Content uploaded by Eric Leibing
Author content
All content in this area was uploaded by Eric Leibing
Content may be subject to copyright.
10.1192/bjp.190.1.80Access the most recent version at doi:
2007, 190:80.BJP
F. Leichsenring and E. Leibing
behavioural therapy for avoidant personality disorder−Cognitive
permissions
Reprints/
permissions@rcpsych.ac.ukwrite to
To obtain reprints or permission to reproduce material from this paper, please
to this article at
You can respond
http://bjp.rcpsych.org/cgi/eletter-submit/190/1/80
from
Downloaded
The Royal College of PsychiatristsPublished by
on September 19, 2011http://bjp.rcpsych.org/
http://bjp.rcpsych.org/site/subscriptions/
go to: The British Journal of PsychiatryTo subscribe to
Suicide trendsSuicide trends
and antidepressantsand antidepressants
The suicide rate in Sweden decreased byThe suicide rate in Sweden decreased by
25% during the 1990s. After analysing25% during the 1990s. After analysing
trends for the years 1978–96, I proposedtrends for the years 1978–96, I proposed
that the cause might be the concurrent in-that the cause might be the concurrent in-
creased use of antidepressants, and datacreased use of antidepressants, and data
from Norway, Denmark, Finland, Hungaryfrom Norway, Denmark, Finland, Hungary
and the USA supported this hypothesisand the USA supported this hypothesis
(Isacsson, 2000). However, naturalistic stu-(Isacsson, 2000). However, naturalistic stu-
dies do not allow definite conclusions,dies do not allow definite conclusions,
which is why the importance of testing thiswhich is why the importance of testing this
hypothesis in other studies was stressed.hypothesis in other studies was stressed.
ReselandReseland et alet al (2006) recently published(2006) recently published
an ‘extended’ (1961–2001) analysis of sui-an ‘extended’ (1961–2001) analysis of sui-
cide and the use of antidepressants in thecide and the use of antidepressants in the
four Nordic countries. They interpretedfour Nordic countries. They interpreted
two non-significant findings as ‘contrast-two non-significant findings as ‘contrast-
ing’ with my data: in Sweden and Den-ing’ with my data: in Sweden and Den-
mark, the decrease in suicide startedmark, the decrease in suicide started
before the introduction of selective seroto-before the introduction of selective seroto-
nin reuptake inhibitors (SSRIs); the suicidenin reuptake inhibitors (SSRIs); the suicide
rate in Norway later stabilised despite in-rate in Norway later stabilised despite in-
creased use of SSRIs.creased use of SSRIs.
However, the classification of deathsHowever, the classification of deaths
was changed in 1969 with the introductionwas changed in 1969 with the introduction
of the category of uncertain cause of deathof the category of uncertain cause of death
in ICD–8. Statistics before 1969 are there-in ICD–8. Statistics before 1969 are there-
fore not comparable with later data.fore not comparable with later data.
Furthermore, forensic pathologists onlyFurthermore, forensic pathologists only
gradually became accustomed to the newgradually became accustomed to the new
classification and the proportions of sui-classification and the proportions of sui-
cides to uncertain cases appear to have sta-cides to uncertain cases appear to have sta-
bilised first in 1979 (Fig. 1). Thus, thebilised first in 1979 (Fig. 1). Thus, the
decrease found by Reselanddecrease found by Reseland et alet al in ‘certainin ‘certain
suicides’ in 1969–79 may be an artefact. Asuicides’ in 1969–79 may be an artefact. A
better way of handling the uncertain casesbetter way of handling the uncertain cases
might be to add them to the certain suicidesmight be to add them to the certain suicides
(Linsley(Linsley et alet al, 2001). This would mean that, 2001). This would mean that
the Swedish suicide rates increased inthe Swedish suicide rates increased in
1969–79, decreased in 1979–89 and de-1969–79, decreased in 1979–89 and de-
creased rapidly in 1989–99. The ‘pre-SSRI’creased rapidly in 1989–99. The ‘pre-SSRI’
decrease in 1979–89 may be a result of thedecrease in 1979–89 may be a result of the
increased use of tricyclic antidepressants.increased use of tricyclic antidepressants.
Stabilisation in suicide rates is to be ex-Stabilisation in suicide rates is to be ex-
pected. Antidepressants cannot save peoplepected. Antidepressants cannot save people
who avoid doctors, who have treatment-who avoid doctors, who have treatment-
refractory depression, schizophrenia orrefractory depression, schizophrenia or
substance misuse from dying by suicide.substance misuse from dying by suicide.
I conclude that the data of ReselandI conclude that the data of Reseland etet
alal do not challenge the hypothesis that thedo not challenge the hypothesis that the
increased use of antidepressants is the causeincreased use of antidepressants is the cause
of the prominent decrease in suicide rateof the prominent decrease in suicide rate
since 1990. Moreover, some ten studiessince 1990. Moreover, some ten studies
provide strong evidence to support the hy-provide strong evidence to support the hy-
pothesis (Isacsson & Rich, 2005; Ludwigpothesis (Isacsson & Rich, 2005; Ludwig
& Marcotte, 2005).& Marcotte, 2005).
