← Experimental Work

Pelotas Birth Cohorts · Brazil · 1982–2015

A new health practice arrives in a city. Then what?

A birth cohort on the other side of the Atlantic, detailed enough to watch a benefit spread — and to see the order in which people reach it.

Kiel Pelotas

The prediction

Predict before you look.

Exclusive breastfeeding for the first months of life is one of the most cost-free, evidence-backed things a family can do for a baby. It needs no money, no equipment, and no clinic — only support and information.

In Pelotas, it spread from almost nobody to 45% of babies between 1993 and 2015. A genuine public-health success, by any measure.

But a success spreading through a city doesn't reach everyone on the same day. It has a path — and the order in which people reach it tells you something about how health itself is distributed.

When a free, purely beneficial health practice spreads through a city, which families take it up first?

  • Fifths, not fixed groups. The richest and poorest fifths are recomputed within each cohort. The families in the 2015 “poorest fifth” are not the same households as in 1982, and the whole income distribution shifted upward over the period.
  • “Inverse equity” is a hypothesis, well-supported here but not a law. It describes a common diffusion pattern; specific interventions can and do reach the poor first when they are actively targeted, as some Brazilian programmes were.
  • “Helps” and “harms” are deliberate simplifications. Caesarean sections are life-saving when indicated; the harm is in the excess, not the procedure. The tag flags direction of concern, not a verdict on any individual birth.
  • Cross-sectional snapshots. As with the companion pieces, each cohort is a different set of families measured once — a picture of a population changing, not of individuals moving through life.