top government scientist who concluded
last year that most antidepressants are too dangerous for children
because of a suicide risk wrote in a memo this week that a new study
confirms his findings.
The official, Dr. Andrew D. Mosholder, a senior epidemiologist at
the Food and Drug Administration who assesses the safety of
medicines, found last year that 22 studies showed that children
given antidepressants were nearly twice as likely to become suicidal
as those given placebos.
His bosses, however, strongly disagreed with his findings, kept
his recommendations secret and initiated a new analysis.
In his memo, dated Monday, Dr. Mosholder said that the results of
the new analysis, undertaken in part at Columbia University, matched
his own. Though the two studies used different methods and different
numbers, they came to similar conclusions, Dr. Mosholder wrote in
the internal memo. A copy of the memo was made available to The New
In the new analysis, Paxil, which is manufactured by
GlaxoSmithKline, and Effexor, made by Wyeth, have been found to be
even more likely to lead children to become suicidal than Dr.
Mosholder's original analysis found, his memo says.
The findings add to the debate over whether the government should
ban prescribing the pills to children. Dr. Graham Emslie, a
researcher whose studies of the drugs in depressed children have
been paid for by both drug makers and the National Institutes of
Health, said he still thinks the benefits of the medicines outweigh
"Limiting doctors' choices in treating depressed kids is not a
good thing," Dr. Emslie said.
Officials at the Food and Drug Administration have struggled to
explain why it has acted so differently from British health
authorities, who last year banned the use of all antidepressants but
Prozac in children.
The F.D.A. is scheduled to hold an advisory committee hearing on
the issue next month. According to people inside and outside the
agency, the F.D.A. may next week make public the results of the
The controversy had its start two years ago when Dr. Mosholder
was reviewing data submitted by GlaxoSmithKline regarding studies of
Paxil in children. Dr. Mosholder noticed that children given the
pill suffered more emotional "lability" or vulnerability, than those
given placebos. He asked the company for more specifics about what
it meant by "lability."
In May 2003, the company submitted a new report showing that
children given Paxil were more likely to become suicidal than those
given placebos. Also, the drug did not improve their depression any
better than the placebo.
Dr. Mosholder asked for similar data from other drug companies.
By last fall, he was looking at the results of 22 studies involving
4,250 children. His analysis of the combined results suggested that
children given the drugs were 1.89 times more likely to become
suicidal than those given placebos. He recommended that the agency
ban doctors from prescribing all but Prozac to children, the only
pill that had proven beneficial against childhood depression.
His bosses, however, suppressed his report and hired researchers
at Columbia to re-analyze the underlying data that Dr. Mosholder had
used, saying that some events labeled by drug-company researchers as
suicidal did not seem worrisome.
Though the original studies had identified just 108
suicidal-related adverse events, the Columbia researchers expanded
their inquiry to include about 400 adverse events, many of which had
been originally labeled as "accidental."
The risk of a suicidal event among those given antidepressants in
the trials was 1.78, only slightly less than the risk Dr. Mosholder
A spokesman for the F.D.A. did not return phone messages. A
spokesman for Wyeth said that Effexor is not approved for use in
children. A spokeswoman for GlaxoSmithKline declined to comment.
Senator Charles E. Grassley of Iowa, who has been pushing an
investigation into the F.D.A.'s handling of the controversy, said
through a spokeswoman that the new memo from Dr. Mosholder
"underscores what my committee investigation is finding as far as
the strength of Dr. Mosholder's original analysis about
antidepressants and kids."