Isacsson, G. (2000)Isacsson, G. (2000) Suicide prevention ^ a medi calSui c ide preven tion ^ a medical
breakthrough?breakthrough? Acta Psych iatr i c a Scandinav i c aActa Psychiatr ica Scandinav ica,, 102102,113^,113^
117.117.
Isacsson, G. & Rich, C. L. (2005)Isacsson, G. & Rich, C. L. (2005) Antidepressant drugAntidepressant drug
use and su i ci de preven t ion .use and su i ci de preven t i on. International Review ofInternational Review of
PsychiatryPsychiatry,, 1717 ,153^162.,153^162.
Linsley, K. R., Schapira, K. & Kelly, T. P. (2001)Linsley, K. R., Schapira, K. & Kelly, T. P. (2 001) OpenOpen
ver d i ctver d i ct
v.v.
suicide ^ importance to researchsuicide ^ i mportance to resear ch Bri t ishBr i ti sh
Journal of PsychiatryJournal of Psychiatry,, 178178, 465^4 68., 4 65^4 68.
Ludwig,J.&Marcotte,D.E.(2005)Ludwig,J.&Marcotte,D.E.(2005)Anti-depressants,Anti-depressants,
suicide, and drug regulation.suicide, and drug regulation. Journal of Policy Analysis andJournal of Policy Analysis and
ManagementManagement,, 2424, 249^272.,249^272.
Reseland, S., Bray, I. & Gunnell, D. (2006)Reseland, S., Bray, I. & Gunnell, D. (2006)
Re l a tionship between an tidepressant sales and secul arRelationship between antidepressant sales and secular
trends in suicide in the Nordic count ries.trends in suicide in the Nordic countries. Br i ti sh Journal ofBr it ish Jou r nal of
PsychiatryPsychiatry,, 188188, 354^358.,354^358.
G. IsacssonG. Isacsson Huddinge University Hospital , M59 ,Huddinge University Hospital , M59 ,
S-141 86 Hudd i nge, Sweden .S-141 86 Hudd inge, Sweden.
Ema il:g oran .isacssonEmail:goran.i sacsson@@sll.sesll.se
doi: 10.1192/ bjp.19 0.1.79doi: 10.1192/bjp.190. 1.79
Author’ s reply:Author’s reply: Professor Isacsson raisesProfessor Isacsson raises
an important issue concerning the interpret-an important issue concerning the interpret-
ation of national suicide data before andation of national suicide data before and
after the introduction (in 1969) of a newafter the introduction (in 1969) of a new
classification of deaths, ‘injury undeter-classification of deaths, ‘injury undeter-
mined whether accidentally or purposelymined whether accidentally or purposely
inflicted’ (ICD–8). The points he raises doinflicted’ (ICD–8). The points he raises do
not, however, invalidate our conclusions.not, however, invalidate our conclusions.
The exclusion of pre-1969 or even pre-The exclusion of pre-1969 or even pre-
1979 (the period when the use of suicide1979 (the period when the use of suicide
and undetermined categories had stabilisedand undetermined categories had stabilised
in Sweden) data from our analyses does notin Sweden) data from our analyses does not
alter our main finding that suicide reduc-alter our main finding that suicide reduc-
tions in three of the four Nordic countriestions in three of the four Nordic countries
preceded the widespread use of SSRIs inpreceded the widespread use of SSRIs in
the early 1990s. With the exception ofthe early 1990s. With the exception of
Sweden, suicide rates continued to increase,Sweden, suicide rates continued to increase,
rather than decrease, in the period 1969–79rather than decrease, in the period 1969–79
in the Nordic countries, indicating that thein the Nordic countries, indicating that the
changed classification had a minor impactchanged classification had a minor impact
on apparent trends in these countries.on apparent trends in these countries.
There are well-recognised problemsThere are well-recognised problems
with interpreting ecological data to inferwith interpreting ecological data to infer
causal effects. Isacsson cites data from acausal effects. Isacsson cites data from a
number of countries where declines in sui-number of countries where declines in sui-
cide rates have coincided with increasedcide rates have coincided with increased
antidepressant prescribing. However, dataantidepressant prescribing. However, data
from other countries, such as England andfrom other countries, such as England and
Wales, Ireland and Italy, demonstrate theWales, Ireland and Italy, demonstrate the
opposite pattern (Gunnell & Ashby,opposite pattern (Gunnell & Ashby,
2004). Professor Isacsson suggests that the2004). Professor Isacsson suggests that the
reduction in suicide rate in Sweden inreduction in suicide rate in Sweden in
1979–89, prior to the use of SSRIs, may1979–89, prior to the use of SSRIs, may
be a result of the increased use of tricyclicbe a result of the increased use of tricyclic
antidepressants. This is possible, but dataantidepressants. This is possible, but data
from Norway suggest that increased usefrom Norway suggest that increased use
of non-SSRI antidepressants in the 1970sof non-SSRI antidepressants in the 1970s
and 1980s was associated with increasesand 1980s was associated with increases
in suicide rates.in suicide rates.
