Intuitive Eating: A Revolutionary Anti-Diet Approach

Overall score

Scientific accuracy

Reference accuracy

Healthfulness

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

Book published in 2020

Published by St. Martin's Essentials

Fourth Edition, Paperback

Review posted October 27, 2024

Primary reviewer: Stephan Guyenet

Peer reviewer: Brooke Nissim-Sabat

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

Weight loss diets cause psychological harm

Supporting quote(s) and page number(s)

 

On pages 19-21, the book lists “symptoms of diet backlash” that it says are “caused by the cumulative effects of dieting”. These include a number of distressing psychological effects and behaviors, like binge eating, “feeling like you don’t deserve to eat”, “social withdrawal”, and developing eating disorders.

On page 73-4, the book has a bulleted list titled “Damage from Dieting: Psychological and Emotional”. This list includes “dieting is linked to eating disorders”, “dieting may cause stress or make the dieter more vulnerable to its effects”, “dieting is correlated with feelings of failure, lowered self-esteem, and social anxiety”, and “dieting gradually erodes confidence and self-trust”.

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

2 out of 4

This claim received a score of 2 out of 4, meaning it’s weakly supported by current evidence. The idea makes sense, and observational studies tend to support it, but interventional studies tend not to (in fact, they often report that dieting has psychological benefits). While we would normally favor the results of intervention studies, in this case we think they might miss long-term effects of dieting that happen over many years. That said, the evidence is conflicting and we aren’t sure if this is the right way to interpret it.

“Psychological harm” is a broad term that could encompass many outcomes, so we focus on evidence linking weight loss dieting with the risk of disordered eating and overall quality of life or well-being.

First, we’d like to acknowledge that it makes sense that dieting could cause psychological harm in some people. As the book argues, dieting encourages people to be vigilant about what they eat, and most people eventually regain most of the weight they lose on any particular diet. It’s not hard to imagine how, in some people, this might result in disordered eating and feelings of frustration and failure.

What does the evidence say? Broadly, there are two types of evidence: observational and interventional. Observational studies examine people in the context of their normal lives and see if there are links between dieting and eating disorders. Intervention studies assign people to diets and see if they cause eating disorders.

Each is informative in its own way. Observational studies can be more relevant to peoples’ typical lives and can cover multiple years to look for relationships that take a long time to emerge. However, they sometimes have a hard time telling if something is really a cause-and-effect relationship (“correlation doesn’t necessarily imply causation”). For example, maybe people with certain personality traits tend to diet and to develop eating disorders, but one doesn’t really cause the other. Intervention studies are better at detecting cause-and-effect relationships but they may miss longer-term effects.

There are a large number of observational studies that have asked this question. To make our investigation more manageable, we focus on a 2016 review paper by a respected researcher in the field, Eric Stice. The paper focuses on studies that follow people over time (prospective studies), which can tell us whether dieting or eating disorders come first, in turn providing a clue about which may have caused the other. The paper reports that dieting is a risk factor for the development of eating disorders overall, and specifically for bulimia nervosa, binge eating disorder, and purging disorder. However, people who diet are less likely to develop anorexia nervosa. These studies tend to report that dieting behavior precedes eating disorders more often than would be expected by chance, consistent with the possibility that the former causes the latter.

In contrast, interventional studies are generally not supportive of the idea that dieting is psychologically harmful. As discussed in our review of Everything Fat Loss, systematic reviews (studies that review findings from many individual studies) of interventional studies generally report that dieting doesn’t increase eating disorder symptoms or reduce quality of life, and in some cases it even improves them. In adolescents, professionally administered weight loss interventions typically do not increase eating disorder risk, and the same is true in studies of adults. That said, a 2023 review paper does remark that while there was no increase in disordered eating in dieters on average, “a small subset of participants may experience disordered eating”. When it comes to quality of life and well-being, studies tend to suggest that professionally administered diet interventions improve it.

Overall, we think this conflicting research is pretty hard to interpret. In most contexts, we’d say interventional research trumps observational research, but here we aren’t sure. Although interventional studies don’t tend to find that professionally administered diets cause disordered eating or reduce quality of life, most people with obesity/overweight who enroll in such studies may already have dieted several times (probably less so for studies of adolescents). That means that if dieting causes eating disorders or reduces quality of life, those people may already have been harmed, and one additional diet may not make much difference.

That said, we think the psychological benefits associated with dieting in intervention studies significantly weaken the claim. One possibility is that dieting doesn’t cause eating disorders on average, but it does cause them in a small subset of people. These may be the people who come to the authors of Intuitive Eating for help. Given the conflicting evidence, we think a score of 2 out of 4 (weakly supported) makes the most sense, but we’re uncertain about it.

Overall (average) score for claim 1

 

2 out of 4

Claim 2

Intuitive Eating has psychological benefits

Supporting quote(s) and page number(s)

Front cover: “Make peace with food. Free yourself from chronic dieting forever. Rediscover the pleasures of eating.”

