Children born preterm stayed leaner through early childhood, a big Japan cohort finds

Much of the worry about how preterm babies grow up has focused on the risk of later obesity, on the theory that an early, catch-up growth spurt can tip into excess weight. But the evidence has always been mixed, with some studies finding the opposite — lower body mass that lingers. A study published in the European Journal of Pediatrics on October 13 uses one of the world's largest birth cohorts to ask which pattern dominates in early childhood, and lands firmly on the side of leaner, not heavier [s1].

What the researchers did

The analysis drew on the Japan Environment and Children's Study, a nationwide birth cohort [s1]. From 100,121 live births, 93,406 children were included [s1]. Preterm birth was defined as delivery before 37 weeks of gestation [s1]. The outcome was thinness, defined using World Health Organization criteria as a body mass index for age more than two standard deviations below the median — a marker of being underweight for one's age, not merely slim [s1].

The researchers compared preterm-born and term-born children at ages 3, 4, 5 and 6, using multivariable logistic regression to estimate the odds of thinness at each age, and linear mixed-effects models to track how BMI itself changed over time [s1].

What they found

Preterm birth was associated with higher odds of thinness at most ages [s1]. The adjusted odds ratios were 1.35 at age 3 (95% confidence interval 0.94 to 1.93), 1.43 at age 4 (1.02 to 1.99), 1.70 at age 5 (1.29 to 2.25), and 1.52 at age 6 (1.20 to 1.92) [s1]. The estimate at age 3 did not reach statistical significance — its interval crosses 1.0 — but from age 4 onward the association was consistent and significant [s1].

The trajectory analysis told the same story from a different angle. Mean BMI was consistently lower among children born preterm, with differences of 0.11 to 0.19 kg/m² across ages 3 to 6, and — the detail that matters most — no evidence of convergence over time [s1]. The gap did not close as the children grew older [s1]. A sensitivity analysis using an alternative definition of thinness gave consistent results [s1].

How much weight the numbers carry

The strengths are scale and consistency. Nearly 93,000 children, four separate ages, two different analytic approaches and a sensitivity check all point the same way, which is hard to dismiss as chance. The effect sizes are moderate — an odds ratio around 1.5 means roughly half-again the odds of thinness, not a transformation — and the BMI gap itself is small in absolute terms, a fraction of a kilogram per square metre. What makes the small gap notable is that it persists rather than fades.

The usual cautions for observational work apply. This is an association, not proof that prematurity causes thinness; preterm birth travels with many other factors — gestational complications, feeding, underlying health — and the researchers could adjust only for what was measured. The study also stops at age 6, so it cannot say whether the pattern holds, closes or reverses later in childhood or adolescence. And the cohort is Japanese, where the baseline distribution of body size differs from other populations, so the exact numbers may not transfer.

Why it matters

The practical point is about how children born preterm are monitored. Growth charts and clinical attention have historically leaned toward spotting faltering height or, more recently, the risk of excess weight. A finding that preterm-born children run persistently leaner — and do not catch up by age 6 — argues for watching body mass as well as linear growth, and for factoring gestational age into how a child's size is interpreted rather than reading a single chart in isolation.

What to watch

Whether the lower-BMI trajectory continues beyond age 6, whether it reflects lower muscle, fat or both, and whether it carries any consequence for later health — or is simply a durable difference in size. Replication in other national cohorts, with different baseline body-size distributions, would show how far the Japanese findings generalise.

This article describes a single observational cohort study and is informational only. It is not medical advice.