Does Dieting Lead to Disordered Eating? What the Research Shows

Quick answers
- What is the difference between dieting and a diagnosable eating disorder?
- Anorexia nervosa, bulimia nervosa and binge-eating disorder are diagnosable and uncommon: a U.S. national sample interviewed in 2012-2013 put lifetime prevalence at 0.80%, 0.28% and 0.85% respectively [2]. I use "disordered eating" here to mean chronic dieting, rigid food rules and preoccupation with weight or shape, a broader pattern that does not require a diagnosis. In longitudinal studies, early dieting predicted later dieting or extreme weight-control behavior [1], and weight and shape preoccupation predicted later dieting and unhealthy weight-control behavior in women [6].
- Can severe calorie restriction cause food preoccupation and binge eating in healthy people?
- Some of the most direct evidence is the Minnesota Starvation Experiment, in which 36 conscientious objectors were semi-starved and most lost more than a quarter of their body weight [3]. A review of the research on food restriction reports that starvation and self-imposed dieting appear to result in eating binges once food is available, and in preoccupation with food, dysphoria and distractibility [4]. That degree of restriction is far more severe than a typical diet, and the experiment had no separate comparison group, so it shows what can happen, not how often.
- Is orthorexia (an obsession with "clean" or "pure" eating) a real diagnosis?
- Not formally. It is not a recognized diagnosis, and a systematic review of 62 studies notes there is still no consensus on diagnostic criteria, and reported prevalence ranged from 2.6% to 87.7% depending on the questionnaire used and the group studied [9][10].
Short answer: some dieting-related patterns predict later extreme weight-control behavior and binge eating, but the research does not show that dieting by itself causes eating disorders, and researchers disagree about which part of dieting matters. A national longitudinal study following 14,322 young Americans from adolescence into adulthood found that early dieting, depression and body image distortion each predicted dieting or extreme weight-loss behavior years later. Early dieting and depression were also associated with binge eating in young adulthood. It was early body image distortion, not dieting, that was linked to an eating-disorder diagnosis. Dieting and eating pathology were more common in women than in men, and rose over time in both sexes [1]. That is an uncomfortable finding to lead with on a site built around root-cause nutrition work. I'm leading with it because it's what the data show, and because the honest follow-up, that researchers disagree about why, is further down.
Disordered eating and a diagnosable eating disorder are not the same thing
Anorexia nervosa, bulimia nervosa and binge-eating disorder are diagnosable and uncommon. The largest nationally representative study of DSM-5-defined eating disorders in U.S. adults as of its 2018 publication, drawn from 36,306 structured interviews conducted in 2012 and 2013, put lifetime prevalence at 0.80% for anorexia, 0.28% for bulimia and 0.85% for binge-eating disorder, with women carrying significantly higher odds than men for all three [2].
I use "disordered eating" in this article to mean the pattern that sits underneath that threshold: chronic dieting, rigid food rules, a mental preoccupation with weight and shape that doesn't let up. In a 15-year follow-up of 1,830 people in the Project EAT cohort, preoccupation with weight and shape was the strongest predictor, among the body-image measures studied, of dieting and unhealthy weight-control behavior in female participants [6]. That study tracked behaviors, not diagnoses, so it tells us who goes on to diet and restrict more, not who goes on to develop an eating disorder.
What severe restriction did to men who weren't known to have eating problems
Some of the most direct evidence that severe restriction can change a person's relationship with food comes from a study designed to understand starvation and refeeding, not dieting. Physiologist Ancel Keys and colleagues at the University of Minnesota recruited 36 conscientious objectors to study semi-starvation and the problem of refeeding civilians who had been starved during the war. Most participants lost more than a quarter of their body weight, and many developed anemia, fatigue, apathy, extreme weakness, irritability, neurological deficits and lower-extremity edema [3]. That is far more severe than a typical diet.
A peer-reviewed review of the research on food restriction, drawing on starvation studies and on self-imposed dieting, reports that both appear to result in eating binges once food is available, and in preoccupation with food and eating, increased emotional responsiveness and dysphoria, and distractibility [4]. The Minnesota experiment had no separate comparison group, so it can't say how many people would respond this way. What the research shows is that severe, prolonged restriction was followed by food preoccupation and binge eating once food returned, and the review's author cautions that the negative effects of restricting may outweigh its benefits.