 

Back cover: “With this updated edition of the classic bestseller, Evelyn Tribole and Elyse Resch teach readers how to:

…

• Prevent or heal the wounds of an eating disorder

• Respect their bodies and make peace with food– at any age, weight, or stage of development”

Page 1: “Today, the research on Intuitive Eating itself is robust, with more than 125 published studies showing benefits– with the growing recognition that IE is an adaptive eating style, which influences positive psychological and physical well-being.”

Page 5: “A recent meta-analysis review of twenty-four studies published between 2005 and 2015 found that Intuitive Eating was associated with the following benefits (Ricciardelli 2016):

• Greater body appreciation and satisfaction

• Positive emotional functioning

• Greater life satisfaction

• Unconditional self-regard and optimism

• Psychological hardiness”

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

2 out of 4

This claim received a score of 2 out of 4, meaning that it’s weakly supported by current evidence. We found ten relevant randomized controlled trials. The highest-quality trial we found supports the book’s claim, but some other trials are unsupportive.

We begin by reviewing the reference cited in the quote above (referred to as Ricciardelli 2016), which is one of the studies that’s most prominently featured in chapter 1. The book says the study reports that Intuitive Eating is associated with many psychological benefits: greater body appreciation and satisfaction, positive emotional functioning, greater life satisfaction, unconditional self-regard and optimism, and psychological hardiness. The paper reviews 24 cross-sectional studies that measured the correlation between Intuitive Eating and various measures of psychological health. Although the study reports what the book says it does, this study doesn’t provide much support for the claim that Intuitive Eating has psychological benefits because:

• The people in these studies were mostly college students who probably weren’t practicing Intuitive Eating on purpose. Most of them probably just happened to be eating in a way that aligns with the book’s principles. Based on this study, it’s not clear that someone who currently doesn’t do Intuitive Eating and decides to apply the principles in the book will reap these benefits.

• Cross-sectional studies look for associations between variables of interest at a specific point in time, and as a result they aren’t very good at telling what caused what. Based on this study, we don’t know whether Intuitive Eating caused psychological benefits, psychological health led to Intuitive Eating, or neither caused the other. The paper emphasizes this: “However, given that all studies used cross-sectional designs, no conclusions regarding the direction of the relationship between intuitive eating and psychosocial correlates can be drawn.”

Intuitive Eating relies heavily on cross-sectional studies and related types of studies that aren’t very good at telling whether Intuitive Eating actually caused the benefits that are observed, although it does cite other types of studies as well. The book doesn’t explain that these results and other similar findings don’t offer much support for its claims.

That said, let’s review studies that are better designed for evaluating the book’s claims. What we’re looking for are randomized controlled trials (RCTs)– studies that randomly assigned people to Intuitive Eating vs. something else, and measured what happened to psychological health and well-being. When well-conducted, these types of studies are the best tool we have for evaluating the effectiveness of an intervention like this.

[Technical details– feel free to skip this paragraph] We performed a scientific literature search for RCTs of Intuitive Eating using the biomedical research database PubMed. We searched on June 19, 2024 with the search term “intuitive eating” and the RCT filter selected. This returned 28 results, all of which we reviewed for relevance. We also considered RCTs identified in the intuitive eating review paper Dyke 2014. Studies had to be RCTs, have Intuitive Eating or something similar as the main intervention, have a comparator group that didn’t receive an Intuitive Eating intervention, and report at least one measure of psychological health or well-being.

We found ten RCTs that meet these criteria:

• Manana et al. 2023 educated 195 Lebanese adults about Intuitive Eating principles using a series of text messages over 5 weeks, and measured several outcomes, including intuitive eating beliefs/practices (IES-2 scale), self-compassion, and perceived stress. Volunteers were divided into three groups: (1) an active IE group that received IE text messages and practice exercises, (2) a passive IE group that only received text messages, and (3) a control group that received “general health-related tips”. The intervention had little effect on intuitive eating beliefs/practices, so it doesn’t appear to have been very effective. There were improvements in self-compassion and perceived stress over time in the active IE group but not the passive IE group. However, intervention groups were not directly compared against the control group in statistical tests, which means these findings are not reliable. Due to the low-intensity intervention and inappropriate data analysis, this study doesn’t tell us much about the effectiveness of Intuitive Eating.

• Berman et al. 2022 randomized 19 women with obesity to one of two interventions: (1) A weight loss program (Weight Watchers), or (2) a self-acceptance, non-weight-loss intervention called Accept Yourself! (AY). This was a small pilot study. The Weight Watchers arm was a 12-week subscription and participants were told to attend weekly meetings. The AY program included 11 weekly meetings plus 9 one-hour “physical movement sessions” (videos with larger-bodied models or a Health At Every Size trainer). Participants were followed up after the intervention and every three months for a year. Self-reported depression (PHQ-9) improved significantly in the AY group relative to the Weight Watchers group immediately after the intervention, but the difference faded quickly and wasn’t consistently there from 3 to 12 months. There was no significant difference between groups in clinician-rated depression (HAM-D), although there was also a trend toward improvement immediately after the intervention. There was no significant difference in weight-related quality of life, and some significant improvement in body image acceptance. This study provides some evidence that AY has psychological benefits, but most of the effects seem to fade quickly after the active intervention is over, and the small size of the study limits how confident we can be about the findings.

