Bech.qxp 30/6/08 02:16 Page 17
The Use of Rating Scales in Affective Disorders
quantitative response, for example in dose–response trials or plasma
concentration–response, only scales that have been accepted by item
response theory models (total score a sufficient statistic) are recommended.
XIV World Congress of Psychiatry
Self-rating Scales versus Clinician-administrated Scales 20–25 September 2008
The use of self-reported rating scales within affective disorders has
Prague Congress Centre, Czech Republic
become more and more important in depression. Scales such as the Beck
Depression Inventory (BDI) or the Zung Self-rating Depression Scale (SDS)
I would like to extend to you a warm invitation to join us at the
were previously often used as supplements to the HAM-D. However,
XIV World Congress of Psychiatry, taking place in Prague in
recently self-reported HAM-D versions have been published.
29
In order to
the autumn of 2008.
have a closer face validity with the ICD-10 or DSM-IV than that of the BDI,
the Major Depression Inventory (MDI) has been developed.
30
The self-
All of our 130 Member Societies and 60 Scientific Sections are
rated version of the HAM-D
expected to be present, along with a full representation of all
6
and the MDI fulfil the item response theory
models, i.e. the total score is a sufficient statistic.
of our partners in mental healthcare, from our patients (who
are at the centre of our work) to health professionals, health
The use of self-reported rating scales to measure manic states is limited
planners and relevant industry.
to hypomanic states because in more severely manic states the response
pattern reflects that the patients are playing ‘the manic game’. Recently,
You would not want to miss this special encounter, where new
patient-reported questionnaires have been published to identify previous
psychiatric findings and perspectives will be discussed, and where
episodes of hypomania in depressed patients in order to test for the
current partnerships will be strengthened and new ones forged.
existence of bipolar II disorder. It seems that the Hypomanic Checklist is
superior to the Mood Disorder Questionnaire in this respect.
31
The most
Professor Juan E Mezzich
comprehensive self-reported scale covering depression and hypomania
President, XIV World Congress of Psychiatry
(aggression) as well as the different anxiety disorders is the SCL-90.
13
From the SCL-90, a subscale analogue to the self-rated HAM-D
Topics include: addiction, affective disorders, animal models in
6
has
been derived containing the following items: feeling blue, blaming
psychiatry, anxiolytics, biological markers, clinical psycho-
yourself for things, feeling no interests in things, feeling that everything
pharmacology, emergency psychiatry, public health and
is an effort, worrying too much about things and feeling low in energy.
psychiatry, somatoform disorders, urban mental health,
Like the HAM-D
cultural psychiatry and molecular neurobiology.
6
, this depression SCL-90 subscale fulfils the items
response theory model and the items are included in the SCL-90 factor
solution by Lipman.
32
Honorary and Supervisory Committee includes: Jaroslav Blahos
(Chair, Czech Medical Association), Jan Bríza (Director, General
Standardisation of Rating Scales for Affective Disorders
Faculty Hospital, Prague), Milan Kubek (President, Czech
As mentioned above, the HAM-D was originally developed to measure a
Medical Chamber) and Mario Maj (Chair, Scientific Committee,
global impression of depressive states. The CGI-S
13
is probably the most
World Congress of Psychiatry). ■
frequently used global assessment of depressive or manic states. This
scale has a score range from one (not at all ill) to seven (among the most
Contact:
extremely ill patients suffering from depression or mania).
13
However,
E: wpa@guarant.cz
the inter-rater reliability of this CGI-S has not been found to be high,
33
T: +420 284 001 444
and in most trials with antidepressants a cut-off score on HAM-D
F: +420 284 001 448
17
of
≤7 has been used to define remission (which implies that the signs and
www.wpa-prague2008.cz
symptoms of illness are absent or almost absent). This cut-off score for
remission was introduced by Reisby et al.
34
and equals a CGI-S score of
one and two (minimally depressed/manic symptomatology). A was recently confirmed for bipolar depression by Vieta et al.
44
). For
modification of the CGI scales for use in bipolar illness (the CGI-BP) has unipolar depression a cut-off score on MADRS of nine or lower has
been suggested by Spearing et al.
33
The CGI-BP gives a severity rating of recently been suggested.
35
mania, depression and overall bipolar illness during treatment of an
acute episode as well as in long-term prophylaxis. The report by the American College of Neuropsychopharmacology
(ACNP) task force on response and remission in major depressive
The cut-off levels are based on both unipolar patients with depression disorders
35
suggests a special focus on the HAM-D symptoms of
and bipolar patients with depression and mania. When making such a depressed mood when defining remission or the core items of the
standardisation for remission and relapse with reference to CGI-S or scale being used. For manic episodes the definition of remission should
CGI-BP, all of the symptom rating scales in Table 5 and the CGI versions focus on the MAS or YMRS symptoms of elevated mood and/or
themselves refer to a global impression of illness. Therefore, the total hostility (destructive behaviour). In an analysis using the HAM-D item
score does not need to be a sufficient statistic. It is of interest that, while of depressed mood as index of validity, the results showed that
the HAM-D
17
remission score of seven or lower has been found to be a score of zero (depressed mood absent) corresponded to an HAM-D
17
adequate in most studies, it is still problematic for the MADRS to establish score of seven or less.
36
A score on depressed mood of one (doubtful
a consensus for the cut-off score of 12 (which is a rather high score but or minimal) corresponded to an HAM-D
17
score of 13 or less, while a
EUROPEAN PSYCHIATRIC REVIEW 17