Testing Psychotropic Drugs in Children
April 30, 2002
David Healy, MD
University of Wales College of Medicine
April 30th 2002.
The background against which this paper is prepared is that in 1997 I organized, chaired and wrote up the recommendations from an International Roundtable Meeting for the British Association for Psychopharmacology on the use of psychotropic drugs in children. This meeting involving senior regulators from the United States and Europe as well as professors of child psychiatry from a number of European countries, North America and the UK. My concern in promoting this meeting was to ensure that children who could benefit from psychotropic drug treatment would be enabled to gain access to treatment. Only six years ago the climate in the United Kingdom was such that children were at a real risk of not getting effective drug treatment for nervous problems.
The reason for this failure to use psychotropic drugs to treat some children that might have benefited had nothing however to do with a lack of studies on these drugs in children. The proceedings of the meeting make it clear that there is in principle no need for any drug studies in children for either anti-psychotics, antidepressants or for treatments for OCD for example (1). Research conducted in children or adolescents with such conditions will in fact only produce a situation in which companies gain a license to vigorously promote their treatments for these conditions.
Treating children for conditions also present in adults proceeds normally on the basis of studies undertaken with adults. There are only two things that clinicians could conceivably learn from studies undertaken in children of drugs already licensed for adults. First, they might learn that paradoxically a treatment, which works in adults doesn’t work in children. Second, they might learn that there are particular toxicities in children that need to be factored in to any risk benefit assessment as regards treatment in children.
All of the major pharmaceutical companies were invited to send representatives to the meeting and many did. Nowhere in the course of the meeting did it become clear that any of the major companies had already undertaken or were then conducting studies on children. In return for this right to create the conditions in which children who may well not need the treatments but who are much more likely to end up on drug treatment as a result of pharmaceutical companies being licensed to promote their treatments to children, the very least that clinicians and the public could expect from market authorisation holders would be to make available critical safety data that arise from such studies. Few if any of the non-company people at that 1997 meeting will have thought that results indicating that the toxicity of SSRI antidepressants is indeed greater in children were in fact at the time of that meeting being actively suppressed.
Pfizer for example had completed by 1996 a series of studies in children who had OCD and depression. These formed the basis for Pfizer’s application for a license to market sertraline for OCD in children in the United States and in many European countries.
Pfizer enrolled 248 children into these studies, 187 in an OCD arm of the studies and 61 in an allied mixed depression/OCD arm.
The scientific literature in which these studies were reported contain only one reference to one suicidal act on sertraline. However Pfizer’s expert report, submitted to the FDA in response to FDA questioning about rates of suicidal acts in these trials, makes it clear that there were in fact at least six children who made suicidal acts while on sertraline and more who became suicidal (2).
Pfizer go to great efforts to explain away these six suicidal acts. First they claim that four of these suicidal acts occurred in the 44 children who were apparently depressed. This would give a 1 in 11 rate of suicidal acts on sertraline in children who were depressed, a 20-fold higher rate of suicidal acts than appear in the sertraline-treated adult literature. This rate remains unavailable to investigators who are at present actively recruiting children to studies with SSRIs for other SSRI companies.
Pfizer claim suicidal acts are common in children who are depressed anyway. They are not this common. But if suicidal acts were this common in depressed teenagers, a conundrum arises. One of the justifications that Pfizer offer for treatment with sertraline is that treatment will reduce suicide rates – there is no evidence for this claim from Pfizer’s studies of sertraline in either children or adults. If, however, there were any cases of suicidal acts averted by treatment with sertraline in these trials in children, then given the figures for suicidal acts that come out of these trials, there must logically have been an even higher rate of suicide provocation that is initially apparent from the data.
The data from these studies also reveal a dose-dependent production of agitation in teenage and preteen children. This dose-dependent agitation, amounting to suicidality in some instances, was not an accidental by-product of the studies but was in fact designed into the study. Doses were pushed up to find out if they would become intolerable – despite evidence published 10 years earlier from adults that sertraline induces dose dependent agitation and the dose to which sertraline was pushed in these children could have reliably been expected to produce agitation.
Furthermore it is clear that in their adult studies Pfizer have been prepared to go to extraordinary lengths to conceal evidence of suicidality. They have mislabeled as placebo suicidal acts, acts that did not occur on placebo. Against this backdrop there is every reason to believe that the true picture revealed in these studies is even worse than a reading of Pfizer’s expert’s report would indicate.
At much the same time Lilly began studies on Zyprexa in children. There are parallel problems to Pfizer’s studies. The studies in adults with Zyprexa that Lilly submitted to the FDA demonstrate, as far as I can establish, a higher death rate on Zyprexa than on any other anti-psychotic ever recorded. In addition to this Lilly have suppressed data on suicidal acts on Zyprexa from these trials. The data are not available in the scientific literature, nor from FOI requests to the FDA, nor from enquiries to the company.
Despite this Lilly are engaged in trials with this agent in children. These trials will underpin vigorous company promotion for this drug in a wide swathe of children with hyperactivity, neurosis, bipolar and psychotic disorders – they won’t enable clinicians to use the drug if needed.
1 Healy D, Nutt D (1997). Consensus Statement on Childhood and Learning Disabilities Psychopharmacology. J Psychopharmacology 11, 291-294
2 Pfixer Expert Report (1997). Sertraline hydrochloride for obsessive compulsive disorder in paediatric patients. Approved Oct 20th 1997.