Restriction can also leave a metabolic footprint. Researchers who followed 14 contestants from "The Biggest Loser" for six years found that resting metabolic rate was still about 500 kilocalories per day lower than their body composition and age predicted [5]. That is a small group with no control arm, and the authors describe the adaptation as proportional to the weight loss but incomplete. This is about metabolism, not eating behavior, so it does not show that dieting leads to disordered eating. I include it because it shows that extreme, competition-style weight loss can leave a lasting physiological mark, not that every diet will.
Is it dieting, or is it restraint? Researchers disagree
The strongest honest caveat to everything above is that the field has not settled which part of dieting is the problem. A 2021 review traces a long-running dispute between two frameworks. It was written by proponents of one of them, the Three-Factor Model of Dieting, so read it as one side's account. Herman and Polivy's restrained-eating model strongly influenced the cognitive-behavioral model of bulimia nervosa. The Three-Factor Model separates restraining eating to prevent over-consumption from dieting to lose weight, and the review's authors argue that three historical trends led restraint theory to unfairly impugn dieting for weight control [7]. That leaves open a possibility the review does not itself claim, which I'd treat as a hypothesis rather than a finding: that rigid rule-keeping (my own gloss, not a term either model uses) matters more than the act of eating less to lose weight. It is a live argument, not a footnote.
Newer observational data add a complication rather than a resolution. In a study following 491 Chinese adults for eight months, people who had ever tried intermittent fasting scored significantly higher on eating-disorder psychopathology and lower on intuitive eating than people who never had, at the start. Eight months later, after adjusting for baseline scores, both current and past intermittent fasting at baseline were associated with higher eating-disorder psychopathology, and past fasting with a higher BMI [8]. Because this was observational, it can't tell us whether fasting contributed to those scores or whether people with more disordered-eating tendencies were drawn to fasting in the first place, although the adjustment makes the second explanation less complete.
When "eating clean" becomes its own problem, and how unsettled that science still is
Orthorexia, a pathological preoccupation with "extreme dietary purity," has its conceptual roots in a definition from 1997 [9][21]. It is still not a formally recognized diagnosis, and its status remains contentious [9]. A systematic review of 62 studies notes there is still no consensus on diagnostic criteria, and prevalence estimates ranged from 2.6% to 87.7%, with the spread tied to the questionnaire and cutoff used and to the group studied, which included cancer survivors and postpartum women among others [10]. I'd rather say plainly that the measurement tools aren't good yet than borrow a number that sounds precise and isn't.
Is midlife actually a higher-risk window? The evidence is genuinely mixed
Given who reads this site, the honest answer matters more than a tidy one. Two studies found no link to menopausal stage: 230 women aged 40 to 60 showed no significant association between WHO-defined menopausal stage and eating or body-image measures [11], and 310 women aged 40 to 68 showed no significant differences across pre-, peri- and postmenopausal groups [12]. Two others did find differences. In 709 women aged 30 to 71, orthorexia scores were higher in the peri- and postmenopausal groups than in the premenopausal group [22]. In 1,001 Lebanese women aged 40 and over, postmenopause was associated with more orthorexic tendencies on one scale and perimenopause with more binge eating, although perimenopausal women's scores on the Düsseldorf orthorexia scale, the one [22] used, trended lower than premenopausal women's, not higher (p = 0.051) [23]. The scale behind the postmenopause finding, the ORTO-15, is one whose use a review of reviews describes as discouraged [9]. All four were cross-sectional surveys, so none can show that the transition caused anything. The first of them also found something different: the severity of menopausal symptoms, though not the stage a woman was in, was strongly associated with most eating-disorder measures [11].
There is a physiological piece underneath the survey data that's worth naming on its own, separate from whether it leads to disordered eating in any given woman. A 2026 narrative review of 56 studies, including animal models, describes perimenopause as characterized by increased visceral fat, loss of lean mass and reduced fat oxidation. The authors say these changes can occur even without a major change in body weight, which may delay their recognition in clinical settings [18]. A retrospective records study of 325 women measured body composition by bioelectrical impedance. Within its normal-weight subgroup, mean visceral fat area and body-fat percentage were higher in the perimenopausal and postmenopausal groups than in the premenopausal group; the pattern was weaker in the overweight group and did not hold in the obesity group [19]. The groups' average ages were roughly 29, 46 and 54, and the analyses were not adjusted for age, so the study can't separate menopausal stage from age. Still, a woman whose shape is changing while the scale barely moves may have a physiological basis for that mismatch.
A separate line of evidence, from treatment data, comes with limits. A chart review at one specialty treatment center found that women 40 and older made up an average of 4.7% of women's inpatient eating-disorder admissions from 1989 to 2001 and 11.6% from 2002 to 2006 [13]. That is one center's admissions, not a measure of how many women in the population are affected, and the data end in 2006, so they can't tell us whether the climb continued.
