How often do diets fail - and what does failure mean?

What does diet failure mean?

There is no single defensible percentage for how many diets fail. Leaving a study, struggling with a prescribed diet, regaining some weight, returning to baseline and not reaching a target are different outcomes. Missing follow-up does not establish either success or failure. A weight-loss percentage also says little about the burden required to achieve it.

EPK aims to support adequate fuelling for daily function and reduce the harshness of trying to sustain too little food. That is an explicit aim, not yet a demonstrated advantage over other approaches. Its evaluation should include hunger, food preoccupation, sleep, energy, social participation and quality of life alongside body mass.

What the studies actually report

One-year completion: in Dansinger's 160-person trial, 35% of the Weight Watchers and Zone groups, 47.5% of the Atkins group and 50% of the Ornish group did not complete follow-up. Overall adherence was low and greater self-reported adherence was associated with greater loss. Non-completion does not establish that hunger caused withdrawal, and completion does not establish full adherence. [1]

Two-year commercial programme: Heshka's randomised trial reported average losses of 2.9 kg with the commercial programme and 0.2 kg with self-help in an intention-to-treat analysis. These are group means, not the proportions who succeeded or failed. Modest average results and individual variability should both be communicated. [3]

Five-year selected successful members: Lowe's survey reported that 42.6% of the five-year group maintained at least 5% loss, 18.8% maintained at least 10%, and 70.3% remained below initial weight. The sample was selected from people who had already reached goal weight. It cannot estimate the success rate of everyone who joined. [2]

Five-year WRAP trial: weight was ascertained for 871 of 1,267 eligible participants, or 69%. Average changes were -0.46 kg after brief advice, -1.95 kg after a 12-week WW referral and -2.67 kg after a 52-week referral. Between-group differences were not statistically significant at five years. The missing 31% cannot simply be labelled dietary failures. [5]

Eight-year intensive support: Look AHEAD reported at least 5% weight loss in about half of its intensive-lifestyle participants at year eight. This demonstrates that maintenance is achievable for a substantial proportion in a supported programme for adults with type 2 diabetes. It does not establish that all ordinary diets are easy or sustainable. [4]

Who was included in the Weight Watchers research?

The Weight Watchers follow-up studies examined members who had already reached their goal weight and maintained it long enough to qualify for lifetime membership. People who left the programme or never reached that milestone were outside this selected group. The findings describe maintenance among successful members; they cannot tell someone joining the programme their overall chances of success. [2,7]

The later study describes an additional benefit available to this group: after reaching goal weight and completing six weeks of maintenance, lifetime members were exempt from registration fees and received services at no cost while remaining within 0.9 kg of their goal weight. Their outcomes therefore reflect both prior success and access to continued support. [7]

To assess a programme as a whole, research needs to follow people from the point they join, report how many leave and why, and distinguish maintained weight loss from reaching a particular goal weight. For EPK, it should also ask what sustaining the programme costs people in hunger, fatigue, food preoccupation and everyday functioning.

Why sustaining restriction can be difficult

Sumithran and colleagues enrolled 50 adults in a ten-week very-low-energy programme and found that several appetite-related hormonal changes and increased hunger were still present a year after initial weight loss. This supports a physiological contribution to the effort of maintenance. It is not evidence that every participant must regain weight or that every moderate deficit causes the same burden. [6]

An association between adherence and weight loss does not explain why adherence is difficult. Hunger, cost, time, social circumstances, expectations and programme fit need direct assessment. Treating every disappointing result as poor willpower skips that question.

How EPK should be tested

The practical question is whether an approach can improve functioning and reduce persistent hunger and food preoccupation while meeting the person's health goals. A prospective comparison should prespecify these outcomes, report everyone enrolled, record reasons for stopping, and track body mass and function at 12 months and beyond.

This supports a critique of restrictive dieting that disregards its human cost. It does not require claiming that all weight loss is harmful or impossible. EPK must demonstrate its own benefits and trade-offs; limitations of other programmes do not establish its effectiveness.

Related working papers

Read the relevant PDFs: EPK–001 to 007 are the revised version 1.2 papers; EPK–008 to 010 are version 1.0 supplements. Earlier PDFs and DOI records remain on the paper archive.