The Trial Where Only the Parents Turned Up
Nearly every parent of an anxious child does it: the extra check, the fourth answer to the same question, the night spent on their bedroom floor. Then a team at Yale ran a trial in which they treated only the parents, and the children never met a therapist at all.
There is a version of the evening that parents of an anxious child will recognise. You check the wardrobe. Then you check it again, because they asked. The question about whether the front door is locked gets answered, and then answered twice more. You lie on the floor of their bedroom until their breathing slows, because it is faster than the alternative and everyone needs to sleep.
Psychologists have a name for all of that. They call it accommodation: any change you make to your own behaviour to spare your child from feeling anxious. When researchers put the standard questionnaire in front of parents of anxious children, more than 97 per cent reported doing at least some of it. Not a subset of parents. Nearly all of them.
The scale of the thing is not in dispute. One in five children aged eight to sixteen in England had what the NHS calls a probable mental disorder in 2023. In 2024-25, more than a million children, about one in ten, had a referral to mental health services, close to double the figure from 2018-19. Waiting lists being what they are, most of the actual work happens at home, at ten at night, with no manual.
Researchers who tracked families week by week through treatment found the loop runs in both directions. Accommodating a fear one week nudged the child's anxiety up the next. The stronger pull went the other way, though. A more anxious child drew more accommodation out of their parents. Nobody decides to do this. It accumulates.
A study of 425 mothers and their children tried to separate the damage done by the anxiety from the damage done by the response to it. Somewhere between a fifth and a half of the trouble these children had at school and at home traced back to the accommodating rather than the anxiety itself. The children could see this more clearly than their parents could.
One group got the standard treatment: twelve weekly sessions of cognitive behavioural therapy, child and therapist, parents left out of it. The other group got twelve sessions of a programme called SPACE, short for Supportive Parenting for Anxious Childhood Emotions. In that group the child never met a therapist. Only the parents turned up.
The parent-only group did at least as well. Around 88 per cent of the children whose parents were treated were rated as improved, against 76 per cent of those who had the therapy themselves. Roughly six in ten in each group no longer met the criteria for the disorder they had arrived with. The difference between the two treatments was not statistically meaningful, which is exactly what the trial was built to test.
One measure did separate them. Accommodation fell further in the parent group, which you would hope, given that bringing it down was the entire intervention.
This is a single trial, and the wider evidence is thinner than that headline makes it sound. A review of parent-focused anxiety programmes published in Behaviour Research and Therapy in early 2025 found real promise and then said plainly that the quality of evidence across the field is low. One pattern held up. Every approach that worked involved changing what parents actually did. The ones that only changed what parents understood did not get the same results.
Eli Lebowitz, the Yale psychologist behind the programme, reduces the alternative to two ingredients: acceptance and confidence. Acceptance means saying the fear out loud. I know you're frightened of sleeping on your own, I know it feels horrible. Confidence is the second half. And I know you can do it.
Most of us manage one or the other. 'There's nothing to worry about' is confidence with no acceptance. 'Fine, I'll stay' is acceptance with no confidence. Neither gives a child anything to stand on.
The practical version is smaller than people expect. Choose one accommodation, not all of them. Tell your child in advance what you are changing and why, in daylight, not in the middle of a bad night. Then hold it, and say the two sentences as many times as it takes.
Expect it to get louder before it gets quieter. A child used to five reassurances will ask for the sixth more insistently before they stop asking. That is not a sign you have made a mistake.
None of which makes accommodation a parenting failure. It is what love does when it is handed a distressed child and no instructions. Every parent in that Yale study was doing it, including the ones whose children got better.
Somewhere in the next few weeks there will be a party your child does not want to go to, or a question that needs answering for the fourth time. The question underneath it is not really whether they can handle it. It is whether we can bear to watch them find out.
Psychologists have a name for all of that. They call it accommodation: any change you make to your own behaviour to spare your child from feeling anxious. When researchers put the standard questionnaire in front of parents of anxious children, more than 97 per cent reported doing at least some of it. Not a subset of parents. Nearly all of them.