• Katcher et al. 2022 randomized 14 “female-identifying undergraduates” in the US to two groups: (1) a 5-week Intuitive Eating intervention administered by Registered Dietitians, followed by a 5-week follow-up period with no further intervention; (2) no intervention for the first 5 weeks and then the same Intuitive Eating intervention for the second 5 weeks. The only measurement in the paper that’s relevant to this section is changes in body appreciation, and the intervention had no effect on it. This was a very small study so it’s not clear that it would have been able to detect an effect if one occurred.

• Resende et al. 2021 randomized 74 Brazilian women with body dissatisfaction to one of two groups: (1) a four-session intuitive eating intervention, or (2) a control group with no intervention. Participants were followed up at 4 and 24 weeks, and dropout rates were acceptable (<15% in both groups at 24 weeks). The trial reports measures of body dissatisfaction (BSQ-8), negative affect (PANAS), self-esteem (RSES), and body appreciation (BAS-2). All of these measures were improved by the intuitive eating intervention relative to the control group, and the size of the improvements was substantial. This trial has higher-quality methods than most of those we have reviewed on Intuitive Eating, and so we find it more convincing.

• Wilson et al. 2020 randomized 94 women college students in the US to two groups: (1) an intervention group that received “a brief dissonance-based non-dieting intervention designed to help college women reject unhealthy dieting behaviors, accept their bodies, and increase healthy eating”; (2) a control group that received “two educational brochures on body image and the USDA MyPlate guidelines”. Group 1 received two 90-120-minute sessions designed to foster body acceptance, encourage healthy eating patterns, and reject “unhealthy dieting and body ideals that are unattainable”, including by encouraging them to “critique unhealthy dieting and unreasonable body ideals”. Immediately after the intervention and one month later, researchers measured a number of outcomes related to Intuitive Eating and mental health. Relative to the control group, the intervention improved some measures related to Intuitive Eating and reduced body dissatisfaction, but had no impact on overall mental health-related quality of life.

• Webber et al. 2018 randomized 33 women and men with obesity to one of two 14-week programs: (1) a stress reduction program, or (2) an Intuitive Eating program based directly on the book Intuitive Eating. Outcomes were measured at 7 and 14 weeks. The study reports no statistically significant differences between groups at 7 or 14 weeks. The Intuitive Eating group didn’t show changes in measures of stress, depression, or food addiction over time, although it’s worth noting that within-group changes over time are less informative than between-group comparisons in trials like this. This trial would be more informative for our purposes if it had a control group with an intervention that wasn’t expected to have some of the same effects as the Intuitive Eating program.

• Mensinger et al. 2016 randomized 80 women with obesity to 6 months of weekly meetings teaching one of two “healthy living programs”: (1) a weight loss program, or (2) a weight-neutral program. The weight-neutral program was the HUGS Program, which “emphasized the principles of eating for well-being and pleasure, size acceptance, and the importance of engaging in physical activity for personal enjoyment and fulfillment”. The women were followed up at 6 months and two years. The HUGS Program increased a measure of intuitive eating relative to the weight loss group, but there were no differences between groups in quality of life, self-esteem, or distress. The HUGS Program group did see improvement over time in quality of life and self-esteem, but within-group changes like this aren’t as informative as between-group differences. Additionally, the weight loss diet group saw a similar improvement in self-esteem, which undermines the argument that Intuitive Eating is an antidote to low self-esteem caused by weight loss dieting.

• Cole and Horacek 2010 randomized 61 military spouses to an intuitive eating intervention vs. a control group. The military spouses were presumably mostly or all women, though this isn’t stated. The paper doesn’t describe the intervention (or lack thereof) given to the control group. Many outcomes are reported, including some related to psychological health, but the paper doesn’t report statistical tests for the most relevant outcomes so we don’t know whether the intervention had a meaningful effect on them.

• Crerand et al. 2007 randomized 123 women with obesity to one of three 40-week interventions: (1) a (initially) very-low calorie meal replacement diet, (2) a more standard weight loss diet, or (3) a “non-dieting program” with similar principles to Intuitive Eating. It was an intensive program with weekly 90-minute meetings for the first 20 weeks and biweekly meetings from week 22-40. The researchers ended up lumping the two dieting groups together for data analysis purposes. The results show that the diet groups experienced a large and significant reduction in a measure of depression relative to the non-dieting group at 20 and 40 weeks. There were no differences between groups in measures of self-esteem or body image at 20 or 40 weeks.

• Bacon et al. 2005 randomized 78 chronically dieting women to one of two 6-month interventions: (1) a typical diet and exercise weight loss diet, (2) a “health at every size” program with principles similar to Intuitive Eating. It was an intensive program with 24 weekly meetings lasting 90 minutes each, followed by another 6 months of optional monthly meetings. Dropout at 6 months was much higher in the diet group (41%) than the health at every size group (8%). 50% of both groups dropped out at 2 years, and the researchers only report data from the people who didn’t drop out for the entire study, even at the earlier time points when there were more participants. This is an unusual choice that seriously undermines the study’s reported findings. The results suggest that body dissatisfaction and a measure of depression significantly declined, and self-esteem significantly increased in the health at every size group relative to the diet group. However, the very high dropout rate could have biased results and this makes the findings hard to interpret.