A separate small qualitative study asks a different question, whether clinicians recognize eating disorders in older women. It interviewed nine Austrian general practitioners about diagnosing and managing eating disorders in patients 65 and older, using a vignette of an older woman with anorexia symptoms. The doctors reported little familiarity with eating disorders in older patients and tended to attribute appetite or weight loss to aging or physical illness. The authors also describe gendered stereotypes, including a view of older women as less concerned with appearance, and note the absence of validated screening tools for older patients [20]. That study is small, from one country, and about patients a decade or more past perimenopause, so I'm not stretching it into a claim about your own doctor.
In the Study of Women's Health Across the Nation, 589 pre- and early perimenopausal women were surveyed. Regular binge eating was reported by 11%, dissatisfaction with eating patterns by 29.3% and marked fear of weight gain by 9.2%. Higher body mass index or waist circumference, current depressive symptoms, past depression and a history of childhood or adolescent abuse were each significantly associated with higher scores on the binge-eating and preoccupation measures [14]. Separately, in 1,647 adults aged 27 to 33 in Project EAT, any childhood maltreatment was associated with more than 60% greater risk of chronic dieting and overeating; emotional neglect was the type most consistently related to disordered eating outcomes by point estimate, and gender differences were not statistically significant [15]. That study is of adults in their late twenties and early thirties, not midlife. I would only say it is one reason a difficult relationship with food can have older roots than the decade it surfaces in.
A brief word on GLP-1 drugs
I've written about appetite-suppressing medication elsewhere on this site, so I'll keep this short. A rapid review of 25 studies (about 9,000 people) of GLP-1 medications in people treated for obesity or type 2 diabetes found limited data overall. Binge-eating episodes fell with liraglutide and binge-eating prevalence with semaglutide, one study found eating-disorder risk scores fell with liraglutide, with no difference from its comparison group, and two studies reported eating-disorder adverse events. The review's conclusion was that for most people, eating behaviors may improve or remain unchanged [16]. Separately, eating-disorder specialists writing a perspective piece note that eating-disorder symptoms were underrecognized in early bariatric-surgery practice until screening protocols were built, and that similar concerns are arising with GLP-1 drugs [17]. The evidence is thin in both directions.
What actually separates caution from a problem
Not every restriction is disordered. A managed elimination diet for a documented intolerance or a medical reason to track intake is not what the research above is about. The pattern worth bringing to a clinician is my own plain-language summary, not a checklist drawn from any one study: food or weight taking up much of the day's mental space, real distress when a rule is broken, and restriction that doesn't ease even after the original reason for it has passed. If that sounds like you or someone you work with, the useful next step is a therapist or physician who treats eating disorders, not another protocol. I'm not that provider. What I can do is keep this page honest about what the evidence shows and doesn't, which is the whole point of how I research anything that goes on this site.
Sources
- Liechty JM, Lee MJ. Longitudinal predictors of dieting and disordered eating among young adults in the U.S. Int J Eat Disord. 2013;46(8):790-800. pubmed.ncbi.nlm.nih.gov/23983018
- Udo T, Grilo CM. Prevalence and Correlates of DSM-5-Defined Eating Disorders in a Nationally Representative Sample of U.S. Adults. Biol Psychiatry. 2018;84(5):345-354. pubmed.ncbi.nlm.nih.gov/29859631
- Kalm LM, Semba RD. They starved so that others be better fed: remembering Ancel Keys and the Minnesota experiment. J Nutr. 2005;135(6):1347-1352. pubmed.ncbi.nlm.nih.gov/15930436
- Polivy J. Psychological consequences of food restriction. J Am Diet Assoc. 1996;96(6):589-592. doi.org/10.1016/S0002-8223(96)00161-7
