Quick Answer
Counting calories can support weight loss, but the numbers are less exact than they look. A 2011 review found self-monitoring consistently linked to weight loss, though the evidence was weak. UK menu figures may legally differ from the food by up to 20%. Also, a US study found restaurant foods averaged 18% above their stated calories. Counting is not suitable for everyone, especially people with an eating disorder history or who are pregnant.
This guide explains why calorie numbers disagree and how US and UK labelling differs. It sets no calorie goals, because needs vary and a clinician or dietitian should advise. For wider habits, see sustainable weight loss tips.
How Accurate Are Calorie Numbers?
| Source | What it found | Why it matters |
|---|---|---|
| Tufts study, 2010 (29 restaurant foods) | Measured calories averaged 18% above stated values | Menu numbers can understate |
| Tufts study (10 frozen meals) | Averaged 8% above label; some items up to twice the stated figure | Packaged foods vary too |
| USDA almond research | 129 calories per serving measured versus 168 to 170 predicted by Atwater factors | Some foods give you fewer calories than the label |
| UK out-of-home rules | A tolerance of plus or minus 20% is acceptable | Label figures are averages |
| Lichtman et al., 1992 (10 diet-resistant adults) | Underreported food intake by about 47%; overreported activity by about 51% | Self-reporting can be a large source of error |
Sources: Tufts study summary, USDA, GOV.UK guidance and PubMed abstract. The Lichtman group was only 10 people within a 224-person study.
What Counting Calories Means
Calories measure the energy in food and drink. The NHS notes that intake matching expenditure keeps weight steady. Eating more than you use can lead to weight gain, and eating less can lead to loss (NHS).
Needs vary with age, weight, height and activity. Online calculators give estimates only, so we do not print a personal figure. A GP or registered dietitian can advise. Our guide to macronutrients shows how protein, carbs and fat each contribute energy.
Does Counting Calories Work?
A 2011 systematic review looked at 22 studies on self-monitoring in weight loss, and 15 focused on dietary monitoring. A significant association between self-monitoring and weight loss was consistently found. However, the authors judged the evidence weak. Samples were mostly white women, and studies relied on self-reported data (PubMed).
So counting may help some people notice what they eat. It does not guarantee loss. Other habits matter too, including sleep and stress. After the loss, maintaining weight loss raises separate questions.
Why Calorie Counting Numbers Disagree
The same food can show different calories in three places. A label gives an average. A lab gives a measured value. Your body absorbs something else again. These gaps explain why two apps, or an app and a label, rarely match.
First, rules allow tolerance. The Tufts team noted that US FDA rules permit up to 20% excess energy. In the UK, 20% either way is acceptable for out-of-home food. Second, general factors can overstate. USDA found almonds gave about 32% fewer usable calories than the Atwater factors predict. Preparation also matters: raw, roasted or ground nuts, and how well they are chewed, change what is absorbed (USDA).
Third, people misjudge portions. In the Lichtman study, diet-resistant adults underreported intake by about 47%. Finally, apps depend on what you enter. In one forum analysis, users often distrusted app data and some checked entries themselves (National Elf Service summary).
As a result, treat any figure as an estimate. Trends over weeks tell you more than one day’s total.
Calorie Labels in the US and UK
| Feature | United States | United Kingdom |
|---|---|---|
| Who must show calories | Chain restaurants with 20 or more locations | Businesses with 250 or more employees; delivery apps regardless of size |
| Since | 7 May 2018 | 6 April 2022 |
| What is shown | Calories on menus; fuller nutrition on request | kcal per portion, plus portion size or number served |
| Accuracy | FDA rules allow up to 20% excess energy (per Tufts report) | Plus or minus 20% tolerance acceptable |
| General-advice statement | Required, noting needs vary | Required, noting adult needs |
Sources: FDA menu labeling requirements and GOV.UK implementation guidance. We did not find Canadian or Australian rules in our sources, so check local labels there.
If You Choose to Track
Tracking is optional. If you try it, a few habits reduce error. Check the serving size on the label, since the NHS points to the energy line on packaging (NHS). Remember that drinks count as well as food. Also, planning meals can help manage intake, and our guide to balancing protein, carbs and fats helps with that.
Treat restaurant figures as approximate, given the tolerances above. Do not chase small daily differences. If tracking raises stress or rigid rules, pause it and speak to a clinician. A related option is eating by hunger cues, covered in emotional eating causes and coping.
