. This would help establish whether the worsening of symptoms was related to anticipating negative effects (a nocebo effect) or a worsening of symptoms that happen to coincide with the study itself (e.g., due to regression to the mean).

Reference 8

Reference

Chapter 3, reference 15.  Traci Mann et al. “Medicare’s search for effective obesity treatments: diets are not the answer.” Am Psychol. Apr;62(3):220-33. 2007

Associated quote(s) and page number(s)

Page 91: “Across all the studies, they found, one-third to two-thirds of dieters gained back significantly more weight than they’d lost.”

Criterion 2.1. Does the reference support the claim?

2 out of 4

This reference received a score of 2, indicating that it offers weak support for the claim. The review article cited does reference several observational studies in which people who dieted eventually ended up weighing more than they did before they first embarked on their diet. To this end the claim does appear well-supported.

But it can’t be ignored that this claim is made in the context of a passage claiming dieting causes weight gain. With this context in mind, then, the claim is weakly supported. Some studies are cited which found a compelling association between dieting and eventual weight gain, but it is still difficult to say if this is because dieting causes weight gain or whether it is due to some characteristic of people who tend to diet. This latter possibility seems more likely given that long-term controlled trials, including several referenced in the cited review, do not support the claim that dieting actually causes weight gain.

Reference 9

Reference

Chapter 6, reference 9.  Evars et al. “Feeling bad or feeling good, does emotion affect your consumption of food? A meta-analysis of the experimental evidence.” Neurosci Biobehav Rev Sep;92:195-208. 2018.

Associated quote(s) and page number(s)

Page 178–179: “With chronic dieters, though, something contrary happens: negative emotions provoke an increased appetite and desire to eat.”

Criterion 2.1. Does the reference support the claim?

3 out of 4

This reference received a score of 3, indicating it offers moderate support for the claim. The cited paper references various studies showing that negative emotions lead “restrained eaters” to increase their food intake, whereas many studies find no such increase among people overall. So while this appears in support of the claim, it is not necessarily the case that “restrained eaters” (the term used in the study) and “chronic dieters” (the term used in the text) are the same. While there seems to be a fair amount of overlap between the two, studies have found that many restrained eaters seek to prevent weight gain, rather than produce weight loss (as is the case for dieters).

Reference 10

Reference

Chapter 10, reference 2.  Tomiyama et al. “Misclassification of cardiometabolic health when using body mass index categories in NHANES 2005-2012” Int J Obes (Lon). May;40(5):883-6. 2016.

Associated quote(s) and page number(s)

Page 255: “Don’t use body mass index (BMI) as a measure of health; using this measure is likely to lead to misdiagnosis for both larger- and smaller-bodies people.”

Criterion 2.1. Does the reference support the claim?

2 out of 4

This reference received a score of 2, indicating that it offers weak support for the claim. This study found that BMI was a predictor of metabolic health, in the sense that the majority of people classified as having overweight (BMI of 25 to >30) or obesity (BMI ≥30) were considered metabolically unhealthy, whereas the majority of people classified as having “normal” weight (BMI 18.5 to <25) or underweight (BMI <18.5) were considered metabolically healthy. And while there was inconsistency (some people with overweight/obesity were considered metabolically healthy and some people with “normal” weight were considered metabolically unhealthy), it seems unreasonable to expect any singular measure of health to be perfect. Additionally, some research suggests people with obesity considered metabolically healthy are nonetheless at a higher risk of developing health issues later on than people of a “normal” BMI.

Overall (average) score for reference accuracy

3.0 out of 4

Summary of the health-related intervention promoted in the book

AD promotes various concepts with the goal of improving health, mostly in the realm of mental health. In addition to not restricting food intake for the purpose of losing weight, AD advocates for intuitive eating and body acceptance.

