Upload
t-e
View
217
Download
1
Embed Size (px)
Citation preview
10.1192/bjp.184.6.541-aAccess the most recent version at DOI: 2004, 184:541-542.BJP
P. E. Holtzheimer, D. Avery and T. E. SchlaepferstimulationAntidepressant effects of repetitive transcranial magnetic
Referenceshttp://bjp.rcpsych.org/content/184/6/541.2#BIBLThis article cites 4 articles, 1 of which you can access for free at:
permissionsReprints/
[email protected] to To obtain reprints or permission to reproduce material from this paper, please
to this article atYou can respond http://bjp.rcpsych.org/letters/submit/bjprcpsych;184/6/541-a
from Downloaded
The Royal College of PsychiatristsPublished by on October 19, 2014http://bjp.rcpsych.org/
http://bjp.rcpsych.org/site/subscriptions/ go to: The British Journal of PsychiatryTo subscribe to
Analysing the efficacy of clozapineAnalysing the efficacy of clozapine
Dr Moncrieff (2003) has suggested thatDr Moncrieff (2003) has suggested that
the advantage of clozapine inthe advantage of clozapine in treatment-treatment-
resistant schizophrenia, when comparedresistant schizophrenia, when compared
with conventional antipsychotics, maywith conventional antipsychotics, may
not be substantial. This appears to benot be substantial. This appears to be
discordant with an earlier meta-analysisdiscordant with an earlier meta-analysis
(Wahlbeck(Wahlbeck et alet al, 2000). As clozapine’s, 2000). As clozapine’s
advantage in treatment-resistant schizo-advantage in treatment-resistant schizo-
phrenia is well accepted in psychiatry andphrenia is well accepted in psychiatry and
is reflected in most practice guidelines,is reflected in most practice guidelines,
any questions about its validity need carefulany questions about its validity need careful
scrutiny. Clues to the disagreement betweenscrutiny. Clues to the disagreement between
meta-analyses on the same topic can oftenmeta-analyses on the same topic can often
be found in the studies that are includedbe found in the studies that are included
or excluded, the ways in which the dataor excluded, the ways in which the data
are abstracted and in the interpretation ofare abstracted and in the interpretation of
the results (Jadadthe results (Jadad et alet al, 1997)., 1997).
Dr Moncrieff included two studies inDr Moncrieff included two studies in
her analysis that were not in the earlierher analysis that were not in the earlier
meta-analysis: Essockmeta-analysis: Essock et alet al (1996) and Kane(1996) and Kane
et alet al (2001).(2001).
The EssockThe Essock et alet al (1996) study was a(1996) study was a
naturalistic study with serious method-naturalistic study with serious method-
ological deficiencies from the perspectiveological deficiencies from the perspective
of determining efficacy of clozapine treat-of determining efficacy of clozapine treat-
ment. The randomisation was imperfect.ment. The randomisation was imperfect.
The study was not blinded. The studyThe study was not blinded. The study
population was poorly defined in terms ofpopulation was poorly defined in terms of
diagnosis. Later application of the Struc-diagnosis. Later application of the Struc-
tured Clinical Interview for DSM–III–Rtured Clinical Interview for DSM–III–R
Personality Disorders to a subgroup of thePersonality Disorders to a subgroup of the
study population picked up diagnosesstudy population picked up diagnoses
including bipolar disorder, organic moodincluding bipolar disorder, organic mood
disorder and one case of ‘no disorder’.disorder and one case of ‘no disorder’.
‘Crossovers’ were allowed, with nearly‘Crossovers’ were allowed, with nearly
66% of the control group receiving cloza-66% of the control group receiving cloza-
pine at some time. There was no restrictionpine at some time. There was no restriction
on the prescription of other medications,on the prescription of other medications,
with patients in both groups receiving otherwith patients in both groups receiving other
psychotropic medications, including otherpsychotropic medications, including other
antipsychotics. An intention-to-treat analy-antipsychotics. An intention-to-treat analy-
sis would be meaningless given the numbersis would be meaningless given the number
of crossovers. Also, analysis of data withof crossovers. Also, analysis of data with
crossovers excluded is unlikely to be infor-crossovers excluded is unlikely to be infor-
mative as it would end up comparing amative as it would end up comparing a
small subgroup of responders in eithersmall subgroup of responders in either
arm of the study. The validity of includingarm of the study. The validity of including
this study in the meta-analysis is question-this study in the meta-analysis is question-
able. This is particularly relevant as theable. This is particularly relevant as the
‘forest plot’ in Moncrieff’s analysis reveals‘forest plot’ in Moncrieff’s analysis reveals
that this is the only study where the effectthat this is the only study where the effect
size is in the opposite direction (i.e. un-size is in the opposite direction (i.e. un-
favourable to clozapine). Thus, inclusionfavourable to clozapine). Thus, inclusion
of this study would dilute the effect size ofof this study would dilute the effect size of
clozapine and vice versa.clozapine and vice versa.
Moncrieff’s handling of the data fromMoncrieff’s handling of the data from
the Kanethe Kane et alet al (2001) study also raises ques-(2001) study also raises ques-
tions. In this longer-duration study, patientstions. In this longer-duration study, patients
in both the control and experimental groupsin both the control and experimental groups
were allowed to drop out if they were notwere allowed to drop out if they were not
responding to the given treatment. A non-responding to the given treatment. A non-
intention-to-treat analysis, as Dr Moncrieffintention-to-treat analysis, as Dr Moncrieff
has done, would end up comparing a smallhas done, would end up comparing a small
subgroup of responders in either group. Ansubgroup of responders in either group. An
intention-to-treat analysis would have cap-intention-to-treat analysis would have cap-
tured clozapine’s strength; that is, showingtured clozapine’s strength; that is, showing
that more patients on clozapine respondedthat more patients on clozapine responded
in comparison with the control group.in comparison with the control group.
Despite these observations, Moncrieff’sDespite these observations, Moncrieff’s
analysis produced an effect size of 0.38analysis produced an effect size of 0.38
(0.44 using a random effects model). In(0.44 using a random effects model). In
my opinion, this is not unimpressive givenmy opinion, this is not unimpressive given
that clozapine is being compared with otherthat clozapine is being compared with other
medications with proven efficacy and notmedications with proven efficacy and not
placebo.placebo.
Declaration of interestDeclaration of interest
I have attended local educational meetingsI have attended local educational meetings
sponsored by Novartis.sponsored by Novartis.
Essock, S. M., Hargreaves,W. A.,Covell,N.H.,Essock, S. M.,Hargreaves,W. A., Covell,N.H., et alet al(1996)(1996) Clozapine’s effectiveness for patients in stateClozapine’s effectiveness for patients in statehospitals: results from a randomized trial.hospitals: results from a randomized trial.Psychopharmacology BulletinPsychopharmacology Bulletin,, 3232, 683^697., 683^697.
Jadad, A. R., Cook, D. J. & Browman,G. P. (1997)Jadad, A. R., Cook, D. J. & Browman,G. P. (1997) AAguide to interpreting discordant systematic reviews.guide to interpreting discordant systematic reviews.Canadian Medical Association JournalCanadian Medical Association Journal,, 156156, 1411^1416., 1411^1416.
Kane, J., Marder, S. R., Schooler, N. R.,Kane, J., Marder, S. R., Schooler, N. R., et alet al (2001)(2001)Clozapine and haloperidol in moderately refractoryClozapine and haloperidol in moderately refractoryschizophrenia: a 6-month randomized and double-blindschizophrenia: a 6-month randomized and double-blindcomparison.comparison. Archives of General PsychiatryArchives of General Psychiatry,, 5858, 965^972., 965^972.
Moncrieff, J. (2003)Moncrieff, J. (2003) ClozapineClozapine vv. conventional. conventionalantipsychotic drugs for treatment-resistantantipsychotic drugs for treatment-resistantschizophrenia: a re-examination.schizophrenia: a re-examination. British Journal ofBritish Journal ofPsychiatryPsychiatry,, 183183, 161^166., 161^166.
Wahlbeck, K.,Cheine, M., Essali, A.,Wahlbeck, K., Cheine, M., Essali, A., et alet al (2000)(2000)Clozapine versus typical neuroleptic medication forClozapine versus typical neuroleptic medication forschizophrenia.schizophrenia. Cochrane LibraryCochrane Library, issue 3.Oxford: Update, issue 3.Oxford: UpdateSoftware.Software.
S. KarunakaranS. Karunakaran Kirwan Rehabilitation &Kirwan Rehabilitation &Extended Care Unit,138 Thuringowa Drive,Kirwan,Extended Care Unit,138 Thuringowa Drive,Kirwan,Townsville,Queensland 4817, AustraliaTownsville,Queensland 4817, Australia
Dr Moncrieff (2003) re-analysed the dataDr Moncrieff (2003) re-analysed the data
of a Cochrane meta-analysis by Wahlbeckof a Cochrane meta-analysis by Wahlbeck
et alet al (2000) on the comparison between(2000) on the comparison between
clozapine and conventional antipsychoticclozapine and conventional antipsychotic
drugs for treatment-resistant schizophrenia.drugs for treatment-resistant schizophrenia.
