And in a second study released Friday, the researchers tried to replicate those results with other foods known to produce allergies in children, including milk, eggs, fish, wheat and sesame. They again produced evidence that the approach might work, but because so few families stuck to the difficult feeding regimen, the outcome cannot be considered conclusive.
Yet collectively, the studies offer more evidence that some medical authorities' long-established infant feeding recommendations may need to be revised. The World Health Organization, for example, recommends only breastfeeding for the first six months of life; following that guideline would keep a parent from introducing foods that might stave off allergies. In the United States, federal guidelines on the diagnosis and management of food allergies already are being reconsidered because of the results announced last year.
The research team, led by Gideon Lack, head of the Department of Pediatric Allergy at King's College London, plans to continue following the hundreds of children to determine how durable the protection may be. Extended research is needed to determine how long the protection might continue, he said.
"It would seem plausible, and the immunologic changes that we observed in [the study] suggest but do not demonstrate, that this may be a longer-term process," he said.
Anthony S. Fauci, director of the National Institute of Allergy and Infectious Diseases, which helped to sponsor the research, said proof of ongoing protection is valuable given most people's difficulty in avoiding exposure to peanuts and peanut butter. He said that while researchers can't assume that the protection will last beyond a year until that is proven, the result is a good indication of long-lasting safety.
"A year is a pretty good time," Fauci said.
For some people who are allergic to them, peanuts can cause a severe, sometimes rapid whole-body reaction called anaphylaxis that includes constriction of the airways, which can be life-threatening. They must be vigilant about consuming even trace amounts of peanuts or peanut oil and may have to carry injectible epinephrine to counter the effects of a reaction.
Milder symptoms can include hives, other kinds of skin rash, digestive problems, shortness of breath or wheezing.
The prevalence of peanut allergies in the United States has risen sharply in recent years, from 0.4 percent in 1997 to 2 percent in 2010. That has prompted widespread changes in food consumption rules in day care centers, elementary schools and on airlines, to prevent individuals with an allergy from accidental exposure.
The 2015 research that first suggested the unorthodox idea of exposing high-risk infants to peanut protein was hailed as a major step forward and a challenge to conventional wisdom. The New England Journal of Medicine called it a "landmark study" in an accompanying editorial.
In that experiment, high-risk infants aged 4 to 11 months were fed tiny amounts of peanut protein each week until they turned 5. (One sign of high risk for peanut allergy was severe eczema.) Just 1.9 percent of the children developed peanut allergies, compared with 13.7 percent of high-risk children who avoided peanut protein.
At the end of that study, the researchers told all the children to stop eating peanut products for a year. After that period, they found that a much larger group of children from the original peanut-avoidance group (18.6 percent) had developed allergies when compared to the children exposed to peanuts as infants (4.8 percent).
To ensure that parents were following the regimen, researchers collected dust from the children's beds and analyzed it for evidence of peanut protein.They found that a high percentage of participants were following instructions.
In the second study, researchers compared infants who were exclusively breast-fed for their first six months with others who were introduced to yogurt, milk, peanuts, eggs, sesame, whitefish and wheat at the same age. There appeared to be fewer allergies among those children, but only 42.8 percent of the parents were able to adhere to the study's instructions, too few to provide conclusive results.
Lack said the primary problem was that many parents dropped out at the first sign of allergy symptoms in their children. Nevertheless, the results were sufficient to indicate that the impact may be related to the amount of each food given to the infants, the study showed.