From autism to migraines, birth order may have wide-reaching effects

In a family of two children, whether you were born first or second may shape your likelihood of developing more than 150 conditions, according to the largest study of its kind. Researchers analysed data from more than 10 million siblings and found associations between birth order and everything from autism and anxiety to hay fever.

Birth order has fascinated scientists for more than a century, with attempts being made to identify links with personality traits and IQ. However, many older studies have been criticised for cherry-picking data or failing to control for confounding factors.

A landmark study in 2015 led by Julia Rohrer at Leipzig University in Germany analysed data from 20,000 children, accounting for these issues. It found that birth order had almost no bearing on personality and only a small association with IQ – a drop of about 1 to 2.5 IQ points between oldest and youngest siblings.

The new analysis took a different approach, focusing on the likelihood of developing different conditions. Benjamin Kramer at the University of Chicago in Illinois and his colleagues first compared 1.6 million pairs of siblings by coupling firstborns from one family with those born second from another family, matched on sex, birth year, parental age and sibling age gap. This allowed them to mitigate some confounding factors, such as the influence of how parents might treat their first and second children differently. They also compared genetically related siblings from 5.1 million families.

Across 418 medical conditions, 150 showed an association with birth order. Of these, 79 were more common in firstborns and 71 were more common in those born second.

The strongest associations for firstborns included several neurodevelopmental conditions, such as autism and Tourette syndrome, and incidents of childhood psychosis. They also had a greater risk of acne, allergies, hay fever and anxiety disorders. In contrast, those born second showed an elevated risk of substance abuse, shingles, biliary tract disease (such as gallstones), gastritis (inflammation of the stomach lining) and migraine.

“Overall, this seems like a really rigorous study,” says Rohrer. But she cautions that the associations are modest – firstborns seem to have a 3.6 per cent relative increased risk of depression, for example – and by no means do the results allow for deterministic statements. “We will only observe every person in one birth-order position. We will never know how their life would have played out differently in another position,” she says.

The team also explored some potential mechanisms that could support the findings. For instance, firstborns’ increased risk of allergies and hay fever might be explained by the “friendly foe” hypothesis, in which later-born children experience greater early-life microbial exposure from older siblings, promoting increased immune tolerance. In line with this idea, it became less likely that only the oldest of the two children would have allergies or hay fever as the age gap between them and their younger sibling increased.

A similar pattern emerged for substance abuse, with the elevated risk in those born second diminishing as age gaps widened. The authors link this finding with research showing that later-born children display increased risk-taking behaviour, but Rohrer says much of this evidence has proven shaky and suggests it might instead be because those born later may be over-represented in careers that can expose people to more drug-taking environments.

The team also suggests that the higher prevalence of autism in firstborns may reflect a combination of biological and environmental factors. Some research suggests that the mother’s immune system reacts more strongly in first pregnancies, for instance, which can affect the fetus’s developing brain. Sometimes, parents with one autistic child stop having more children, meaning there may be a potential unknown bias within the families that chose to have another child after the first one has been diagnosed.

Another interpretation is “diagnostic substitution”, says Rohrer. “ADHD and autism diagnoses are partially dependent on psychometrically assessed intelligence [such as through IQ tests] – the same [autistic] behaviour may be diagnosed as intellectual disability when combined with reduced intelligence, but as ADHD when combined with average intelligence,” she says. Because firstborns tend to have slightly higher IQs, they may be “more likely to end up with one diagnosis and later-borns more likely to end up with a different diagnosis, despite the same symptoms”.

Ray Blanchard at the University of Toronto in Canada suggests the results might differ if sibling sex or only-children were considered. For instance, his research suggests that older brothers slightly increase the odds of homosexuality in later-born boys. This may be explained by mothers developing antibodies to male-specific proteins in first pregnancies that change the immune system and subtly influence sexual orientation in later-born sons. “These obvious distinctions have proven very important in research on birth order and sexual orientation,” says Blanchard, adding that he thinks the order of the sex of siblings should be included in research like Kramer’s too.