Weight, Metabolism, and Why Diets Fail by Month Nine

If you plot the weight curve from almost any diet trial ever run, you get the same picture. A steady fall for about six months. A plateau somewhere around month six to nine. Then a slow climb back, often to within a few kilograms of where things started. It happens with low-carbohydrate, low-fat, Mediterranean, intermittent fasting, meal replacements and commercial programs alike. When a pattern is that consistent across that many different approaches, the explanation is not in the approaches. It is in the body.

What the body is actually defending#

Body weight is regulated, not chosen. The hypothalamus receives constant signals about energy stores, from fat tissue, the gut, the pancreas and the liver, and adjusts appetite and energy expenditure to keep the system stable.

The system evolved under conditions where losing weight was a threat and gaining it was an opportunity. It is therefore asymmetric. It defends far more vigorously against loss than against gain. Push weight down and the response is prompt and powerful. Push it up and the response is weak and easily overridden.

This is not a metaphor. It has been measured in controlled feeding studies, in hormone assays, and in brain imaging after weight loss. The defence is quantifiable, and it does not care how motivated you are.

The two feedback loops#

Loop one: energy expenditure falls further than it should#

Some drop in energy expenditure after weight loss is simply arithmetic. A smaller body costs less to run and less to move.

But careful metabolic ward studies, where people are fed measured amounts and their expenditure is measured directly, find that expenditure falls below what the new body size predicts. This is called adaptive thermogenesis. In classic studies, a 10 per cent reduction in body weight produced a fall in total energy expenditure of roughly 20 to 25 per cent, several hundred kilocalories a day more than the smaller body accounted for.

Follow-up of participants from a televised extreme weight-loss competition found resting expenditure still substantially below prediction six years later, in people who had largely regained. That study is small and unusual, and it should not be treated as the universal case. But it demonstrates that adaptation can persist far longer than "a few weeks of eating normally".

Loop two: appetite hormones shift, and stay shifted#

This is the larger effect, and the less discussed one.

• Leptin, produced by fat tissue, signals sufficiency. It falls disproportionately with weight loss, and the brain reads that fall as scarcity.
• Ghrelin, from the stomach, drives hunger. It rises after weight loss and stays raised.
• Peptide YY, GLP-1 and cholecystokinin, released from the gut after eating, signal fullness. All fall.

In a landmark Australian study, people who completed a ten-week weight-loss program had these hormones measured at intervals afterwards. At twelve months, ghrelin was still elevated and the satiety hormones still suppressed, and subjective hunger was higher than before the diet ever started. In people who had already regained a substantial share of the weight. The signal did not switch off when the diet did.

Modelling work has quantified the relative size of the two loops. For every kilogram of weight lost, the metabolic feedback increases the energy gap by roughly 20 to 30 kilocalories a day, while the appetite feedback increases the drive to eat by around 100 kilocalories a day. The hunger response is about three times stronger than the metabolic slowdown.

That single fact reframes the plateau. It is not that your metabolism has broken. It is that your appetite has been turned up, quietly and continuously, and holding it down requires effort that increases the longer you sustain it.

Why this is physiology, not weakness#

It is worth being explicit, because most people carry the opposite belief and have carried it for years.

If a treatment for high blood pressure worked for nine months and then stopped working in most people, nobody would conclude that the patients lacked discipline. We would conclude the treatment does not hold long-term for that condition, and we would look at the underlying biology.

Weight regain is the same situation with a different moral framing attached. The framing is not harmless. Experiencing weight stigma is independently associated with higher cortisol, worse mental health, avoidance of medical care and worse physical health outcomes. Effects that persist after accounting for body weight itself. People who feel judged by clinicians attend less, get screened less, and are diagnosed later. Shame is not a motivator; measured as an intervention, it performs badly.

What the long-term studies actually found#

Two conclusions come out of this table. The first is that the health benefits and the weight change are not the same variable. Look AHEAD improved a long list of hard outcomes without changing cardiovascular event rates, and the Diabetes Prevention Program's benefit outlasted the weight difference between groups. The second is that structured, continued human support is the intervention with the best evidence, not the specific dietary composition.

What predicts maintenance#

Drawing across the maintenance literature, the factors that consistently separate people who hold a change from people who do not are these.

• Continued contact. Regular check-ins with a person, clinician, group, coach, is the most consistent predictor in trials. Support that stops is followed by regain, in study after study.
• High volume of routine movement. Maintainers are usually active at a level well above the general population, often around an hour most days, much of it ordinary walking rather than structured exercise.
• Flexible rather than rigid rules. Rigid all-or-nothing rules predict worse outcomes, because a single deviation reads as total failure. Flexible patterns tolerate real life.
• Sleep. Short and irregular sleep raises ghrelin, lowers leptin, and measurably increases hunger and preference for energy-dense food within days. Sleep is the most underrated lever in this entire subject.
• Treating what is actually driving things. Untreated obstructive sleep apnoea, PCOS, an underactive thyroid, chronic pain, depression, and some prescribed medicines, including certain antipsychotics, antidepressants, steroids, insulin and some anticonvulsants: all affect weight regulation directly. It is always worth asking whether one of these is in play before assuming the problem is behavioral.
• A tolerable food environment. What is within arm's reach at 11 p.m. shapes behavior more than intentions formed at 9 a.m.
• Adding rather than removing. Protein and fiber at meals increase fullness through the same satiety hormones that weight loss suppresses. Framing this as adding food, not subtracting it, is both kinder and more sustainable.

