You Are Literally What You Eat — Half of You Renews in Three Weeks, the Rest Takes Decades

“You are what you eat.” The German philosopher Feuerbach said it in 1850, presumably not expecting anyone to take it quite this literally. But run the numbers and the phrase turns out to be less poetic metaphor and more biochemical fact — with one caveat: completion takes a while.

The reversal comes in two layers. First, the fast part: half your body (50%) is replaced with new raw material in as little as two to three weeks. Most of that is water and blood. Second, the slow part: getting to 90% takes decades, because muscle and bone are in no hurry. And the neurons making up roughly 2.8% of your total body mass — the ones reading this sentence right now — were assembled before you were born and will never be replaced.

Today’s hamburger is being incorporated into your body at this very moment. Let’s calculate how quickly that process fills you up — and how many burgers it takes to get there.

Leonardo da Vinci's study of human proportions, the Vitruvian Man
Leonardo da Vinci, Vitruvian Man (c. 1490). Even this perfectly proportioned body is, in reality, having its materials swapped out continuously — over weeks for some tissues, decades for others. Source: Wikimedia Commons (photo by Luc Viatour, CC BY-SA 3.0 · original artwork in the public domain)

This article is a calculation for fun, not medical advice. For actual health decisions, please consult a qualified professional.


INPUT

Variable 1: Reference body mass and water content

M=70,000  g  (70  kg)M = 70{,}000 \; \text{g} \; (70 \; \text{kg})

The average adult male body mass in South Korea (ages 20–60) was approximately 71.5 kg per the 2022 Korea National Health and Nutrition Examination Survey.[1] We use 70 kg as a round figure that also aligns with international references. This is a central estimate.

Separating out water:

About 60% of the human body is water.[2] Water turns over on a timescale of days. Whole-body water turnover is roughly 8–10% per day,[3] which in an exponential-decay model gives a water replacement half-time of τw10\tau_w \approx 10 days (≈ 0.027 years). (Back-calculated from ~42 L total body water and a daily exchange of ~3–4 L.) That means 90% of your body water is replaced within about 23 days.

Mwater=70,000×0.60=42,000  gM_{\text{water}} = 70{,}000 \times 0.60 = 42{,}000 \; \text{g}

Mstruct=MMwater=70,00042,000=28,000  gM_{\text{struct}} = M - M_{\text{water}} = 70{,}000 - 42{,}000 = 28{,}000 \; \text{g}

In the turnover model that follows, water (60%) gets its own exponential term; the remaining structural mass (40%, 28 kg) is broken down tissue by tissue.

Variable 2: Tissue fractions and replacement half-times

We decompose the 28 kg structural mass by tissue. The fractions sum to exactly 1.000 — verified in-text.[4]

Tissue Symbol Mass (g) Structural fraction fif_i Half-time τi\tau_i Renewable? Confidence
Skeletal muscle fμf_{\mu} 11,200 0.400 5–15 yr Yes Key uncertainty
Adipose (fat cells) ffatf_{\text{fat}} 4,200 0.150 9 yr Yes Solid
Bone fbonef_{\text{bone}} 4,200 0.150 10 yr Yes Solid
Red blood cells frbcf_{\text{rbc}} 1,680 0.060 120 days Yes Solid
Other tissue (gut wall, lung, vascular endothelium, etc.) fotherf_{\text{other}} 3,920 0.140 5 yr (assumed) Yes Needs assumption
Epidermis (stratum corneum) fskinf_{\text{skin}} 560 0.020 21 days Yes Solid
Small-intestine epithelium fepf_{\text{ep}} 280 0.010 4 days Yes Solid
Brain / spinal cord (neurons) fneuronf_{\text{neuron}} 1,512 0.054 ∞ (lifetime) No Solid
Cardiac muscle fcardiacf_{\text{cardiac}} 336 0.012 Near-lifetime No (approx.) Solid
Ocular lens (both eyes) flensf_{\text{lens}} 112 0.004 ∞ (lifetime) No Solid
Total 28,000 1.000

fi\sum f_i check: 0.400+0.150+0.150+0.060+0.140+0.020+0.010+0.054+0.012+0.004=1.0000.400 + 0.150 + 0.150 + 0.060 + 0.140 + 0.020 + 0.010 + 0.054 + 0.012 + 0.004 = 1.000

The 40% skeletal muscle figure comes from DEXA (dual-energy X-ray absorptiometry, the gold standard for body composition) measurements on adult males.[4] The 15% fat fraction is an average for adult men; you can adjust it in the calculator below.

