New
rating system for depression based on fuzzy
logic
The
current depression rating scales weight all
symptoms equally. They usually depend on
the expression of subjective feelings by the
patient. These feelings may be expressed
differently and therefore perceived differently
by the clinicians depending on a large number of
interacting factors including the symptoms
themselves. Two posters presented at the recent
ECNP meeting in Prague by Japanese researchers
proposed a new scale based on fuzzy logic and
multivariate theory to evaluate depressive
symptoms.
Six questions such as "Do you have a good
appetite?" were selected from the criteria for
major depressive episode in the DSM-IV. The
physician or patient answers by marking a visual
analogue scale VAS at the appropriate position.
The symptom severity is determined by the
distance from the end of the VAS corresponding
to the symptomatic state. The symptom range
is divided equally into three parts, good,
moderate and bad. The clinical weight of the six
symptoms and the categorical weight for each
degree (good, moderate and bad) were determined
using the data of the interval by multivariate
analysis in a group of 30 outpatients who met
DSM-IV criteria for major depressive disorder
(22 with moderate and 8 with mild severity).
Validity of this new evaluation tool was tested
in comparison with the Hamilton Depression
Rating Scale (HAM-D17) in 10
depressed patients treated with the serotonin
and noradrenaline reuptake inhibitor (SNRI),
milnacipran. Severity scores were evaluated
before treatment and after 10 and 30 days
treatment. According to the new scale, after 10
and 30 days of milnacipran treatment the
severity of depressive symptoms in all patients
had improved from moderate to mild. Using the
HAM-D, however, the symptoms were still
considered to be moderate in 6 patients after 10
days and on one patient after 30 days.
This study shows that it is possible to express
the clinical weight of each symptom of
depression and the categorical weight for each
degree. The mathematical analysis allows a more
precise estimation of the severity of symptoms.
The application of this new scale may lead to a
better evaluation of the evolution of depressive
symptoms during treatment. ECNP
Congress, Prague, September 2003, Posters
P.6.301, P.6.302