Despite the fact that at least ten RCTs have been published on Intuitive Eating or related interventions, the evidence remains weak. Most of these trials have serious limitations that make their findings hard to interpret. Of the trials that had better methods, the results are mixed. However, we do think Resende et al. 2021 stands out for its higher-quality methods, and it does support the claim in Intuitive Eating.

There is some evidence supporting the claim in Intuitive Eating, but it’s very inconsistent so we score this claim a 2 out of 4.

Overall (average) score for claim 2

2 out of 4

Claim 3

Intuitive Eating helps people recover from eating disorders

Supporting quote(s) and page number(s)

Back cover: “With this updated edition of the classic bestseller, Evelyn Tribole and Elyse Resch teach readers how to:

…

-Prevent or heal the wounds of an eating disorder”

Chapter 17 is titled “The Ultimate Path Toward Healing from Eating Disorders” and describes using Intuitive Eating as part of a recovery program for eating disorders.

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

2 out of 4

This claim received a score of 2 out of 4, meaning it’s weakly supported by current evidence. We found four relevant randomized controlled trials. Most of them weren’t very informative, and their findings were inconsistent. However, the strongest study we found does support the book’s claim.

Chapter 1, “The Science Behind Intuitive Eating”, cites one trial that tested the book’s advice for treating binge eating disorder (BED). The trial is referred to as “Smitham 2008” in the text, but there is no corresponding citation for it in the references section. We were able to find a dissertation on Intuitive Eating written by a person named Delores Smitham, describing what we think is the trial the book refers to, but we weren’t able to find a version of it that has been published in a peer-reviewed journal.

The electronic version of the dissertation only includes the first few pages, and the abstract provides little detail. It says it was an eight-week treatment program and implies that it was based at least in part on Intuitive Eating. The study doesn’t appear to have had a control group, which is mentioned as a limitation in Intuitive Eating. The abstract provides little information on the results, but implies that effectiveness was similar to other BED treatments. This implies that the treatment was effective, but not better than other treatments. However, this claim is hard to evaluate since we don’t have any details about the trial. Overall, this citation is not very informative.

Chapter 17 of Intuitive Eating discusses the use of Intuitive Eating (and other strategies) for addressing eating disorders. Although the approach seems logical, supporting evidence is not provided and most of the chapter is patient anecdotes. Due to the limited evidence in the book, we did our own literature searches.

[Technical details– feel free to skip this paragraph] To find relevant studies, we did scientific literature searches in PubMed with the RCT filter selected on August 25, 2024. Search terms were ““eating disorder” AND (intuitive eating OR health at every size OR HUGS or non-dieting)”, “binge AND (intuitive eating OR health at every size OR HUGS OR non-dieting)”,  “bulimia AND (intuitive eating OR health at every size OR HUGS OR non-dieting)”, and “anorexia AND (intuitive eating OR health at every size OR HUGS or non-dieting)”. Studies had to be RCTs, have Intuitive Eating or something similar as the main intervention, have a comparator group that didn’t receive an Intuitive Eating intervention, and report at least one measure of disordered eating.

These searches returned four RCTs:

• Berman et al. 2022 randomized 19 women with obesity to one of two interventions: (1) A weight loss program (Weight Watchers), or (2) a self-acceptance, non-weight-loss intervention called Accept Yourself! (AY). This was a small pilot study. The Weight Watchers arm was a 12-week subscription and participants were told to attend weekly meetings. The AY program included 11 weekly meetings plus 9 one-hour “physical movement sessions” (videos with larger-bodied models or a Health At Every Size trainer). Participants were followed up after the intervention and every three months for a year. The results show that a measure of eating disorder severity (EDDS) improved significantly more in the AY group than the Weight Watchers group. However, the two groups were already fairly different at baseline, raising questions about whether the two groups were really comparable. This study provides weak evidence that AY lowers eating disorder risk.

• Resende et al. 2021 randomized 74 Brazilian women with body dissatisfaction to one of two groups: (1) a four-session intuitive eating intervention, or (2) a control group with no intervention. Participants were followed up at 4 and 24 weeks, and dropout rates were acceptable (<15% in both groups at 24 weeks). Disordered eating was measured by the Eating Attitudes Test-26 (EAT-26). The results show that at 4 and 24 weeks, disordered eating scores were meaningfully reduced in the intuitive eating group relative to the control group. This trial has higher-quality methods than most of those we have reviewed on Intuitive Eating, and so we find it more convincing.

• Bacon et al. 2005 randomized 78 “chronically dieting” women to one of two 6-month interventions: (1) a typical diet and exercise weight loss diet, (2) a “health at every size” program with principles similar to Intuitive Eating. It was an intensive program with 24 weekly meetings lasting 90 minutes each, followed by another 6 months of optional monthly meetings. Dropout at 6 months was much higher in the diet group (41%) than the health at every size group (8%). 50% of both groups dropped out at 2 years, and the researchers only report data from the people who didn’t drop out for the entire study, even at the earlier time points when there were more participants. This is an unusual choice that seriously undermines the study’s reported findings. The results suggest that there was no difference between groups in a measure of bulimia. However, the very high dropout rate could have biased results and this makes the findings hard to interpret.