Isacsson suggests that the stabilisationIsacsson suggests that the stabilisation
in the decline in suicide rates is expected be-in the decline in suicide rates is expected be-
cause not all people with depression consultcause not all people with depression consult
doctors and conditions other than depres-doctors and conditions other than depres-
sion contribute to overall suicide numbers.sion contribute to overall suicide numbers.
7979
BRITISH JOURNAL OF PSYCHIATRYBRITISH JOURNAL OF PSYCHIATRY (2007), 190, 79^82(2007), 190, 79^82
CorrespondenceCorrespondence
EDIT ED BY KIRIAKOS XENIT IDIS an d COL IN C AMPB ELLEDIT ED BY KIRIAKOS XENIT IDIS an d COLIN CAMPBELL
ContentsContents
&&
Suicide trends and an tidepressantsSuicidetrends and antidepressants
&&
Cognitive^behaviouraltherapyCognitive^behaviouraltherapy
foravoidantpersonalitydisorderfor avoidantpersonali tydisorder
&&
Anti-phospholipidantibodies,antipsychotictreatmentAnti-phospholipidantibo dies,antipsychotic treatment
and cardiovascu lar morbidityand cardiovascu lar morbidity
&&
Letters to the EditorLetters to the Editor
AUTHOR ’ S P ROOFAUTHOR ’ S P ROOF
Fig. 1Fig. 1 Suicide rates in Sweden1969^2003. Percentages refer to the total (suicides + uncertain cases).Suicide rates in Sweden1969^2003. Percentages refer to the total (suicides + uncertain cases).
CORRESPONDENCECORRESPONDENCE
We agree with this analysis. Nevertheless,We agree with this analysis. Nevertheless,
the continued year-on-year rise in antide-the continued year-on-year rise in antide-
pressant use in the study period does indi-pressant use in the study period does indi-
cate a wider population of individuals,cate a wider population of individuals,
presumably some of whom are at risk ofpresumably some of whom are at risk of
suicide, being treated by these drugs.suicide, being treated by these drugs.
Our assessment of suicide and antide-Our assessment of suicide and antide-
pressant prescribing in the Nordic countriespressant prescribing in the Nordic countries
was more comprehensive than Isacsson’swas more comprehensive than Isacsson’s
original analysis and in our view providesoriginal analysis and in our view provides
weaker evidence than that originally pre-weaker evidence than that originally pre-
sented (Isacsson, 2000). Nevertheless thesented (Isacsson, 2000). Nevertheless the
most comprehensive assessment of the eco-most comprehensive assessment of the eco-
logical data to date (Ludwig & Marcotte,logical data to date (Ludwig & Marcotte,
2005) does support Isacsson’s view. In an2005) does support Isacsson’s view. In an
area where the influence of the pharmaceu-area where the influence of the pharmaceu-
tical industry is widespread we favour atical industry is widespread we favour a
more cautious interpretation of the ecologi-more cautious interpretation of the ecologi-
cal data.cal data.
Declaration of interestDecl a ration of interest
D.G. was an independent advisor to theD.G. was an independent advisor to the
Medicines and Healthcare Products Regu-Medicines and Healthcare Products Regu-
latory Agency Expert Working Group onlatory Agency Expert Working Group on
the Safety of SSRIs, receiving expensesthe Safety of SSRIs, receiving expenses
and an attendance fee.and an attendance fee.
Gunnell, D. & Ashby, D. (2 004)Gunnell, D . & Ashby, D . (2004) Ant idepressan ts andAntidepressants and
suicide: what is the balance of benefit and harm?sui cide: what is the balance of benefit and harm? BMJBMJ,,
329329, 34^38., 34^38.
Isacsson, G. (2000)Isacsson, G. (2000) Suicide prevent ion ^ a medicalSui ci de preven t ion ^ a medical
breakthrough?breakthrough? Acta Psychiatr ica Scand inav i caActa Psychiatri ca Scand inav i ca,, 102102,113^,113^
117.117.
Ludwig, J. & Marcot te, D. F . (2005)Ludwig,J.&Marcotte,D.F.(2005)Anti-depressants,Anti-depressants,
suicide, and drug regulation .suicide, and drug regulation. Journal of Policy Analysis andJournal of Pol icy Analysis and
ManagementManagement,, 2424, 249^272., 249^272.
