The Nutrition Paradox of Pelotas

Why — the evolutionary reading

The same body, dialed for a different century

A body can be perfectly adapted to a world that no longer exists.

This is the core idea of evolutionary mismatch: a trait that raised fitness in the environment where it evolved can lower health once the environment changes faster than biology can follow. Metabolism is the textbook case — and Pelotas caught it happening in real time.

The mismatch, drawn

Two clocks running at different speeds. The environment races ahead; the evolved body barely moves. The widening gap between them is where mismatch disease lives.

The food environment (fast) Evolved metabolism (slow) The mismatch

Pelotas didn't just get richer. It got richer in a particular way — and the three steps below are what turned social progress into a new disease of abundance.

Calibrated for scarcity

A metabolism shaped over deep time — and tuned again in early life — expects an uncertain, often lean food supply. It is built to capture and store energy efficiently.

The world flips

In one or two generations the supply becomes ultra-processed, calorie-dense and constant. The scarcity the body prepared for simply doesn't arrive.

Thrift becomes risk

The same efficient storage that once protected against famine now stockpiles a surplus that never ends. Undernutrition recedes; overweight rises to meet it.

DOHaD

Peter Gluckman and Mark Hanson gave this its developmental form: early life doesn't just build a body, it forecasts the world that body will meet. When the forecast is wrong — a lean prediction, an abundant reality — the mismatch is set before a child can walk.

Gluckman & Hanson, Mismatch (2006); Predictive adaptive responses and human evolution, Trends Ecol Evol (2005)

Comparative

Barbara Natterson-Horowitz calls obesity a mismatch disorder and widens the lens across species. Her useful complication: even wild animals gorge and fatten when abundance and access coincide. The problem isn't that we left some effortlessly lean "natural" state — it's that constant plenty is an environment few animals, us included, are built for.

Natterson-Horowitz & Bowers, Zoobiquity (2012) & Wildhood (2019)

This isn't only a story about bodies changing — it's a story about who changes first.

In 1982, overweight in Pelotas was a rich child's condition. By 2015, poor and rich one-year-olds looked almost identical on this measure — not because the wealthy improved, but because the poorest caught up to them, fast. When a health signal spreads that way, equity can improve even as the underlying problem gets worse for everyone. And if mismatch is the diagnosis, the treatment Natterson-Horowitz points to is not willpower but re-alignment: rebuilding an environment the evolved body can actually read.

Nutrition figures: Gonçalves H, et al. "Infant nutrition and growth: trends and inequalities in four population-based birth cohorts in Pelotas, Brazil, 1982–2015." Int J Epidemiol 2019;48(Suppl 1):i80–i88. doi.org/10.1093/ije/dyy233

Cohort methodology: Bertoldi AD, et al. "Trends and inequalities in maternal and child health in a Brazilian city." Int J Epidemiol 2019;48(Suppl 1):i4–i15. doi.org/10.1093/ije/dyy170

How honest is this chart

Four snapshots, not one trajectory. The 1982, 1993, 2004 and 2015 Pelotas cohorts each enrolled a new set of newborns. No single infant appears in more than one of these points. What the chart shows is how the population of one-year-olds changed across generations — not how any one child's metabolism shifted over their own lifetime.
The mixing board is a model, not a measurement. Pelotas measured length, weight, household income and maternal history — not methylation, cortisol, or HPA-axis reactivity. The "calibration" language on this page illustrates a plausible mechanism from the broader DOHaD and mismatch literature. It is not something these cohorts directly tested.
Mismatch is one lens among several. The same 33 years saw Brazil's universal health system created (1988), extreme poverty fall, and conditional cash transfer programmes expand nationally. Any of these could independently explain part of the shift. The evolutionary framing adds something — a reason to expect divergence rather than uniform improvement — but it doesn't replace the social and political history.
Pelotas is not Brazil, and Brazil is not everywhere. The city's economy underperformed the national average over this period, which may have shaped the pace of these changes in either direction.

One city. Thirty-three years. Four generations of newborns.

Predict before you look.

Undernutrition fell — and overweight rose to meet it

The same body, dialed for a different century

Calibrated for scarcity

The world flips

Thrift becomes risk