Efficacy and safety of antidepressants for children and adolescents
BMJ 2004; 328 doi: https://doi.org/10.1136/bmj.328.7444.879 (Published 08 April 2004) Cite this as: BMJ 2004;328:879All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
A correspondent has submitted an inquiry about my relationship to the
non-profit organization, the Law Project for Psychiatric Rights, located
in Anchorage, AK.
While I do not think that my participation as a consultant
for the Law Project for Psychiatric Rights meets either the technical
requirements, nor spirit, of disclosure, I am happy to discuss the
relationship.
In my April 2004 posting, I revealed that I am a "contract
psychiatrist." This is exactly my participation with the Law Project for
Psychiatric Rights, where I had been involved in two legal cases at the
time of my Rapid Response to the Jureidini article.
I do not see how the Law Project would gain or lose financially from
the results of my letter, although I suppose someone might decide to
retain me as a consultant, as a result. Any financial gain or loss by the
Law Project for Psychiatric Rights would seem to be a very
tenuous connection, at best.
With regards to non-financial conflicts, which
were apparently also an issue for the same querying
correspondent --- I am neither the member of any political organization,
nor any religious organization, whose interests may be affected by the
publication of my letter. I am not a member of the Church of Scientology,
nor am I am affiliated with the Citizens Commission for Human Rights.
While I do hold deep convictions, it is to academic
and scientific integrity; to the protection of fundamental human rights
and liberties; and to the revelation of distortions arising from concealed
trial results, fallacious research methodologies, and file drawer effects
-- all of which are increasingly and appropriately drawing the attention
of many physicians in the USA and abroad.
Respectfully,
Grace E. Jackson, MD
Competing interests:
I have served as an expert witness for psychiatric consumers facing involuntary medication
in situations where they posed no imminent danger to self or others.
I have been an invited lecturer of the British Psychological Society, the University of Central England, and the Overload Network.
Competing interests: No competing interests
Cathyrn Clary, a Pfizer Inc. employee, in a rapid response letter to
the bmj [1] chides Jureidini for his criticism of the Pfizer-sponsored
trial of sertaline for major depression in children/adolescents, which was
originally reported by Wagner in JAMA [2].
Regarding the efficacy of sertaline, Clary criticises Jureidini as
follows-:"Dr. Jureidini and his co-authors do not specifically summarize
the efficacy results in the JAMA-sertraline article, except to note that
they “doubt Wagner’s claim, and Varley’s editorial support for that claim,
that their results are clinically, as well as statistically, significant.”
Clary claims that the Wagner study's results are statistically and
clinically significant. In particular, Clary states that the Wagner study
demonstrated that sertaline produced a 69% response rate (in the primary
efficacy endpoint) and that it was significantly better than placebo,
which had a 59% response rate. That absolute risk difference (ARR) is
equal to an absolute benefit of 10%, which Clary initially acknowledges is
modest. However, at a later stage in her rapid response letter, she argues
that the efficacy result is clinically significant.
This is Clary's argument-:
"Thus, a response rate of 69% on sertraline (despite only a 10%
advantage over placebo) makes it understandable why physicians in the past
decade might have made the clinical judgment that the benefit-risk ratio
was tipped in favor of SSRI treatment, even though regulatory approval is
lacking for most SSRIs (including sertraline) in childhood depression.
When one puts the issue of childhood depression in context, then we
feel less able than the authors are to dismiss a 69% response rate on
sertraline, even if it is only 10% superior to placebo. Cumulative rates
of major depression and suicide in childhood and adolescence are
staggering. In Western Europe alone, an estimated 10.1 million children
will develop major depression by age 18, and more than 50,000 will have
committed suicide (Birmaher et al, 1996; Olfson et al, 2003; US Census
Bureau).
While a 10% advantage over placebo may seem trivial to the authors,
the sheer magnitude of the suffering makes us wonder about their
unwillingness to “lower the threshold for accepting a new intervention.”
Even 10% improvement in response to this devastating illness, while
small in absolute terms, translates into substantial relief of suffering
for hundreds of thousands of children and potentially saving the lives of
some patients."
Before I discuss whether the 10% efficacy result is clinically
significant, consider whether one can be confident that the 10% efficacy
result is the "true value". Sackett, the guru of EBM, stated that
confidence in a RCT's results is proportional to the RCT's signal/noise
ratio multiplied by the square root of the sample size [3]. The Wagner
study consisted of only 189 sertaline treated children. That is a small
study when compared to cardiovascular RCTs that routinely enroll thousands
of patients. Wagner reported [2] that-: "Based on a 40% decrease in the
adjusted CDRS-R total score at study end point, 69% of sertraline-treated
patients compared with 59% of placebo patients were considered responders
(P = .05)". Is that result statistically significant and can one be
confident in the internal/external validity of the study's results?
Mathews questioned the statistical validity of the Wagner study's
result in a letter to JAMA [4]. Mathew stated-:"The authors used P values
to assess statistical significance. While this is at times useful, it is
an arbitary value and the consensus now is that it does not convey much.
The use of confidence intervals (CIs) would have conveyed more information
about the clinical utility of the results. Based on the data in the
article, we calculated that sertraline produced an absolute risk reduction
for depression of 10% (95% CI, 0.15%-19.8%); the corresponding number
needed to treat of 10 has a 95% CI of 5 to 540." In other words, Mathews
is pointing out that the Wagner study was so small in size that one cannot
be certain that the 10% absolute risk reduction is the "true value" and
the ARR could be anywhere between 0.15-19.8%. I think that it would be
difficult to cogently argue that an absolute risk reduction in the range
of 0.15-10% (lower half of the estimated ARR range) is either
statistically or clinically significant.