S. ReselandS. Reseland KD-G Cons ul ti n g,H osletoppen 56,K D-G Consulting, Hosletoppen 56,
1362 Hosle, Norway. Email: sreselan1362 Hosle, Norway. Email: sreselan@@on line .noonline.no
I. Bray, D. GunnellI. Bray, D. Gunnell Department of SocialDepartment of Social
Med i cine,Un iversi ty of Bristo l , Brist ol ,UK.Medicine, University of Bristol, Bristol , U K.
doi: 10.1192/bjp.190. 1.79adoi: 10.1192/bjp.190.1.79a
Cognitive^ behavioural therapyCognitive^behavioural therapy
for avoidant personality disorderfor avoidant personality disorder
EmmelkampEmmelkamp et alet al (2006) reported that cog-(2006) reported that cog-
nitive–behavioural therapy (CBT) wasnitive–behavioural therapy (CBT) was
more effective than brief dynamic therapymore effective than brief dynamic therapy
(BDT) for the treatment of avoidant per-(BDT) for the treatment of avoidant per-
sonality disorder. However, the study hassonality disorder. However, the study has
several methodological shortcomings.several methodological shortcomings.
In the BDT group it is not clear whetherIn the BDT group it is not clear whether
and to what extent a manualised treatmentand to what extent a manualised treatment
was realised. The article includes non-specificwas realised. The article includes non-specific
references to several psychodynamic manualsreferences to several psychodynamic manuals
and it is not clear what therapeutic proce-and it is not clear what therapeutic proce-
dures were actually carried out. Furthermore,dures were actually carried out. Furthermore,
no disorder-specific treatment manual wasno disorder-specific treatment manual was
used. In contrast, in the CBT group theused. In contrast, in the CBT group the
manual of Beck & Freeman (1990) formanual of Beck & Freeman (1990) for
avoidant personality disorder was applied.avoidant personality disorder was applied.
No data with regard to adherence and com-No data with regard to adherence and com-
petence were reported and thus it is notpetence were reported and thus it is not
clear whether both treatments were carriedclear whether both treatments were carried
out with equal competence.out with equal competence.
Besides the presence or absence of theBesides the presence or absence of the
diagnosis according to the Structural Clini-diagnosis according to the Structural Clini-
cal Interview for DSM–IV Axis II Disorderscal Interview for DSM–IV Axis II Disorders
(SCID–II) several self-report measures were(SCID–II) several self-report measures were
applied as ‘primary outcome measures’.applied as ‘primary outcome measures’.
However, the authors focus on a specificHowever, the authors focus on a specific
measure that they regarded as primary. Inmeasure that they regarded as primary. In
addition to other outcome measures,addition to other outcome measures,
EmmelkampEmmelkamp et alet al used the Personality Dis-used the Personality Dis-
order Belief Questionnaire (PDBQ; Arntzorder Belief Questionnaire (PDBQ; Arntz etet
alal, 2004). Arntz, 2004). Arntz et alet al (2004) explicitly in-(2004) explicitly in-
cluded items from Beck & Freemancluded items from Beck & Freeman
(1990) and hence the PDBQ is specifically(1990) and hence the PDBQ is specifically
tailored to the effects of CBT. Possibly thetailored to the effects of CBT. Possibly the
most convincing difference between CBTmost convincing difference between CBT
and BDT was found with regard to theand BDT was found with regard to the
number of patients still fulfilling thenumber of patients still fulfilling the
SCID–II criteriaSCID–II criteria at follow-up (9at follow-up (9 v.v. 36%).36%).
However, it is notHowever, it is not clear whether the ‘inde-clear whether the ‘inde-
pendent assessor’ was masked to the treat-pendent assessor’ was masked to the treat-
ment group.ment group.
In two outcome measures that referIn two outcome measures that refer
more specifically to the features of avoidantmore specifically to the features of avoidant
personality disorder, the Social Phobiapersonality disorder, the Social Phobia
Anxiety Inventory (SPAI) and the Avoid-Anxiety Inventory (SPAI) and the Avoid-
ance Scale, another measure developed byance Scale, another measure developed by
the authors (Emmelkamp, 1982), boththe authors (Emmelkamp, 1982), both
CBT and BDT achieved large and nearlyCBT and BDT achieved large and nearly
identical pre-/post-treatment effect sizes:identical pre-/post-treatment effect sizes:
0.920.92 v.v. 0.82 (SPAI) and 1.880.82 (SPAI) and 1.88 v.v. 1.751.75
(Avoidance Scale). Emmelkamp(Avoidance Scale). Emmelkamp et alet al re-re-
ported that ‘CBT was significantly superiorported that ‘CBT was significantly superior
on all primary outcome measures.’ How-on all primary outcome measures.’ How-
ever, for the difference between the CBTever, for the difference between the CBT
and BDT groups in SPAI score theand BDT groups in SPAI score the PP waswas
0.09, which is not significant at the level0.09, which is not significant at the level
ofof aa¼0.01 set by the authors. Furthermore,0.01 set by the authors. Furthermore,
at follow-up, there were no differences be-at follow-up, there were no differences be-
tween CBT and BDT groups in SPAI andtween CBT and BDT groups in SPAI and
Avoidance Scale scores. Differences wereAvoidance Scale scores. Differences were
only reported for the PDBQ and for twoonly reported for the PDBQ and for two
scales that refer to other personality disor-scales that refer to other personality disor-
ders. For BDT, ‘no significant differenceders. For BDT, ‘no significant difference
was found between BDT and control’ butwas found between BDT and control’ but
no data are reported. Compared with theno data are reported. Compared with the
waiting list control, CBT was only superiorwaiting list control, CBT was only superior
in two of six measures but the sample sizein two of six measures but the sample size
of the waiting list control was smallof the waiting list control was small
((nn¼1515 v.v. 26 for CBT and 28 for BDT26 for CBT and 28 for BDT
post-treatment). The fact that almost nopost-treatment). The fact that almost no
differences were found between the waiting-differences were found between the waiting-
list control and both BDT and CBT is (at leastlist control and both BDT and CBT is (at least
in part) a result of the insufficient sample size.in part) a result of the insufficient sample size.