• Campos et al. 2022 randomized 58 people with obesity (mostly women) to three groups: (1) an individualized eating plan, (2) an intuitive eating intervention, and (3) an intuitive eating and nutritional guidelines intervention. We haven’t been able to access the full text of this study so we don’t know important trial details, like how long or how intensive the intervention was, but the abstract says the study lasted six months. The abstract says there was no difference between groups in binge eating scores. However, we don’t place much weight on this finding because we haven’t seen the full text.

Our main conclusion from these studies is that the scientific evidence on the impact of Intuitive Eating on eating disorders is weak. There are few randomized controlled trials, most don’t seem very informative, and their findings are inconsistent. However, the strongest study we found does support the book’s claim, so we give it a 2 out of 4, meaning it’s weakly supported overall.

Overall (average) score for claim 3

 

2 out of 4

Overall (average) score for scientific accuracy

2 out of 4

Reference 1

Reference

Chapter 7, reference 7.  Ciampolini, Lovell-Smith, and Sifone Nutr Metab 7:4. 2010

Associated quote(s) and page number(s)

Page 100: “Two other studies using this initial hunger recognition technique demonstrated that people trained in this method had improved insulin sensitivity, compared to the nontrained controls.”

This quote is in the context of a box labeled “Studies show: eating in response to initial hunger improves health”. It argues that “restoring and validating hunger and training people to recognize initial hunger could help in the prevention and treatment of diabetes and associated disorders.”

Criterion 2.1. Does the reference support the claim?

2 out of 4

This reference received a score of 2 out of 4, meaning it weakly supports the associated claim in the book.

The reference is used to support a narrow claim and a broad claim. The narrow claim is that the cited study showed that eating in response to initial hunger (but only when blood glucose is below 85 mg/dL) improved insulin sensitivity. The broad claim is that this study supports the argument that “restoring and validating hunger and training people to recognize initial hunger could help in the prevention and treatment of diabetes and associated disorders.”

The study trained 181 lean and overweight people to recognize symptoms of initial hunger and eat in response to it, but only if blood glucose levels were less than 85 mg/dL (described in this previous paper).

The study doesn’t support the narrow claim that the intervention improved insulin sensitivity, because it doesn’t report effects on insulin sensitivity. However, another closely related study cited in the same sentence in Intuitive Eating does support this claim. The error here is that Intuitive Eating says both studies support the claim, when only one of them does. We don’t think this is an important error since it doesn’t meaningfully impact the book’s argument, so we don’t detract points for it.

That said, we don’t think this study offers much support to the broad claim either: that this study supports the argument that “restoring and validating hunger and training people to recognize initial hunger could help in the prevention and treatment of diabetes and associated disorders.”

The main reason is that the study trained people not to eat when their blood sugar wasn’t below 85 mg/dL, even when feeling hungry. In addition, people in the intervention group were encouraged to eat up to 1 kg of vegetables or fruit to “prevent distress from excessive hunger when [initial hunger] appeared half an hour or more before mealtimes.” It’s not hard to imagine that this became a form of calorie restriction, at least in some people.

The intervention caused weight loss, but it’s not clear that this was due to “restoring and validating hunger and training people to recognize initial hunger” rather than training people not to eat under some circumstances, or eating vegetables or fruit instead of more calorie-rich foods if they feel hungry. Given the well-established impact of weight loss on insulin sensitivity, it’s also not surprising that insulin sensitivity improved.

Reference 2

Reference

Chapter 13, reference 16.  Tomiyama et al. Int J Obes 40:883. 2016.

Associated quote(s) and page number(s)

Page 209: “A research team from UCLA found that 54 million Americans were labeled “obese or overweight” but were in fact healthy according to metabolic indicators.”

Criterion 2.1. Does the reference support the claim?

4 out of 4

This reference received a score of 4 out of 4, meaning it strongly supports the associated claim in the book.

The study is based on a nationally representative survey of weight and health in the US, the NHANES survey (2005-2012). It uses data on blood pressure, triglyceride, cholesterol, glucose, insulin resistance and C-reactive protein to determine the cardiovascular and metabolic health of people of different body mass index (BMI– an indirect measure of body fatness). People were considered “metabolically healthy” if 0 or 1 risk factor was outside the healthy range.

Table 1 reports that among people classified as having overweight or obesity by BMI, about 54 million are classified as “metabolically healthy”, while about 93 million are classified as “metabolically unhealthy”.

The point Intuitive Eating is making in that passage is that BMI often misclassifies heavier people as unhealthy when they actually aren’t. We think this study is consistent enough with that point to receive full marks. However, we do want to note two additional points:

• The definition of “metabolically healthy” varies and is debated, and this study allowed people with one unhealthy risk factor to be classified as such. We think the definition used in this study is fine, but the numbers would change if a different definition were used.

• The probability of being “metabolically unhealthy” did rise quite a bit with BMI. Moving from lean to overweight, to obesity type I, to obesity type II/III, the probability of being “metabolically unhealthy” increased from 31%, to 53%, to 71%, to 84%. So while BMI is certainly a crude tool for diagnosing metabolic health, it’s not totally uninformative.

