If your child has a food allergy, you have probably had this moment. Another kid at the playground is holding a peanut butter sandwich. Someone spills milk at the next table. A birthday party comes with a cake you did not pick. Your stomach drops, and you start doing the math on how close is too close.
A study published in Acta Paediatrica in 2026 looked directly at that fear, and the results are reassuring. They also come with limits that matter more than the headline does.
What the researchers did
The team, led by Dr. Idit Lachover-Roth at Rambam Health Care Campus in Haifa, recruited 293 children aged 1 to 18 from allergy clinics at Meir Medical Center and Schneider Children’s Medical Center. Of those, 192 had a confirmed food allergy. The rest served as controls, including children with other allergic conditions but no food allergy.
Each child had allergenic food applied to the skin of the forearm for 15 minutes under supervision. The foods tested were milk, peanut, egg, sesame and tree nuts. In total the researchers performed 465 of these patch tests, 263 in the food-allergic group.
This is one of the larger prospective studies to test the question directly, and it took three years to complete.
What they found
In 83.7% of exposures, nothing happened at all. No redness, no swelling, no reaction even at the spot where the food sat on the skin.
About 15% of children had a mild local reaction, meaning redness or slight swelling confined to the contact site. Three children, just over 1%, had a more extensive skin reaction, but it stayed limited to the arm that was tested and cleared after treatment.
No child had anaphylaxis. No child had a systemic reaction of any kind.
Two patterns showed up in who reacted. Younger children were more likely to develop a mild local reaction than older ones. Children with milk allergy also had slightly more local reactions than the rest, which Dr. Lachover-Roth has suggested may come down to milk being a liquid, so the proteins sit against the skin differently than a solid food does.
The part that gets lost in the headline
Read quickly, this study sounds like permission to relax about food allergy generally. It is not that, and the researchers say so themselves. Their paper explicitly does not change any current recommendation about avoidance or emergency treatment.
Four limits are worth stating plainly.
It tested skin contact, not eating. Eating the food remains the way food allergy reactions turn severe. Nothing here changes that.
It did not test the route most school rules exist to interrupt. Peanut butter on a child’s hands, then on a shared toy, a table edge, a doorknob, and eventually in another child’s mouth. That is ingestion, arriving by an indirect path. The study looked at food sitting on skin, not food travelling from skin to mouth.
It tested intact skin only. Children with active eczema or broken, cracked or inflamed skin were not what this study measured, and the researchers note the risk may be different there. If your child has eczema, this is the line to hold onto, because a compromised skin barrier lets protein through in a way normal skin does not.
It did not test eyes, nose or mouth. Mucous membranes behave differently to skin. A child who rubs peanut butter into their eye is not covered by these findings.
What it does change, practically
The realistic gain here is a smaller circle of things to worry about, which is not nothing when the worry is constant.
If a food your child is allergic to lands on their arm, their leg, or their hand, the evidence says a systemic rea no ction from that contact alone is very unlikely. Wash the area with soap and water, watch them, and carry on. Panic and an immediate trip to the emergency room are not the required response to a smear of yogurt on a sleeve.
That has knock-on effects on daily life. Birthday parties, restaurants, playdates and shared spaces get somewhat easier to say yes to. Families who have been keeping all allergens out of the house entirely, or declining social invitations on contact grounds alone, have evidence to work with when they talk to their allergist about loosening that.
None of it replaces the plan you already have. Avoidance of eating the food, an epinephrine auto-injector within reach, and a school that knows what to do all stay exactly where they were.
Telling a contact reaction apart from a real emergency
This is the practical skill worth having, because the two look nothing alike once you know what you are looking at.
A local contact reaction stays where the food touched. Redness, mild swelling, itching, sometimes a few hives right at that spot. It does not spread to other parts of the body, and the child is otherwise fine. An oral antihistamine is a reasonable response here, and washing the area helps.
Anaphylaxis involves more than one body system, or the airway or circulation on their own. Trouble breathing, wheezing, throat tightness, a hoarse voice, swelling of the lips, tongue or throat, repeated vomiting, widespread hives, sudden pallor, dizziness or collapse. Symptoms in more than one system at once, or any breathing or circulatory symptom, means anaphylaxis until proven otherwise.
For anaphylaxis, epinephrine goes first and it goes immediately. Do not wait to see whether an antihistamine works, because antihistamines cannot reverse airway swelling or a drop in blood pressure. They have a role after epinephrine has been given, for the itching and hives, and no role as a substitute for it.
Epinephrine now comes in more than one form. EpiPen and Auvi-Q are auto-injectors, and neffy is an intranasal spray that does not require a needle. Whichever your child has been prescribed, the rule is the same: give it early, then call 911, because a reaction can rebound hours later even after the first dose works. We compare the options in more detail on our page on Auvi-Q vs EpiPen.
On school and daycare policies
Classroom allergen rules are not really about skin contact, and this study is not an argument against them. They exist because young children put their hands in their mouths, share food, and cannot reliably read a label. Cutting the number of allergen-containing items in a room full of three-year-olds is a way of reducing accidental eating, which is the thing that actually causes severe reactions.
What the study can do is take some heat out of the conversation. A parent whose child has an allergy does not need to treat every crumb on a table as a potential emergency, and a parent whose child does not have an allergy can understand that the point of the packed lunch rules is ingestion, not proximity.
If you are heading into a new school year, the more useful conversation with the school is about the emergency plan. Where the epinephrine is stored, who is trained to give it, who is authorized to give it when that person is off site, and what happens on field trips.
Testing tells you what you are actually dealing with
A lot of food allergy anxiety comes from not having a clear picture. Plenty of children carry an allergy label from a single reaction years ago, or from a broad panel of blood tests that flagged foods they have eaten without a problem since.
Proper evaluation, meaning history, targeted testing, and where appropriate a supervised oral food challenge, sorts out what is a true allergy from what is not. Some children turn out to have outgrown a milk or egg allergy. Others find out the label was never accurate. That is a bigger change to a family’s daily life than any single study.
Our allergists see children at all our locations across Manhattan, Glendale, Staten Island, Garden City and New Hyde Park.
Book online or call (212) 686-6321.