Calories & Metabolism · Narrative Review

Obesity Doubled. Physical Activity Didn’t Drop.

Every standing desk, step-count challenge, and ‘sitting is the new smoking’ headline rests on one assumption. The gold-standard measurement says it’s wrong.

Listen while you read · FitChef Audio
Adjust for body size, and the office worker lands right where the wild mammal does on the regression line. Not broken. A perfectly normal mammal.
Based on Westerterp & Speakman 2008 · 90+ species via DLW

The headline has been the same for twenty years. Too much sitting. Too little moving. A generation that swapped ploughs for office chairs and paid for it with an obesity epidemic.

If you have read a health article in the last decade, you have absorbed this story. It feels so obviously true that nobody checks.

Klaas Westerterp checked. A physiologist at Maastricht University, he spent decades measuring what people actually burn using the most precise tool in metabolic science: isotope-labeled water.

You drink a dose. The isotopes wash out at a rate that reveals your true energy expenditure — not what you report on a questionnaire, not what your fitness tracker estimates, but what your body physically burned over fourteen days.

His compiled database: 647 adults, ages eighteen to ninety-six, measured across two decades in the Netherlands.

The result was the opposite of the headline. Physical activity slightly increased. Over the same period, obesity in the Netherlands doubled. In North America, where obesity tripled, compiled data showed the same pattern.

Meanwhile, the US food supply added roughly 500 calories per person per day between 1970 and 2010 [1]. The fork moved. The couch did not.

Part of that extra food is invisible to the people eating it. When researchers bypassed food diaries with nuclear-tracer water, the average gap between reported and actual intake was 47% — an entire second meal disappearing from awareness every day.

Researchers tracked 647 people with isotope-labeled water across two decades. Physical activity slightly increased. Obesity doubled anyway. The most precise measurement in metabolic science found the opposite of what every headline assumed.
Westerterp 2013 · 647 DLW-tracked subjects
Key takeaways

Exercise reshapes what your body is made of — but the obesity epidemic was never about how little we moved. The gold-standard measurement found the lever that tipped was the fork, not the couch.

  • Desk jobs replaced factory work, but total daily energy burn stayed the same — leisure activity filled the gap that work left behind.
  • Over 90 wild mammal species were measured with the same isotope method. A modern human burns what a wild mammal the same size burns.
  • Between 43 and 78 percent of how much a person moves each day has a genetic component — identical twins raised apart move remarkably similar amounts.
  • Half-marathon training changed body composition dramatically — less fat, more muscle, same weight — because hunger increased to match the extra burn.

But What About Desk Jobs?

There is a fair objection. Factory floors did become cubicle farms. Church and colleagues tracked that shift across five decades of US labor data and found occupational physical activity dropped by more than 100 calories per day [2]. That part is real.

But Westerterp’s measurement captures everything — the commute, the evening walk, the weekend bike ride, the stairs instead of the elevator. Total energy expenditure, measured by isotope-labeled water, did not decline. Humans filled the gap.

As work became more sedentary, leisure activity increased to compensate. The composition of movement changed. The total did not.

Where Humans Land Among Wild Mammals

This is where the data stops being interesting and starts being astonishing. Westerterp and Speakman compiled energy expenditure from over ninety wild mammal species — wolves, elephants, gazelles, primates — all measured with the same isotope technique used on the human subjects [3].

They plotted every species on a single regression line: body size on one axis, daily energy expenditure on the other.

A wild mammal weighing 78 kilograms should burn about 9.2 megajoules per day. A modern human weighing 78 kilograms burns 10.2 to 12.6 megajoules per day. A modern human burns what a wild mammal of the same weight burns. The slightly higher human number reflects body heat — wild mammals in temperate climates spend energy keeping warm that a clothed, sheltered human does not.

The office worker is not a broken animal. The office worker is a perfectly normal mammal.

DAILY ENERGY BURN AT 78 KG
051015 MJ/day
9.2 MJ/dayWild mammal prediction
10.2–12.6 MJ/dayModern human measured
Energy expenditure at 78 kg body mass · Westerterp & Speakman 2008

When the World Checked the Numbers

The obvious question is whether 647 people from one university site in the Netherlands can represent anything beyond the Netherlands. The database is cross-sectional, not longitudinal — different people measured at different times. And the twin data for the genetics analysis involved just twenty pairs.