Reference 3

Reference

Chapter 15, reference 24. Rozin. Social Research 66:9. 1999.

Associated quote(s) and page number(s)

Page 233: “Interestingly, research has shown that the medical practices of French versus American physicians reflect their culture. Consequently, American physicians prescribe more medication, while the French physicians are inclined to suggest rest, vacations, or a stay at a spa.”

Criterion 2.1. Does the reference support the claim?

2 out of 4

This reference received a score of 2 out of 4, meaning that it weakly supports the claim. While the reference is accurately cited, we don’t think the reference itself is very convincing.

The reference is a light-hearted discussion paper published in 1999 by Paul Rozin, titled “Food is Fundamental, Fun, Frightening, and Far-reaching”. The quote Intuitive Eating draws from is this: “French medicine conceives of disease as  some kind of internal imbalance, as a weakness in what is called the “terrain.” American medicine is more inclined to think of disease as a harmful external influence, such as germs or toxins. As a result of this distinction, American medicine prescribes more antibiotics, while the French are much more inclined  to suggest rest, vacations, or a stay at a spa.”

To support this statement, the paper cites the 1988 book Medicine and Culture, by Lynn Payer. We haven’t been able to access this book electronically.

We’re skeptical of this claim, for two reasons:

1. The original claim comes from a book published 36 years ago. Assuming it was true when it was written, it’s not clear that it remains true nearly four decades later.

2. We aren’t sure what this statement was originally based on, but it seems unlikely to us that French doctors regularly prescribe vacation and spa stays as a treatment for health conditions for the average person.

Reference 4

Reference

Chapter 14, reference 12.  Garber et al. Med Sci Sports Exerc 43:1334. 2011.

Associated quote(s) and page number(s)

Page 226: “The American College of Sports Medicine recommends that strength training and flexibility become an integral part of a fitness program for all healthy adults. Specifically, they recommend:

• Strength training at least twice a week.

• One set of eight to twelve repetitions of eight to ten exercises for conditioning each major muscle group.

• Stretch at least two or three days per week”

Criterion 2.1. Does the reference support the claim?

3 out of 4

This reference received a score of 3 out of 4, meaning that it moderately supports the claim.

The reference is an evidence review and position statement from the American College of Sports Medicine that recommends types and amounts of exercise for improving and maintaining fitness and health.

It recommends strength training 2-3 times per week, using 8-12 repetitions per exercise to “improve strength and power in most adults” (although it has other ranges for people who are older or untrained). It also recommends stretching 2-3 times per week. This is consistent with the statement in Intuitive Eating.

However, we weren’t able to find a place in the paper where it recommends “eight to ten exercises for conditioning each major muscle group”. The recommendations in table 1 and elsewhere for resistance training simply recommend training “each major muscle group” without mentioning a specific number of exercises. Eight to ten exercises seems unnecessary and potentially daunting for someone who is just looking for basic fitness.

Reference 5

Reference

Chapter 2, reference 3.  Field et al. Pediatrics 112:900. 2003.

Associated quote(s) and page number(s)

Page 29: “Here are some sobering studies indicating that dieting promotes weight gain…

• Research on nearly seventeen thousand kids ages nine to fourteen concluded that “in the long term, dieting to control weight is not only ineffective, it may actually promote weight gain.”

Criterion 2.1. Does the reference support the claim?

3 out of 4

This reference received a score of 3 out of 4, meaning that it moderately supports the claim in the book.

The study measured the eating and dieting behavior of boys and girls 9-14 years old and followed them for three years. Over that period, the study reports that boys and girls who dieted were more likely to gain weight. The study does come to the conclusion described in Intuitive Eating, although the quote they use in the book isn’t exactly as it appears in the paper (the difference doesn’t change its meaning): “in the long-term, data suggest that dieting to control weight is not only ineffective, it may actually promote weight gain”.

Although the study is described accurately in the book, it doesn’t offer much support to the claim the book is making, due to the study’s limitations. Because it’s an observational study, it can’t really tell us whether dieting caused weight gain, people more prone to weight gain are more likely to diet, or some other explanation caused the association.

Another minor point that doesn’t affect our score for this reference: the study included 14,972 children. We don’t think “nearly seventeen thousand kids” is an accurate way to describe that figure. Since this doesn’t really affect the point the book is making, we think it’s probably a typo.

How we view this reference overall depends on whether we focus on how accurately it describes the study (basically accurately) or how well the study itself supports the larger point the book is making (not well). We split the difference and give it a score of 3 out of 4, meaning the claim is moderately supported.

 

Reference 6

Reference

Chapter 13, reference 4. Daníelsdóttir, O’Brien, and Ciao. Obes Facts 3:47. 2010.

Associated quote(s) and page number(s)

Page 211: “The incidence of weight stigma has escalated 66 percent with the rise in public health campaigns to prevent ‘obesity’”

Criterion 2.1. Does the reference support the claim?

3 out of 4

This reference received a score of 3 out of 4, meaning that it moderately supports the claim in the book.

