Reverse Dieting: Coming Out of a Calorie Deficit

When a calorie deficit ends, body weight rises within days. The rise is not fat. It is glycogen and the water bound to it, plus gut content, plus sodium-linked fluid. Fat regain, when it happens, is slower and smaller than the first-week jump suggests. The arithmetic below separates the two.

What maintenance means after a deficit

Maintenance is the energy intake at which stored energy is unchanged over a period long enough to average out day-to-day noise. It is not a fixed number. It falls as body mass falls, because moving a smaller body costs less and because the thermic cost of digesting food falls with intake.

Mifflin-St Jeor (1990) estimates resting metabolic rate (RMR):

• Men: RMR = 10 × weight(kg) + 6.25 × height(cm) − 5 × age + 5
• Women: RMR = 10 × weight(kg) + 6.25 × height(cm) − 5 × age − 161

Total daily energy expenditure (TDEE) is RMR multiplied by an activity factor. Published activity factors range from about 1.2 (sedentary) to about 1.9 (very active). The Mifflin-St Jeor equation carries a standard error of roughly 10% for RMR in adults, so any single figure is a midpoint with a band.

A 40-year-old woman, 165 cm, 70 kg, moderately active (factor 1.55):

RMR = 10 × 70 + 6.25 × 165 − 5 × 40 − 161
RMR = 700 + 1031.25 − 200 − 161
RMR = 1370.25 kcal/day
TDEE = 1370.25 × 1.55
TDEE = 2123.9 kcal/day
Band at ±10% on RMR: 1233–1507 kcal RMR
Band on TDEE: 1911–2336 kcal/day

If that person loses 10 kg and reaches 60 kg, the same equation gives a lower RMR:

RMR = 10 × 60 + 6.25 × 165 − 5 × 40 − 161
RMR = 600 + 1031.25 − 200 − 161
RMR = 1270.25 kcal/day
TDEE = 1270.25 × 1.55
TDEE = 1968.9 kcal/day

The drop from 2124 to 1969 kcal/day is about 155 kcal/day, or roughly 7%. That is the mass effect alone. Adaptive thermogenesis — the additional fall in expenditure beyond what mass loss predicts — is reported in some studies but its size is contested and varies between individuals. Treat it as a range, not a fixed penalty.

Why the scale rises in the first week

Glycogen is stored carbohydrate. Each gram of glycogen is bound to roughly 3–4 g of water in muscle and liver. A typical adult stores 400–600 g of glycogen. Restoring depleted glycogen after a deficit can therefore add 1.5–2.5 kg of water-associated weight within days. Sodium and gut content add more.

The Atwater factors (Atwater, 1900) assign 4 kcal/g to carbohydrate and protein, 9 kcal/g to fat. A 500 g glycogen restoration at 4 kcal/g represents 2000 kcal of stored energy. If that energy came from food eaten in surplus, it is not fat. If it came from a return to maintenance eating, it is refilling a tank that was emptied.

A 2 kg rise on the scale in week one is therefore consistent with zero fat gain. Fat gain requires a sustained energy surplus. The 7,700 kcal per kilogram figure often quoted for fat describes adipose tissue, not body weight, and it is wrong in both directions: early weight change includes water and glycogen, and later maintenance falls as mass falls, so the same deficit produces a slowing rate.

What a maintenance phase is for

A maintenance phase — sometimes called a diet break — is a period at estimated maintenance intake. Its purposes are practical:

• Establish the intake at which weight is stable, which is the only empirical test of maintenance.
• Restore glycogen and fluid, so subsequent measurements are less confounded.
• Allow training performance and hormonal markers to recover, where they were suppressed.
• Practise the eating pattern that will be used long term.

There is no requirement to reverse diet in small increments. The evidence for a gradual ramp — adding 50–100 kcal per week — is weak. What matters is arriving at maintenance and staying there long enough to measure it.

Estimating maintenance after weight loss

Use the current weight, not the starting weight. Recalculate after every 2–3 kg of change. The table below shows how the figure moves with the condition that changes it.

Activity factors are multipliers, not measurements. The 1.2–1.9 range is a convention, not a finding. A person who is unsure should start at a factor near 1.4 and adjust from observed weight over 2–3 weeks.

What people get wrong

The common error is treating the first-week rise as fat regain and cutting intake again. The mistake is natural because the scale is the only feedback most people have, and it responds to water within days. Fat change is slow and hidden beneath fluid shifts. A person who panics at a 2 kg rise and cuts 500 kcal/day is responding to glycogen, not adipose tissue.

The second error is assuming maintenance is a single number that holds for months. It falls as mass falls. A maintenance figure calculated at 70 kg overestimates by roughly 150 kcal/day at 60 kg in the example above. Recalculating is not optional if the goal is accuracy.

The third error is expecting a linear rate of loss. The 7,700 kcal/kg figure is a simplification. It describes adipose tissue, not body weight, and it fails early (water and glycogen dominate) and late (maintenance falls). A deficit that produced 0.5 kg/week at the start will produce less later at the same intake, because the same intake is now a smaller deficit.

The not-medical-advice line

This page describes arithmetic, not treatment. It does not diagnose, treat, or assume any clinical situation. Anyone with a history of disordered eating, or who is unsure whether their eating pattern is safe, should speak to a clinician. In the United States, the National Eating Disorders Association (NEDA) helpline is available by text at 988 (Suicide & Crisis Lifeline) or through the NEDA website. In the United Kingdom, Beat provides support at 0808 801 0677. These are routes to help, not substitutes for individual medical advice.

Worked example: adjusting to maintenance

A 35-year-old man, 180 cm, 85 kg, lightly active (factor 1.375). He has been eating 1800 kcal/day for 12 weeks and has lost 8 kg. He wants to return to maintenance.

Step 1: current weight = 77 kg
Step 2: RMR = 10 × 77 + 6.25 × 180 − 5 × 35 + 5
Step 3: RMR = 770 + 1125 − 175 + 5
Step 4: RMR = 1725 kcal/day
Step 5: TDEE = 1725 × 1.375
Step 6: TDEE = 2371.9 kcal/day
Step 7: Band at ±10% on RMR: 1552–1898 kcal RMR
Step 8: Band on TDEE: 2135–2609 kcal/day

He increases intake from 1800 to about 2370 kcal/day. The first week, weight may rise 1–2 kg from glycogen and water. That is expected. After 2–3 weeks at a stable intake, if weight is stable, that intake is his maintenance. If weight is still rising after week three, the estimate is high by roughly the amount of the rise divided by the weeks, converted at 7,700 kcal/kg — remembering that figure is approximate and applies to adipose tissue.

Summary of figures and their sources

• Mifflin-St Jeor (1990): RMR equation, standard error about 10%.
• Activity factors 1.2–1.9: convention, not measurement.
• Atwater factors (1900): 4 kcal/g carbohydrate and protein, 9 kcal/g fat.
• Glycogen water binding: roughly 3–4 g water per gram glycogen, from muscle biopsy studies.
• 7,700 kcal/kg: approximate energy density of adipose tissue, not body weight.
• NEDA and Beat: helpline routes, not clinical advice.

Every figure here is an estimate with a band. The band is the point.