I am particularly bothered by the fact that the Wagner study mainly
recruited patients with moderate depression, and one cannot therefore be
sure that the Wagner study's marginal efficacy results are applicable to
children with severe depression. I personally think that it is difficult
to accurately determine whether a drug is truly efficacious if the tested
population of "sick" patients has a placebo response rate of 59%. I think
that a high placebo response rate probably reflects a sample population of
"less sick" MDD patients and I believe that one cannot automatically
extrapolate the Wagner study's results to severely depressed children in
clinical practice.
Both Wagner and Clary argue that the 10% ARR is not only
statistically significant; they also argue that a 10% efficacy result is
clinically significant.
Spielman in a letter to JAMA [5] questions the clinical significance
of a 10% ARR. He states-: "Dr Wagner and colleagues stated that the
"significance of the results is clinically as well as statistically
relevant," an assertion that reaches well beyond the trial's results. The
average improvement in the main outcome measure—the CDRS-R—was 22.84
points for the sertraline group vs 20.19 points for the placebo group. The
authors did not report the standard deviation, however; thus, an effect
size cannot be calculated. However, simple division shows that the gains
made by the placebo group were 88.4% of those in the active medication
group, leaving the active medication superior by less than a 12% margin
compared with an inert placebo. The slight difference between active and
inactive treatment in the main outcome measure, while statistically
significant, is of questionable clinical relevance. Statistically
significant differences were also found on the Clinical Global Impression
(CGI) Improvement and Severity of Illness scales. While these scales are
popular in clinical trial research, there is scant evidence about their
validity. On the remaining 3 outcome measures, no statistically
significant differences were observed."
I have never seen a "gold standard" definition of clinical
significance, but I find it difficult to understand how one can justify a
10% ARR as being clinically significant if the single RCT demonstrating a
10% efficacy result has a low signal/noise ratio (small signal of 10%, and
high noise level due to the recruitment of patients with high placebo
response rates which could theoretically vary from trial-to-trial, a
subjective and questionably valid primary endpoint, and a questionable ITT
analysis) AND a small sample size (189 treated patients), AND if one
cannot be certain that more severely depressed children in clinical
practice will respond as well as the moderately depressed children in the
Wagner study (questionable external validity).
Clary states that the cumulative suicide rate in childhood depression
is staggering and she quotes an expected figure of 50,000 suicides among
10 million depressed children under the age of 18 in the USA. She
therefore implies that sertaline would be of particular value in severely
depressed children, who presumably are more likely to commit suicide.
However, we have no guarantee that sertaline would have a 10% efficacy
result in preventing suicide in severely depressed children. In fact, a
number of people believe that sertaline may actually increase the suicide
rate. Clary states that there was no difference in suicide rate between
sertaline and placebo patients in the Wagner study. However, the Wagner
study only consisted of a few hundred patients and was of limited duration
(10 weeks). Clary also states that a trial of sertaline in OCD patients
demonstrated that sertaline was not associated with an increased suicide
risk.
However, and by contrast, the Committee on Safety of Medicines [6]
stated that sertaline is associated with a higher suicide risk. In
particular, the Committee stated-:
"Only one placebo-controlled trial has been performed with sertraline
in children and adolescents with obsessive compulsive disorder (OCD), who
were treated for 12 weeks. There were no reports of suicide, suicidal
thoughts, or self-harm in patients treated with sertraline in this trial,
and one report in a placebo-treated patient. There were, however, two
reports of suicidal thoughts and one report of self harm in children and
adolescents treated with sertraline for OCD in a 52-week open-label trial
(giving a rate of 1.4 % (3/208)).
In all controlled trials in MDD and OCD the rate of suicidal thoughts
and self-harm in the sertraline group is 1.8% (5/281), the rate in the
placebo group is 1.1% (3/279).
In all sertraline-treated patients the rate of suicidal thoughts and
self-harm in the sertraline group is 3.7% (23/616). This estimate takes
into account the fact that in the extension trials only those patients who
switched from placebo to sertraline were new to sertraline. Patients are
only counted once in the denominator regardless of how many trials they
may have participated in.
In sertraline-treated patients with MDD the rate of suicidal thoughts
and self-harm in sertraline-treated patients is 4.6% (20/435). Extension
trials handled as above.
In sertraline-treated patients with OCD the rate of suicidal thoughts
and self-harm in sertraline-treated patients is 1.4% (3/208). Extension
trials handled as above.
The data show a consistently higher incidence of suicidal thoughts
and self-harm in children and adolescents with depression treated with
sertraline, which is in the region of twice the apparent placebo rate.
Data on the risk of suicidal behaviour in OCD trials do not appear to
indicate any unusual risk."
I have personally never studied the relevant suicide literature, so I
cannot personally judge whether sertaline actually increases the
likelihood of suicide in severely depressed children. However, if
sertaline actually increases the suicide rate, this would be a matter of
great concern.