After selecting nine randomised controlledAfter selecting nine randomised controlled
trials and analysis she concluded that thetrials and analysis she concluded that the
Cochrane review might have overestimatedCochrane review might have overestimated
the effects of clozapine as she found a lowerthe effects of clozapine as she found a lower
overall effect. This was explained by the useoverall effect. This was explained by the use
of data from intention-to-treat analysis inof data from intention-to-treat analysis in
the largest included study by Rosenheckthe largest included study by Rosenheck etet
alal (1997) and inclusion of the large study(1997) and inclusion of the large study
by Essockby Essock et alet al (1996), which was excluded(1996), which was excluded
in the Cochrane review.in the Cochrane review.
There are good reasons for reporting theThere are good reasons for reporting the
results from the studies by Rosenheckresults from the studies by Rosenheck et alet al
(1997) and Essock(1997) and Essock et alet al (1996) separately(1996) separately
from the other seven studies rather than giv-from the other seven studies rather than giv-
ing the overall results. These two studies areing the overall results. These two studies are
long-term studies with durations of 1 and 2long-term studies with durations of 1 and 2
years, respectively. The study populationsyears, respectively. The study populations
were much larger than most of the otherwere much larger than most of the other
studies, which were short-term studies last-studies, which were short-term studies last-
ing 6–29 weeks. The two long-term studiesing 6–29 weeks. The two long-term studies
found a small to no difference in treatmentfound a small to no difference in treatment
effect between clozapine and the conven-effect between clozapine and the conven-
tional antipsychotic. These results have ational antipsychotic. These results have a
large negative impact on the overall effectlarge negative impact on the overall effect
because of the large study populations.because of the large study populations.
However, the use of intention-to-treat ana-However, the use of intention-to-treat ana-
lysis will result in smaller differences be-lysis will result in smaller differences be-
tween the clozapine and control group thetween the clozapine and control group the
longer the study lasts, because drop-outslonger the study lasts, because drop-outs
are classified as relapses irrespective of theare classified as relapses irrespective of the
reason for discontinuation. Longer studiesreason for discontinuation. Longer studies
tend to have larger drop-out rates, as is alsotend to have larger drop-out rates, as is also
apparent in this meta-analysis, resulting inapparent in this meta-analysis, resulting in
smaller differences between study groups.smaller differences between study groups.
Reporting the results from the short-Reporting the results from the short-
term and long-term studies separately willterm and long-term studies separately will
probably show that clozapine has a higherprobably show that clozapine has a higher
treatment effect than that reported bytreatment effect than that reported by
Moncrieff. Short-term studies explore theMoncrieff. Short-term studies explore the
pharmacological efficacy of a medicinepharmacological efficacy of a medicine
whereas long-term studies explore the treat-whereas long-term studies explore the treat-
ment effect in daily practice and can bement effect in daily practice and can be
influenced by the patient’s willingness toinfluenced by the patient’s willingness to
continue treatment. These results shouldcontinue treatment. These results should
be reported separately.be reported separately.
53 953 9
BR I T I SH JOURNAL OF P SYCHIATRYBR IT I SH JOURNAL OF P SYCHIATRY ( 2 0 0 4 ) , 1 8 4 , 5 3 9 ^ 5 4 6( 2 0 0 4 ) , 1 8 4 , 5 3 9 ^ 5 4 6
CorrespondenceCorrespondence
EDITED BY KHALIDA ISMAILEDITED BY KHALIDA ISMAIL
ContentsContents && Analysing the efficacyof clozapineAnalysing the efficacyof clozapine && Parental age difference andParental age difference and
schizophreniaschizophrenia && Physical illness and schizophreniaPhysical illness and schizophrenia && Antidepressanteffects of repetitiveAntidepressanteffects of repetitive
transcranialmagnetic stimulationtranscranialmagnetic stimulation && Evidence in cannabis researchEvidence in cannabis research && MRCPsych examsMRCPsych exams
CORRESPONDENCECORRESPONDENCE
Essock, S. M.,Hargreaves,W. A., Covell,N.H.,Essock, S. M.,Hargreaves,W. A.,Covell,N.H., et alet al(1996)(1996) Clozapine’s effectiveness for patients in stateClozapine’s effectiveness for patients in statehospitals: results from a randomized trial.hospitals: results from a randomized trial.Psychopharmacology BulletinPsychopharmacology Bulletin,, 3232, 683^697., 683^697.
Moncrieff, J. (2003)Moncrieff, J. (2003) ClozapineClozapine vv. conventional. conventionalantipsychotic drugs for treatment-resistantantipsychotic drugs for treatment-resistantschizophrenia: a re-examination.schizophrenia: a re-examination. British Journal ofBritish Journal ofPsychiatryPsychiatry,, 183183, 161^166., 161^166.
Rosenheck, R.,Cramer, J., Xu,W.,Rosenheck, R., Cramer, J., Xu,W., et alet al (1997)(1997) AAcomparison of clozapine and haloperidol in hospitalizedcomparison of clozapine and haloperidol in hospitalizedpatients with refractory schizophrenia.Department ofpatients with refractory schizophrenia.Department ofVeterans Affairs Cooperative Study Group onVeterans Affairs Cooperative Study Group onClozapine in Refractory Schizophrenia.Clozapine in Refractory Schizophrenia. New EnglandNew EnglandJournal of MedicineJournal of Medicine,, 337337, 809^815., 809^815.
Wahlbeck, K., Cheine, M., Essali, M. A.,Wahlbeck, K.,Cheine, M., Essali, M. A., et alet al (2000)(2000)Clozapine versus typical neuroleptic medication forClozapine versus typical neuroleptic medication forschizophrenia.schizophrenia. Cochrane LibraryCochrane Library, issue 3.Oxford: Update, issue 3.Oxford: UpdateSoftware.Software.
K.H.KhoK.H. Kho GGZDelfland, St Jorisweg 2, 2612 GAGGZDelfland, St Jorisweg 2, 2612 GADelft,The NetherlandsDelft,The Netherlands
Author’s reply:Author’s reply: Dr Karunakaran rightlyDr Karunakaran rightly
points out some problems with the in-points out some problems with the in-
terpretation of the Essockterpretation of the Essock et alet al (1996)(1996)
naturalistic study of clozapine. However,naturalistic study of clozapine. However,
despite its imperfections, that study de-despite its imperfections, that study de-
serves some attention, both because it wasserves some attention, both because it was
a large study and because its naturalistic de-a large study and because its naturalistic de-
sign attempted to replicate the conditions insign attempted to replicate the conditions in
which clozapine would be given in real clin-which clozapine would be given in real clin-
ical practice. The randomisation was notical practice. The randomisation was not
imperfect but unbalanced. The study wasimperfect but unbalanced. The study was
indeed not blinded, but this usually favoursindeed not blinded, but this usually favours
the experimental treatment, in this casethe experimental treatment, in this case
clozapine. Application of the Structuredclozapine. Application of the Structured
Clinical Interview for DSM–IV confirmedClinical Interview for DSM–IV confirmed
that 95% of cases had a diagnosis ofthat 95% of cases had a diagnosis of
schizophrenia or schizoaffective disorder.schizophrenia or schizoaffective disorder.
It is indeed difficult to decide what outcomeIt is indeed difficult to decide what outcome
data to use, as I mention in my paper. How-data to use, as I mention in my paper. How-
ever, despite the number of crossovers, anever, despite the number of crossovers, an
intention-to-treat analysis in such a largeintention-to-treat analysis in such a large
sample would be expected to show somesample would be expected to show some
difference if the effect of clozapine is sub-difference if the effect of clozapine is sub-
stantial. In the Kanestantial. In the Kane et alet al (2001) study I(2001) study I
did use intention-to-treat data, but also re-did use intention-to-treat data, but also re-
peated the analysis withpeated the analysis with non-intention-to-non-intention-to-
treat data, because of thetreat data, because of the curiously highcuriously high
drop-out rate in the comparison group.drop-out rate in the comparison group.
My analysis was meant to draw atten-My analysis was meant to draw atten-
tion to the fact that results of differenttion to the fact that results of different
studies are quite discrepant. The largeststudies are quite discrepant. The largest
study to date, and one that appears tostudy to date, and one that appears to
be methodologically robust, found onlybe methodologically robust, found only
slight differences between clozapine andslight differences between clozapine and
haloperidol, which are of doubtful clinicalhaloperidol, which are of doubtful clinical
relevancerelevance (Rosenheck(Rosenheck et alet al, 1997). In this, 1997). In this
situation simply quoting the results of asituation simply quoting the results of a
meta-analysis may be misleading.meta-analysis may be misleading.
Dr Kho is right to point out that long-Dr Kho is right to point out that long-
term studies find smaller effects. Thisterm studies find smaller effects. This
cannot be attributed to drop-out rates incannot be attributed to drop-out rates in
the Rosenheckthe Rosenheck et alet al (1997) study, at least,(1997) study, at least,
where the higher drop-out rate in the halo-where the higher drop-out rate in the halo-
peridol group would tend to produce anperidol group would tend to produce an
inflated difference between clozapine andinflated difference between clozapine and
the comparator drug. We also cannotthe comparator drug. We also cannot
assume that short-term studies simply mea-assume that short-term studies simply mea-
sure pharmacological effects and long-termsure pharmacological effects and long-term
studies are confounded by non-compliance.studies are confounded by non-compliance.
Drugs may have different short- and long-Drugs may have different short- and long-
term pharmacological effects. Short-termterm pharmacological effects. Short-term
studies might be more likely to be con-studies might be more likely to be con-
founded by non-specific factors such asfounded by non-specific factors such as
differential expectations of treatments.differential expectations of treatments.