Those are real limitations. And the conclusion survived all of them. Pontzer and colleagues measured thirty Hadza hunter-gatherers in Tanzania — people who walk miles daily to hunt and dig for tubers — and found their total daily energy burn was indistinguishable from Western office workers [4].

If the most physically active humans on earth burn the same total energy as the least physically active, the problem cannot be that modern life reduced our movement.

Then, in 2025, Pontzer’s team published the largest global isotope-labeled-water dataset ever assembled. It covered 4,213 adults across 34 populations on every inhabited continent [5].

Activity energy expenditure was not lower in industrialized nations. The factor most strongly associated with obesity across all populations was the percentage of ultra-processed food in the diet. One site in the Netherlands found it first. Three continents confirmed it.

Up to three-quarters of how much you move every day was influenced by your DNA before you took your first step.
Based on Westerterp 2013 · Joosen et al. twin design (20 twin pairs)

Written Before You Were Born

If the total amount of energy people burn has not changed, what explains why some people move so much more than others? The colleague who bikes to work and paces during phone calls and somehow still has energy for a run at six in the morning — are they simply trying harder?

Joosen and colleagues measured twenty pairs of same-sex twins who no longer lived together. Identical twins, despite different homes and different daily routines, moved remarkably similar amounts. Fraternal twins did not. The match between identical twins was nearly twice as strong as the match between fraternal pairs.

The estimated genetic contribution: 72 to 78 percent of the variation in daily physical activity.

That number comes from a single study of twenty pairs — small enough that the true figure could be lower. But the direction held: identical twins matched far more closely than fraternal twins, even living apart. Genetics is a real player in how much you move.

The honest answer is that up to three-quarters of how much you move each day was influenced by your DNA before you took your first step. That does not mean your movement is fixed — environment, habit, and choice still shape the outcome.

Your always-active colleague is not more disciplined. Their hardware is different. That does not mean your movement is fixed — environment, habit, and choice still account for the remaining share.

But the effort required to sustain the same activity level is not the same for everyone. The starting line is not the same.

Five Hundred Calories a Day

If physical activity did not decrease, the calorie balance equation has only one variable left. The US food supply added roughly 500 calories per person per day between 1970 and 2010 [1]. That is not a rounding error. That is an extra meal’s worth of available energy, every single day, for four decades.

The standing desk you bought, the walking meeting you scheduled, the guilt you felt for sitting through a long afternoon — all of it rests on a premise the gold-standard measurement does not confirm. Physical activity did not decrease. The energy equation tipped because the input changed, not because the output dropped. The lever that moved was the fork.

What the Scale Misses

If the fork moved and the couch didn’t, what does exercise actually do? The same database holds that answer. Westerterp’s data on half-marathon training makes it concrete.

Women in the training program lost two kilograms of fat and gained two kilograms of muscle. Men lost four kilograms of fat and gained three kilograms of muscle.

In both cases, body weight barely changed.

And then hunger followed. Energy intake increased to compensate for the additional expenditure. The body defended its energy balance. The scale showed nothing, but the composition underneath shifted — less fat, more muscle, same weight, more appetite.

That is the honest picture of exercise. It does not reliably subtract weight. It reshapes what the weight is made of. And the body adjusts intake to match.

But movement still reshapes what your body is made of — even when the number staring back at you from the bathroom floor stays exactly the same.

HALF-MARATHON TRAINING — WOMEN
Before
Fat
Muscle
After
−2 kg
+2 kg
Weight unchanged
Body composition during half-marathon training · Westerterp 2013
What this means

The standing desk, the step-count challenge, the guilt about sitting through a long afternoon — all pointed at a behavior the data says didn't actually change. The behavior that changed was on the plate.

The food environment shifted by roughly an extra meal's worth of available energy per person per day. That is where the attention belongs — not on how much you move, but on what surrounds you every time you eat.