The scale of the thing is not in dispute. One in five children aged eight to sixteen in England had what the NHS calls a probable mental disorder in 2023. In 2024-25, more than a million children, about one in ten, had a referral to mental health services, close to double the figure from 2018-19. Waiting lists being what they are, most of the actual work happens at home, at ten at night, with no manual.
The kindness that costs later
Accommodation works. That is the trap. Answer the fifth 'are you sure?' and the worry lifts, for about ten minutes. Then it comes back a little heavier, because something has quietly been confirmed: the danger was real enough to need checking, and your child could not have got through it without you.Researchers who tracked families week by week through treatment found the loop runs in both directions. Accommodating a fear one week nudged the child's anxiety up the next. The stronger pull went the other way, though. A more anxious child drew more accommodation out of their parents. Nobody decides to do this. It accumulates.
A study of 425 mothers and their children tried to separate the damage done by the anxiety from the damage done by the response to it. Somewhere between a fifth and a half of the trouble these children had at school and at home traced back to the accommodating rather than the anxiety itself. The children could see this more clearly than their parents could.
The trial where the children stayed at home
In 2020 a team at Yale published a study that should have changed more minds than it has. They took 124 children aged seven to fourteen, all with a diagnosed anxiety disorder, and split them into two groups.One group got the standard treatment: twelve weekly sessions of cognitive behavioural therapy, child and therapist, parents left out of it. The other group got twelve sessions of a programme called SPACE, short for Supportive Parenting for Anxious Childhood Emotions. In that group the child never met a therapist. Only the parents turned up.
The parent-only group did at least as well. Around 88 per cent of the children whose parents were treated were rated as improved, against 76 per cent of those who had the therapy themselves. Roughly six in ten in each group no longer met the criteria for the disorder they had arrived with. The difference between the two treatments was not statistically meaningful, which is exactly what the trial was built to test.
One measure did separate them. Accommodation fell further in the parent group, which you would hope, given that bringing it down was the entire intervention.
This is a single trial, and the wider evidence is thinner than that headline makes it sound. A review of parent-focused anxiety programmes published in Behaviour Research and Therapy in early 2025 found real promise and then said plainly that the quality of evidence across the field is low. One pattern held up. Every approach that worked involved changing what parents actually did. The ones that only changed what parents understood did not get the same results.
What goes in place of the reassurance
The obvious misreading is 'stop being soft'. That is not what any of this says, and getting it wrong makes things worse. Pull the help away while telling a child there is nothing to be frightened of and you have added loneliness to the fear.Eli Lebowitz, the Yale psychologist behind the programme, reduces the alternative to two ingredients: acceptance and confidence. Acceptance means saying the fear out loud. I know you're frightened of sleeping on your own, I know it feels horrible. Confidence is the second half. And I know you can do it.
Most of us manage one or the other. 'There's nothing to worry about' is confidence with no acceptance. 'Fine, I'll stay' is acceptance with no confidence. Neither gives a child anything to stand on.
The practical version is smaller than people expect. Choose one accommodation, not all of them. Tell your child in advance what you are changing and why, in daylight, not in the middle of a bad night. Then hold it, and say the two sentences as many times as it takes.
Expect it to get louder before it gets quieter. A child used to five reassurances will ask for the sixth more insistently before they stop asking. That is not a sign you have made a mistake.
The harder half
Very little of this is actually about the child. The difficult part of dropping an accommodation is not their distress. It is sitting inside your own while they have theirs, and not reaching for the fix.None of which makes accommodation a parenting failure. It is what love does when it is handed a distressed child and no instructions. Every parent in that Yale study was doing it, including the ones whose children got better.
Somewhere in the next few weeks there will be a party your child does not want to go to, or a question that needs answering for the fourth time. The question underneath it is not really whether they can handle it. It is whether we can bear to watch them find out.
Sources & Further Reading