Essock, S. M.,Hargreaves,W. A.,Covell,N.H.,Essock, S. M.,Hargreaves,W. A., Covell, N.H., et alet al(1996)(1996) Clozapine’s effectiveness for patients in stateClozapine’s effectiveness for patients in statehospitals: results from a randomized trial.hospitals: results from a randomized trial.Psychopharmacology BulletinPsychopharmacology Bulletin,, 3232, 683^697., 683^697.
Kane, J., Marder, S. R., Schooler, N. R.,Kane, J., Marder, S. R., Schooler, N. R., et alet al (2001)(2001)Clozapine and haloperidol in moderately refractoryClozapine and haloperidol in moderately refractoryschizophrenia: a 6-month randomized and double-blindschizophrenia: a 6-month randomized and double-blindcomparison.comparison. Archives of General PsychiatryArchives of General Psychiatry,, 5858, 965^972., 965^972.
Rosenheck, R., Cramer, J., Xu,W.,Rosenheck, R.,Cramer, J., Xu,W., et alet al (1997)(1997) AAcomparison of clozapine and haloperidol in hospitalizedcomparison of clozapine and haloperidol in hospitalizedpatients with refractory schizophrenia.Department ofpatients with refractory schizophrenia.Department ofVeterans Affairs Cooperative Study Group onVeterans Affairs Cooperative Study Group onClozapine in Refractory Schizophrenia.Clozapine in Refractory Schizophrenia. New EnglandNew EnglandJournal of MedicineJournal of Medicine,, 337337, 809^815., 809^815.
J. MoncrieffJ. Moncrieff Mascalls Park,Rehabilitation Unit,Mascalls Park,Rehabilitation Unit,Mascalls Lane,Brentwood, Essex CM14 5HQ,UKMascalls Lane,Brentwood, Essex CM14 5HQ,UK
Parental age differenceParental age differenceand schizophreniaand schizophrenia
To offer hypotheses based simply on clini-To offer hypotheses based simply on clini-
cal experience is pathetically out of date.cal experience is pathetically out of date.
Perhaps it may be allowed, for a moment,Perhaps it may be allowed, for a moment,
in deference to my advancing years.in deference to my advancing years.
Fifty years ago, with some other pur-Fifty years ago, with some other pur-
pose in mind, I surveyed some 370 casespose in mind, I surveyed some 370 cases
of schizophrenia in young men. It struckof schizophrenia in young men. It struck
me that, with mild but undue frequency,me that, with mild but undue frequency,
there was a tendency for their parents’ agesthere was a tendency for their parents’ ages
to be unusual in one of two ways – eitherto be unusual in one of two ways – either
by there being aby there being a 4410-year age difference10-year age difference
in the couple, or by the mother being olderin the couple, or by the mother being older
than the father. In decades of practice since,than the father. In decades of practice since,
my impression has remained that this asso-my impression has remained that this asso-
ciation with schizophrenia occurs a littleciation with schizophrenia occurs a little
too often to be accidental. Of course, totoo often to be accidental. Of course, to
prove that would have required time,prove that would have required time,
money, thousands of cases, and the inclina-money, thousands of cases, and the inclina-
tion to undertake a major statistical enter-tion to undertake a major statistical enter-
prise, and none of those was in my reach.prise, and none of those was in my reach.
It is therefore gratifying now to findIt is therefore gratifying now to find
that, at long last, my hypothesis has beenthat, at long last, my hypothesis has been
solidly supported, albeit inadvertently, bysolidly supported, albeit inadvertently, by
ZammitZammit et alet al (2003). They demonstrate,(2003). They demonstrate,
in a 26-year follow-up of some 50 000 teen-in a 26-year follow-up of some 50 000 teen-
agers, that advancing paternal age is a riskagers, that advancing paternal age is a risk
factor for schizophrenia, while maternalfactor for schizophrenia, while maternal
age is not – the latter being a significantage is not – the latter being a significant
negative finding to which, however, theynegative finding to which, however, they
pay no further attention. Since this meanspay no further attention. Since this means
that, compared with the normal popula-that, compared with the normal popula-
tion, people with schizophrenia tend totion, people with schizophrenia tend to
have fathers who are older but mothershave fathers who are older but mothers
who are not, it follows necessarily that thewho are not, it follows necessarily that the
age difference between the parents alsoage difference between the parents also
tends to be greater than in the generaltends to be greater than in the general
population.population.
This does away with ZammitThis does away with Zammit et alet al’s’s
hypothesis that advancing paternal age ishypothesis that advancing paternal age is
pathogenic for schizophrenia by virtue ofpathogenic for schizophrenia by virtue of
increasing germ cell mutations. There isincreasing germ cell mutations. There is
no need to invoke genetic mutation withno need to invoke genetic mutation with
age, given the linkage they have uncovered,age, given the linkage they have uncovered,
in passing, between parental age differencein passing, between parental age difference
and schizophrenia. A more economicaland schizophrenia. A more economical
hypothesis is that to be born to a statisticallyhypothesis is that to be born to a statistically
off-centre parental couple is a risk factor foroff-centre parental couple is a risk factor for
schizophrenia – or, in more ordinary lan-schizophrenia – or, in more ordinary lan-
guage, there is some psychological risk inguage, there is some psychological risk in
being the child of an odd couple.being the child of an odd couple.
Are there other social oddities waitingAre there other social oddities waiting
to be identified statistically in schizophre-to be identified statistically in schizophre-
nogenic couples?nogenic couples?
Zammit, S., Allebeck, P., Dalman,C.,Zammit, S., Allebeck, P., Dalman,C., et alet al (2003)(2003)Paternal age and risk for schizophrenia.Paternal age and risk for schizophrenia. British Journal ofBritish Journal ofPsychiatryPsychiatry,, 183183, 405^408., 405^408.
H. BourneH. Bourne Via P.De Cristofaro 40, 00136 Roma,Via P.De Cristofaro 40, 00136 Roma,ItalyItaly
Authors’ reply:Authors’ reply: Dr Bourne suggests that asDr Bourne suggests that as
advancing paternal, but not maternal ageadvancing paternal, but not maternal age
is associated with schizophrenia, thenis associated with schizophrenia, then
people with schizophrenia tend to havepeople with schizophrenia tend to have
fathers who are older than the normalfathers who are older than the normal
population, but mothers who are not. Thispopulation, but mothers who are not. This
is incorrect. In our study, as others haveis incorrect. In our study, as others have
previously shown, advancing maternalpreviously shown, advancing maternal
ageage isis associated with schizophrenia, butassociated with schizophrenia, but
this association can be explained bythis association can be explained by
paternal age, a consequence of the fact thatpaternal age, a consequence of the fact that
there is strong correlation between parentalthere is strong correlation between parental
ages.ages.
Dr Bourne makes an interesting point,Dr Bourne makes an interesting point,
however, based on his observations in clin-however, based on his observations in clin-
ical practice that large differences in paren-ical practice that large differences in paren-
tal ages may result in some sort oftal ages may result in some sort of
psychological risk factor for schizophreniapsychological risk factor for schizophrenia
in the offspring. In fact, the absolute differ-in the offspring. In fact, the absolute differ-
ence between parental ages in our study isence between parental ages in our study is
associated with schizophrenia in the crudeassociated with schizophrenia in the crude
analysis, but this association is eliminatedanalysis, but this association is eliminated
after adjusting for the effects of paternalafter adjusting for the effects of paternal
age (Table 1). As paternal age increases,age (Table 1). As paternal age increases,
5 4 05 4 0
CORRESPONDENCECORRESPONDENCE
the difference between maternal and pater-the difference between maternal and pater-
nal ages must also increase given thenal ages must also increase given the
biological age threshold for motherhood.biological age threshold for motherhood.
However, in younger fathers with olderHowever, in younger fathers with older
mothers, even large differences in parentalmothers, even large differences in parental
ages is not associated with increasing riskages is not associated with increasing risk
of schizophrenia. In contrast, the associa-of schizophrenia. In contrast, the associa-
tion between advancing paternal age andtion between advancing paternal age and
risk of developing schizophrenia is notrisk of developing schizophrenia is not
altered by adjusting for parental differences.altered by adjusting for parental differences.
The hypothesis of increasing germ cell mu-The hypothesis of increasing germ cell mu-
tations remains the most likely explanationtations remains the most likely explanation
for this association between advancingfor this association between advancing
paternal age and risk of schizophrenia.paternal age and risk of schizophrenia.