Furthermore, at least in some measures, theFurthermore, at least in some measures, the
waiting-list group achieved medium or evenwaiting-list group achieved medium or even
large effect sizes.large effect sizes.
The results reported by EmmelkampThe results reported by Emmelkamp etet
alal (2006) are at variance with those re-(2006) are at variance with those re-
ported by Svartbergported by Svartberg et alet al (2004), who(2004), who
found BDT and CBT to be equally effectivefound BDT and CBT to be equally effective
for cluster C personality disorders.for cluster C personality disorders.
Overall, the design, statistical analysesOverall, the design, statistical analyses
and reporting of the results raise seriousand reporting of the results raise serious
concerns about an investigator allegianceconcerns about an investigator allegiance
effect (Luborskyeffect (Luborsky et alet al, 1999)., 1999).
Arntz, A., Dreessen, L., Schouten, E.,Arntz, A., Dreessen, L., Schouten, E., et alet al (2004)(2004)
Beliefs in personality disorders: a test with theB eliefs in personality disorders: a test with the
personality disorder belief questionnaire.personality disorder belief questionnaire. BehaviorBehavior
Resear ch and The rapyResearch and Therapy,, 4242,1215^1225.,1215^12 2 5.
Beck, A. R. & Freeman, A. (1990)Beck, A. R. & Freeman, A. (1990) Cogn itive TherapyCogni tive Therapy
for Personality Disordersfor Personality Disorders. Guilford.. Guilford.
Emmelkamp, P. (1982)Emmelkamp, P. (1 982) Phobic and Obsessive^Phobic and Obsessi ve^
Compulsive Disorders:Theory, Research and PracticeCompulsive Disorders: Theory, Research and Practice..
Plenum.Plenum.
Emmelkamp, P. M. G., Benner, A. , Kuipers, A.,Emmelkamp, P. M. G., Benner, A., Kuipers, A. , et alet al
(2006)(2006)
Comparison of brief dynamic and cognitive^Compari son of brief dynami c and cogn i t i ve^
behavioural therapies in avoidant personality disorder.behavioural therapies in avoidant personality disorder.
British Journal of PsychiatryBritish Journal of Psychiatry,, 189189, 6 0^64., 6 0^64.
Luborsky, L., Diguer, L., Seligman, D. A.,Luborsky, L., Diguer, L., Seligman, D. A., et alet al (1999)(1999)
The researcher ’s own allegiances: ‘wi ld’card inThe researcher’s own allegiances:‘wild’card in
compari son of treatmen t efficacy.comparison of treatment efficacy. Clinical Psychology:Clinical Psychology:
Sci ence and Pr acti c eSc ienc e and Pr a cti ce,, 66,95^106.,95^106.
Svartberg, M., Stiles, T. & Seltzer, M. H. (20 04)Svartberg, M., Stiles, T. & Seltzer, M. H. (20 04)
Randomiz ed , con trolled trial of the effectiveness ofRandomized, controlled trial of the effectiveness of
short-term dynamic psychotherapy and cognitiveshort-term dynamic psychotherapy and cognitive
therapy for Cl ust e r C personal i ty disor ders.therapy for Cluster C personality disorders. AmericanAmerican
Journal of PsychiatryJournal of Psychiatry,, 161161, 810^817., 810^817.
F. LeichsenringF. Leichsenring Centre for PsychosocialCentre for Psychosocial
Med ici ne, von-S iebold-S trasse 5, D-37075Med i c ine, von-S i ebold-S t rasse 5,D-3 7075
Gotti ng en,Germany. Email: Fle i chsGo
«
t t i ngen,German y. Emai l: Flei chs@@gwdg.degwdg.de
E. LeibingE. Leibing Centre for Psychosocial Medicine,Centre for Psychosocial Medicine,
Gotti ngen ,Ger many.Go
«
t t i ngen,German y.
