Insight into schizophrenia: the effects of cognitive behavioural
therapy on the components of insight and association with
sociodemographics—data on a previously published
randomised controlled trial
3. Method
This has been fully described in the first publication
by this research group (Turkington et al., 2002)
and so relevant details only are given here.
3.1. Subjects
The study population for this randomised controlled
trial was drawn from six centres in the United
Kingdom. All patients between the ages of 18 and 65
years who met International Classification of Diseases
Version10 research criteria for schizophrenia (WorldHealth Organisation, 1992) were eligible for participation.
Lists were drawn from sources such as
inpatient and outpatient lists, depot and clozapine
clinics, community mental health team lists and case
registers. As a result, participants had varying lengths
of illnesses and numbers of previous relapses. Patients
were not recruited to the study if they had a coexisting
brain disease, a primary diagnosis of drug or
alcohol dependence or a learning disability severe
enough to interfere with rating. Any patient who was
deteriorating to such a degree as to be under
consideration for a period of inpatient care or
intensive home treatment was not included in the
study until stabilized
3.2. Insight in Schizophrenia Nurse (ISN) and
Assessors
Each centre was allocated an Insight into Schizophrenia
nurse (ISN) who worked with community
mental health teams. The ISN received 10 days’
intensive training in the CBT of schizophrenia and
was followed by weekly supervision. Each centre also
had an assessor. The assessors trained together at the
start of the study for a week, at the end of which they
were examined on the outcome measures for the main
insight study, i.e., Comprehensive Psychopathological
Rating Scale (CPRS), Schedule for Assessment of
Insight (SAI) and Health of Nation Outcome Scale
(HONOS) to ensure inter-rater reliability across the
sites. The intra-class correlation calculated using
CPRS was 0.71.
Ratings were carried out with the assessors being
blind to the patients’ allocation group as far as
possible. Assessors were informed that some study
material was to be left with some treatment as usual
(TAU) participants to help preserve the blindness of
the ratings and to avoid assessment bias. Each
participant met the ISN at the point of recruitment
to the study so that mention that they had met the
nurse did not necessarily mean they were in the
intervention arm of the study. All assessors noted any
perceived breaches of blindness as the study proceeded
and attempts to guess the participant’s study
group. The blindness of the assessors appeared to be
satisfactory as in the three circumstances where this
happened; identification of allocated group was no
more than expected by chance.