Finally, should one prescribe sertaline because 69% of moderately
depressed children may respond to the drug? Clary argues that one should
prescribe the drug because physicians are not likely to prescribe a
placebo agent. Clary acknowledges that 85% (59% divided by 69%) of the
drug's antidepressant effect is due to a high "expectancy effect" in a
highly suggestible group of patients. Then, I presume that Clary would
agree that Pfizer should inform prospective pediatric MDD patients upfront
that 85% of sertaline's antidepressant action is due to an "expectancy
effect", and I think that this specific information should be included in
the informational package insert document supplied with the drug, and in
Pfizer's sertaline advertisments. I also think that many people may agree
that Pfizer is ethically obliged to inform pediatric MDD patients that
alternative psychiatric treatments may offer an equivalent "expectancy
effect" (equal to 85% of sertaline's total therapeutic effect), and that
use of those alternative treatments may avoid some of the adverse events
associated with the use of sertaline.
References:
1. Clary C. Sertaline Pediatric Depression Trial Met Highest
Standards of Study Design and Reporting. bmj rapid response letter. April
27 2004.
http://bmj.bmjjournals.com/cgi/eletters/328/7444/879#57741.
2. Wagner KD, Ambrosini P, Rynn M, Wohlberg C, Yang R, Greenbaum MS,
Childress A, Donnelly C, Deas D. Efficacy of sertraline in the treatment
of children and adolescents with major depressive disorder: two randomized
controlled trials. JAMA 2003;290:1033-1041.
3. Sackett, David L. Why randomized controlled trials fail but
needn't: 2. Failure to employ physiological statistics, or the only
formula a clinician-trialist is ever likely to need (or understand!) CMAJ:
165(9):1226-1237, October 30, 2001.
Available online at http://www.cmaj.ca/cgi/content/full/165/9/1226.
4. Mathews M. Efficacy of Sertaline in the Treatment of Children and
Adolescents With Major Depression. JAMA. 2004;291:40.
5. Spielmans G. Efficacy of Sertaline in the Treatment of Children
and Adolescents With Major Depression. JAMA. 2004;291:41.
6. Committee on Safety of Medicines. Selective Serotonin Reuptake
Inhibitors (SSRIs) - overview of regulatory status and CSM advice relating
to major depressive disorder (MDD) in children and adolescents: Summary of
clinical trials.
Available at
http://medicines.mhra.gov.uk/ourwork/monitorsafequalmed/safety messages/ssrioverviewclintrialdata_101203.htm#sertraline
Competing interests:
None declared
Competing interests: No competing interests
Dear sir:
Notably, there is a converging evidence that depression occurs in all
age groups including children and adolescents (CAs) and for that matter
CAs need proper diagnostic assessment and thereafter appropriate
treatments, which may include either suitable drugs or psychotherapies or
their combination. From symptoms and diagnostic perspectives, depression
in CAs is often characterized by atypicalities such as decreased school
performance, antisocial behaviors, withdrawal from friends, withdrawal
from daily activities, excessive weight gain, excessive weight loss,
aggressive spells, remarkable agitation, somatic symptoms and use of
alcohol/other substances. In addition, recent research conducted among CAs
documents alarmingly that the incidence of suicidal attempts and complete
suicide is staggeringly high in this population and is on the increase.
Notably, DSM-IV-defined depression could be found in children as young as
six years of age.
However, when aforesaid depressive psychopathological symptoms among
CAs are recognized, they should have proper screening and some important
scales are developed for this purpose; Children's Depression Inventory for
7-17 years; The Center for Epidemiological Studies Depression Scale for
Children; Child Depression for 8-12 years; Adolescent Depression Scale for
13-18 years; and The Beck Depression Inventory for individuals 14 and
older. The use of screening tools helps in the precise diagnosis of
depression, guides its severity, and tracks therapeutic changes following
the use of treatments and its outcome. Moreover, precise diagnosis of
depression in CAs has major impact on the treatment outcome.
It is documented that preschoolers (<7 years) also suffer from
depression and DSM-IV criteria for depression in preschoolers may not be
adequate, duration of symptom criteria, i.e, of two weeks may need
modification. The most common symptoms among children of 2-5 years of age
include irritability, angry spells, resentfulness, death/suicidal theme
in play, feeling lonely, and high level of frustration. When children
communicate death wishes, their parents must be interviewed. These
children are reported to have high rate of comorbid disorders and moreover
about 5% of them do have 4 or more of those aforesaid symptoms.
The specific objective of presenting these defining
psychopathological symptoms in CAs and preschoolers is to generate
confidence among primary care physicians, pediatricians and mental health
professionals who relatively lack adequate skills to recognize depression
in CAs and preschoolers. At the same time, reportedly about 45% of
physicians lack confidence in treating depression in this psychiatric
population.
In an interesting study of adolescents (9-13 years), heterotypic
(versus homotypic) trajectory (continuity from one diagnosis to another
diagnosis) was significant to depression to anxiety and from anxiety to
depression and all heterotypic continuity was seen in girls (9 through 16
years).
Now after introducing these clinical facts, I posit one question
which is " Does precisely diagnosed depression in CAs need proper
treatment?" The answer to this question is certainly "yes". If yes, are
then the available drug treatments such as TCAs, SSRIs, mood stabilizers,
and others really effective in CAs? Are these drugs fairly safe and cause
no harmful effects in CAs? Are these drugs less expansive and easily
accessible to all CAs globally? These questions are highly complex and if
the answers to them are "yes", use them. But if the answers to these
questions are "no", use them sparingly and keep on developing better drugs
with better clinical profile to be used in children and adolescents with
depression.
Notably, we are beginning to develop and understand the complexities
of pediatric psychopharmacology, and research efforts should continue in
order to develop better psychotropic drugs for the treatment of depression
in CAs, which is a very devastating mental disorder
Finally, it looks fruitless to enter in the poignant debate related
to the findings of Jureidini et al (2004) because supporters and opponents
to those authors have already expressed their respective concerns.