doi: 10.1192/bjp.190.1.80doi: 10.1192/bjp.190. 1.8 0
Author’s reply:Author ’s reply: Our study was designedOur study was designed
in close cooperation with full-time clini-in close cooperation with full-time clini-
cians and in both groups (CBT and BDT)cians and in both groups (CBT and BDT)
application of manuals was highly flexibleapplication of manuals was highly flexible
to be representative of the respective thera-to be representative of the respective thera-
pies as they are carried out in clinical prac-pies as they are carried out in clinical prac-
tice and to enhance the external validity oftice and to enhance the external validity of
the study. Sessions were audiotaped andthe study. Sessions were audiotaped and
scored using the Coding System of Thera-scored using the Coding System of Thera-
peutic Focus on Action and Insight (CFAI;peutic Focus on Action and Insight (CFAI;
SamoilovSamoilov et alet al, 2000) by two independent, 2000) by two independent
raters who were masked to the treatmentraters who were masked to the treatment
group (interrater reliability (Kendall’sgroup (interrater reliability (Kendall’s WW))
ranged from 0.86 to 0.91). In general, re-ranged from 0.86 to 0.91). In general, re-
sults revealed that therapists adhered tosults revealed that therapists adhered to
the respective therapies (Emmelkampthe respective therapies (Emmelkamp etet
alal, 2004)., 2004).
8080
AUTHOR ’ S P ROOFAUTHOR’S PROOF
CORRESPONDENCECORRESPONDENCE
To the best of our knowledge there areTo the best of our knowledge there are
no measures of ‘psychodynamic origin’ spe-no measures of ‘psychodynamic origin’ spe-
cifically related to avoidant personalitycifically related to avoidant personality
disorder and hence we used the PDBQ.disorder and hence we used the PDBQ.
Furthermore, it was not feasible to keepFurthermore, it was not feasible to keep
the independent assessors who completedthe independent assessors who completed
the SCID–II unaware of the treatmentthe SCID–II unaware of the treatment
group in a number of instances.group in a number of instances.
Post-treatment CBT was significantlyPost-treatment CBT was significantly
superior to BDT on all ‘primary’ outcomesuperior to BDT on all ‘primary’ outcome
measures. A significance level ofmeasures. A significance level of aa¼0.1 set0.1 set
rather than 0.01 as claimed by Leichsenringrather than 0.01 as claimed by Leichsenring
& Leibing. Even if we exclude the SPAI& Leibing. Even if we exclude the SPAI
scores (scores (PP¼0.09), this still leaves superior0.09), this still leaves superior
outcome for CBT on three out of four out-outcome for CBT on three out of four out-
come variables. The lack of power to detectcome variables. The lack of power to detect
differences between the waiting-list controldifferences between the waiting-list control
group and the active treatments is acknowl-group and the active treatments is acknowl-
edged as a limitation.edged as a limitation.
There are important differences be-There are important differences be-
tween our study and that of Svartbergtween our study and that of Svartberg etet
alal (2004). Svartberg(2004). Svartberg et alet al included all typesincluded all types
of cluster C and self-defeating personalityof cluster C and self-defeating personality
disorders, rather than limiting their studydisorders, rather than limiting their study
to avoidant personality disorder. Two-to avoidant personality disorder. Two-
fifths of their sample did not fulfil criteriafifths of their sample did not fulfil criteria
for avoidant personality disorder treatmentfor avoidant personality disorder treatment
and treatment consisted of 40 rather thanand treatment consisted of 40 rather than
of 20 sessions. Furthermore, outcome withof 20 sessions. Furthermore, outcome with
respect to personality disorders (sic) wasrespect to personality disorders (sic) was
only assessed with the Millon Clinicalonly assessed with the Millon Clinical
Multiaxial Inventory (Millon, 1994), ratherMultiaxial Inventory (Millon, 1994), rather
than with the gold standard SCID–II. Final-than with the gold standard SCID–II. Final-
ly, the lack of a control group in the studyly, the lack of a control group in the study
of Svartbergof Svartberg et alet al renders the results diffi-renders the results diffi-
cult to interpret.cult to interpret.
In contrast to most other psychotherapyIn contrast to most other psychotherapy
studies, we did our utmost to prevent an ef-studies, we did our utmost to prevent an ef-
fect of investigator allegiance. The studyfect of investigator allegiance. The study
was designed in close cooperation withwas designed in close cooperation with
two psychodynamic therapists (G.F. andtwo psychodynamic therapists (G.F. and
H.K.) and two cognitive–behavioural thera-H.K.) and two cognitive–behavioural thera-
pists (A.B. and A.K.), who all fully partici-pists (A.B. and A.K.), who all fully partici-
pated in the design of the study, selection ofpated in the design of the study, selection of
measures, treatment manuals (including de-measures, treatment manuals (including de-
gree of flexibility) and therapists.gree of flexibility) and therapists.
Emmelkamp, P. M. G., Grauwelman, I. & Rengers, L.Emmelkamp, P. M. G., Grauwelman, I. & Rengers, L.
(2004)(2004)
Onderzoek naar cognitieve gedragstherapie enOnderzoek naar cognitieve gedragstherapie en
psychodynamische therapie bij de ontwijkendeps ychod ynamische the rapi e bij de ontwijkende
persoonlijkheidsstoornis:De construct validiteit van depersoonli jkhe i dsstoor nis: De constru ct valid iteit van de
behandel ingen . [Research i nto CBTand BDTof thebehandelingen. [Research into CBTand BD T of the
avoidant personality disorder: the construct validity ofav oidant personality d i so r der: the constru ct validity of
the treatments ] Inthe treatments] In Psychoanalytische PsychotherapiePsychoanalytische Psychotherapie
Vergel ijkenderwi jsVergel ijkenderwi js (eds W. B.C. Hoenink,M. J. Rexwinkel(eds W. B.C.Hoenink,M. J. Rexwinkel
& W.Roelofsen), pp. 37^45. van Gorcum.& W.Roelofsen), pp. 37^45. van Gorcum.