Bar chart of cell turnover periods by tissue on a logarithmic time axis
Turnover periods span orders of magnitude. Gut lining renews in days, red blood cells in about four months, muscle and bone over roughly a decade — and neurons never. Source: Original work (CC0)

Variable 3: The non-renewable fraction — “the 2.8% paradox”

Non-renewable tissues (neurons, cardiac muscle, ocular lens) as a share of total body mass:

fnon-renewable=1,512+336+11270,000=1,96070,0002.8%f_{\text{non-renewable}} = \frac{1{,}512 + 336 + 112}{70{,}000} = \frac{1{,}960}{70{,}000} \approx 2.8\%

No diet, no matter how long you follow it, will ever replace this 2.8%.[5][6][7] The neurons alone account for 1,512 g — assembled in utero, they have been processing your thoughts ever since.

The maximum achievable replacement ceiling is 100%2.8%=97.2%100\% - 2.8\% = 97.2\%. Note that this 2.8% is measured against total body mass (70 kg). If you use structural mass (28 kg, i.e. excluding water) as the denominator instead, the non-renewable fraction becomes 1,960/28,0007.0%1{,}960/28{,}000 \approx 7.0\% and the ceiling drops to 93.0% — but the headline uses the more intuitive total-body-mass basis.

Stacked bar of a 70 kg body's composition with the 97.2% replacement ceiling marked
About 60% of a 70 kg body is water and 40% is structural mass. Only the 2.8% of neurons, lens, and cardiac muscle never turns over — which is what fixes the replacement ceiling at 97.2%. Source: Original work (CC0)

Variable 4: Half-time ranges

Tissue τ\tau (optimistic) τ\tau (conservative) Key reference
Water 10 days 10 days Yamada et al. 2022 — large-scale doubly-labeled water study[3]
Small-intestine epithelium 4 days 5 days Clevers 2013[8]
Epidermis 14 days 28 days Weinstein & Frost 1968[9]
Red blood cells 115 days 125 days Franco 2012[10]
Adipose cells 5 yr 10 yr Spalding et al. 2008, central value 9 yr[11]
Bone 8 yr 12 yr Raggatt & Partridge 2010[12]
Skeletal muscle 5 yr 15 yr Spalding 2005 ¹⁴C range; dominant uncertainty[13]
Other tissue 3 yr 7 yr Weak assumption; low sensitivity

The 5–15 year range for muscle τ\tau is the dominant sensitivity driver for this article’s headline range. That’s a direct consequence of muscle being the largest single fraction (fμ=0.400f_{\mu} = 0.400).


FORMULA

Step 1: The turnover model — why tissues differ

Cell replacement in each tissue follows an exponential-decay curve. The fraction of “original material” still present after tt years is et/τie^{-t/\tau_i}; the fraction replaced with new material is 1et/τi1 - e^{-t/\tau_i}.

Why do tissues turn over at different rates? The answer is their cell-division cycle.

  • Small-intestine epithelium (τ=4\tau = 4 days): Front-line exposure to food, stomach acid, and digestive enzymes — the harshest environment in the body, hence the fastest turnover.
  • Red blood cells (τ=120\tau = 120 days): Circulate hundreds of thousands of times through capillaries and physically wear out. They also lack a nucleus, so they cannot self-replicate.
  • Adipose cells (τ=9\tau = 9 years): Store and release energy in a relatively stable mechanical environment — slow and steady.
  • Skeletal muscle cells (τ=5\tau = 5–15 years): The cell itself (myonuclei, myofibers) turns over far more slowly than the protein inside it (which cycles on a scale of days to weeks).[13] An extra gym session does not speed up cell replacement. The turnover clock is set by biological programming, not by nutritional input.
  • Neurons (τ=\tau = \infty): Adult neocortical neurons are essentially not generated after the developmental period.[5] They are repaired across a lifetime, but never replaced.