Intuitive eating is a strategy based around listening to your body rather than external pressures and ideas about food. This involves paying close attention to your physiological cues, like eating mostly when hungry and stopping when full and satiated. Intuitive eating further involves giving yourself unconditional permission to eat, meaning no food is forbidden, nor should eating any food be a source of guilt. With intuitive eating, food choices will be guided by cravings, preferences, and how different foods make you feel.

Body acceptance, meanwhile, refers to a psychological framework in which a person accepts their body as it is and without judgment. A multitude of psychological strategies are typically employed in the process of pursuing weight acceptance. AD doesn’t always use or fully endorse the term body acceptance, using other terms like body appreciation, body peace, and ultimately preferring the term body liberation. Nonetheless, the recommendations made in AD are largely in line with body acceptance as it is defined in the scientific literature.

Some specific advice given in AD, most of which are components of either intuitive eating or body acceptance, include not considering any food good or bad (page 241), having a pleasurable relationship with food (page 246), getting rid of your scale (page 250), not feeling bad about your body (page 250), and performing physical activity according to intrinsic motivation, like having fun, rather than according to extrinsic motivators, like trying to lose weight (page 252–253). AD also suggests finding weight-neutral (as in, without weight loss) solutions to concerns potentially related to weight, like building strength and stamina if you struggle with physical exertion (page 249).

Condition targeted by the book, if applicable

AD is primarily focused on negative mental health outcomes related to dieting and weight stigma, including disordered eating and poor self-esteem.

Apparent target audience of the book

The apparent target audience of AD is people who have struggled with dieting for weight loss, people who are unhappy with their body, and people who are unhappy with their relationship with food.

Criterion 3.1. Is the intervention likely to improve the target condition?

3 out of 4

The intervention received a score of 3, indicating it is likely to moderately improve the target conditions in the medium-to-long-term.

A number of studies have found that people who adhere more to the tenets of intuitive eating tend to have better self-esteem, self-compassion, and general well-being, are less likely to display emotional eating, and display less binge-purge behavior. Studies also found that greater body acceptance is associated with less depression and anxiety as well as greater self-esteem, self-compassion, and sexual satisfaction. These findings come primarily from cross-sectional studies, though, meaning it is hard to say for sure whether these variables are responsible for these outcomes or merely associated with them.

Thankfully, a number clinical trials have tested the effect of recommendations akin to what are found in AD, often utilizing the Health at Every Size (HAES) principles, with many reporting benefits to psychological health:

• An 18-month trial of women with binge eating disorder observed a reduction in binge-eating behavior with a non-diet intervention involving psychological support, behavior strategies, and elements of body acceptance. Interestingly, participants assigned to a similar intervention but with the addition of diet advice (e.g., limiting fat intake) also saw reductions in binge-eating behavior, suggesting the diet avoidance was not essential for benefit to occur.

• One randomized controlled trial of women with morbid obesity found that a 3-month non-diet intervention reduced stress and improved aspects of mental health (like emotional control and stability), with improvements sustained 9 months after the end of the intervention. These benefits were also seen in a control group given diet and exercise advice.

• A randomized controlled trial of women with obesity and metabolic syndrome examined the effect of a 3-month HAES-related non-diet intervention. Compared to a wait-list control group, the non-diet group experienced improvements in their general psychological well-being. Note that this study recommended participants obtain at least 4 hours of physical activity weekly, which is not a recommendation in AD.

• A randomized controlled trial (conducted by well-known HAES researcher Lindo Bacon) looked at the effect of a non-diet intervention (involving body acceptance and intuitive eating principles) compared to a diet and exercise intervention, with both interventions including a social support element via regular classes lasting 6 months. At the end of the interventions, both groups experienced similar benefits in some domains (like reduced depression and bulimia symptoms) and the non-diet group experienced improvements in body image avoidance and body dissatisfaction (the latter being not quite statistically significant compared to the diet group; p=0.078). Finally, more participants in the non-diet group reported the program made them feel better about themselves than did participants in the diet group.