Millon, T. (19 94)Millon, T. (19 94) Mi llon Clinical Multiaxial Inventory ^ IIIMil lon Cl inical Multiax ial Inventory ^ II I..
Dicandrien.Dicandrien.
Samoilov , A., Goldfried, M. R. & Shapiro, D. A.Samoilov, A., Goldfried, M. R. & Shapiro, D. A.
(2000)(2000)
Coding system of therapeutic focus on actionCodi ng system of therapeuti c focus on acti on
and insight.and insight. Journal of Consu lting and Clinical PsychologyJourna l of Consulting and Clinical Psychology,,
6868,513^514.,513^514.
P. M. G . Em m e l k a m pP. M . G . E m m e l k a m p Department of ClinicalDepartment of Clinical
Psychology,University of Amsterdam, RoetersstraatPsychology,University of Amsterdam, Roetersstraat
1 5 ,1018 WB Amst erdam ,The Netherlands. Emai l:1 5 , 1018 WB Amsterdam,The Netherlands. Email:
P.M.G.EmmelkampP.M.G.Emmelkamp@@uva.nluva.nl
doi: 10.1192/bjp.190. 1 .80adoi: 10. 1192/bjp.190. 1 .80a
Anti-phospholipid antibodies,Anti-phospholipid antibodies,
neuroleptic treatment andneuroleptic treatment and
cardiovascular morbiditycardiovascular morbidity
JoukamaaJoukamaa et alet al (2006) reported a clear re-(2006) reported a clear re-
lationship between the number of neurolep-lationship between the number of neurolep-
tic drugs prescribed and mortality of peopletic drugs prescribed and mortality of people
with schizophrenia. The more importantwith schizophrenia. The more important
causes of death were cardiovascular diseasecauses of death were cardiovascular disease
and unspecified respiratory disease. More-and unspecified respiratory disease. More-
over, the authors postulated that over-over, the authors postulated that over-
looked venous thrombosis or pulmonarylooked venous thrombosis or pulmonary
embolism accounted for some respiratoryembolism accounted for some respiratory
deaths.deaths.
OomenOomen et alet al (1995) documented in-(1995) documented in-
creased vascular morbidity at 2-year fol-creased vascular morbidity at 2-year fol-
low-up in patients with anti-phospholipidlow-up in patients with anti-phospholipid
antibodies who were newly admitted forantibodies who were newly admitted for
psychiatric treatment. These patientspsychiatric treatment. These patients
showed a range of cardiovascular accidentsshowed a range of cardiovascular accidents
(arterial or venous thrombosis, pulmonary(arterial or venous thrombosis, pulmonary
embolism and myocardial infarction). Theembolism and myocardial infarction). The
negative control group without anti-phos-negative control group without anti-phos-
pholipid antibodies had no vascular com-pholipid antibodies had no vascular com-
plications during follow-up.plications during follow-up.
Vascular events associated with suchVascular events associated with such
autoantibodies range from superficial toautoantibodies range from superficial to
life-threatening multiple organ thrombosislife-threatening multiple organ thrombosis
developing over a short period (‘cata-developing over a short period (‘cata-
strophic’ anti-phospholipid syndrome).strophic’ anti-phospholipid syndrome).
Thrombosis in anti-phospholipid syndromeThrombosis in anti-phospholipid syndrome
appears to be a ‘two-hit’ phenomenon.appears to be a ‘two-hit’ phenomenon.
Autoantibodies (the first ‘hit’) are continu-Autoantibodies (the first ‘hit’) are continu-
ally present in the circulation, yet a localally present in the circulation, yet a local
trigger (the second ‘hit’) is required to in-trigger (the second ‘hit’) is required to in-
duce thrombus formation. Erkan & Lock-duce thrombus formation. Erkan & Lock-
shin (2006) recently suggested theshin (2006) recently suggested the
elimination of reversible thrombosis riskelimination of reversible thrombosis risk
factors and heparin prophylaxis duringfactors and heparin prophylaxis during
high-risk periods in people with persistenthigh-risk periods in people with persistent
anti-phospholipid antibodies. Chengappaanti-phospholipid antibodies. Chengappa
et alet al (1991) and Schwartz(1991) and Schwartz et alet al (1998) de-(1998) de-
monstrated a high prevalence of anti-phos-monstrated a high prevalence of anti-phos-
pholipid antibodies (about 30%) inpholipid antibodies (about 30%) in
patients. A prospective study is ongoing inpatients. A prospective study is ongoing in
our departments to confirm the prevalenceour departments to confirm the prevalence
of anti-phospholipid antibodies with a firstof anti-phospholipid antibodies with a first
episode of acute psychosis before and afterepisode of acute psychosis before and after
neuroleptic treatment. If historical data areneuroleptic treatment. If historical data are
confirmed, more attention should be paidconfirmed, more attention should be paid
to the fact that up to one-third of patientsto the fact that up to one-third of patients
presenting with psychosis have anti-phos-presenting with psychosis have anti-phos-
pholipid antibodies and are at risk of cardi-pholipid antibodies and are at risk of cardi-
ovascular or respiratory morbidity/ovascular or respiratory morbidity/
mortality when neuroleptic treatment ormortality when neuroleptic treatment or
physical restraint are used.physical restraint are used.