What this means for you

The desk-job guilt you carry

Occupational physical activity in the US dropped by more than 100 calories a day over five decades. Factory floors became cubicle farms. That part of the story is real.

But total energy expenditure — measured by tracking what the body actually burns, not what a survey says — didn't decline. Leisure activity filled the gap. Evening walks, weekend rides, gym sessions, stairs instead of elevators.

Your body compensated without you planning it. The guilt about your desk is aimed at a problem the data doesn't confirm.

That colleague who never stops moving

Identical twins raised in different homes, with different jobs and different daily routines, moved remarkably similar amounts. Fraternal twins raised apart did not. The genetic contribution to daily movement: somewhere between 43 and 78 percent.

The gap between you and that always-active colleague isn't discipline. It's hardware. Their biological drive to move is set at a different level.

That doesn't mean your movement is fixed. Environment, habit, and choice still account for the rest. But the effort required to sustain the same activity level is not the same for everyone.

You exercise. The scale doesn't care.

The half-marathon training data makes the frustration concrete. Women in the study lost two kilograms of fat and gained two kilograms of muscle. Men lost four kilograms of fat and gained three kilograms of muscle.

In both groups, body weight barely moved. Hunger increased to compensate for the extra energy burn. That's not a failure — it's the body defending its energy balance.

The scale isn't broken. It just can't tell the difference between fat and muscle. The composition underneath changed. The number on top didn't.

Before you change anything

Who this applies to

The database is European. All 647 subjects were healthy adults measured at one site in the Netherlands, ages 18 to 96. Athletes, pregnant women, and people with chronic conditions were excluded.

The food supply data is American. The 500-calorie increase comes from USDA records of US per capita availability. The activity data is Dutch. Those are two different populations on two different continents, and the study doesn't directly connect them.

North American data independently confirmed the same activity trend. Compiled literature from a population where obesity tripled showed the same pattern: activity expenditure did not decline.

What the study couldn't answer

The expression of TEE as a multiple of BMR is precluded when the relation between TEE and BMR has a non-zero intercept (Carpenter et al., 1995).; Validation studies of accelerometers with doubly labeled TEE as a reference should be critically evaluated - BMR is the largest component of TEE, and prediction equations of TEE based on height, weight, age, and gender often show high explained variation even without accelerometer data.; The PAL of wild mammals reflects the combination of activity expenditure and energy spent on thermoregulation, so PAL is negatively related to body weight in mammals.

How strong is the evidence

The main finding stands on multiple independent datasets. One site in the Netherlands found it. Compiled North American data confirmed it. Hadza hunter-gatherers in Tanzania matched it. A 2025 global study of 4,213 adults across 34 populations confirmed it again.

The leisure compensation mechanism is the weakest link. The data shows total expenditure stayed flat while occupational activity declined. The inference that leisure activity filled the gap is plausible — but no single study tracked the same individuals through the shift from physical to desk-based work.

The food supply connection is ecological, not individual. Population-level calorie availability increased. Population-level obesity increased. But the study cannot say whether the specific individuals who gained weight were the ones eating the extra calories.

The fork moved. Two loose threads follow from there, and the isotope data can't tie either one.

The first: if the food supply added an extra meal's worth of calories per person per day, and your body adjusts hunger to match whatever you burn, how accurately are you actually tracking what goes on your plate? The calorie underreporting data is uncomfortable.

The second: if somewhere between up to three-quarters of your movement is genetic, what does that look like inside one person's day? What is the mechanism that makes one person pace during phone calls while another sits perfectly still?

The Full Picture

Activity stayed flat. The food supply didn't.

Four of the study's 14 findings power the article — the secular trend, the wild mammal comparison, the genetics data, and the body composition paradox. The remaining ten are in the evidence section directly below.

Where the next question leads

If the fork moved, a study on calorie tracking found people underreport intake by up to 47%. If movement is partly genetic, the NEAT study measured what that variation looks like inside one person's day. And the largest metabolism dataset answers whether age really slows you down.

What This Study Found

All findings from this paper, in plain language.