S. ZammitS. Zammit Department of PsychologicalDepartment of PsychologicalMedicine,University of Wales College of Medicine,Medicine,University of Wales College of Medicine,Heath Park,Cardiff CF14 4XN,UKHeath Park,Cardiff CF14 4XN,UK
P. AllebeckP. Allebeck Department of Social Medicine,Department of Social Medicine,Gothenburg University, SwedenGothenburg University, Sweden
C. DalmanC. Dalman Psychiatric Epidemiology, StockholmPsychiatric Epidemiology, StockholmCentre of Public Health, SwedenCentre of Public Health, Sweden
I. Lundberg,T.HemmingsonI. Lundberg,T.Hemmingson Department ofDepartment ofPublic Health Sciences,Karolinska Institute,Public Health Sciences,Karolinska Institute,Stockholm, SwedenStockholm, Sweden
M. J.OwenM. J.Owen Department of PsychologicalDepartment of PsychologicalMedicine,University of Wales College of Medicine,Medicine,University of Wales College of Medicine,Cardiff,UKCardiff,UK
G. LewisG. Lewis Division of Psychiatry,University ofDivision of Psychiatry,University ofBristol,Bristol,UKBristol,Bristol,UK
Physical illness and schizophreniaPhysical illness and schizophrenia
I read with interest the report by McCrea-I read with interest the report by McCrea-
die (2003), which concludes that thedie (2003), which concludes that the
lifestyle of people with schizophrenia mustlifestyle of people with schizophrenia must
give cause for concern in relation to coron-give cause for concern in relation to coron-
ary heart disease. Despite being at anary heart disease. Despite being at an
increased risk of developing various physi-increased risk of developing various physi-
cal health problems, the detection rate andcal health problems, the detection rate and
treatment of physical illness among peopletreatment of physical illness among people
with mental illness is very poor (Koranwith mental illness is very poor (Koran etet
alal, 1989). The reasons why this vulnerable, 1989). The reasons why this vulnerable
group of patients do not receive the physi-group of patients do not receive the physi-
cal health care they deserve are manifoldcal health care they deserve are manifold
and need to be addressed. They range fromand need to be addressed. They range from
physical symptoms being misinterpreted asphysical symptoms being misinterpreted as
part of psychiatric illness by professionals,part of psychiatric illness by professionals,
to poor social skills, lack of motivation,to poor social skills, lack of motivation,
cognitive impairment and social isolationcognitive impairment and social isolation
occurring as part of mental illness makingoccurring as part of mental illness making
individuals with schizophrenia less likelyindividuals with schizophrenia less likely
to report symptoms and adhere to treat-to report symptoms and adhere to treat-
ment. When they do present themselves,ment. When they do present themselves,
their lack of social skills and the stigma oftheir lack of social skills and the stigma of
mental illness may also make it less likelymental illness may also make it less likely
that they receive good care (Phelanthat they receive good care (Phelan et alet al,,
2001).2001).
Services focusing on lifestyle changesServices focusing on lifestyle changes
geared to the particular needs of peoplegeared to the particular needs of people
with severe mental illness should bewith severe mental illness should be
planned. Periodic medical reviews by gener-planned. Periodic medical reviews by gener-
al practitioners using essential checklistsal practitioners using essential checklists
should be mandatory. Inability to clearlyshould be mandatory. Inability to clearly
appreciate or describe a medical problem,appreciate or describe a medical problem,
compounded by a reluctance to discusscompounded by a reluctance to discuss
such problems, contributes to the lack ofsuch problems, contributes to the lack of
attention to medical problems in patientsattention to medical problems in patients
with schizophrenia. Thorough, routinewith schizophrenia. Thorough, routine phy-phy-
sical examination whenever a patient issical examination whenever a patient is
seen is the best way forward but it is doubt-seen is the best way forward but it is doubt-
ful whether psychiatric services have theful whether psychiatric services have the
resources and time to implement this. It isresources and time to implement this. It is
necessary for a medical orientation on thenecessary for a medical orientation on the
part of psychiatrists while evaluating allpart of psychiatrists while evaluating all
patients. Refresher training should be regu-patients. Refresher training should be regu-
larly provided for psychiatrists and keylarly provided for psychiatrists and key
members of multidisciplinary communitymembers of multidisciplinary community
psychiatric teams. This could encompasspsychiatric teams. This could encompass
elements of detection, management andelements of detection, management and
preventive counselling (Lambertpreventive counselling (Lambert et alet al,,
2003). To ensure appropriate care for2003). To ensure appropriate care for
comorbid medical problems there shouldcomorbid medical problems there should
be active efforts on the part of generalbe active efforts on the part of general
practitioners as well as mental healthpractitioners as well as mental health
teams.teams.
Koran, L. M., Sox,H. C. Jr, Marton, K. I.,Koran, L. M., Sox,H. C. Jr, Marton, K. I., et alet al (1989)(1989)Medical evaluation of psychiatric patients. I. Results in aMedical evaluation of psychiatric patients. I. Results in astate mental health system.state mental health system. Archives of GeneralArchives of GeneralPsychiatryPsychiatry,, 4646, 733^740., 733^740.
Lambert,T. J. R.,Velakoulis, D. & Pantelis, C. (2003)Lambert,T. J. R.,Velakoulis, D. & Pantelis, C. (2003)Medical comorbidity in schizophrenia.Medical comorbidity in schizophrenia. Medical Journal ofMedical Journal ofAustraliaAustralia,, 178178 (suppl. 5), S67^S70.(suppl. 5), S67^S70.
McCreadie, R. G. (2003)McCreadie, R. G. (2003) Diet, smoking andDiet, smoking andcardiovascular risk in people with schizophrenia.cardiovascular risk in people with schizophrenia.Descriptive study.Descriptive study. British Journal of PsychiatryBritish Journal of Psychiatry,, 183183,,534^539.534^539.
Phelan, M., Stradins, L. & Morrison, S. (2001)Phelan, M., Stradins, L. & Morrison, S. (2001)Physical health of people with severe mental illness.Physical health of people with severe mental illness. BMJBMJ,,322322, 443^444., 443^444.
C.T. Sudhir KumarC.T. Sudhir Kumar The Maudsley Hospital,The Maudsley Hospital,Denmark Hill, London SE5 8AZ,UKDenmark Hill, London SE5 8AZ,UK
Antidepressant effects of repetitiveAntidepressant effects of repetitivetranscranial magnetic stimulationtranscranial magnetic stimulation
The report by MartinThe report by Martin et alet al (2003) seems(2003) seems
in conflict with previous meta-analysesin conflict with previous meta-analyses
ofof repetitive transcranial magnetic stimula-repetitive transcranial magnetic stimula-
tion (rTMS) (Holtzheimertion (rTMS) (Holtzheimer et alet al, 2001;, 2001;
McNamaraMcNamara et alet al, 2001; Burt, 2001; Burt et alet al, 2002)., 2002).
We wish to provide a broader context forWe wish to provide a broader context for
interpreting these results.interpreting these results.
The analysis by MartinThe analysis by Martin et alet al waswas
designed to minimise type 1 error – to iden-designed to minimise type 1 error – to iden-
tify the level of confidence that can betify the level of confidence that can be
placed in purported antidepressant effectsplaced in purported antidepressant effects
of rTMS. It combined only studies withof rTMS. It combined only studies with
similar methodologies, included only stu-similar methodologies, included only stu-
dies that met high standards of randomis-dies that met high standards of randomis-
ation and blinding, and analysed onlyation and blinding, and analysed only
end-point depression ratings (rather thanend-point depression ratings (rather than
analysing change scores or controlling foranalysing change scores or controlling for
baseline depression severity). With thisbaseline depression severity). With this
approach, the review found a statisticallyapproach, the review found a statistically
significant effect size for high-frequencysignificant effect size for high-frequency
((441 Hz) rTMS applied to the left prefron-1 Hz) rTMS applied to the left prefron-
tal cortex (tal cortex (770.35, 95% CI0.35, 95% CI 770.66 to0.66 to
770.04,0.04, PP¼0.03), but did not find evidence0.03), but did not find evidence
that antidepressant effects werethat antidepressant effects were clinicallyclinically
significant or that they persisted over time.significant or that they persisted over time.
The other meta-analyses attempted toThe other meta-analyses attempted to
minimise type 2 error – to identify whetherminimise type 2 error – to identify whether
there is reason to believe that rTMS mightthere is reason to believe that rTMS might
have significant antidepressant propertieshave significant antidepressant properties
warranting further investigation. Theywarranting further investigation. They
combined studies with different methodolo-combined studies with different methodolo-
giesgies and calculated effect sizes based onand calculated effect sizes based on
changes in depression severity over time.changes in depression severity over time.
Such a technique can be important whenSuch a technique can be important when
analysing studies where different treatmentanalysing studies where different treatment
arms may start at different baselines. Usingarms may start at different baselines. Using
these analytic techniques, priorthese analytic techniques, prior meta-meta-
analysesanalyses found effect sizes for high-found effect sizes for high-
frequency,frequency, left prefrontal rTMS rangingleft prefrontal rTMS ranging
from 0.5 to 0.8, suggesting that rTMS doesfrom 0.5 to 0.8, suggesting that rTMS does
5 415 41
Table 1Table 1 Crude and adjusted odds ratios (ORs) and 95% confidence intervals (95% CIs) for developingCrude and adjusted odds ratios (ORs) and 95% confidence intervals (95% CIs) for developing
schizophrenia according to maternal^paternal age differenceschizophrenia according to maternal^paternal age difference
Parental age difference (years)Parental age difference (years) Number inNumber in
cohort (%)cohort (%)
Number withNumber with
schizophrenia (%)schizophrenia (%)
Crude ORCrude OR
(95% CI)(95% CI)
Adjusted ORAdjusted OR
(95% CI)(95% CI)11
0^10^122 10 757 (23)10757 (23) 64 (0.59)64 (0.59) 1.01.0 1.01.0
2^32^3 12 711 (27)12 711 (27) 85 (0.67)85 (0.67) 1.1 (0.8^1.6)1.1 (0.8^1.6) 1.1 (0.8^1.5)1.1 (0.8^1.5)
4^54^5 9345 (20)9345 (20) 69 (0.74)69 (0.74) 1.2 (0.9^1.7)1.2 (0.9^1.7) 1.2 (0.8^1.7)1.2 (0.8^1.7)
6^96^9 9260 (20)9260 (20) 77 (0.83)77 (0.83) 1.4 (1.0^2.0)1.4 (1.0^2.0) 1.2 (0.9^1.8)1.2 (0.9^1.8)
10^4710^47 4332 (10)4332 (10) 40 (0.92)40 (0.92) 1.6 (1.0^2.3)1.6 (1.0^2.3) 1.2 (0.8^1.9)1.2 (0.8^1.9)
Linear trend across categoriesLinear trend across categories 1.12 (1.03^1.21)1.12 (1.03^1.21) 1.06 (0.96^1.16)1.06 (0.96^1.16)33
1. Adjusted for paternal age.1. Adjusted for paternal age.2. Baseline comparison group.2. Baseline comparison group.3.3. PP¼0.249.0.249.