Step 2: Whole-body replacement function R(t)R(t)

Water now gets its own exponential term rather than being treated as instantly replaced. This corrects a bug in simpler models where 50% targets yielded zero or negative times.

R(t)=0.60(1et/τw)+0.40irenewablefi(1et/τi)R(t) = 0.60\left(1 - e^{-t/\tau_w}\right) + 0.40 \sum_{i \in \text{renewable}} f_i\left(1 - e^{-t/\tau_i}\right)

Where:

  • 0.600.60: water’s share of total body mass
  • τw=10/3650.027\tau_w = 10/365 \approx 0.027 yr: water turnover half-time[3]
  • 0.400.40: structural mass as a share of total body mass (28/7028/70)
  • fif_i: each tissue’s fraction within structural mass (summed over renewable tissues only)

Renewable tissue fractions sum to: 0.400+0.150+0.150+0.060+0.140+0.020+0.010=0.9300.400 + 0.150 + 0.150 + 0.060 + 0.140 + 0.020 + 0.010 = 0.930

Convergence ceiling:

R()=0.60×1+0.40×0.930=0.60+0.372=0.972=97.2%R(\infty) = 0.60 \times 1 + 0.40 \times 0.930 = 0.60 + 0.372 = 0.972 = 97.2\%

No diet ever crosses this ceiling. The exact figure shifts slightly with body fat percentage (as the fat and muscle fractions redistribute), but the non-renewable core — neurons, cardiac muscle, ocular lens — is fixed and is what anchors the ceiling.

Important note: R(0)=0R(0) = 0 (you start at zero). Older models that treated water as instantaneously replaced set R(0)=0.60R(0) = 0.60, which caused sub-60% targets to collapse to zero or negative times. The corrected model handles any target normally.

Curve of replacement fraction R(t) rising fast then converging on the 97.2% ceiling
Replacement fraction R(t) over time. Half fills in within weeks (water and blood), but reaching 90% takes 10–18 years because of muscle and bone, and it stalls at 97.2%. Source: Original work (CC0)

Step 3: Reversal 1 — “half your body in a few weeks”

We back-calculate when R(t)=0.50R(t) = 0.50 (50% replacement).

At low targets the water term dominates. Water alone:

0.60(1et/0.027)=0.50    et/0.027=16    t=0.027×ln60.027×1.7920.049  yr0.60\left(1 - e^{-t/0.027}\right) = 0.50 \implies e^{-t/0.027} = \frac{1}{6} \implies t = 0.027 \times \ln 6 \approx 0.027 \times 1.792 \approx 0.049 \; \text{yr}

0.049 yr × 365 days = about 17.9 days (water alone). Adding blood, intestinal epithelium, and epidermis, the full-model back-calculation puts 50% replacement at roughly 2–3 weeks (~17 days).

Intuition check for English-speaking readers: The popular belief that “the body renews itself slowly” is reasonable in everyday terms — you don’t feel different week to week. But when you measure body composition by mass, more than half of it is water, and water refreshes in three weeks. The Yamada et al. 2022 doubly-labeled water study — 5,604 participants across 23 countries — confirmed average daily water turnover of ~3.2 L/day for an adult male of this size, which is exactly what this 10-day half-time is built on.[3] The fast half exists, and it is literally half of you.

Step 4: Debunking the “7-year myth”

The folk claim that “the body completely renews every 7 years” comes from misreading the fast tissues. Reaching 95% replacement in each tissue requires roughly t95%3τit_{95\%} \approx 3\tau_i:

Tissue τi\tau_i Time to 95% Multiple of 7 years
Water 10 days ~30 days 0.011×
Intestinal epithelium 4 days 12 days 0.005×
Epidermis 21 days 63 days 0.024×
Red blood cells 120 days 360 days 0.14×
Adipose cells 9 yr 27 yr 3.9×
Bone 10 yr 30 yr 4.3×
Skeletal muscle (conservative) 15 yr 45 yr 6.4×

“Seven years” captures the fast-cycling tissues but completely ignores muscle (fμ=0.400f_\mu = 0.400), the single largest fraction in the body. It’s a bit like timing a cross-country road trip by only counting the pit stops.