• A randomized controlled trial of women with overweight or obesity compared a HAES-related intervention to either a control group given similar social support or a wait-list control group. Compared to the control groups, the HAES group experienced improvements in body esteem and reduced binge-eating behavior which were maintained one year later.

• One quasi-experimental (nonrandomized) trial looked at the effects of a 4-month HAES-based program in a group of women. Compared to a wait-list control group, the HAES group experienced improvements in self-esteem, body-esteem, and depression symptoms at the end of the program. These differences were no longer present one year later.

• In one randomized controlled trial, a short intervention designed to promote intuitive eating and body acceptance reduced body image dissatisfaction among female university students compared to a control group given informational brochures (related to dietary guidelines and body image).

• A 10-week trial examined the effect of an intervention, delivered via weekly group sessions, promoting body acceptance and compassion among women with overweight or obesity seeking a weight-loss treatment. Compared to a control group, participants in the intervention group experienced reductions in self-directed weight stigma, emotional eating, and self-hatred, as well as increases in quality of life, overall mental health, and physical activity.

Overall, these findings provide support for the idea that the interventions promoted in AD will improve psychological health, especially as it relates to body image. But there are limitations worth mentioning. For one thing, all clinical trials we identified included exclusively women, meaning the extent to which it applies to men is unknown. Also, most of these clinical trials utilized psychoeducation overseen by mental health professionals and included social support in the form of group sessions and classes. The results seen may therefore not be entirely applicable to someone pursuing the book’s advice on their own.

Finally, it’s interesting to note that some of the studies found that benefits were equally improved in a control group given diet and exercise advice, raising the question of whether or to what extent other elements (like social support and feelings derived from participation in a program) play a significant role in the mental health benefits.

Criterion 3.2. Is the intervention likely to improve general health in the target audience?

2 out of 4

The intervention received a score of 2, indicating it is likely to slightly improve the general health of the target audience.

There have been over a dozen randomized controlled trials examining the health effect of “non-diet” interventions, often utilizing strategies related to intuitive eating and/or weight acceptance. Of the trials assessing body weight, most observed no statistically significant change in weight or BMI compared to a control group not given an active intervention. And when examining participants’ body weight changes compared to the start of the study, results are decidedly mixed, with trials reporting weight loss, no change in weight, or weight gain. On the whole, these studies suggest the recommendations found in AD will have a neutral effect on body weight, on average.

A few “non-diet” trials have also looked at changes in cardiometabolic risk markers, including blood lipids, blood pressure, and glucose levels. When excluding trials advocating for exercise (which is not a specific recommendation in AD), most “non-diet” studies observe no apparent effect on these metrics compared to a control group. In one study, a non-diet intervention (involving intuitive eating and body acceptance) resulted in improvements in systolic blood pressure and LDL compared to baseline, despite body weight being non-significantly increased (by 0.3 kg/0.7 lbs). However, given the evidence overall, it seems premature to say whether the recommendations in AD will improve these aspects of cardiometabolic health.

Finally, a number of “non-diet” trials have assessed changes in psychological health measures, with benefits seen to depression, anxiety, and stress, although evidence is limited.

Criterion 3.3. Does the diet portion of the intervention promote an adequate nutrient intake for general health in the target audience?

1 out of 4

The book received a score of 1, indicating it is likely somewhat nutritionally inadequate.

AD does not recommend any specific foods or dietary pattern, making it difficult to make claims about how the intervention will impact adequate nutrient intake. As a substitute, we decided to look at the typical dietary intake of nutrients in the general population, since this could reflect how someone will eat who adopts intuitive eating (based on current research it does not appear intuitive eating alters dietary intake much). According to data from the National Health and Nutrition Examination Survey (NHANES), high numbers of Americans do not meet the the Estimated Average Requirement (EAR) or Adequate Intake (AI) for a number of nutrients, including vitamin E, potassium, calcium, vitamin K, and magnesium. Based on this, AD may promote an inadequate intake of some nutrients, since this is true of the average diet.

Overall (average) score for healthfulness

2.0 out of 4