CORRESPONDENCECORRESPONDENCE
indeed haveindeed have statisticallystatistically significant antide-significant antide-
pressant effects. However, these analysespressant effects. However, these analyses
all agree that theall agree that the clinicalclinical significance ofsignificance of
these effects is not yet established.these effects is not yet established.
The results of the MartinThe results of the Martin et alet al reviewreview
do not suggest at all that rTMS has no anti-do not suggest at all that rTMS has no anti-
depressant effects. On the contrary, thisdepressant effects. On the contrary, this
methodologically rigorous review identifiesmethodologically rigorous review identifies
statistically (but not clinically) significant,statistically (but not clinically) significant,
short-term antidepressant effects for 2short-term antidepressant effects for 2
weeks of high-frequency, left prefrontalweeks of high-frequency, left prefrontal
rTMS and recommends further studies torTMS and recommends further studies to
establish efficacy and identify optimalestablish efficacy and identify optimal
parameters. Even more importantly,parameters. Even more importantly,
numerous studies have shown that rTMSnumerous studies have shown that rTMS
alters brain functioning, with effectsalters brain functioning, with effects
ranging from altered gene expression inranging from altered gene expression in
animals to modified cerebral perfusion inanimals to modified cerebral perfusion in
humans; in many cases, these effects arehumans; in many cases, these effects are
very similar to those seen with establishedvery similar to those seen with established
antidepressant treatments.antidepressant treatments.
With these points in mind, we offer theWith these points in mind, we offer the
following recommendations to help guidefollowing recommendations to help guide
use of rTMS in clinical and researchuse of rTMS in clinical and research
settings.settings.
(a)(a) Given the small clinical effects seenGiven the small clinical effects seen
with rTMS in studies to date, it doeswith rTMS in studies to date, it does
not seem that rTMS is appropriate fornot seem that rTMS is appropriate for
widespread clinical use at this time.widespread clinical use at this time.
(b)(b) Large, multi-site trials are warranted toLarge, multi-site trials are warranted to
clarify the antidepressant effects ofclarify the antidepressant effects of
rTMS.rTMS.
(c)(c) Future studies of rTMS should incorpo-Future studies of rTMS should incorpo-
rate several improvements in studyrate several improvements in study
design, including appropriate (anddesign, including appropriate (and
well-documented) randomisation, ade-well-documented) randomisation, ade-
quate blinding of subjects and thera-quate blinding of subjects and thera-
pists (probably requiring an improvedpists (probably requiring an improved
sham condition), and better assessmentsham condition), and better assessment
of the duration of any antidepressantof the duration of any antidepressant
effects.effects.
(d)(d) More research should be directed atMore research should be directed at
clarifying which patient and treatmentclarifying which patient and treatment
characteristics might lead to greatercharacteristics might lead to greater
antidepressant effects with rTMS.antidepressant effects with rTMS.
(e)(e) More research should be directed atMore research should be directed at
identifying and testing potentialidentifying and testing potential
mechanisms by which rTMS producesmechanisms by which rTMS produces
antidepressant effects.antidepressant effects.
Declaration of interestDeclaration of interest
P.E.H. and D.A. have received researchP.E.H. and D.A. have received research
support from the US National Institute ofsupport from the US National Institute of
Mental Health. D.A. has received researchMental Health. D.A. has received research
support from Philips, Inc., belongs tosupport from Philips, Inc., belongs to
speakers’ bureaux at Cephalon, Wyethspeakers’ bureaux at Cephalon, Wyeth
Pharmaceuticals and Pfizer, and is aPharmaceuticals and Pfizer, and is a
member of consultancy or advisory boardsmember of consultancy or advisory boards
at Bristol-Myers-Squibb, Cyberonics, Eliat Bristol-Myers-Squibb, Cyberonics, Eli
Lilly, Inc., Forest Laboratories, Inc.,Lilly, Inc., Forest Laboratories, Inc.,
GlaxoSmithKline, Janssen PharmaceuticaGlaxoSmithKline, Janssen Pharmaceutica
Products, Inc., Neuronetics, Inc. and UBCProducts, Inc., Neuronetics, Inc. and UBC
Pharma, Inc. T.E.S. has received researchPharma, Inc. T.E.S. has received research
support from NV Organon, USA, Cybero-support from NV Organon, USA, Cybero-
nics, Inc. and Magstim, Inc., UK; belongsnics, Inc. and Magstim, Inc., UK; belongs
to speakers’ bureaux at NV Organon,to speakers’ bureaux at NV Organon,
USA, Eli Lilly and Company, SwitzerlandUSA, Eli Lilly and Company, Switzerland
and Pfizer, Inc., Switzerland; and is a mem-and Pfizer, Inc., Switzerland; and is a mem-
ber of advisory boards at Otsuka Pharma-ber of advisory boards at Otsuka Pharma-
ceuticals, USA and Janssen AG, Switzerland.ceuticals, USA and Janssen AG, Switzerland.
Burt,T., Lisanby, S.H. & Sackeim,H. A. (2002)Burt,T., Lisanby, S.H. & Sackeim,H. A. (2002)Neuropsychiatric applications of transcranial magneticNeuropsychiatric applications of transcranial magneticstimulation: a meta-analysis.stimulation: a meta-analysis. International Journal ofInternational Journal ofNeuropsychopharmacologyNeuropsychopharmacology,, 55, 73^103., 73^103.
Holtzheimer, P., Russo, J. & Avery, D. (2001)Holtzheimer, P., Russo, J. & Avery, D. (2001) Ameta-Ameta-analysis of repetitive transcranial magnetic stimulation inanalysis of repetitive transcranial magnetic stimulation inthe treatment of depression.the treatment of depression. PsychopharmacologyPsychopharmacologyBulletinBulletin,, 3535, 149^169., 149^169.
Martin, J. L. R., Barbanoj, M. J., Schlaepfer,T. E.,Martin, J. L. R., Barbanoj, M. J., Schlaepfer,T. E., etetalal (2003)(2003) Repetitive transcranial magnetic stimulationRepetitive transcranial magnetic stimulationfor the treatment of depression: systematic review andfor the treatment of depression: systematic review andmeta-analysis.meta-analysis. British Journal of PsychiatryBritish Journal of Psychiatry,, 182182, 480^491., 480^491.
McNamara, B., Ray, J. L., Arthurs, J.,McNamara, B., Ray, J. L., Arthurs, J., et alet al (2001)(2001)Transcranial magnetic stimulation for depression andTranscranial magnetic stimulation for depression andother psychiatric disorders.other psychiatric disorders. Psychological MedicinePsychological Medicine,, 3131,,1141^1146.1141^1146.
P. E.Holtzheimer, D. AveryP. E.Holtzheimer, D. Avery HarborviewHarborviewMedical Center,University of Washington MedicalMedical Center,University of Washington MedicalCenter,Department of Psychiatry and BehavioralCenter,Department of Psychiatry and BehavioralSciences, 325 Ninth Avenue,Box 359896, Seattle,Sciences, 325 Ninth Avenue,Box 359896, Seattle,WA 98104-2499,USAWA 98104-2499,USA
T. E. SchlaepferT. E. Schlaepfer Department of Psychiatry,Department of Psychiatry,University Hospital,Bonn,GermanyUniversity Hospital,Bonn,Germany
Evidence in cannabis researchEvidence in cannabis research
The article by CoffeyThe article by Coffey et alet al (2003) regarding(2003) regarding
adolescent precursors of cannabis depen-adolescent precursors of cannabis depen-
dence has a number of substantial problemsdence has a number of substantial problems
in the measures used, the analysis of datain the measures used, the analysis of data
and the reporting and discussion of theirand the reporting and discussion of their
findings. One of the study’s major findingsfindings. One of the study’s major findings
is that the ‘relationship between cannabisis that the ‘relationship between cannabis
dependence and persistent frequent drink-dependence and persistent frequent drink-
ing in adolescence changed direction, froming in adolescence changed direction, from
a risk association in the univariate modela risk association in the univariate model
to ato a protectiveprotective association in the adjustedassociation in the adjusted
model’ (Coffeymodel’ (Coffey et alet al, 2003: p. 333, empha-, 2003: p. 333, empha-
sis added). The use of the term protectivesis added). The use of the term protective
implies causality, rather than the negativeimplies causality, rather than the negative
correlation which more accurately portrayscorrelation which more accurately portrays
the statistical relationship. It also tacitly im-the statistical relationship. It also tacitly im-
plies a value judgement that heavy drinkingplies a value judgement that heavy drinking
is preferable to cannabis dependence.is preferable to cannabis dependence.