Reference:
Jon N Jureidini, Christopher J Doecke, Peter R Mansfield, Michelle M
Haby, David B Menkes, and Anne L Tonkin
Efficacy and safety of antidepressants for children and adolescents. BMJ
2004; 328: 879-883
Competing interests:
None declared
Competing interests: No competing interests
Sir:
On March 9, the Committee on Safety of Medicines (CSM) advised
avoiding olanzapine and risperidone in the treatment of patients with
dementia (1), reflecting concern over excess stroke risk.
Anti-psychotics have long been used (off licence) in dementia to
treat neuro-psychiatric disturbance and are highly effective (2).
Atypicals have recently been favoured over older preparations to avoid
anti-cholinergic and Parkinsonian side effects, but a recent study of
elders on typical and atypical anti-psychotics has shown no significant
difference in risk of stroke between pharmacological groups (3). Stroke,
while potentially catastrophic, is a quantifiably rare "excess risk" event
in an "already at risk" group (4), demanding balanced thinking against
real reductions in agitation and aggression.
We are concerned that anti-psychotic therapy will be unnecessarily
withheld, and also that GPs will now adopt widespread usage of
conventional anti-psychotic drugs. The Adults with Incapacity (Scotland)
Act 2000 specifies that informed proxy decisions should made based on
sound generalist and specialist advice, and in conjunction with relatives
or carers (5). We regularly meet such carers in clinical practice who feel
well served by medication, who are willing to engage in life’s realities
of assessing and accepting risk, and who make those choices in conjunction
with informed medical opinion. It seems to us that the blanket injunction
issued by the CSM is oversimplified, and may prove detrimental to patient
care.
Yours sincerely
Donald Mowat
Douglas Fowlie
Tom MacEwan
Consultants in Old Age Psychiatry
Royal Cornhill Hospital, ABERDEEN AB25 2ZH
Email: donald.mowat@gpct.grampian.scot.nhs.uk
References
www.mca.gov.uk/aboutagency/regframework/csm/csmhome.htm
Brodaty, H. et al. (2003): "A Randomised Placebo-Controlled Trial of
Risperidone for the Treatment of Aggression, Agitation and Psychosis of
Dementia". J Clin Psychiatry 64:2 134 – 143.
Herrmann, N. (2003) Data on poster presented at International
Conference on Geriatric Psychopharmacology
Schneider, L. (2003) " Meta – Analysis of Atypical Anti-psychotics
for Dementia Patients: Balancing Efficacy and Adverse Events". Proceedings
of Eleventh Congress of the Interrnational Psychogeriatric Association,
Symposium 078-004.
Adults with Incapacity (Scotland) Act (2000):
www.scotland.gov.uk/about/JD/CL/00016360/home.aspx
Competing interests:
TM has received support from Janssen Cilag to attend an international meeting; our Department has received limited educational support from various pharmaceutical companies.
Competing interests: No competing interests
Alan Wade in his rapid response letter to the bmj [1] states "The
conclusion in this case must be that the evidence of benefit does not
exist, not that lack of benefit exists. It is important that we do not
needlessly dismiss potentially useful drugs when viable alternatives are
thin on the ground."
I cannot understand how the second sentence follows logically from
the first sentence. If there is no RCT evidence of benefit of SSRIs in
major depression in children/adolescents, then one cannot conclude that
the drug is potentially useful unless there is other EBM evidence that
proves that the drug is effective and clinically useful. The author
supplies no additional (ancillary) EBM evidence that proves that SSRI's
are beneficial for major depression in children/adolescents. Also, the
lack of viable alternative drugs does not alter the EBM
validity/invalidity of EBM studies of SSRIs in major depression in
children and adolescents.
I think that it is irrational to question Jureidini's agenda, when
one should really question the scientific valididy of his particular
analysis. Wade also suggests that the benefits have been exaggerated with
respect to other therapies (other than SSRIs), and that there is a large
placebo effect in other diseases, but those two comments have no
pertinence with respect to the validity, or invalidity, of Jureidini's
analysis, which should be criticised on its own merit (or lack of merit).
Chris Manning in his rapid response letter [2] states with respect to
the fact that a SSRI drug was of benefit in his personal case of
depression-:"As a patient, and service user, I am also pleased that my own
narrative has value, whether it is supportive of the evidence gleaned from
the highly artifical environment of RCTs or flies in the face of it." I
cannot readily understand how narratives of the anecdotal value of SSRIs
can be of EBM evidentiary value when it comes to scientifically
determining whether SSRIs are clinically effective in major depression. If
a number of anecdotal reports suggest that SSRIs are of clinical value in
major depression, then the psychiatric community has to devise a
scientically valid method of determining whether the drugs really are of
value by designing/performing a clinical research study that minimises
systematic bias. If the evidence from RCTs does not provide significantly
positive evidence of SSRI's efficacy, and mental health professionals
regard those RCTs as being artificial (or otherwise invalid), then they
should design and perform studies that can provide a higher level of EBM
evidence. If they cannot design and perform a more scientically valid
study than a RCT, then they are logically obliged to accept the results of
RCTs as being the "best" EBM evidence presently available. This logic
doesn't depend on any person's agenda, because the scientific validity of
an EBM study should theoretically be independent of the personal bias of
the person performing the study, or the personal biases of the community
of mental health professionals who interpret the study.
