A Small Case Series, and the Limits of What It Can Claim

The design, and what it permits

A case series is a written description of a small number of patients treated at a single clinic, published because something about them resembled each other. There is no control group, no randomisation and no statistical test. The patients arrive by whatever route brings people to that clinic, rather than by sampling from a population. Such a report can propose a hypothesis; testing one is beyond it.

The constraint bites hardest when the resemblance is strong. A similarity among three patients at one hospital may be a property of the disorder or a property of the hospital, and referral routes decide who ever appears. A department that receives adolescents with severe behavioural problems will meet many who were hyperactive as small children, and will never meet the hyperactive children who went on to no behavioural trouble. Nothing in the design separates those two accounts.

Sayanti Ghosh and Mausumi Sinha, psychiatrists at R.G. Kar Medical College in Kolkata, published three such descriptions in 2012 in the journal Case Reports in Psychiatry. The boys were 13, 14 and 16. All had been diagnosed with conduct disorder, the term for repeated violation of rules and of other people’s rights, by experienced psychiatrists using a structured clinical interview against DSM-IV-TR criteria. All three scored in the normal range of intelligence on the WISC-IV, between 90 and 95.

Each had been hyperactive in early childhood, from the age of three in two of the boys and from six in the third, with inattention emerging in the first school years. Defiance followed: temper tantrums, rudeness, refusal to comply with rules at home and at school. That intermediate stage is oppositional defiant disorder, which describes persistent defiance toward adults without the stealing, truancy and aggression that define the later diagnosis. In all three the conduct problems arrived last, in adolescence, and had worsened sharply in recent months.

Two of the boys had been diagnosed with ADHD, combined type, in childhood. One improved partially with behaviour therapy and parental counselling before treatment was stopped two years later; the other received methylphenidate and behaviour therapy at about nine, abruptly discontinued after six months. The third had never been treated. All three households were under considerable pressure, from parental alcohol use, marital conflict, parental depressive and bipolar illness, separation and a hostile stepfather.

Ghosh and Sinha are explicit that larger samples and more detailed study are needed. The domains of overlap they emphasise are aggression, hostility and emotionality, and setting the cases beside published twin and family research, they raise the possibility of a common psychopathological spectrum: one underlying liability, genetic or environmental or both, appearing at three ages under three names.

What this means for you and your child

The idea is old and the field has not settled it. The report’s own discussion lays out the disagreement plainly. Some researchers hold oppositional defiant disorder to be a discrete condition with a good outcome that follows a course of its own, while others treat it as a milder version of conduct disorder, with only a proportion of children moving from one to the other. Three boys at one hospital do not adjudicate between those positions, and the paper does not claim they do.

The firmer evidence in the report comes from the cohort studies it cites rather than from the cases. In one of them, oppositional defiant disorder in children with ADHD predicted conduct disorder in adolescence independent of how severe the ADHD was, with the risk roughly three times higher. A tripled risk is a statement about the odds facing a group and not a forecast for any child inside it.

Each of these three households was difficult in ways a clinician would record, and the report records them. Weighing them is another matter. With three patients and no comparison, there is no way to determine whether the progression the authors describe belongs to the disorders, to the family circumstances, or to the particular route that brought these boys to a psychiatric department in the first place.

Two of the three began ADHD treatment in childhood and stopped it early. Reading the later conduct disorder back onto that discontinuation is exactly the inference this design forbids, and the third boy, who received no treatment at any point, arrived at the same diagnosis regardless. Whether early and sustained treatment changes the trajectory is a real question, and it is one that only a controlled study can answer.

A case series earns its place by naming a possibility clearly enough that a larger study can go and look for it. It has no authority past that point, and nothing here establishes that hyperactivity in a three-year-old is the opening move of a sequence ending in adolescence. When a report of three boys is offered to you as evidence of what your child will become, the first question is how many boys there were.