This study utilises logistic regression forThis study utilises logistic regression for
the majority of its statistical analysis with-the majority of its statistical analysis with-
out adequately considering some importantout adequately considering some important
caveats. First and foremost, as alreadycaveats. First and foremost, as already
mentioned, correlation does not equalmentioned, correlation does not equal
causality. This is particularly the casecausality. This is particularly the case
when there are a substantial number ofwhen there are a substantial number of
independent variables associated with theindependent variables associated with the
dependent variable. In the case of cannabisdependent variable. In the case of cannabis
use, as the authors point out, there areuse, as the authors point out, there are
many independent variables related to can-many independent variables related to can-
nabis use, such as socio-economic statusnabis use, such as socio-economic status
(not discussed), parental drug use patterns(not discussed), parental drug use patterns
(not discussed), antisocial behaviour, cigar-(not discussed), antisocial behaviour, cigar-
ette smoking and level of education, toette smoking and level of education, to
name a few that are known. Statistical textsname a few that are known. Statistical texts
(e.g. Gravetter & Wallnau, 1996) point out(e.g. Gravetter & Wallnau, 1996) point out
that to gain the best measure from the usethat to gain the best measure from the use
of logistic regression, there should be fewof logistic regression, there should be few
independent variables that are unrelatedindependent variables that are unrelated
to each other and that ‘a regression solutionto each other and that ‘a regression solution
is extremely sensitive to the combination ofis extremely sensitive to the combination of
variables that is included in it’ (Tabachnickvariables that is included in it’ (Tabachnick
& Fidell, 1996: p. 126).& Fidell, 1996: p. 126).
These issues are particularly concerningThese issues are particularly concerning
when such papers can be reported in thewhen such papers can be reported in the
mass media (as this study was) on a topicmass media (as this study was) on a topic
such as cannabis use, which generatessuch as cannabis use, which generates
strong public responses and is the forumstrong public responses and is the forum
for a great deal of misinformation andfor a great deal of misinformation and
manipulation of research results to suitmanipulation of research results to suit
political and ideological agendas. The sim-political and ideological agendas. The sim-
ple acknowledgement of study limitationsple acknowledgement of study limitations
would substantially improve the quality ofwould substantially improve the quality of
the debate surrounding such a complexthe debate surrounding such a complex
social, psychological and medical problem.social, psychological and medical problem.
Coffey,C., Carlin, J. B., Lynskey, M.,Coffey,C.,Carlin, J. B., Lynskey, M., et alet al (2003)(2003)Adolescent precursors of cannabis dependence: findingsAdolescent precursors of cannabis dependence: findingsfrom theVictorian Adolescent Health Cohort Study.from theVictorian Adolescent Health Cohort Study.British Journal of PsychiatryBritish Journal of Psychiatry,, 182182, 330^336., 330^336.
Gravetter, F. J. & Wallnau, L. B. (1996)Gravetter, F. J. & Wallnau, L. B. (1996) Statistics for theStatistics for theBehavioral Sciences: A First Course for Students ofBehavioral Sciences: A First Course for Students ofPsychology and EducationPsychology and Education (4th edn).Minneapolis, MN:(4th edn).Minneapolis,MN:West.West.
Tabachnick, B. G. & Fidell, L. S. (1996)Tabachnick, B. G. & Fidell, L. S. (1996) UsingUsingMultivariate StatisticsMultivariate Statistics (3rd edn).NewYork:(3rd edn).NewYork:HarperCollins.HarperCollins.
P.MillerP. Miller Faculty of Arts,Deakin University, andFaculty of Arts,Deakin University, andTurning Point Alcohol and Drug Centre, 54^62Turning Point Alcohol and Drug Centre, 54^62Gertrude Street, Fitzroy,Victoria 3065, Australia.Gertrude Street, Fitzroy,Victoria 3065, Australia.E-mail: millerpE-mail: millerp@@pipeline.com.aupipeline.com.au
The media response to CoffeyThe media response to Coffey et alet al (2003)(2003)
was predictable. ‘Anti-drug campaignerswas predictable. ‘Anti-drug campaigners
say new research, showing one in threesay new research, showing one in three
teenagers who smokes cannabis weekly be-teenagers who smokes cannabis weekly be-
comes hooked by their early 20s, provescomes hooked by their early 20s, proves
that it should not be treated as a ‘‘soft’’that it should not be treated as a ‘‘soft’’
drug. The shocking study found teens whodrug. The shocking study found teens who
used cannabis every week were at high riskused cannabis every week were at high risk
of addiction’ (Lawrence, 2003). Coffey isof addiction’ (Lawrence, 2003). Coffey is
quoted as saying, ‘The message here is thatquoted as saying, ‘The message here is that
5 4 25 4 2
CORRESPONDENCECORRESPONDENCE
cannabis is not as harmless as we hadcannabis is not as harmless as we had
thought earlier’ – an amazing conclusionthought earlier’ – an amazing conclusion
from a study where only 1% of the respon-from a study where only 1% of the respon-
dents identified as dependent reporteddents identified as dependent reported
social consequences of their use, while thesocial consequences of their use, while the
most prevalent symptom (10%) was persis-most prevalent symptom (10%) was persis-
tent desire. In everyday parlance, theytent desire. In everyday parlance, they
smoked because they liked it.smoked because they liked it.
Use of the very broad categorisations ofUse of the very broad categorisations of
the DSM is especially worrisome. Cliniciansthe DSM is especially worrisome. Clinicians
using these guidelines apply them to peopleusing these guidelines apply them to people
presenting with problems. The use of suchpresenting with problems. The use of such
categorisations in research, however, con-categorisations in research, however, con-
stitutes imprecise criteria to determine astitutes imprecise criteria to determine a
person’s dependence, resulting in theperson’s dependence, resulting in the
phenomenon being grossly overreported.phenomenon being grossly overreported.
Researchers have been able to generateResearchers have been able to generate
dependency by applying these same criteriadependency by applying these same criteria
to behaviours as diverse as jogging, shop-to behaviours as diverse as jogging, shop-
ping, sex, prayer and mountain climbing.ping, sex, prayer and mountain climbing.
In fact, these activities were found to beIn fact, these activities were found to be
as addictive as cannabis (Franklin, 1990).as addictive as cannabis (Franklin, 1990).
Problems include the disjunctive natureProblems include the disjunctive nature
of the criteria (dependency can be ascribedof the criteria (dependency can be ascribed
to two people with absolutely no symptomsto two people with absolutely no symptoms
in common), and the essentially subjectivein common), and the essentially subjective
way in which the characteristics areway in which the characteristics are
defined. The lack of specificity in the mea-defined. The lack of specificity in the mea-
surement of cannabis dependence results insurement of cannabis dependence results in
subjective measures being presented assubjective measures being presented as
objective and an over-reliance on the inter-objective and an over-reliance on the inter-
pretive framework brought to bear. Howpretive framework brought to bear. How
did the authors differentiate betweendid the authors differentiate between
‘wants’ and what DSM characterises as‘wants’ and what DSM characterises as
‘needs’? Was this differentiation communi-‘needs’? Was this differentiation communi-
cated to respondents? The study fails to dif-cated to respondents? The study fails to dif-
ferentiate respondents with no dysfunctionferentiate respondents with no dysfunction
associated with their dependence from thoseassociated with their dependence from those
with significant cannabis-related problems.with significant cannabis-related problems.
Finally, the only index of consumptionFinally, the only index of consumption
employed is frequency of use. This is mostemployed is frequency of use. This is most
unsatisfactory; a ‘smoke’ is not a standard-unsatisfactory; a ‘smoke’ is not a standard-
ised measure and the consequent lack ofised measure and the consequent lack of
any demonstrable association between tet-any demonstrable association between tet-
rahydrocannabinol consumption and therahydrocannabinol consumption and the
dependence syndrome begs the question,dependence syndrome begs the question,
dependent on what? Preparing a joint?dependent on what? Preparing a joint?
Inhaling deeply?Inhaling deeply?
Coffey,C.,Carlin, J. B., Lynskey, M.,Coffey,C., Carlin, J. B., Lynskey, M., et alet al (2003)(2003)Adolescent precursors of cannabis dependence: findingsAdolescent precursors of cannabis dependence: findingsfrom theVictorian Adolescent Health Cohort Study.from theVictorian Adolescent Health Cohort Study.British Journal of PsychiatryBritish Journal of Psychiatry,, 182182, 330^336., 330^336.
Franklin, D. (1990)Franklin, D. (1990) Hooked: not everyone becomesHooked: not everyone becomesaddicted.How come?addicted.How come? HealthHealth,, 44, 38., 38.
Lawrence, J. (2003)Lawrence, J. (2003) Teenagers addicted to pot.Teenagers addicted to pot. SundaySundayMailMail (Queensland), 6 April.(Queensland), 6 April.
T. PalmerT. Palmer Youth Substance Abuse Service, LevelYouth Substance Abuse Service, Level1/131Johnstone Street, Fitzroy,Victoria 3065,1/131Johnstone Street, Fitzroy,Victoria 3065,Australia. E-mail: tpalmerAustralia. E-mail: tpalmer@@ysas.org.auysas.org.au
Authors’ reply:Authors’ reply: In response to Dr Miller weIn response to Dr Miller we
would like to state some general principles,would like to state some general principles,
to clarify our methodology and provideto clarify our methodology and provide
some additional results. First, we have nosome additional results. First, we have no
argument with the truism that causalityargument with the truism that causality
cannot be inferred from correlation. Drcannot be inferred from correlation. Dr
Miller seems to overlook the fact that,Miller seems to overlook the fact that,
despite widespread awareness of the dan-despite widespread awareness of the dan-
gers of determining causality, the termsgers of determining causality, the terms
‘risk’ and ‘protective’ are commonly used‘risk’ and ‘protective’ are commonly used
to describe associations identified in longi-to describe associations identified in longi-
tudinal studies. Indeed, identifying andtudinal studies. Indeed, identifying and
interpreting such associations is theinterpreting such associations is the
primary reason for conducting cohortprimary reason for conducting cohort
studies. The reiteration of standard caveatsstudies. The reiteration of standard caveats
should not be necessary in every articleshould not be necessary in every article
arising from these studies and would makearising from these studies and would make
for very tedious reading indeed.for very tedious reading indeed.