References:
1. Wade. Alan G. What's your agenda. bmj rapid response letter.
http://bmj.bmjjournals.com/cgi/eletters/328/7444/879#58312
2. Manning Chris. Everybody has an agenda. bmj rapid response letter.
http://bmj.bmjjournals.com/cgi/eletters/328/7444/879#58145
Competing interests:
None declared
Competing interests: No competing interests
Depression in children is a serious and life threatening illness
justifying aggressive and sustained treatment.
Most medications used for children are only tested in adults and
therefore used “off-label”. In general it is accepted that benefit in
adults is assumed to confer benefit in children. This is at least in part
due to difficulties of obtaining consent and conducting studies in
children. In the absence of evidence to the contrary this should also
apply to antidepressants.
At present there is an apparent campaign against SSRI antidepressants and
this paper, if not in actual content, then certainly in implication, seems
to follow this trend. It is worth remembering that the prescribing
information clearly states that these drugs have not been tested in
children and to our knowledge no attempt has been made to licence these
drugs for children. While in the light of commercial interests a raised
bar in term of judgement is necessary, we have to be careful not to raise
the bar so high that nothing gets over
What do Jureidini and colleagues find is wrong with the evidence
available?
1. Exaggeration of benefits – selective publication of positive
results is well recognised throughout medicine – this applies equally to
psychotherapy so strongly recommended with no evidence in this paper. We
strongly believe that the scientific process of peer review, publication
and replication or lack of it will eventually bring out the truth.
2. Improvement in control groups – a large placebo response is not
only seen in treatments of depression – including psychotherapy – but is
common to many areas of medicine such as asthma, pain and hypertension
3. Adverse effects – these cannot be taken in isolation, risk/benefit
must be assessed and depression in unstable adolescents is a serious
illness with potential lethal outcomes
4. Antidepressants cannot reliably be recommended (in children) –
based on published evidence fluoxetine can, other SSRI’s simply have
insufficient data for us to make such a judgement. This recommendation is
reinforced by a recent review in the Lancet.
5. A more critical approach to published data is needed – no one
could possibly disagree with this sentiment - this paper too?
It is important to recognize that different agendas play a role in
psychiatric pharmacotherapy: companies want to sell their drugs, patients
want to get efficacious and safe treatments, third parties want to get
acceptable effects at a minimum of costs. It’s a difficult task for
physicians to mediate between these interests and to stay away from
additional less legitimate ideological agendas. The conclusion in this
case must be that the evidence of benefit does not exist, not that lack of
benefit exists. It is important that we do not needlessly dismiss
potentially useful drugs when viable alternatives are thin on the ground.
Reference
Whittington et al Lancet 2004; 363: 1341-45
Competing interests:
Both authors heave received research grants, consultancy fees and support to attend conferences from a number of pharmaceutical companies which produce antidepressant drugs.
Competing interests: No competing interests
Dear Sir
I defer to the analytical abilities of those who have contributed to
this debate; they all clearly demonstrate that this subject area, and more
importantly, those affected by it, are demanding of far greater focus
and attention than they have received to date. Perhaps, it is time to turn
down the heat on some old favourites, like diabetes, asthma and CHD, and
put the brain highest on the list for deserving our attentions?
As a former front-line "jobbing Joe", I was relieved to be able to
start prescribing non-cardiotoxic products in the treatment of depression,
and can also report that taking an SSRI, after having taken a TCA, has
given me a life free of the latter's paralysing and life-enfeebling side-
effects. As a patient, and service user, I am also pleased that my own
narrative has value, whether it is supportive of the evidence gleaned from
the highly artifical environment of RCTs or flies in the face of it. I am
also convinced that the term depression does not, in any way, do justice
to the complexity of the brain and that, attempting to strain everyone
through such an outwardly-derived diagnosis, is like trying to claim that
the Amazonian Forest complexity of the brain can be understood using 4
theories stating that it has at least 7 moving parts.
Furthermore, as I have witnessed the various UK and US actings-out on
this, admittedly, crucial matter, I have realised that a lot of what has
happened to date is also about the varied agendas of all of those
involved, not just the industry. I am also sure that most of us now know
what we are AGAINST in all of this: the potential for the over-marketing
of any medicine, however achieved, and especially to vulnerable groups and
desperate practitioners. The challenge now is surely how we all work
together TOWARDS achieving the best possible outcomes for those who suffer
from a genuine illness ("if you can describe it, then you have never had
it" Prof. Lewis Wolpert), which itself has high comorbidities and often
tragic outcomes.
I was well trained by a Teutonic Professor of Pharmacology who
instilled in all of us the varying degress of toxicity associated with ALL
medicinal products. We need to move forwards in a both...and way now, not
an either...or one.
Taking sides, for whatever reason, either personally or organisationally,
is the worst possible way forward and compounds the damage already done to
the community we are tasked with serving here.
Many of us spend every day working to drive these issues up the
national agenda in the UK. It is still highly confounding, however, that
mental health is taken to be synonymous, at worst, with mental illness
(or, at best, its absence)and is seen as a 'soft-touch' in terms of
funding. It figures, for example, next to nowhere in the Government's
current consultation on public health and services for CAMH, especially in
primary care, have their own extremely chequered history of low-level
provision, long-waiting times and people being referred back as
"inappropriate referrals".
By way of closure, I can well remember psychotherapists and
counsellors phoning me up about clients/patients who had just started in
therapy (either referred by me or not)and who had become agitated and
suicidal, sometimes at the very first session. If we cannot appreciate
that taking a meta/unstable situation and throwing ANY intervention at it
(including 'just' a listening ear)is going to be fraught with some
difficulties (and I defer to the analysts to tease that one out), then we
will simply continue in our either..or shismatic thinking and behaviours.