Step 5: Reversal 2 — “90% takes a very long time”

We back-calculate when R(t)=0.90R(t) = 0.90. At this level, fat, bone, and muscle must be substantially replaced — that takes decades. There’s no closed-form inverse, so we use bisection.

Standard adult (15% body fat, conservative τμ=15\tau_\mu = 15 yr):

ffat=(0.15/0.40)×0.40=0.15f_{\text{fat}} = (0.15/0.40) \times 0.40 = 0.15, fμ=0.40f_\mu = 0.40 (baseline).

Verification at t=17t = 17 yr:

e17365100(water term:1)e^{-17 \cdot \frac{365}{10}} \approx 0 \quad (\text{water term}: \approx 1)

e17/90.153,e17/100.184,e17/150.322,e17/50.034e^{-17/9} \approx 0.153, \quad e^{-17/10} \approx 0.184, \quad e^{-17/15} \approx 0.322, \quad e^{-17/5} \approx 0.034

R(17)=0.60×1+0.40×(0.010+0.020+0.060+0.150×0.847+0.150×0.816+0.400×0.678+0.140×0.966)R(17) = 0.60 \times 1 + 0.40 \times (0.010 + 0.020 + 0.060 + 0.150 \times 0.847 + 0.150 \times 0.816 + 0.400 \times 0.678 + 0.140 \times 0.966)

=0.60+0.40×(0.010+0.020+0.060+0.127+0.122+0.271+0.135)= 0.60 + 0.40 \times (0.010 + 0.020 + 0.060 + 0.127 + 0.122 + 0.271 + 0.135)

=0.60+0.40×0.745=0.60+0.298=0.898= 0.60 + 0.40 \times 0.745 = 0.60 + 0.298 = 0.898

At t=17t = 17 yr, R89.8%R \approx 89.8\% — just shy of 90%. Bisection gives t17t^* \approx 17–18 yr.

Optimistic scenario (τμ=5\tau_\mu = 5 yr):

e10/5=e20.135,e10/90.329,e10/100.368e^{-10/5} = e^{-2} \approx 0.135, \quad e^{-10/9} \approx 0.329, \quad e^{-10/10} \approx 0.368

R(10)=0.60+0.40×(0.010+0.020+0.060+0.150×0.671+0.150×0.632+0.400×0.865+0.140×0.865)R(10) = 0.60 + 0.40 \times (0.010 + 0.020 + 0.060 + 0.150 \times 0.671 + 0.150 \times 0.632 + 0.400 \times 0.865 + 0.140 \times 0.865)

=0.60+0.40×(0.010+0.020+0.060+0.101+0.095+0.346+0.121)= 0.60 + 0.40 \times (0.010 + 0.020 + 0.060 + 0.101 + 0.095 + 0.346 + 0.121)

=0.60+0.40×0.752=0.60+0.301=0.90190.1%= 0.60 + 0.40 \times 0.752 = 0.60 + 0.301 = 0.901 \approx 90.1\%

Optimistic scenario: ~10 years to 90% (bisection: t9.9t^* \approx 9.9 yr).

Step 6: Scenario comparison

Scenario Muscle τ\tau Body fat Time to 90% Plausibility
Optimistic (5 yr, 15% fat) 5 yr 15% ~10 yr Optimistic lower bound
Conservative (15 yr, 15% fat) 15 yr 15% ~18 yr Supported by mainstream literature
Conservative + aging (15 yr, age 50) 15 yr+ 15% ~20 yr+ Conservative upper bound

Headline: For a typical adult, optimistic ~10 years, conservative ~18 years. In every scenario, neurons, the ocular lens, and cardiac muscle (~2.8%) are never replaced.