Risks and Who Should Skip Counting Calories
Duke Psychiatry reports that calorie apps can trigger, maintain or worsen disordered eating symptoms. In one study cited, 26.1% of participants said fitness or weight-loss apps perpetuated their disordered eating. Among people with eating disorder diagnoses who used an app for counting, 73.1% said it contributed to symptoms. Not everyone who counts develops an eating disorder (Duke Psychiatry).
A forum analysis found mixed effects: some people in recovery felt the app helped, while others reported relapse. It did not test cause and effect (summary).
- Eating disorder history, or feeling prone to one: skip counting and talk to a clinician.
- Pregnancy: NHS guidance says not to diet or try to lose weight, because nutrient intake may fall (NHS trust).
- Under 18: we found no NHS guidance recommending counting for children, so ask a GP or paediatrician.
- Warning signs: rigid thinking about food, growing preoccupation, or feeling the app is making things worse.
In the US, the National Alliance for Eating Disorders helpline is 866-662-1235, weekdays (the Alliance). The old NEDA line is disconnected. Call or text 988 in a crisis (SAMHSA). In the UK, Samaritans is free on 116 123.
Evidence Summary for Calorie Counting Claims
| Claim | Where it stands | Reader guidance |
|---|---|---|
| Calorie apps are accurate | Overstated | They depend on your entries and on database values. Verify entries against labels. |
| Counting guarantees weight loss | Not supported | The review found an association with weak evidence, not a guarantee. |
| Online calculators give your exact needs | Overstated | They are estimates. A dietitian can personalise advice. |
| Restaurant labels are exact | Overstated | Tolerance is up to 20%, and a US study found 18% higher on average. |
| Every calorie on the label reaches your body | Not supported | USDA found almonds gave about 32% fewer usable calories. |
| Dietitians always prescribe a calorie count | Not supported | We found no source for this. Advice depends on the person. |
| Success stories show it works for everyone | Weak evidence | Stories are not trials. Results vary, and some people are harmed. |
| Counting is safe for anyone | Not supported | Apps can worsen disordered eating in some people. |
Counting Calories FAQ
They are only as accurate as the entries and portions you log. Label values can differ from the food by up to 20%, so treat totals as estimates.
A 2011 review consistently found an association between self-monitoring and weight loss, but rated the evidence weak.
Labels use averages and general factors. FDA rules allow up to 20% excess energy, and UK rules accept a 20% tolerance either way.
Yes. The NHS tips for losing weight do not rely on counting calories. See our sustainable weight loss tips for other habits.
Watch for rigid food rules or growing preoccupation. Duke Psychiatry advises being a careful observer and switching strategies if an app does not help.
No. People with an eating disorder history, who are pregnant, or who are under 18 should speak to a clinician first.
US chains with 20 or more locations must. UK businesses with 250 or more employees must, and delivery apps too.
USDA researchers measured about 129 calories per serving versus 168 to 170 from standard factors, around 32% fewer.
NHS guidance says not to diet or try to lose weight in pregnancy. Speak to your midwife or doctor about nutrition.
Related Reading
- Intermittent fasting for weight loss: what the studies show
- Top 10 weight loss myths checked against the evidence
- What is BMI and is it a good indicator of success
- How to spot a fad diet
Sources
- PubMed: Self-monitoring in weight loss: a systematic review of the literature (2011)
- PubMed: Discrepancy between self-reported and actual caloric intake and exercise in obese subjects (1992)
- ScienceDaily: Tufts study on energy content of restaurant and frozen foods (J Am Diet Assoc, 2010)
- USDA: Going nuts over calories
- GOV.UK: Calorie labelling in the out of home sector, implementation guidance
- FDA: Menu labeling requirements
- NHS: Understanding calories
- NHS: Tips to help you lose weight
- Duke Psychiatry: The trouble with tracking
- The Mental Elf: MyFitnessPal and people with eating disorders
- NHS Derbyshire Family Health Service: Managing your weight during pregnancy
Written and fact-checked by the HealthCoachJP editorial team. No clinician has reviewed this article. See our sources cited page. Last reviewed and rewritten October 2026.
Medical disclaimer: this article is general information, not medical advice. Talk to a doctor or registered dietitian before changing how you eat, especially if you are pregnant or have had an eating disorder.