Many of us go into health and social care to make a positive
difference to the lives of those who suffer; there is no valid point in
continued witch-hunts and professional bashing. We need to start working
now on agreeing the way forwards, not endlessly re-iterating where we have
been.
Yours Faithfully
Dr Chris Manning
CE Primhe
Hon Vice President Depression Alliance
National MH Taskforce
Competing interests:
I have depression. I have treated depression in primary care in adolescents and two children. I have designed and run educational and training programmes for primary and secondary care professionals, at which whole-systems and whole-person approaches have been advocated and, as part of which ,medical treatment may or may not be appropriate. These programmes have, to date, been funded mostly by the pharma. industry in the absence of any coherent or available funding from elsewhere. At all times, a line has been drawn, and that line is the drug product and its marketing. The activities of Depression Alliance and Primhe are funded from supportership fees, the Department of Health and the industry. Both charities look forward to the time when England is not at war and can spend time focussing on the 'crying' needs of its citizens.
Competing interests: No competing interests
We wish to address the methods used by Dr. Jureidini and his co-
authors in their recent review (Jureidini et al, 2004) of SSRIs in
children, as well as to take issue with one of the important conclusions
they make relating to the potential public health value of drugs with
modest effect sizes in the treatment of pediatric depression, a serious
illness.
Specifically, we would like to disagree with several misstatements
that are made in the Jureidini article about our report of 2 placebo-
controlled clinical trials of sertraline in pediatric depression (Wagner,
JAMA, 2003), and with the overall conclusion, which suggests that this
article exhibited “disturbing shortcomings in the methods and reporting…”
Methods used in the clinical review
The objective of the article by Dr. Jureidini and colleagues is to
critically review the “quality of methods and reporting of…published
trials of newer antidepressants in children.” However, the authors
provide no summary of the criteria they used to assess either the
methodology or the reporting, though standard criteria are readily
available. For example, the Cochrane Database, as well as other
systematic reviews, has routinely employed a standardized checklist to
assess the quality of the methodology of clinical trials (Jadad, 1996).
Application of this checklist to our trial that studied the efficacy and
safety of sertraline results in a Jadad score of 5, which is the highest
possible score.
Similarly, medical journal editors have agreed upon standardized
methods for reporting clinical trials (the Consort Guidelines; Moher et
al, 2003). A review of the sertraline study against the Consort reporting
guidelines (Moher et al, 2003) indicates that we complied with all 22 of
the reporting guidelines. Thus, objective consensus guidelines indicate
that the quality of the sertraline clinical trial (Wagner et al, 2003), as
well as the quality of how it was reported, both met the highest consensus
standards.
Use of double-blind methodology
Dr. Jureidini and colleagues criticize the psychopharmacology trials in
children because (they speculate) adverse events might have unblinded the
children as to treatment assignment, and this “suggests that part of the
impact of an active drug might be due to unblinding.” While this is
always a concern in double-blind trials of any medicine, we note that Dr.
Jureidini and co-authors do not express this concern when it comes to
their recommendation of more effective psychological treatments with “no
known adverse effects.” In fact, the psychological treatment studies that
Dr. Jureidini and colleagues cite (Harrington et al, 1998) were neither
double-blind nor SSRI-comparator studies, thereby prohibiting the
conclusion that psychological treatments are “more effective with no
known adverse effects”. Furthermore, most of the studies are good
examples of the very under-reporting of adverse events that Dr. Jureidini
(mistakenly) objects to in our sertraline treatment study (Wagner et al,
2003). The lack of double-blind methodology, and the lack of appropriate
sample sizes in psychotherapy treatment studies helps to explain why
reviews (Gaffan et al, 1995) have shown that the therapeutic “allegiance”
of the principal investigator is the strongest predictor of a positive
outcome in a psychotherapy studies.
Sertraline: Review of Safety and Tolerability
Dr Jureidini and his co-authors make several misstatements about our
clinical trial with sertraline in depressed pediatric patients. The
authors question the tolerability of sertraline (on p.880) because 9% of
patients discontinued due to adverse events compared to 3% on placebo.
The authors conclude this section by stating: “Other sources of data
support the view that adverse effects might be more frequent than the
authors of these studies imply.” The authors then go on to cite only a
small 42 patient study of fluoxetine in OCD (King et al, 1991). Dr.
Jureidini and his co-authors chose to ignore a double-blind, placebo-
controlled study of sertraline in childhood OCD that provides a much
larger (N=187) and more systematic assessment of an SSRI (sertraline) in
OCD. (March et al, 1998). It should be noted that 137 patients in this
study elected to continue treatment for an additional 12 months, for a
total of 15 months of treatment. Sertraline was well-tolerated during
this course of treatment. (Cook et al, 2003). Not to mention such a large
study creates the unfortunate impression that the authors have their own
bias, and that they are selectively reporting only the studies that
support (albeit weakly) their own views. It should also be noted that
this sertraline OCD study forms the basis of the conclusion by the UK
Committee on Safety of Medicines and many other regulatory agencies around
the world that sertraline is safe and effective in the treatment of
pediatric OCD. As such, the callout box on page 879 which states that the
UK Committee on Safety of Medicines has banned all remaining selective
serotonin reuptake inhibitors except fluoxetine for use in patients under
18 years of age, is inaccurate.
