Supporting someone with anxiety: why presence works and rescue doesn't
- Matthew Hallam

- Apr 23, 2025
- 6 min read
Updated: Apr 27

There is a moment, when someone you love is anxious, that almost everyone knows. Their breathing changes. They start to look for an exit. The thing they were going to do tonight feels suddenly impossible. And in that moment, if you are the person standing next to them, your nervous system does what it is built to do. It moves to fix it. To soothe. To reassure. To make the moment smaller so that they can manage.
Most of the time, this is the right instinct. It is the same instinct that settles a crying child or steadies a friend at a funeral. The trouble is that anxiety is not, strictly speaking, a problem to be soothed away. It is a system trying to learn that something it has flagged as dangerous is, in fact, survivable. And the kind of help that makes the moment smaller, repeated often enough, can quietly stop that learning from happening.
What follows is a different way to think about supporting someone with anxiety. The shift is small, but it matters. You cannot make someone feel safe inside their own nervous system. Your nervous system is not theirs. What you can do is stay steady while they learn safety in theirs, and sometimes that means letting the moment feel, to both of you, like failing.
Researchers have a word for the kind of help that families, partners and friends offer in response to a loved one's anxiety. They call it family accommodation. Despite the name, it is not only parents who do it. Anyone who shares a life with an anxious person tends to do it. The term refers to the changes the supporter makes to their own behaviour, in the moment, to reduce the anxious person's distress.
It can look like a thousand things. Answering the same worried question for the fifth time. Making the phone call so they do not have to. Driving the long way around to avoid the bridge. Cancelling on the dinner because the room will be too crowded. Letting them sleep in your bed when the panic comes. Going into the shop with them so they do not go alone. Most of these gestures are loving. Many of them are exactly what a thoughtful person would do.
The pattern is common. Early research found that around nine in ten close family members do at least some of it. Hermida-Barros and colleagues (2024) looked at this more recently in Neuroscience and Biobehavioral Reviews. They pooled more than one hundred studies and over eight thousand people. Higher levels of accommodation tracked with more severe symptoms over time. Reducing accommodation was linked to large gains in functioning.
It is worth pausing on what this evidence is, and what it is not. It does not say that supporters cause their loved one's anxiety. The arrow does not run that way. What the research suggests is that the most natural, loving response to seeing someone you care about distressed can, when it becomes the pattern, be one of the things that holds the anxiety in place.
It helps to look at what happens in the anxious person's nervous system when they are distressed. Anxiety is the system flagging something as dangerous and pushing for it to be avoided. When the supporter steps in and removes the trigger, the distress drops quickly.
That drop in distress is the trap. The system has just learned, very efficiently, that the threat was real and that escape was the right answer. It has also learned that the way to escape is through you. The next time the trigger appears, the urge to avoid will be a little stronger, and the pull toward your help will be a little stronger, and the chance to find out that the situation was actually survivable is now a little further away.
This pattern is not limited to childhood. It used to be assumed that family accommodation was mostly a parenting issue. The lifespan research has changed that picture. Sperling and colleagues (2025) studied nearly two hundred people in intensive treatment, ranging in age from eight to seventy-six. The study, published in Behavior Therapy, found the same dynamic at work in both children and adults. The accommodating relative could be a parent, a partner, an adult child, a sibling, or a friend. The mechanism was the same.
The clinical implication is straightforward, and is the heart of why this post exists. Helping someone feel less anxious in the moment is not the same as helping them with their anxiety. Sometimes those two things line up. Often they do not.
If less helping in the moment is part of the answer, the obvious follow-up is what the supporter is meant to do instead. The most studied answer comes from a programme called SPACE. It was developed by Eli Lebowitz at the Yale Child Study Center. SPACE stands for Supportive Parenting for Anxious Childhood Emotions, but the framework translates to adult relationships.
The 2020 trial by Lebowitz and colleagues was published in the Journal of the American Academy of Child and Adolescent Psychiatry. They randomised one hundred and twenty-four anxious children into two groups. One group did individual cognitive behavioural therapy. The other did SPACE, in which only the parents went to sessions. The child had no direct contact with a therapist at all. The two approaches did about as well as each other. The children whose parents shifted how they responded improved at a similar rate to the children in therapy themselves.
Berger, Silverman and Lebowitz (2024) ran a more recent trial of an adapted version of the same programme, this one for parents of adult children who had become highly dependent. The supporter-focused approach showed promising signal on the same outcomes. The point worth taking from this is not that adults need their parents to fix things. It is that supporting differently, regardless of the relationship, tends to be more useful than supporting harder.
What SPACE actually teaches supporters has two pillars. The first is more support, expressed in a particular way. It means acknowledging the reality of the distress and showing confidence in the person's ability to cope. The second is less accommodation, reduced gradually. The person is told ahead of time what the supporter will and will not do.
What follows is a structure for those moments where you can feel the pull to step in and rescue, and you would like to respond differently. It is not a script. The actual words will be yours. It is a way of giving yourself four moves to make, in roughly this order, when the urge to fix the moment is strongest.
Take a worked example. A partner with social anxiety has agreed to come to a friend's gathering. As the time approaches, they begin to spiral. The old pattern would be to call ahead, make excuses for them, and stay home. The new pattern looks like this.
The fourth move is the one supporters find hardest, and it is the one the research keeps pointing to. Letting them face the feared situation, while you stay nearby and steady, is what gives the system a chance to learn. Stepping in to remove the situation is, kindly, what keeps the system stuck where it is.
Two practical notes about doing this in real life. First, telling the person what you are doing helps. "I love you, and I am not going to make the call for you anymore. I will sit with you while you make it." Surprise withdrawal of accommodation can feel like punishment. Planned withdrawal, with warmth, does not. Second, the change is gradual. You are choosing one or two patterns to shift first, not dismantling the whole scaffolding at once.
There is a thing that often happens when supporters read material like this for the first time. They begin to feel, sometimes acutely, that all the help they have given over the years has been wrong. That if only they had known, the person they love would not be where they are now. This response is understandable. It is also not what the evidence supports.
Accommodation is not a moral failing. It is what loving people do when they watch someone they love suffer. Anthes and Dreisoerner (2026), in a review of self-compassion research published in Mindfulness, note that harsh self-criticism switches on the body's threat system. It activates the same physiology that real danger does. A supporter who responds to learning about accommodation by judging themselves harshly is now in exactly the state that makes thoughtful, steady responses harder.
Two things often help here. The first is acknowledging that supporting someone with anxiety is genuinely tiring, and that the supporter has their own nervous system that needs looking after. The second is using the support that already exists. In Australia, family members and carers may be eligible for up to two Medicare-rebated sessions per calendar year as part of the anxious person's mental health treatment plan, under the Better Access initiative. The Carer Gateway (carergateway.gov.au) offers practical, peer and counselling support for anyone in a caring role.
If there is one message worth carrying out of this, it is the one that gives the post its title. Presence works. Rescue does not. Staying calm and warm and close while the person you love faces something hard is one of the most useful things a supporter can do, even, and especially, when in the moment it feels like you are doing nothing at all.
.png)