Case study: breaking an OCD checking ritual with a paradoxical prescription

The case below is a real, previously published one, summarised from O’Connor’s (1983) account in Family Process. It is not a patient of this author. It repays reading in full for anyone treating obsessional rituals in children, and it appears here because the mechanism recurs constantly in reassurance-driven presentations that look nothing like it on the surface.

The presenting problem

Michael was ten when his paediatrician referred him for a preoccupation with vomiting. The word alone could flood him with panic, followed by stomach ache, headache, cold sweat and one to three hours of incapacity, most often at school.

A precise ritual had formed around the fear. Each morning he sought reassurance from both parents that he would not vomit that day, and neither parent was permitted to use the word while reassuring him. His father called from work partway through the morning to reassure him again. Michael said that if his father told him he would not vomit, he felt sure he would not.

School refusal had become common. His parents had learned not to go out together for more than two hours.

Why reassurance had become the problem

By referral, the household ran on managing the fear. Mother, father, the school nurse and a babysitter were all supplying reassurance in one form or another, and none of it was reducing the fear for long.

Mapped as an attempted solution, this is a textbook case of a family’s care becoming the machinery holding a child’s distress up. Each reassurance relieved him briefly, then required a slightly more elaborate version the following morning.

Two directives on different parts of the loop

The first was a symptom prescription aimed at the fear. Michael was asked to sit in a kitchen chair for one hour a day and deliberately try to make himself vomit, and to defer any vomiting-related worry arising at other times to that scheduled hour.

As is usual with this kind of prescription, he could not produce the symptom on demand. O’Connor’s point is that the attempt mattered more than the failure. A worry that had always arrived uninvited was now something he was being asked to summon, which changes what it is.

The second targeted the ritual. Michael’s father gave up a button from his sweater, and Michael was told that while he kept it, his father’s strength and reassurance travelled with him. His father was no longer to reassure him verbally in the mornings, but could ask whether he still had the button.

O’Connor’s own retrospective analysis reads the shift through Madanes’ model: the meaning of the symptom moved from evidence of parental overinvolvement toward evidence of a ten-year-old developing age-appropriate independence. That is the clinician’s later formulation of what happened, not the explanation given to the family at the time.

Two years later

A follow-up call found Michael doing well academically, with a wider circle of friends he stayed overnight with, and no longer raising the old fears.

Asked what she thought had helped, his mother named family support and time, and said it took time for him to grow up. She did not credit the technique.

The paper reports this without smoothing it over, which is the right decision and a useful discipline. A clinician’s formulation and a family’s account of what helped do not always agree, even when the outcome is good.

What the case generalises to

The specific fear matters less than the structure. A scripted request for reassurance, a relief that never quite lasted, two parents positioned as the source of it.

That structure is what a strategic formulation targets, whatever content the obsession has attached itself to, which is why one case built around an unusual-sounding fear transfers so directly to presentations that share nothing with it but the loop.