The authors then go on to deplore “The failure of drug companies to
disclose increased suicidal activity secondary to these drugs…” For
sertraline in the treatment of childhood depression (Wagner et al, 2003)
this statement is false, and we would request that the authors make a
correction in BMJ. The article in JAMA provides the following data on
suicidal related ideation and/or activity (p.1038): “Seven sertraline-
treated patients and 6 placebo patients had adverse events that met the
established criteria for a ‘serious’ adverse event, including suicide
attempt (2 sertraline and 2 placebo), suicidal ideation (3 sertraline),
and aggressive reaction (1 sertraline)…” On p.1039 of the JAMA article,
there is further discussion of the issue of suicide as follows: “Our
trials show a lack of significant difference in suicidal ideation between
sertraline-treated and placebo patients, as measured by the CDRS-R. In
patients who continued the 24-week open-label extension study, only 1
episode of suicidal ideation was reported, and the investigator attributed
this event to teasing by classmates.”
Sertraline: Review of Efficacy
Dr. Jureidini and his co-authors do not specifically summarize the
efficacy results in the JAMA-sertraline article, except to note that they
“doubt Wagner’s claim, and Varley’s editorial support for that claim, that
their results are clinically, as well as statistically, significant.” Dr.
Jureidini and his co-authors then go on, in the next paragraph, to make
the generalization that “the authors of all of the four larger studies
have exaggerated the benefits, downplayed the harms, or both.” We have
already summarized the extent to which the JAMA report of our studies did
not “downplay” the harms. We feel compelled to quote the concluding
sentences of the JAMA article to demonstrate that the authors were not
exaggerating the benefits, but instead were reporting a significant
result, while appropriately qualifying what the study found. After
highlighting the extent to which childhood depression is a major public
health problem, and cautioning that “further research is needed” they
concluded by stating: “Whether lower initial dosages in children would
improve tolerability or long-term sertraline treatment in children and
adolescents would result in maintenance of effect an improvement in
quality of life deserves study. Nonetheless, the results reported here
support the conclusion that sertraline is an effective, safe, and well-
tolerated treatment for children and adolescents with MDD.”
General Comments
Finally, we agree with the authors that the effect size of SSRIs in the
treatment of childhood depression is modest. For example, in our
sertraline pediatric depression study, there was only a 10% difference in
endpoint responder rates between sertraline and placebo.
We are not sure why the non-specific effect of placebo is so high in
childhood depression, though it is likely to be due to a high expectancy
effect in a highly suggestible patient group. We do know, though, that
placebo will not be prescribed by any physician for the treatment of major
depression in children. We also believe that there are sufficient safety
concerns with the TCAs (eg, cardio-toxicity and overdose lethality) that
make them a poor first-line treatment, especially since evidence for their
efficacy is even weaker than the SSRIs. Also, as we have indicated, the
evidence for the efficacy of psychological therapies is based on
relatively small studies, none of which used a double-blind methodology.
Thus, a response rate of 69% on sertraline (despite only a 10% advantage
over placebo) makes it understandable why physicians in the past decade
might have made the clinical judgment that the benefit-risk ratio was
tipped in favor of SSRI treatment, even though regulatory approval is
lacking for most SSRIs (including sertraline) in childhood depression.
When one puts the issue of childhood depression in context, then we
feel less able than the authors are to dismiss a 69% response rate on
sertraline, even if it is only 10% superior to placebo. Cumulative rates
of major depression and suicide in childhood and adolescence are
staggering. In Western Europe alone, an estimated 10.1 million children
will develop major depression by age 18, and more than 50,000 will have
committed suicide (Birmaher et al, 1996; Olfson et al, 2003; US Census
Bureau).
While a 10% advantage over placebo may seem trivial to the authors,
the sheer magnitude of the suffering makes us wonder about their
unwillingness to “lower the threshold for accepting a new intervention.”
Even 10% improvement in response to this devastating illness, while small
in absolute terms, translates into substantial relief of suffering for
hundreds of thousands of children and potentially saving the lives of some
patients. Daily, clinicians face the difficult question of how to treat
these patients, for whom the data are often confusing or inconclusive. We
stand by the reported results from the sertraline study, and welcome
objective and fair discussion on this topic. However, reports containing
inaccurate or misleading statements such as that by Dr. Jureidini and
colleagues only cloud the issue and impede successful patient treatment.
References
Birmaher B, Ryan ND, Williamson DE, Brent DA, Kaufman J, Dahl RE,
Perel J, Nelson B. Childhood and adolescent depression: a review of the
past 10 years. Part I. J Am Acad Child Adolesc Psychiatry 1996;35:1427-
1439.
Cook EH, Wagner KD, March JS, Biederman J, Landau P, Wolkow R, Messig
M. Long-term sertraline treatment of children and adolescents with
obsessive-compulsive disorder. J Am Acad Child Adolesc Psychiatry
2001;40:1175-1181.
King RA, Riddle MA, Chappell PB, Hardin MT, Anderson GM, Lombroso P,
et al. Emergence of self-destructive phenomena in children and adolescents
during fluoxetine treatment. J Am Acad Child Adolesc Psychiatry
1991;30:179-86.
Harrington R, Whittaker J, Shoebridge P, Campbell F. Systematic
review of efficacy of cognitive behaviour therapies in childhood and
adolescent depressive disorder. BMJ 1998;316:1559-1563.
Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of
randomized clinical trials: is blinding necesssary? Control Clin Trials
1996;17:1–12.