The potential for inadequate control ofThe potential for inadequate control of
confounding by unmeasured or omittedconfounding by unmeasured or omitted
confounding factors is always a possibilityconfounding factors is always a possibility
in any multivariate analysis. Researchersin any multivariate analysis. Researchers
are inevitably constrained by the measuresare inevitably constrained by the measures
they have at their disposal which, in turn,they have at their disposal which, in turn,
result from the constraints of researchresult from the constraints of research
directions, design, responder burden anddirections, design, responder burden and
so on. Dr Miller criticises us for omittingso on. Dr Miller criticises us for omitting
socio-demographic measures while includ-socio-demographic measures while includ-
ing correlated behavioural measures. Ining correlated behavioural measures. In
terms of the former, we assessed theterms of the former, we assessed the
influence of both parental education andinfluence of both parental education and
metropolitan residence on cannabis depen-metropolitan residence on cannabis depen-
dence but as there was no evidence ofdence but as there was no evidence of
univariate associations for either measureunivariate associations for either measure
they were unlikely to be confounders (par-they were unlikely to be confounders (par-
ental education, reference group ‘someental education, reference group ‘some
tertiary’: completed secondary school ORtertiary’: completed secondary school OR
0.8 (95% CI 0.5–1.3); incomplete second-0.8 (95% CI 0.5–1.3); incomplete second-
ary OR 1.0 (95% CI 0.6–1.6); school inary OR 1.0 (95% CI 0.6–1.6); school in
metropolitan Melbourne: OR 1.0 (95%metropolitan Melbourne: OR 1.0 (95%
CI 0.6–1.5)). As they were uninformative,CI 0.6–1.5)). As they were uninformative,
these findings were omitted from the articlethese findings were omitted from the article
in the interests of parsimony and con-in the interests of parsimony and con-
serving space. As the report focused onserving space. As the report focused on
adolescent behavioural and mental healthadolescent behavioural and mental health
predictors of cannabis dependence, bothpredictors of cannabis dependence, both
parental substance use and peer substanceparental substance use and peer substance
use, although likely to be predictors, wereuse, although likely to be predictors, were
not considered relevant to the question.not considered relevant to the question.
Indeed, they were omitted from the analysisIndeed, they were omitted from the analysis
as their inclusion could have masked theas their inclusion could have masked the
associations of interest, exactly as Dr Millerassociations of interest, exactly as Dr Miller
describes.describes.
We acknowledge that confoundingWe acknowledge that confounding
occurred between some of the explanatoryoccurred between some of the explanatory
measures included in the multivariatemeasures included in the multivariate
analysis. We illustrated and discussed inanalysis. We illustrated and discussed in
some detail the confounding that occurredsome detail the confounding that occurred
between early-onset cannabis use, ciga-between early-onset cannabis use, ciga-
rrette smoking and antisocial behaviour.ette smoking and antisocial behaviour.
Furthermore, the interaction betweenFurthermore, the interaction between
problematic alcohol use and weekly canna-problematic alcohol use and weekly canna-
bis use to which Dr Miller objects arose asbis use to which Dr Miller objects arose as
post hocpost hoc examination of confounding.examination of confounding.
Mr Palmer misunderstands the denomi-Mr Palmer misunderstands the denomi-
nator of the reported symptom prevalences:nator of the reported symptom prevalences:
we described overall symptom prevalencewe described overall symptom prevalence
in the 1601 participants. Symptom preva-in the 1601 participants. Symptom preva-
lences in participants classified as beinglences in participants classified as being
cannabis dependent were reported in ancannabis dependent were reported in an
earlier publication and were: toleranceearlier publication and were: tolerance
17%, withdrawal 74%, unintentioned use17%, withdrawal 74%, unintentioned use
84%, persistent desire 91%, excessive time84%, persistent desire 91%, excessive time
spent obtaining, using or recovering fromspent obtaining, using or recovering from
use 74%, social consequences of use 18%use 74%, social consequences of use 18%
and continued use despite acknowledgedand continued use despite acknowledged
health problems 63% (Coffeyhealth problems 63% (Coffey et alet al, 2002)., 2002).
Furthermore, participants classified asFurthermore, participants classified as
dependent cannabis users reported compul-dependent cannabis users reported compul-
sive and out-of-control use more frequentlysive and out-of-control use more frequently
than those classified with dependent alco-than those classified with dependent alco-
hol use. That there is gathering evidencehol use. That there is gathering evidence
of social, physical and mental health harm,of social, physical and mental health harm,
including dependence, arising from long-including dependence, arising from long-
term cannabis use is now beyond debate.term cannabis use is now beyond debate.
For a brief and informative review of theFor a brief and informative review of the
current literature on this topic see Ashtoncurrent literature on this topic see Ashton
(2002).(2002).
Mr Palmer debates what really consti-Mr Palmer debates what really consti-
tutes cannabis dependence. That youngtutes cannabis dependence. That young
people ‘are smoking because they like it’people ‘are smoking because they like it’
does not preclude the possibility that theydoes not preclude the possibility that they
may be dependent. Alternatively, theymay be dependent. Alternatively, they
may be using it to stop feeling awful, inmay be using it to stop feeling awful, in
the self-medication paradigm. He quotesthe self-medication paradigm. He quotes
an assertion that other non-challengingan assertion that other non-challenging
behaviours performed persistently maybehaviours performed persistently may
also fit dependence criteria. This may bealso fit dependence criteria. This may be
so, but the harm that arises from theseso, but the harm that arises from these
activities is a moot point. The issue thatactivities is a moot point. The issue that
concerns us, and that we used the currentconcerns us, and that we used the current
gold standard instrument in populationgold standard instrument in population
research to identify, is that cannabis de-research to identify, is that cannabis de-
pendence inevitably prolongs heavy use.pendence inevitably prolongs heavy use.
No measure applied at interview can beNo measure applied at interview can be
considered to be completely sensitive andconsidered to be completely sensitive and
specific for all the reasons that Mr Palmerspecific for all the reasons that Mr Palmer
states but the unreferenced assertion thatstates but the unreferenced assertion that
the ‘phenomena [are] grossly overreported’the ‘phenomena [are] grossly overreported’
is unsupportable in the light of extensiveis unsupportable in the light of extensive
developmental and confirmatory workdevelopmental and confirmatory work
performed in treatmentperformed in treatment and non-treatmentand non-treatment
settings (e.g. Nelsonsettings (e.g. Nelson et alet al, 1999). We do, 1999). We do
not consider it a problem that individualsnot consider it a problem that individuals
can be classified as dependent withcan be classified as dependent with
different combinations of symptoms –different combinations of symptoms –
conversely, we need to increase ourconversely, we need to increase our
understanding of symptomunderstanding of symptom combinationscombinations
and their significance (Nelsonand their significance (Nelson et alet al,,
1999).1999).
5 4 35 4 3
CORRESPONDENCECORRESPONDENCE
The final point that Mr Palmer makes isThe final point that Mr Palmer makes is
to query the validity of our measure ofto query the validity of our measure of
cannabis use. He appears to have misreadcannabis use. He appears to have misread
the definition – we did not ask aboutthe definition – we did not ask about
‘smokes’ at all. We asked participants‘smokes’ at all. We asked participants
how often they ‘used cannabis’ without spe-how often they ‘used cannabis’ without spe-
cifying the method of delivery. We assumecifying the method of delivery. We assume
the word ‘used’ is unambiguous andthe word ‘used’ is unambiguous and
involves ingestion in some manner.involves ingestion in some manner.
Finally, we follow no political agendaFinally, we follow no political agenda
but seek only to inform the general publicbut seek only to inform the general public
and policy makers using sound epidemio-and policy makers using sound epidemio-
logical evidence resulting from good studylogical evidence resulting from good study
design, careful analysis and cautiousdesign, careful analysis and cautious
interpretation. Our article represents a stepinterpretation. Our article represents a step
towards filling the evidence void in thetowards filling the evidence void in the
current polarised debate about importantcurrent polarised debate about important
public health and policy issues surroundingpublic health and policy issues surrounding
cannabis use (Strangcannabis use (Strang et alet al, 2000)., 2000).
Ashton,H. (2002)Ashton,H. (2002) Cannabis or health?Cannabis or health? Current OpinionCurrent Opinionin Psychiatryin Psychiatry,, 1515, 247^253., 247^253.
Coffey,C.,Carlin, J. B., Degenhardt, L.,Coffey,C., Carlin, J. B., Degenhardt, L., et alet al (2002)(2002)Cannabis dependence in young adults: an AustralianCannabis dependence in young adults: an Australianpopulation study.population study. AddictionAddiction,, 9797, 187^194., 187^194.
Nelson,C. B., Rehm, J.,Uston,T. B.,Nelson,C. B., Rehm, J.,Uston,T. B., et alet al (1999)(1999)Factor structures for DSM^IV substance disorderFactor structures for DSM^IV substance disordercriteria endorsed by alcohol, cannabis, cocaine andcriteria endorsed by alcohol, cannabis, cocaine andopiate users: results from theWHO reliability andopiate users: results from theWHO reliability andvalidity study.validity study. AddictionAddiction,, 9494, 843^855., 843^855.