One counter-intuitive side note: higher body fat can actually speed up overall replacement. Adipose cells (τ=9\tau = 9 yr) turn over faster than skeletal muscle (τ=15\tau = 15 yr). If fat makes up a larger fraction, the slow-muscle drag shrinks and the body renews faster overall — not by exercising more, but by having proportionally less of the slowest tissue.

Step 7: How many burgers?

Once you have the time, the burger count is straightforward. Eating nn burgers per day for tt^* years:

Nburgers=t×365×nN_{\text{burgers}} = t^* \times 365 \times n

Conservative scenario (t=18t^* = 18 yr), 5 burgers/day:

N=18×365×5=32,850N = 18 \times 365 \times 5 = 32{,}850

Optimistic scenario (t=10t^* = 10 yr):

N=10×365×5=18,250N = 10 \times 365 \times 5 = 18{,}250

To rebuild 90% of your body, you need roughly 18,000–33,000 burgers at 5 per day. (Why 5? An adult male’s reference intake of ~2,500 kcal/day divided by one Big Mac at 567 kcal[14] gives 2,500/5674.42{,}500/567 \approx 4.4, rounded up to 5.)

One important point: eating 1 burger a day versus 10 does not change the cell turnover rate (τi\tau_i). The biological replacement clock is driven by cell-division programming, not nutritional throughput. The burger count is merely a way of tallying up the meals eaten along the way — not a lever you can pull to speed the process up.

Step 8: Age correction

Cell turnover slows with age.

  • Bone: After age 40, osteoclasts (bone-resorbing cells) begin to outpace osteoblasts (bone-forming cells), extending the remodeling cycle.[15]
  • Epidermis: Keratinocyte replacement slows from ~21 days in young adults to 35–40 days in older skin.[16]
  • Muscle: Satellite cells (the stem cells that repair and regenerate muscle fibers) decline in activity with age.[17]

τiage=τibase×(1+αimax(0,age30)10)\tau_i^{\text{age}} = \tau_i^{\text{base}} \times \left(1 + \alpha_i \cdot \frac{\max(0, \text{age} - 30)}{10}\right)

αi\alpha_i: epidermis 0.15, bone 0.10, muscle 0.05. These are qualitative corrections; individual variation is substantial.



OUTPUT

Half your body (50%) is replaced with new raw material in roughly 2–3 weeks. Mostly water and blood. The next half — going from 50% to 90% — takes another 10–18 years. Mostly muscle and bone. At 5 burgers a day, that works out to somewhere between 18,000 and 33,000 burgers.

And the last 2.8% — the neurons in your brain, your cardiac muscle, your ocular lens — is never replaced. Feuerbach said “you are what you eat,” but his aphorism has a permanent exception clause: the part of you doing the reading. The neurons processing these words were assembled before you took your first breath. They have never eaten a single thing.

Eat clean and you are, eventually, built from clean materials. Eat junk and you are built from junk. It just takes 10–18 years to see the renovation through. Your brain, however, did not sign up for the program. It is keeping the original blueprints.


References

[1]: Korea Disease Control and Prevention Agency (KDCA), “2022 Korea National Health and Nutrition Examination Survey (KNHANES IX-1)”, 2023. https://www.kdca.go.kr/board/board.es?mid=a20501010000&bid=0015

[2]: Watson, P. E. et al., “Total body water volumes for adult males and females estimated from simple anthropometric measurements”, The American Journal of Clinical Nutrition, 33(1), 27–39, 1980. Adult total body water fraction ~60% (males). https://doi.org/10.1093/ajcn/33.1.27

[3]: Yamada, Y. et al., “Variation in human water turnover associated with environmental and lifestyle factors”, Science, 378(6622), 909–915, 2022. Doubly-labeled water measurements across 23 countries (n = 5,604). Mean adult daily water turnover ~3.2 L/day (30 yr male, 70 kg). Against ~42 L total body water, daily exchange rate ~7.6%, giving a water replacement half-time of ~10–13 days. https://doi.org/10.1126/science.abm8668