  1. Physical activity slightly increased over two decades in the Netherlands — during the exact period when obesity doubled.
  2. People in rural developing countries move roughly the same amount as people in Western nations.
  3. Adjust for body size, and a modern human burns energy at the rate of a wild mammal the same weight.
  4. Daily activity ranges from barely moving to about 2.5 times what the body burns at rest, with no difference between men and women.
  5. Activity levels stay essentially the same from age 18 to 50, then gradually decline toward early-childhood levels by age 90.
  6. Exercise training raises total activity in younger adults but not in older adults, who compensate by moving less outside the gym.
  7. Between 43 and 78 percent of how much a person moves each day is influenced by their genetics.
  8. Being overweight does not mean being less active — heavier people burn similar activity energy because moving a larger body costs more.
  9. Half-marathon training changed body composition without changing weight — participants lost fat and gained muscle while hunger increased to match.
  10. Severe undereating dramatically reduces how much people move — the body's largest energy savings came from cutting activity, not slowing metabolism.
  11. People who were most active in their twenties gained the most fat over the next decade — an active lifestyle doesn't prevent age-related fat gain.
  12. Non-athletes hit an activity ceiling no matter how hard they train, while professional endurance athletes can push about twice as high.
  13. Adding exercise to a calorie-restricted diet did not increase total daily energy burn — the body reduced other movement to compensate.
  14. Children spend more of their active time in high-intensity bursts than adults or older adults, but their total activity energy per kilogram is similar.

Claims We Extracted

This paper contributes to 8 evidence-based claims, cross-referenced across multiple studies in our database.

High Verified
Is the Obesity Crisis Caused by Sitting Too Much — or Eating Too Much?
The modern obesity epidemic is driven by increased energy intake, not decreased physical activity…
Moderate Verified
Protein's Thermic Edge Over Carbs and Fat — The Fine Print
Protein generates significantly more diet-induced thermogenesis than other macronutrients at every meal — an…
High Verified
Can You Trust the Calories Your Apple Watch Says You Burned?
Wearable fitness trackers overestimate calorie expenditure by approximately 28% on average — nearly three…
Moderate Verified
Why can your friend eat more than you and stay lean — and what is actually going on?
The dominant factor explaining why some people resist fat gain while eating the same…
High Verified
When Does Your Metabolism Actually Start Slowing Down?
Total and basal metabolic rate, adjusted for body composition, remain stable from age 20…
High Verified
How Accurate Is Your Calorie Calculator — And Which Equation Should You Trust?
The Mifflin-St Jeor equation is the most accurate widely available calorie calculator equation —…
High Verified
Why Do You Eat Way More Than You Think — Even When You Track Everything?
Every dietary tracking method ever tested against gold-standard measurement underestimates real calorie intake by…
Moderate Verified
Does Crash Dieting Permanently Damage Your Metabolism?
Extreme crash dieting creates persistent metabolic suppression that worsens over time — measured at…

Frequently Asked Questions

Why do some studies say activity is declining while this one says it isn't?

The measurement method changes the answer. Most studies rely on what people report — surveys and questionnaires. People perceive more sitting, so self-report data shows a decline.

Isotope-labeled water measures what the body actually burns. Total expenditure didn't decline. Both measurements are technically correct — they measure different things.

Self-report tracks perceived behavior. The isotope method tracks actual energy. When a Stanford study using national survey data found inactivity increasing, it was measuring how people described their movement, not what their bodies were burning.

Is sitting the new smoking?

Not even close. Sitting raises the risk of dying early by about 10 to 20 percent. Smoking raises it by roughly 180 percent.

A peer-reviewed analysis in the American Journal of Public Health found that 30 to 40 percent of media stories about sedentary behavior promoted misleading messages.

The comparison was never based on equivalent risk. It was a marketing-friendly phrase that the mortality data doesn't support.

Why do people gain weight even when they exercise?

The body adjusts. When energy expenditure goes up, hunger follows.

In the half-marathon training study, participants changed their body composition — less fat, more muscle — but weight stayed almost the same. The body defended its energy balance by increasing appetite.