Chengappa, K. N., Carpenter, A. B., Keshavan, M. S.Chengappa, K. N., Carpenter, A. B., Keshavan, M. S.
et alet al (19 91)(19 91)
Elevated IGG and IGM anticardiolipinElevated IGG and IGM anticardiolipin
antibodies in a subgroup of medicated and unmedicatedantibodies in a subgroup of medicated and unmedicated
shizophrenic patients.shizophrenic patients. Biological PsychiatryBiological Psychiatry,, 3030, 731^735., 731^735.
Erkan, D . & Lockshi n, M. D. (2006)Erkan, D. & Lockshin, M. D. (2006) Antiphospho lipidAntiphospholipid
syndrome.syndrome. Current Opinion in RheumatologyCurrent Opinion in Rheumatology,, 1818,242^248., 242 ^248.
Joukamaa, M., Heliovaara, M., Knekt, P.,Joukamaa, M., H elio
«
vaara, M., Knekt, P., et alet al (2006)(2006)
Schizophrenia, neuroleptic medication and mortality.Schizophrenia, neur o l epti c medi cation and morta lity.
British Journal of PsychiatryBritish Journal of Psychiatry,, 18 8188,122^127.,12 2^127.
Oomen, H. A.,Wekking, F. M., de Jong, J.,Oomen, H. A.,Wekking, F. M., de Jong, J., et alet al (19 9 5)(19 95)
Screening psychiatric admissions for anticardiolipinScreening psychiatri c admissions for ant icard iol ipi n
antibody.antibody. Psychiatry ResearchPsychiatry Research,, 5858,83^88.,83^88.
Schwartz, M. D., Rochas, M.,Weller, B.Schwartz,M.D.,Rochas,M.,Weller,B. et alet al (19 9 8)(19 9 8)
High association of anticardiolipin antibodies withH igh association of anticardiolipin antibodies with
psychosis.psychosis. Journal of Clinical PsychiatryJournal of Clinical Psychiatry,, 5959, 2 0^23., 20^2 3.
E. LeuciE. Leuci Psychiatry Department, Parma, Fidenza,Psychiatry Department, Parma, Fidenza,
Italy.Italy.
L. ManentiL. Manenti Psychiatry Department, FidenzaPsychiatry Department, Fidenza
District, ASL Parma,Via Berenini 1 51 , Fidenza (PR ),Distri ct, ASL Parma,Via Berenin i 1 5 1, Fidenza (PR),
43100 Italy. Email: lucio.manenti43100 Italy. Email: lucio.manenti@@aod.itaod.it
C. MagginiC. Maggini Psychiatry Department, Parma,Psychiatry Department, Parma,
Fidenza, Italy.Fidenza, Italy.
doi: 10.1192/ bjp.19 0.1. 81doi: 10.1192/bjp.190. 1 .81
Letters to the EditorLetters to the Editor
In order to speed up the publication ofIn order to speed up the publication of
correspondence and to encourage debatecorrespondence and to encourage debate
among our readers and authors, all lettersamong our readers and authors, all letters
to the Editor must from now on be sub-to the Editor must from now on be sub-
mitted online as eLetters. Hard-copy sub-mitted online as eLetters. Hard-copy sub-
missions will no longer be considered. Tomissions will no longer be considered. To
submit an eLetter, click ‘submit a response’submit an eLetter, click ‘submit a response’
in the box at the top right of the screenin the box at the top right of the screen
when viewing online an article on whichwhen viewing online an article on which
you wish to comment. If your letter is ayou wish to comment. If your letter is a
general one, and not in response to ageneral one, and not in response to a
specific article, please ‘submit a response’specific article, please ‘submit a response’
to this letter (you can also submit a generalto this letter (you can also submit a general
letter from the eLetters home page). Weletter from the eLetters home page). We
aim to publish eLetters online, if accepted,aim to publish eLetters online, if accepted,
within 10 days of submission. A selectionwithin 10 days of submission. A selection
of these letters will be included in subsequentof these letters will be included in subsequent
printed issues.printed issues.
Correspondence EditorsCorrespondence Editors
doi: 10.1192/ bjp.19 0.1. 81adoi: 10. 1192/bjp.190. 1 .81a
8181
AUTHOR ’ S P ROOFAUTHOR ’ S P ROOF