[4]: Heymsfield, S. B. et al., “Human body composition: advances in models and methods”, Annual Review of Nutrition, 17, 527–558, 1997. DEXA-based adult male body composition: ~40% muscle, ~15% fat, ~15% bone. https://doi.org/10.1146/annurev.nutr.17.1.527

[5]: Bhardwaj, R. D. et al., “Neocortical neurogenesis in humans is restricted to development”, PNAS, 103(33), 12564–12568, 2006. Adult neocortical neurons are essentially not generated after the developmental period. https://doi.org/10.1073/pnas.0605177103

[6]: Bergmann, O. et al., “Evidence for cardiomyocyte renewal in humans”, Science, 324(5923), 98–102, 2009. Annual cardiomyocyte turnover rate <1%; only a tiny fraction is ever replaced over a lifetime. https://doi.org/10.1126/science.1164680

[7]: Wride, M. A., “Lens fibre cell differentiation and organelle loss: many paths lead to clarity”, Philosophical Transactions of the Royal Society B, 366(1568), 1219–1233, 2011. Mechanism of lifelong retention of lens nuclear fiber cells. https://doi.org/10.1098/rstb.2010.0324

[8]: Clevers, H., “The intestinal crypt, a prototype stem cell compartment”, Cell, 154(2), 274–284, 2013. Small-intestine epithelial cell turnover cycle: 4–5 days. https://doi.org/10.1016/j.cell.2013.07.004

[9]: Weinstein, G. D. & Frost, P., “Abnormal cell proliferation in psoriasis”, Journal of Investigative Dermatology, 50(3), 254–259, 1968. Epidermal keratinocyte turnover ~21 days (normal skin). https://doi.org/10.1038/jid.1968.36

[10]: Franco, R. S., “Measurement of red cell lifespan and aging”, Transfusion Medicine and Hemotherapy, 39(5), 302–307, 2012. Mean red blood cell lifespan: 120 days. https://doi.org/10.1159/000342232

[11]: Spalding, K. L. et al., “Dynamics of fat cell turnover in humans”, Nature, 453, 783–787, 2008. ¹⁴C dating of adipose cells; mean adipocyte lifespan ~9 years. https://doi.org/10.1038/nature06902

[12]: Raggatt, L. J. & Partridge, N. C., “Cellular and molecular mechanisms of bone remodeling”, Journal of Biological Chemistry, 285(33), 25103–25108, 2010. Bone remodeling cycle ~10 years. https://doi.org/10.1074/jbc.R109.041087

[13]: Spalding, K. L. et al., “Retrospective birth dating of cells in humans”, Cell, 122(1), 133–143, 2005. ¹⁴C radiocarbon dating of cell birth dates; analysis of muscle cell (myonuclei) turnover rates. Protein turnover (days–weeks) is distinct from cell-level turnover (years–decades). https://doi.org/10.1016/j.cell.2005.08.017

[14]: McDonald’s official nutrition information (2024). Big Mac: 567 kcal, 25.5 g protein. https://www.mcdonalds.com/us/en-us/product/big-mac.html

[15]: Parfitt, A. M., “Osteonal and hemi-osteonal remodeling: the spatial and temporal framework for signal traffic in adult human bone”, Journal of Cellular Biochemistry, 55(3), 273–286, 1994. Bone remodeling cycle and age-related changes; osteoclast dominance after age 40. https://doi.org/10.1002/jcb.240550303

[16]: Ghadially, R. et al., “The aged epidermal permeability barrier”, Journal of Clinical Investigation, 95(5), 2281–2290, 1995. Slowing of keratinocyte replacement in aged skin (21 days → 35+ days). https://doi.org/10.1172/JCI117919

[17]: Conboy, I. M. & Rando, T. A., “Heterochronic parabiosis for the study of the effects of aging on stem cells and their niches”, Cell Cycle, 11(12), 2260–2267, 2012. Age-related decline in muscle satellite cell activity. https://doi.org/10.4161/cc.20437

You Might Also Like

This calculation was prepared with the assistance of AI tools and published after the Let's Calc Editorial Team verified the assumptions, formulas, and sources.