top of page
  • Writer's pictureStephanie Fiorentino

The Science of Diet Failure

Updated: Aug 9, 2023

Please note this post contains clinical descriptors of body size and specific numbers and percentages that some may find triggering.

95% of diets fail.

You’ve almost certainly seen that statistic floating around intuitive eating, anti-diet, and eating disorder spaces. But you may be wondering where exactly that number comes from. I had that question myself a few years ago and decided to do some digging. Turns out, the initial research indicating a 95% weight loss failure rate is from way back in 1959.

Yup. The scientific community has known that diets don’t work for sixty years.

“The Results of Treatment for Obesity: A Review of the Literature and Report of a Series” from Albert Stunkard (a psychiatrist specializing in weight loss) and Mavis McLaren-Hume (a dietitian) had some key findings:

1) Diet interventions have a high drop out rate. When Stunkard reviewed research claiming “weight loss success” he found that most studies had high levels of attrition which, as it turns out, is still a well-documented problem in today’s weight loss literature. The estimates on attrition are wide ranging, anywhere from 10 – 80%, and details on attrition are typically underreported as they undermine validity.

But the point is, if people are dropping out of a study, the intervention obviously isn’t working for them and those people should be (but typically aren’t) included in the final data. Let’s say I do a weight loss study with 100 participants. 40 drop out. Of the 60 remaining participants, 20 lose weight in the short term. Does my intervention have a 33% success rate? Or an 80% failure rate?

2) Next, Stunkard found what anyone reading this post probably already knows – that it’s rare for people to lose significant weight dieting, and that those who do typically gain it back. Again, this is an ongoing problem in weight loss research. Many studies report pretty-great-seeming results at three or six months but fail to include meaningful follow up. And it’s worth noting that long term interventions often burry weight regain results in data tables while commenting on “overall” weight loss success in more frequently read abstracts and conclusions.

Let’s go back to my hypothetical weight loss study. Of the 20 participants to lose weight, let’s say they lost an average of 15% total body weight at the six-month mark. But by the two year follow up, they’ve gained back 10%. If I report an average overall 5% total body weight loss, that hardly represents the experience of my participants. As any person struggling with chronic dieting will tell you, losing 30 pounds and gaining 20 back doesn’t feel like a diet “success”.

So ok, this is one study from 60 years ago. But as I’ve highlighted, future literature reviews support Stunkard and McLaren-Hume’s results.

- In 1992, a panel from the National Institutes of Health presented findings that almost all weight loss is regained within five years.

- A 2007 multidisciplinary review found no significant evidence to support dieting and weight loss interventions and additionally noted that the harms of weight cycling are more devastating than any benefits of short-term weight loss.

- A 2015 review found the odds of achieving a “normal” weight were 1 in 124 for women with obesity class one, and 1 in 677 for women with obesity class three.

You’ll find additional literature reviews as well as investigations into problematic weight loss intervention methodologies cited below. But suffice to say the evidence is

clear. Yes, there are a multitude of (often flawed) studies demonstrating short term weight loss. But in the long term, weight loss interventions are not supported by statistical evidence.

Imagine if your doctor prescribed a medicine that was expensive, had lots of negative side effects, and only worked for like 5% of people. Would you take that pill?

As always, this content is for educational purposes only. It should not be used to prevent, diagnose, or treat any medical or psychological condition. It is not an alternative for individualized care.



Batterham, Rachel L. “Switching the Focus from Weight to Health: Canada's Adult Obesity Practice Guideline Set a New Standard for Obesity Management.” EClinicalMedicine, vol. 31 (2021): p. 100636. doi:10.1016/j.eclinm.2020.100636.

Dalle Grave, R., Suppini, A., Calugi, S., & Marchesini, G. “Factors associated with attrition in weight loss programs.” International Journal of Behavioral Consultation and Therapy vol. 2,3 (2006): pp. 341-353.

Fildes, Alison et al. “Probability of an Obese Person Attaining Normal Body Weight: Cohort Study Using Electronic Health Records.” American Journal of Public Health vol. 105,9 (2015): e54-9. doi:10.2105/AJPH.2015.302773

Mann, Traci, et al. “Medicare's Search for Effective Obesity Treatments: Diets Are Not the Answer.” American Psychologist vol. 62,3 (2007): pp. 220–233., doi:10.1037/0003-066x.62.3.220.

Methods for voluntary weight loss and control. NIH Technology Assessment Conference Panel.” Annals of Internal Medicine vol. 116,11 (1992): 942-9. doi:10.7326/0003-4819-116-11-942

Miller, W C. “How effective are traditional dietary and exercise interventions for weight loss?.” Medicine and science in sports and exercise vol. 31,8 (1999): 1129-34. doi:10.1097/00005768-199908000-00008

Rothblum, Esther D. “Slim Chance for Permanent Weight Loss.” Archives of Scientific Psychology, vol. 6,1 (2018): pp. 63–69. doi:10.1037/arc0000043.

Stunkard, Albert. “The Results of Treatment for Obesity: A Review of the Literature and Report of a Series.” A.M.A. Archives of Internal Medicine, vol. 103,1 (1959): p. 79. doi:10.1001/archinte.1959.00270010085011.

Recent Posts

See All




[D]o “95% of diets fail”, as is sometimes claimed?

This, of course, requires defining what is a “diet success”, and there is no universally accepted definition for what this means (and it likely varies according to each person’s needs and preferences).

Nevertheless, some researchers have proposed a weight loss of ≥10% of one’s initial body weight after ≥1 year as successful weight loss.

Using that definition, the percentage of people achieving successful long-term weight loss is variable in studies, with rates of 10%, 20%, 26%, and 33% reported over periods of 1–8 years.

These findings suggest that the failure rate of dieting is high, but not as high as 95%. Of course, it's important to note that the…

bottom of page