The reference is a narrative review paper about “anti-fat prejudice”. It contains the following quote, which is consistent with the book’s claim: “Weight-based discrimination, the behavioral manifestation of anti-fat prejudice, has increased by 66% over the past decade with prevalence rates now comparable to race-based prejudice.” We couldn’t find a place in the paper where it attributes this claimed increase to public health campaigns intended to prevent obesity.

To support that statement, the review paper cites two other studies, one of which is relevant to the 66% figure. That one is a survey study of perceived discrimination in Americans conducted in 1995–1996 and 2004–2006. It reports: “The prevalence of weight/height discrimination in US adults went up from 7.3% in 1995–1996 to 12.2% in 2004–2006, demonstrating a significant increase of 66% (P < 0.01).” While this study is cited accurately, the absolute increase in discrimination was small (4.9%), and we think it would have been appropriate to share that.

While the reference is cited accurately, there are two things that lead us to subtract a point:

1. There’s nothing in the reference supporting the connection with public health campaigns intended to prevent obesity.

2. The 66% increase makes the change sound impressive, but in absolute terms it only changed from 7.3% to 12.2%. We think the 66% figure is a little bit misleading without this additional context.

Reference 7

Reference

Chapter 3, reference 2. Beaver. “Bad diets kill more people around the world than smoking, study says”. Washington Post, April 5, 1999.

Associated quote(s) and page number(s)

Page 39-40: “Messages about eating healthfully are everywhere, from nonprofit health organizations to food companies touting the health benefits of their particular product. The inherent message? What you eat can improve your health. Conversely, take one wrong move (bite) and you’re one step closer to the grave. Is this an exaggeration? No! Here are some recent headlines found in a Google search:

• …

• ‘Bad Diets Kill More People Around the World Than Smoking, Study Says.’”

Criterion 2.1. Does the reference support the claim?

4 out of 4

This reference received a score of 4 out of 4, meaning that it strongly supports the claim.  It’s a news article reporting on research linking diet with the risk of dying prematurely. The article is cited accurately and it supports the book’s argument that the public often receives messages that diet is an important determinant of health, including the risk of dying prematurely.

Reference 8

Reference

Chapter 17, reference 2. Galimache et al. Am J Clin Nutr. 109:1402. 2019.

Associated quote(s) and page number(s)

Page 280: “A systematic review of 94 studies found that the prevalence of eating disorders more than doubled, from 3.5% for the 2000-2006 period to 7.8% for the 2013-2018 period. The bottom line, according to the researchers, is that eating disorders are highly prevalent worldwide.”

Criterion 2.1. Does the reference support the claim?

4 out of 4

This reference received a score of 4 out of 4, meaning that it strongly supports the claim. The paper contains the following quotes:

• “…the most recent studies confirm that EDs are highly prevalent worldwide, especially in women.”

• “…the weighted means of point ED prevalence increased over the study period from 3.5% for the 2000–2006 period to 7.8% for the 2013–2018 period.”

These quotes straightforwardly support the statements in the book, and the study type also seems adequate to support them.

Reference 9

Reference

Chapter 2, reference 10. Neumark-Sztainer et al. J Am Diet Assoc. 106:559. 2006.

Associated quote(s) and page number(s)

Page 29: “Teenage dieters had twice the risk of gaining more weight, compared to nondieting teens, according to a five-year study. Notably, at baseline, the dieters did not weigh more than their nondieting peers. This is an important detail, because if the dieters weighed more, it would be a confounding factor (which would implicate other factors, rather than dieting, such as genetics).”

This quote appears inside a box titled “Dieting increases your risk for weight cycling”

Criterion 2.1. Does the reference support the claim?

2 out of 4

This reference received a score of 2 out of 4, meaning that it weakly supports the claim. Intuitive Eating cites the study fairly accurately, but the study itself offers little support to the claim that dieting causes weight gain.

The study measured dieting behaviors and body weight in 2,516 US adolescents in 1999, then again five years later in 2004. It reports that girls and boys who reported dieting in 1999 had about 1.8 times higher odds of becoming overweight by 2004 than those who didn’t report dieting (table 3). This isn’t quite “twice the risk” reported in Intuitive Eating, but it’s fairly close. We would have preferred “nearly twice the risk”, or even better, the specific figure from the paper.

Intuitive Eating (as well as the study authors) takes this as evidence that dieting behaviors caused weight gain, when this really isn’t clear from the study. It’s possible that people diet because they know they’re more susceptible to weight gain (e.g., due to family history), and they tend to gain more weight for the same reason. The point is not to argue for this particular explanation, but to show that other explanations are possible and not ruled out by this type of study.

A more rigorous way to answer this question is to look for long-term follow-up studies of weight loss randomized controlled trials, where people are randomly assigned to a weight loss diet vs. none. This eliminates the confounding factors that can plague studies like the one cited in Intuitive Eating.

One good example of this is the ten-year follow-up study of the Diabetes Prevention Program trial, a large and rigorous randomized controlled trial that included an intensive weight loss program (maximum average weight loss was about 7 kg, or 15 lbs at one year). The intensive weight loss program lasted an average of 2.8 years, after which the initial study was complete and all participants were offered the weight loss program. This second phase seems to have been very low-intensity because participants in the control group didn’t lose weight. In any case, ten years from the beginning of the study, probably few people were still closely following the original program.