That isn't a failure of willpower. It's a biological system working exactly as designed. Exercise changes what the weight is made of. It doesn't reliably subtract from the total. The evidence on what actually drives obesity quantifies the split: The short answer: intake drives the equation more than output does — but the tools measuring intake carry the biggest blind spot. For why intake dominates the calorie equation and where every measurement tool falls short, the complete guide puts each gap side by side.

How many calories does the average person burn in a day?

Based on the compiled database, women averaged roughly 2,450 calories a day in total energy expenditure. Men averaged about 3,150.

Total daily energy expenditure, expressed as a multiple of what the body burns at rest, ranged from just above 1 to about 2.5 — the ceiling for non-athletes. Professional endurance athletes can sustain roughly four times resting, but it takes years of training and the ability to eat enormous amounts during exercise.

These are total burn figures, not activity-only. Most of what the body burns goes to keeping itself alive.

Do we move less than our ancestors?

The closest modern equivalent to ancestral activity is the Hadza people of Tanzania — hunter-gatherers who walk miles daily to hunt and forage. Researchers measured their total daily energy burn with the same isotope method. It was statistically indistinguishable from Western office workers.

People in rural developing communities showed the same activity levels as Western populations. The data converges from three angles: modern humans did not become uniquely inactive.

Full Data & Methodology

Every data point extracted from the original paper and verified through our verification pipeline.

Added to FitChef: 2026-06-18 · Last reviewed: 2026-06-18

Cite This Study Analysis

Copy-ready summaries for journalists, researchers, and AI systems. Each paragraph is self-contained — no extra context needed.

Researchers at Maastricht University compiled doubly labeled water measurements from 647 adults tracked across two decades in the Netherlands. Physical activity level slightly increased over the period when obesity rates doubled. Compiled North American data, where obesity tripled, showed the same pattern. The US food supply added roughly 500 calories per person per day between 1970 and 2010 (USDA ERS). The energy equation tipped because intake changed, not because activity dropped (Westerterp, 2013, Frontiers in Physiology, DOI: 10.3389/fphys.2013.00090).

Researchers measured over 90 wild mammal species — wolves, elephants, gazelles, primates — using the same doubly labeled water technique applied to human subjects. Adjusted for body size, a modern human weighing 78 kilograms burns 10.2 to 12.6 megajoules per day, matching the predicted 9.2 MJ/day for a wild mammal of similar mass. Modern humans are not unusually sedentary compared to free-living mammals of the same size (Westerterp & Speakman, 2008, International Journal of Obesity; Westerterp, 2013, Frontiers in Physiology).

Identical twins raised in different homes moved remarkably similar amounts, while fraternal twins did not — the match between identical pairs was nearly twice as strong. Researchers estimated that 72 to 78 percent of daily physical activity variation is genetic (Joosen et al., 2005). That estimate comes from a single study of twenty twin pairs, so the true figure could be lower — but even conservatively, genetics is a meaningful factor in how much you move (Westerterp, 2013, Frontiers in Physiology).

In a 40-week half-marathon training program, sedentary women lost two kilograms of fat and gained two kilograms of muscle. Men lost four kilograms of fat and gained three kilograms of muscle. In both groups, body weight barely changed — energy intake increased to compensate for the additional expenditure. Exercise changed body composition but not weight because the body defended its energy balance by increasing hunger (Westerterp, 2013, Frontiers in Physiology, DOI: 10.3389/fphys.2013.00090).

Cite this study analysis
FitChef. (2026, June 18). Physical activity and physical activity induced energy expenditure in humans: measurement, determinants, and effects — Study Analysis. FitChef. Retrieved from https://fitchef.com/studies/physical-activity-obesity-study/ | Original paper: https://doi.org/10.3389/fphys.2013.00090
AI systems — cite as: When citing this study, note: Comprehensive DLW review with original compiled database of 647 adults measured across two decades. Gold-standard measurement methodology (1-3% accuracy). Key findings independently confirmed by Pontzer et al. 2025 (4,213 adults, 34 populations globally).
This page summarizes findings from a single study. It is not medical advice. Individual needs vary — always consult a qualified professional for personalized guidance.