- Fothergill E, Guo J, Howard L, et al. Persistent metabolic adaptation 6 years after "The Biggest Loser" competition. Obesity (Silver Spring). 2016;24(8):1612-1619. pubmed.ncbi.nlm.nih.gov/27136388
- Sharpe H, Griffiths S, Choo TH, et al. The relative importance of dissatisfaction, overvaluation and preoccupation with weight and shape for predicting onset of disordered eating behaviors and depressive symptoms over 15 years. Int J Eat Disord. 2018;51(10):1168-1175. pubmed.ncbi.nlm.nih.gov/30194690
- Chen JY, Singh S, Lowe MR. The food restriction wars: Proposed resolution of a primary battle. Physiol Behav. 2021;240:113530. pubmed.ncbi.nlm.nih.gov/34273346
- He J, Chen X, Cui T, et al. Engagement in Intermittent Fasting is Prospectively Associated With Higher Body Mass Index, Higher Eating Disorder Psychopathology, and Lower Intuitive Eating in Chinese Adults. Int J Eat Disord. 2025;58(1):225-237 (published online 2024). pubmed.ncbi.nlm.nih.gov/39530408
- Ng QX, Lee DYX, Yau CE, et al. On Orthorexia Nervosa: A Systematic Review of Reviews. Psychopathology. 2024;57(4):345-358. pubmed.ncbi.nlm.nih.gov/38432209
- Carpita B, Nardi B, Bonelli C, et al. Prevalence of orthorexia nervosa in clinical and non-clinical populations: a systematic review. CNS Spectr. 2024;29(6):549-569. pubmed.ncbi.nlm.nih.gov/39618419
- Mangweth-Matzek B, Rupp CI, Vedova S, et al. Disorders of eating and body image during the menopausal transition: associations with menopausal stage and with menopausal symptomatology. Eat Weight Disord. 2021;26(8):2763-2769. pubmed.ncbi.nlm.nih.gov/33595812
- Thompson KA, Bardone-Cone AM. Menopausal status and disordered eating and body image concerns among middle-aged women. Int J Eat Disord. 2019;52(3):314-318. pubmed.ncbi.nlm.nih.gov/30702172
- Ackard DM, Richter S, Frisch MJ, Mangham D, Cronemeyer CL. Eating disorder treatment among women forty and older: increases in prevalence over time and comparisons to young adult patients. J Psychosom Res. 2013;74(2):175-178. pubmed.ncbi.nlm.nih.gov/23332534
- Marcus MD, Bromberger JT, Wei HL, Brown C, Kravitz HM. Prevalence and selected correlates of eating disorder symptoms among a multiethnic community sample of midlife women. Ann Behav Med. 2007;33(3):269-277. pubmed.ncbi.nlm.nih.gov/17600454
- Emery RL, Yoon C, Mason SM, Neumark-Sztainer D. Childhood maltreatment and disordered eating attitudes and behaviors in adult men and women: Findings from Project EAT. Appetite. 2021;163:105224. pubmed.ncbi.nlm.nih.gov/33766616
- Jebeile H, Danielsen YS, Sumithran P, et al. GLP-1 Receptor Agonist Medications for Obesity and Type 2 Diabetes Treatment: A Rapid Review of Changes in Eating Behaviors and Eating Disorder Risk. Obes Rev. 2026;27(4):e70049 (published online December 2025). pubmed.ncbi.nlm.nih.gov/41340366
- Herb Neff KM, Dunford A, Goldschmidt AB, Allison KC, Richeson A. Eating Disorder Risk and Screening in Patients Using GLP-1RAs: Lessons Learned From Metabolic and Bariatric Surgery (perspective article). Int J Eat Disord. 2026 (online ahead of print). pubmed.ncbi.nlm.nih.gov/42502918
- Lobato VA, Minari TP, Pisani LP. Perimenopause and metabolic vulnerability: hormones, body composition and lifestyle changes. Climacteric. 2026 (online ahead of print):1-16. pubmed.ncbi.nlm.nih.gov/42634879
- Szeliga A, Chedraui P, Meczekalski B. The Impact of the Menopausal Transition on Body Composition and Abdominal Fat Redistribution. J Clin Med. 2026;15(2):740. pubmed.ncbi.nlm.nih.gov/41598677
- Kohestani T, Otto P, Köttl H. Overlooked and undertreated: gendered ageism in primary care management of eating disorders. Int J Equity Health. 2025;24(1):358. pubmed.ncbi.nlm.nih.gov/41469993
- Valente M, Syurina EV, Donini LM. Shedding light upon various tools to assess orthorexia nervosa: a critical literature review with a systematic search. Eat Weight Disord. 2019;24(4):671-682. pubmed.ncbi.nlm.nih.gov/31228168
- Raynal P, Soccodato M, Fages M, Séjourné N. A comparative study of orthorexia between premenopausal, perimenopausal, and postmenopausal women. Eat Weight Disord. 2022;27(7):2523-2531. pubmed.ncbi.nlm.nih.gov/35357686
- Khalil J, Boutros S, Kheir N, et al. Eating disorders and their relationship with menopausal phases among a sample of middle-aged Lebanese women. BMC Womens Health. 2022;22(1):153. pubmed.ncbi.nlm.nih.gov/35538474