Looking at body weight changes over the ten-year period, what we see is that the weight loss group and the control group gradually converge, ending up at a similar weight. However, the weight loss group never became heavier than the control group. This suggests that weight loss dieting probably doesn’t cause people to gain more weight than they would have otherwise. One caveat is that this study was done in adults, and we can’t rule out the possibility that effects could be different in adolescents.

Reference 10

Reference

Chapter 3, reference 5. Birch et al. N Engl J Med. 324:232. 1991.

Associated quote(s) and page number(s)

Page 38: “A landmark study, led by Leann Birch, PhD, and published in the New England Journal of Medicine, confirmed that preschool-age children have an innate ability to regulate their eating according to what their bodies need for growth. This holds true even when, meal by meal, the little tykes’ eating appears to be a parent’s nightmare. Researchers found that at a given meal, calorie intake was highly variable, but it balanced out over time.”

Criterion 2.1. Does the reference support the claim?

4 out of 4

This reference received a score of 4 out of 4, meaning that it strongly supports the claim.

The reference is a study of 15 preschool children whose food intake at preschool and at home was measured for six days by weighed food records. As described in Intuitive Eating, the results show that calorie intake varied a lot at a single meal, but tended to balance out over the course of the day. In other words, if a child ate a small lunch, they would usually compensate by eating more later. This suggests that they were able to regulate their total calorie intake appropriately.

We do want to caution that just because children are able to regulate their calorie intake to some degree doesn’t mean they will necessarily regulate their calorie intake appropriately over the long run when provided with unhealthy and highly palatable foods, or make good choices about what to eat when offered such foods.

Overall (average) score for reference accuracy

3.1 out of 4

Summary of the health-related intervention promoted in the book

Intuitive Eating recommends that people stop dieting for weight loss, work on accepting their bodies how they are, satisfy hunger and cravings, and build a relationship with food that’s less restrictive and more enjoyable. It provides light advice and encouragement on eating healthy foods and exercising, but focuses on health benefits and feeling good rather than weight management.

Condition targeted by the book, if applicable

Intuitive Eating aims to help prevent and treat eating disorders like binge eating disorder and anorexia, but its main aim is to improve peoples’ general well-being by improving their relationship with food and themselves.

Apparent target audience of the book

The apparent target audience of the book is people who have a history of dieting (“chronic dieters”) and are dissatisfied with it.

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

2 out of 4

The book’s advice received a score of 2 out of 4, meaning that it probably modestly improves well-being and eating disorder severity in the medium to long term on average. It may be more effective when the intervention is more intensive, supported by a skilled practitioner, or in a subset of people who respond particularly well, but the results of studies are inconsistent enough that our best guess is that the average effect isn’t large.

In the scientific accuracy section, we reviewed the evidence that the book’s advice can improve measures of well-being and help people recover from eating disorders (claims 2 and 3; expand scoring section to see our evidence review and discussion). We think the evidence is weak overall, and somewhat contradictory, but the most convincing study we found does support the effectiveness of the book’s advice for well-being and eating disorders among women with body dissatisfaction.

Given how weak the evidence is currently, we think there’s a lot of room for further research to update our thinking on Intuitive Eating.

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

2 out of 4

The book’s advice received a score of 2 out of 4, meaning that it would probably slightly improve general health, mostly through its positive effects on mental health in people who are experiencing distress from a history of dieting. For a detailed discussion of this evidence, expand the scoring section of the scientific accuracy section and see claims 2 and 3.

In addition to its benefits for general well-being and treatment of eating disorders, the book also recommends healthy eating (chapter 15) and physical activity (chapter 14), though the recommendations aren’t very intensive. The impact of these recommendations on physical health may be undermined by the book’s advice to give yourself unconditional permission to eat any food at any time (regardless of its healthfulness), which may increase the intake of less healthy foods. Given how light the advice on exercise is, we doubt it would have much impact (it’s hard to get people to exercise consistently even with more intensive interventions).

A systematic review published in 2021 reviewed the randomized controlled trials that have tested the impact of Intuitive Eating or Mindful Eating on what people eat. Echoing what we found in our research review in the scientific accuracy section, the studies were generally of poor quality. The review concludes: “Results as a whole fail to consistently support the hypothesis that ME and IE interventions alter energy intake or diet quality.”

Given how weak the evidence is currently, we think there’s a lot of room for further research to update our thinking.

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’s advice received a score of 1 out of 4, meaning that the diet a person would eat by following its advice would probably be somewhat nutritionally inadequate. The reason is that typical diets in wealthy countries like those in North America and Europe are already somewhat nutritionally inadequate, and Intuitive Eating would probably have little impact on overall diet quality.

We reviewed the evidence on the impact of intuitive eating on diet quality in the previous section. A systematic review concludes: “Results as a whole fail to consistently support the hypothesis that ME and IE interventions alter energy intake or diet quality.”

Overall (average) score for healthfulness

1.7 out of 4