Notice what is not on that list: a specific diet name, a target number, or a deadline.

What is oversold#

Metabolism-boosting supplements do essentially nothing. Green tea extract, raspberry ketones, garcinia and "fat burners" have either no effect or effects too small to notice, and some carry real liver risk. Detoxes and cleanses remove water and money.

"Starvation mode" as usually described, the claim that eating too little makes you gain weight, is a distortion of a real phenomenon. Adaptive thermogenesis slows loss; it does not create gain from nothing.

Genetic diet-matching has been tested directly, in trials randomising people to diets matched or mismatched to their genotype, and did not outperform chance. Body-type diets and metabolic-typing quizzes have no basis.

The most oversold thing is the twelve-week transformation, presented as an endpoint. Month three is the easy part of the curve.

The GLP-1 medications, factually#

This class has genuinely changed the clinical picture, and it deserves an accurate description rather than either hype or suspicion.

What they are. GLP-1 receptor agonists mimic glucagon-like peptide-1, a hormone released from the gut after eating. A newer subgroup acts on both GLP-1 and GIP receptors. They were developed for type 2 diabetes and were later found to reduce body weight substantially.

How they work. They act on appetite-regulating circuits in the hypothalamus and brainstem, and slow gastric emptying. People describe reduced hunger and, often more strikingly, reduced food preoccupation. The constant background negotiation about food quietens. Mechanistically, this is the first class that acts on the appetite loop described above rather than asking someone to out-discipline it.

What the trials show. In large randomised trials of the leading agents, average weight change over roughly 15 to 18 months has been in the region of 15 per cent of body weight for one agent and around 20 per cent for a dual-receptor agent, alongside lifestyle support, with wide individual variation. A large cardiovascular outcomes trial in people with existing cardiovascular disease and a higher body weight, but without diabetes, found a significant reduction in major cardiovascular events.

What happens on stopping. In extension studies, most of the weight returned within about a year of stopping, and appetite returned with it. This is consistent with a chronic condition being treated rather than cured.

Known downsides. Nausea, vomiting, constipation and diarrhea are common, especially early. Gallbladder problems and pancreatitis are uncommon but recognized. Muscle mass is lost alongside fat, as it is with any weight loss, which is one reason resistance training is discussed alongside them. They are avoided in people with a personal or family history of medullary thyroid carcinoma or MEN2, and in pregnancy. Compounded or online-sourced versions have caused documented harm, including dosing errors.

What I see in clinic#

The most common thing people bring me is not a question. It is an apology. They start by apologizing for their weight, before I have asked anything, often before they sit down. That tells me they have been through this conversation before and it did not go well.

What I have learned to ask instead is what changed. Weight rarely drifts for no reason. A shift to nights. A new medicine. A knee that stopped working. A bereavement. Untreated sleep apnoea, which I now suspect in anyone who snores and wakes tired, because it is common, it wrecks sleep, and treating it changes how people feel more reliably than anything I could say about food.

And the thing I most often end up saying: the fact that a previous attempt was followed by regain is not evidence about you. It is the expected result of a system doing exactly what it evolved to do. That is a different conversation from the one most people are braced for.

At work#

Shift work makes all of this harder, for reasons that are physiological rather than behavioral. Circadian misalignment alters glucose handling and appetite hormones independently of what is eaten, and night shifts shorten and fragment sleep, which itself raises hunger. People on rotating rosters are not making worse choices; they are making choices in a body running on the wrong clock.

Three things are within reach in an industrial setting. Food availability on the night shift, which is a facilities decision, not an individual one. Roster design, since forward-rotating shifts are tolerated better than backward-rotating ones. And how workplace health screening is conducted. A medical that reports a BMI with no context and no route to support does not improve health, and it reliably reduces the chance that person attends the next one.

When to get help now#

Seek urgent care for weight loss you did not intend, particularly more than about 5 per cent of your body weight over six months, and for rapid weight gain over days with leg swelling or breathlessness when lying flat, which can indicate a heart, kidney or liver problem rather than fat gain.

Seek prompt help if eating is followed by vomiting, laxatives, diuretics or compensatory exercise; if you experience episodes of eating that feel out of control and leave you distressed; if food or body image dominates your thinking; or if you have thoughts of harming yourself. Eating disorders occur across every body size and are frequently missed in people who are not underweight.

Book a routine appointment for loud snoring with witnessed pauses in breathing or daytime sleepiness, for new fatigue with cold intolerance or hair thinning, for irregular periods with acne or unwanted hair growth, or if weight changed noticeably after a new medicine was started.

The bottom line#

The nine-month plateau is not a failure of resolve; it is the predictable point where a defended biological system catches up, and the appetite side of that defence is roughly three times the size of the metabolic side. Knowing that changes what a sensible plan looks like: sustainable rather than maximal, supported rather than solitary, built on sleep and movement and treating whatever else is driving things. Health improves with those changes whether or not the scale moves, and those improvements count on their own terms. If you have regained weight before, the useful conclusion is about the strategy and the biology, not about you.