Strang, J.,Witton, J. & Hall,W. (2000)Strang, J.,Witton, J. & Hall,W. (2000) Improving theImproving thequality of the cannabis debate: defining the differentquality of the cannabis debate: defining the differentdomains.domains. BMJBMJ,, 320320, 108^110., 108^110.
C.CoffeyC. Coffey Centre for Adolescent Health, 2Centre for Adolescent Health, 2Gatehouse Street, Parkville 3052,Victoria, Australia.Gatehouse Street, Parkville 3052,Victoria, Australia.E-mail: coffeyE-mail: coffey@@cryptic.rch.unimelb.edu.aucryptic.rch.unimelb.edu.auJ. B. CarlinJ. B. Carlin Murdoch Children’s ResearchMurdoch Children’s ResearchInstitute and University of Melbourne, AustraliaInstitute and University of Melbourne, AustraliaM. LynskeyM. Lynskey Washington University, St Louis,Washington University, St Louis,Missouri,USAMissouri,USAG.C. PattonG.C. Patton Murdoch Children’s ResearchMurdoch Children’s ResearchInstitute, Parkville,Victoria, AustraliaInstitute, Parkville,Victoria, Australia
MRCPsych examsMRCPsych exams
I read with interest the informative editorialI read with interest the informative editorial
on the MRCPsych examination by Dr Tyreron the MRCPsych examination by Dr Tyrer
and Professor Oyebode (2004). I agree withand Professor Oyebode (2004). I agree with
the authors’ view that examinations requirethe authors’ view that examinations require
continuous assessment and refinement andcontinuous assessment and refinement and
also note their admission that politicalalso note their admission that political
and external factors are likely to driveand external factors are likely to drive
further changes.further changes.
However, I am still puzzled to noteHowever, I am still puzzled to note
their ambiguity over defining the directiontheir ambiguity over defining the direction
of change in the future. They give threeof change in the future. They give three
examples of potential future directions:examples of potential future directions:
modularisation of courses with assessmentmodularisation of courses with assessment
at the conclusion of modules; continuationat the conclusion of modules; continuation
of high-stakes tests; and regrading of theof high-stakes tests; and regrading of the
record of in-service training (RITA) as anrecord of in-service training (RITA) as an
exit examination at the completion ofexit examination at the completion of
higher specialist training. However, theirhigher specialist training. However, their
description of these examples is vague.description of these examples is vague.
This is an era of heightened societalThis is an era of heightened societal
expectations, increased regulatory controlexpectations, increased regulatory control
and external scrutiny of professionals.and external scrutiny of professionals.
There remains at least a theoreticalThere remains at least a theoretical
possibility of external quality assurancepossibility of external quality assurance
standards and mechanisms being imposedstandards and mechanisms being imposed
on the medical Royal Colleges, includingon the medical Royal Colleges, including
the Royal College of Psychiatrists.the Royal College of Psychiatrists.
Eraut (1994) has argued that a pro-Eraut (1994) has argued that a pro-
fessional’s competence has at least twofessional’s competence has at least two
dimensions,dimensions, scope and quality. Scope con-scope and quality. Scope con-
cerns what a person is competent in – thecerns what a person is competent in – the
range of roles, tasks and situations forrange of roles, tasks and situations for
which their competence is established orwhich their competence is established or
may be reliably inferred. Quality concernsmay be reliably inferred. Quality concerns
judgements on the quality of that workjudgements on the quality of that work
along a continuum. Determining the ac-along a continuum. Determining the ac-
ceptable and measurable cut-off pointsceptable and measurable cut-off points
on the quality dimension for senior houseon the quality dimension for senior house
officers, specialist registrars and consul-officers, specialist registrars and consul-
tants remains an important task for thetants remains an important task for the
profession.profession.
Schon (1987) has argued that if profes-Schon (1987) has argued that if profes-
sions are blamed for ineffectiveness andsions are blamed for ineffectiveness and
impropriety, their schools are blamed forimpropriety, their schools are blamed for
failing to teach the rudiments of effectivefailing to teach the rudiments of effective
and ethical practice. Greater emphasis onand ethical practice. Greater emphasis on
the processes of training, reflective practice,the processes of training, reflective practice,
training the trainers, continuing profes-training the trainers, continuing profes-
sional development, relevant educationalsional development, relevant educational
research and interprofessional learningresearch and interprofessional learning
would help to sustain and enhance thewould help to sustain and enhance the
profile of psychiatry in the society. The pro-profile of psychiatry in the society. The pro-
fession requires a clear direction from itsfession requires a clear direction from its
leaders.leaders.
Eraut, M. (1994)Eraut, M. (1994) Concepts of competence and theirConcepts of competence and theirlimitations. Inlimitations. In Developing Professional Knowledge andDeveloping Professional Knowledge andCompetenceCompetence, pp.163^181. London: Falmer., pp.163^181. London: Falmer.
Schon, D. A. (1987)Scho« n, D. A. (1987) Preparing professionals for thePreparing professionals for thedemands of practice. Indemands of practice. In Educating the ReflectiveEducating the ReflectivePractitionerPractitioner, pp.1^21. San Francisco,CA; London: Jossey-, pp.1^21. San Francisco,CA; London: Jossey-Bass.Bass.
Tyrer, S. & Oyebode, F. (2004)Tyrer, S. & Oyebode, F. (2004) Why does theWhy does theMRCPsych examination need to change?MRCPsych examination need to change? British JournalBritish Journalof Psychiatryof Psychiatry,, 184184, 197^199., 197^199.
R. A. FaruquiR. A. Faruqui Charing Cross Scheme & ImperialCharing Cross Scheme & ImperialCollege London,Charing Cross Hospital, FulhamCollege London,Charing Cross Hospital, FulhamPalace Road, LondonW6 8RF,UKPalace Road, LondonW6 8RF,UK
Authors’ reply:Authors’ reply:We have noted Dr Faruqui’sWe have noted Dr Faruqui’s
comments on our editorial. Dr Faruquicomments on our editorial. Dr Faruqui
believes we should be more specific aboutbelieves we should be more specific about
recommendations for psychiatry examina-recommendations for psychiatry examina-
tions in the future, and argues that we havetions in the future, and argues that we have
been ambiguous in not defining the formatbeen ambiguous in not defining the format
for future psychiatry examinations in morefor future psychiatry examinations in more
detail.detail.
The Royal College of Psychiatrists isThe Royal College of Psychiatrists is
not able independently to direct the coursenot able independently to direct the course
of examinations in the future. The Post-of examinations in the future. The Post-
graduate Medical Educational and Train-graduate Medical Educational and Train-
ing Board has indicated what principlesing Board has indicated what principles
should hold in postgraduate examinations,should hold in postgraduate examinations,
and the Royal College of Psychiatristsand the Royal College of Psychiatrists
follows these as well as observing thefollows these as well as observing the
practices of the other medical Royalpractices of the other medical Royal
Colleges.Colleges.
There is a move to include workplaceThere is a move to include workplace
assessments as part of the panoply of as-assessments as part of the panoply of as-
sessment of competence. The methods tosessment of competence. The methods to
achieve this have not yet been fully de-achieve this have not yet been fully de-
scribed or, indeed, evaluated. The degreescribed or, indeed, evaluated. The degree
to which this kind of assessment will formto which this kind of assessment will form
part of the assessment of a candidate in apart of the assessment of a candidate in a
future MRCPsych examination has notfuture MRCPsych examination has not
been made explicit.been made explicit.
This is the present state of affairs. WeThis is the present state of affairs. We
are not expressing our own opinions inare not expressing our own opinions in
this part of the editorial; we are indicat-this part of the editorial; we are indicat-
ing the present state of play. We believeing the present state of play. We believe
that competence is judged by publicthat competence is judged by public
examinations and that performance isexaminations and that performance is
measured by workplace assessments thatmeasured by workplace assessments that
approximate to what occurs in the realapproximate to what occurs in the real
world. Quality of work is not assessed inworld. Quality of work is not assessed in
examinations and we do not believe thatexaminations and we do not believe that
this is part of the remit of examinationthis is part of the remit of examination
boards.boards.
Declaration of interestDeclaration of interest
S.T. is the immediate past Chief Exam-S.T. is the immediate past Chief Exam-
iner of the Royal College of Psychia-iner of the Royal College of Psychia-
trists and F.O. is the present Chieftrists and F.O. is the present Chief
Examiner and is a member and exami-Examiner and is a member and exami-
ner of the Professional Licensing Assess-ner of the Professional Licensing Assess-
ment Board of the General Medicalment Board of the General Medical
Council.Council.
S.TyrerS.Tyrer University of Newcastle uponTyne,RoyalUniversity of Newcastle uponTyne,RoyalVictoria Infirmary,Newcastle uponTyne NE14LP,UKVictoria Infirmary,Newcastle uponTyne NE14LP,UK
F.OyebodeF.Oyebode South Birmingham Mental HealthSouth Birmingham Mental HealthNHS Trust,Queen Elizabeth Psychiatric Hospital,NHS Trust,Queen Elizabeth Psychiatric Hospital,MindelsohnWay, Edgbaston,Birmingham B15 2QZ,MindelsohnWay, Edgbaston,Birmingham B15 2QZ,UKUK
5 4 45 4 4