Various theories have been advanced over the years. The “hygiene hypothesis” suggests that early exposure to bugs creates a protective natural immunity, which in turn reduces the risk of allergies. However, the data on which this is based is mainly observational, where people moving from low-allergy countries to high-allergy countries develop the risk of the place they are living in.

Another piece of research highlighted how Jewish children brought up in the UK had 10 times the risk of peanut allergy compared with Jewish children brought up in Israel — where peanuts are routinely in children’s diets before the age of 1. This is clearly interesting but didn’t prove cause and effect, which is why a series of studies done in the UK a decade ago are crucial to understand what might be happening.

These studies took the presumption that avoiding peanuts to avoid allergy was sensible and turned it on its head. Instead they asked whether this avoidance might actually be making things worse. So they carried out a series of randomised controlled trials — a powerful way to try and untangle cause and effect. Noting that when they did the study, peanut allergy had doubled in Western countries over the previous decade, they took 640 infants who had severe eczema or egg allergy, and who would have been at high risk to develop a peanut allergy as well. They split the children into two groups, with one group told to continue to avoid peanuts and the other recommended to eat them.

They were followed up, and 60 months later it was clear that there was a marked difference between the groups. Almost 14% of children avoiding peanuts had a reaction to them on a skin-prick sensitivity test compared to just under 2% of the children who ate them. In other words, the peanut exposure had a protective effect; worldwide, guidelines telling parents about what to feed their children were updated. Now, the NHS recommends that peanuts can be given, ground or as a paste, from six months old.

There are several lessons contained within this policy turnaround. First, where we don’t know what the best advice is, we should admit that uncertainty. It’s easy to assume that what sounds sensible actually is. It might be — but it also might not. The second is admitting that uncertainty enables us to do fair tests of interventions. This in turn can give people accurate advice and enable them to avoid harms. We can’t tell for sure from this data, but the change in policy might be arresting the previously observed rises in peanut allergy.

The third is that high-quality trials are a broader social good. The children who took part in the peanut trials had a good chance of getting the most effective advice rather than the (wrong) advice being given at the time — even though they didn’t know it yet. We should be profoundly grateful to researchers willing to question the status quo: it’s what better healthcare depends on.

---

Margaret McCartney is a GP and broadcaster.

mgtmccartney

---