The instruction to raise independent, mentally strong children almost always lands on the child. Teach them grit. Let them fail. Do not swoop. The work is theirs, and the parent’s job is mostly to stand back and allow it.
A strong test of a parent-focused approach points somewhere else. It examines a set of parental habits so ordinary that many parents would not recognize them as a possible treatment target.
A trial in which the children never attended
Eli Lebowitz and colleagues at the Yale Child Study Center ran a randomized trial, published in the Journal of the American Academy of Child and Adolescent Psychiatry, with 124 children aged 7 to 14 who all had a primary anxiety disorder. Each child was randomly assigned to one of two treatments.
The Artful Age
A weekly letter on aging well, family across generations, and the creative life after the kids leave home.
The first was cognitive behavioral therapy, the standard of care: 12 weekly hour-long sessions between the child and a therapist, with no treatment for the parents. The second was a program called SPACE, which stands for Supportive Parenting for Anxious Childhood Emotions. Also 12 weekly hour-long sessions, with one difference. The child never came. The parents did.
SPACE was noninferior to CBT on the primary and secondary anxiety outcomes, and on ratings from independent evaluators, from the parents, and from the children themselves. Noninferior is a specific and modest word. It does not mean SPACE beat therapy, and it does not mean the two were identical. It means that treating only the parents did not come out meaningfully worse than treating the child, which was not the expected result. Seventy-eight percent of families completed every session and assessment, and dropout did not differ between the two groups.
What the parents were actually asked to change
SPACE targets something called family accommodation. The researchers define it as the many changes parents make to their own behavior and routines in order to help a child avoid or reduce anxiety.
Their examples are recognizable. Sleeping next to a child with separation anxiety. Speaking on behalf of a child with social anxiety. Answering the same reassurance-seeking question for the fifth time that evening. None of this is neglectful or lazy. Accommodation is, as the paper puts it, well-intentioned by definition. It is what love does when a child is frightened. It is also associated with more severe anxiety and greater impairment, and it can predict a poorer response to therapy.
What matters most here is what parents were taught to do in its place, because it is the part that gets lost whenever this research is summarized. They were not taught to withdraw, to toughen the child up, or to let the child sink. They were taught supportive responses that acknowledge the child’s distress while also conveying confidence in the child’s ability to cope. Both halves, together, every time.
Only then did the practical work begin. Parents mapped out their accommodations in detail, selected one to change, built a plan, and told the child what was going to happen and why. When that one was going well, they chose another. The manual includes whole sections on handling what happens when a child responds to the change with distress or anger, because that is expected rather than a sign it has gone wrong.
Keeping the whole thing in proportion
It would be easy to read all that and conclude a child’s anxiety is the parents’ doing. The wider evidence does not support that, and it is worth saying clearly.
A meta-analysis by Bryce McLeod, Jeffrey Wood and John Weisz pooled 47 studies in Clinical Psychology Review. Parenting, taken as a whole, accounted for about 4 percent of the variance in childhood anxiety. Warmth accounted for less than 1 percent, which should retire the idea that anxious children are made by insufficiently affectionate parents. The authors’ own conclusion is that explaining where children’s anxiety comes from will require looking at factors other than parenting.
One dimension stood out from the rest. Autonomy granting, meaning how much room a child is given to do things for themselves, accounted for about 18 percent of the variance, more than three times any other parenting dimension in the analysis. That is the dimension the original instruction was pointing at, and the meta-analysis suggests the instinct behind it is sound even if the rest of the parenting advice around it is doing less work than people think.
What the evidence does not license
Several things, and they matter more than usual here.
The trial had no placebo arm, which is normal for this design but means both groups received a real treatment. Parental involvement in the CBT arm was deliberately restricted so the two approaches could be told apart, and that is not how CBT is usually delivered, so the comparison is cleaner than clinical practice. The sample was 83 percent white and mostly of medium to high socioeconomic status, and the authors say plainly that whether the findings generalize beyond that remains to be established. The meta-analysis is built from cross-sectional studies, so it cannot say whether less autonomy makes children anxious or anxious children are given less room.
The larger caution is about what kind of thing this was. SPACE is a manualized treatment, delivered across twelve supervised sessions by trained therapists, to children with diagnosed anxiety disorders. It is not an instruction to stop comforting a worried child, and a parent who reads it that way has the finding backwards.
What is left
The useful version of “raise them to be independent” turns out not to be a stance toward the child at all. It is a question a parent can ask about their own week: which of the things I am doing were built around a fear, and are they still necessary?
The trial showed that the full, therapist-delivered SPACE program could reduce child anxiety without direct child-therapist contact. It did not isolate the confidence message, reduced accommodation, or any other single component as the mechanism. The authors explicitly called for mechanism research. Family accommodation fell more after SPACE than after CBT, which supports, but does not prove, the proposed pathway.