Jureidini JN, Doecke CJ, Mansfield PR, Haby MM, Menkes DB, Tonkin AL.
Efficacy and safety of antidepressants for children and adolescents. BMJ
2004;328:879-883.
March JS, Biederman J, Wolkow R, Safferman A, Mardekian J, Cook EH,
Cutler NR, Dominguez R, Ferguson J, Muller B, Riesenberg R, Rosenthal M,
Sallee FR, Wagner KD, Steiner H. Sertraline in children and adolescents
with obsessive-compulsive disorder: a multicenter randomized controlled
trial. JAMA 1998;280:1752-1756.
Moher D, Schulz KF, Altman DG; CONSORT Group. The CONSORT statement:
revised recommendations for improving the quality of reports of parallel-
group randomised trials. Clin Oral Investig 2003;7:2-7.
Olfson M, Shaffer D, Marcus SC, Greenberg T. Relationship between
antidepressant medication treatment and suicide in adolescents. Arch Gen
Psychiatry 2003;60:978-982.
US Census Bureau, Global population estimates, 2002, Table A-7a
Wagner KD, Ambrosini P, Rynn M, Wohlberg C, Yang R, Greenbaum MS,
Childress A, Donnelly C, Deas D. Efficacy of sertraline in the treatment
of children and adolescents with major depressive disorder: two randomized
controlled trials. JAMA 2003;290:1033-1041.
Competing interests:
Drs. Clary and Romano are full time employees and shareholders of Pfizer Inc
Competing interests: No competing interests
In their detailed analysis of the efficacy of antidepressants in
juvenile depression(1), Jureidini et al have crucially failed to point out
that most of these trials have been conducted on both children and
adolescents, some as young as 6 years (2). In addition, suicidality is
often an exclusion criteria, as are numerous comorbid disorders.
Therefore, we do not know the effect of these drugs in the depressed,
suicidal adolescent population who frequently present with associated
comorbidity, as this sample have not been analysed separately. This group
would be more representative of the typically depressed adolescent that
may be treated with antidepressants in the UK, and just as antidepressant
drugs cannot 'confidently be recommended as a treatment option for
childhood depression', neither can we confidently conclude that
antidepressants are not effective in this group.
Likewise, cognitive behavioural therapy (CBT) has not been trialled in
severe adolescent depression with associated suicidality and comorbidity.
The quoted study by Harrington et al (3) found that psychological
treatment was an effective intervention for depressive symptoms and mild
depression only, therefore we cannot conclude that this will also be
effective in severe depression, and free of adverse effects.
More non-industry treatment trials of severe adolescent depression are
urgently needed to address these issues.
1. Jureidini et al, Efficacy and safety of antidepressants for
children and adolescents, BMJ, 328: 879-83, 2004.
2.Wagner et al, Efficacy of sertraline in the treatment of children and
adolescents with major depressive disorder. JAMA, 290, 1033-1041, 2003.
3. Harrington et al, Psychological treatment of depression in children and
adolescents. A review of treatment research. BJP, 173, 291-298, 1998.
Competing interests:
BD is a research fellow in a government funded treatment study of adolescent depression, involving antidepressants and cognitive behavioural therapy.
Competing interests: No competing interests
Response to Pfizer
In the hope that late is better than never here is a response to
criticisms of our systematic review made by Clary and Romano <1> who
are employees of Pfizer, the manufacturer of sertraline.
Criticism
We should have used standard quality assessment tools.
Response
These would not have detected many of the problems we found. The
current standard quality assessment tools do not cover many important
potential problems with controlled trials.
Criticism
We should have criticised the trials of cognitive therapy.
Response
This criticism has merit but our review already covered much ground.
Criticism
We should not have accused drug companies of not disclosing suicidal
activity because the report of one published study did disclose it.
Response
This criticism has merit but what we wrote was a general statement
about the disclosure debate. We remain concerned about failure of
disclosure in drug promotion.
Criticism
We should not have claimed that authors had overstated benefits and
understated harms.
Response
Our claim is justified. For example, Wagner et al. claimed that “the
results reported here support the conclusion that sertraline is an
effective, safe, and well-tolerated short-term treatment for children and
adolescents” despite their data showing no significant difference on their
primary endpoint but a significantly higher rate of adverse
effects.<2>
Criticism
We should not have denied that there was a worthwhile benefit from
antidepressants.
Response
We had concluded that a small benefit was likely but a worthwhile
benefit was unlikely. Data that have become available after our review
have supported our conclusions.<3,4>
1. Clary CM, Romano SJ. Sertraline Pediatric Depression Trial Met
Highest Standards of Study Design and Reporting. BMJ Rapid Response 2004
April 27
2. Wagner KD, Ambrosini P, Rynn M, Wohlberg C, Yang R, Greenbaum MS,
Childress A, Donnelly C, Deas D. Efficacy of sertraline in the treatment
of children and adolescents with major depressive disorder: two randomized
controlled trials. JAMA 2003;290:1033-1041.
3. Whittington CJ, Kendall T, Fonagy P, et al. Selective serotonin
reuptake inhibitors in childhood depression: systematic review of
published versus unpublished data. Lancet. 2004 Apr 24;363(9418):1341-
1345.
4. Jureidini J, Tonkin A, Mansfield PR. TADS study raises concerns.
BMJ. 2004;329:1343-1344.
Competing interests:
I am a co-author of the Jureidini et al systematic review and founder of Healthy Skepticism Inc.
Competing interests: No competing interests