healthmedicinet

Treatment setting and location

Treatment is provided at the Psychodynamic Unit at “Thorsberg”, the Residential Facility
at the Drammen District Psychiatric Center, Norway. This unit was developed and founded
with the task of managing difficult-to-treat conditions through intensive, time-limited
in-patient care. The unit has room for a total of six patients, each receiving an
eight week “block” treatment program. Thus, groups of up to six patients enter and
terminate treatment simultaneously. Each year a total of five groups or “blocks” are
treated with a maximum total of 30 patients per annum.

Recruitment procedures

Patients are recruited among referrals for in-patient psychiatric care at the residential
treatment facility of the Drammen District Psychiatric Center. Patients are referred
partly by practitioners at local out-patient psychiatric clinics, nearby psychiatric
hospitals, and by general practitioners in the Drammen area. Patients are initially
screened for the inclusion/exclusion criteria by the intake-team at the residential
facility, before completing an evaluation session with one of the treatment providers
at the psychodynamic ward. The evaluator reaches a final decision regarding inclusion/exclusion
on the bases of prior treatment history, existing diagnostic information, referral
information, and the patients’ responses to intervention in the evaluation session.
Patients who are offered treatment at the ward are then informed about the study and
invited to participate. If patients are willing and eligible to participate, signed
and informed consent is obtained after comprehensive explanation of the procedures
of the project. Participants do not receive compensation in any form for evaluation,
screening, or assessments. Participating in the assessment-procedures is obligatory
as part of internal quality control for patients receiving the treatment program,
however participation in research project is voluntary.

Participating patients

Inclusion criteria

Adult patients (aged 18–70) are eligible to participate if they a.) Fulfill basic
criteria of need for hospitalization for psychiatric treatment including insufficient
general functioning and loss of mastery/function in multiple domains, e.g. inadequate
self-care, substantial breakdown in relational, occupational, and/or personal functioning.
b.) There are clear indications of prior treatment resistance, i.e. previous failure
to respond adequately (in terms of significant symptomatic relief and increased functioning)
in three or more prior treatment attempts for the current episode/ongoing psychiatric
disease. The number of three previous attempts were chosen to increase the likelihood
that included patients had severe and truly treatment resistant disorders, rather
than previous non-response being attributable to chance factors, or the delivery of
poor or unsuitable treatment. And c.) The patient demonstrates signs of capacity for
taking an intrapsychic perspective on own problems during the evaluation session,
i.e. the ability to regard one’s problems as the result of difficulties in dealing
with feelings, thoughts, and reactions to self/others. All three criteria are to be
fulfilled for inclusion. Co morbid axis-I and axis-II disorders are allowed, as is
medication with the exception of continuous intake of sedatives.

Exclusion criteria

Patients are excluded from the study if they suffer from one or more of the following
disorders; a psychotic disorder (except short, reactive psychotic episodes), bipolar
disorder type I, dissociative identity disorder, addiction of such severity that detoxification
is indicated (after which entering treatment is possible), psychiatric disorders secondary
to medical conditions, mental retardation, or insufficient command of the Norwegian
language hindering the utilization of treatment. Also, acute suicide risk and history
of severe acting out/problems with impulse control are considered contraindications
to participation.

Therapists

All therapists providing individual treatment in the program are trained and certified
psychologists. With the exception of one psychologist who left the unit soon after
the program was started, all therapists participate in a three-year core training
in ISTDP (Intensive Short-Term Dynamic Psychotherapy) including supervision of video-taped
therapy sessions. The core training program is delivered in Norway by internationally
renowned tutors organized and directed by Jon Frederickson, Washington School of Psychiatry,
US. In addition to Frederickson the team of tutors consists of Kees Cornelissen, PTC
De Viersprong, Netherlands, and Professor Allan Abbass, Dalhousie University, Canada.
In addition, the therapists take part in bimonthly group supervisions with Professor
Allan Abbass. Finally, peer supervision takes place on a weekly basis in the unit.
The level of experience of individual therapists is highly variable ranging from 2
to over 30 years.

The group-psychotherapists are highly experienced, trained, and certified in traditional
psychodynamic group psychotherapy and have co-developed the synthesis of ISTDP-principles
and traditional group therapeutic ideas adhered to in the program. The body awareness
instructor/physiotherapist is highly experienced, specialized and certified in psychomotor
physiotherapy. Finally, the therapist administering art therapy is certified and highly
experienced in this modality.

The treatment program

The treatment program is an eight week residential intervention, within these eight
weeks a number of treatment components are delivered; the majority of which are actively
integrated within the theoretical frameworks of ISTDP
11] and affect integration theory
12]. Patients enter the program in groups of up to six patients (minimum four), each
member of a group enters treatment at the same time and terminates eight weeks later.
Patients receive either two weekly individual single sessions (45 minutes) or one
weekly double session (90 minutes) depending on their capacity for tolerating intensive,
relatively long-lasting pressure and focus toward the experiencing of warded off feelings.

All individual therapists were educated and trained in the basic principles of ISTDP,
and instructed to adhere to these principles in individual sessions. Treatment integrity
and adherence to ISTDP-principles are continuously checked in the bimonthly group
supervisions with Professor Allan Abbass. Also, videotapes of all individual sessions
are available for systematic adherence and competence evaluation.

Patients also receive two weekly group sessions 90 minutes each. These group sessions
integrate the notion of pressure to feeling, and systematic clarification and challenge
of defenses, with traditional group therapeutic principles
27]. There are also weekly body awareness training groups based on principles from psychomotor
physiotherapy, along with bi-weekly low-intensity physical exercise (walking), and
weekly psycho-educational lectures providing a conceptual understanding of the therapeutic
process. Finally, all patients take part in weekly art therapy groups focusing on
the experience and expression of feelings through creative and artistic displays.
Individual psychomotor therapy is an optional addition offered to all patients at
the start of treatment.

Beyond the individual therapist, all patients are provided a primary treatment contact
among the ward staff with whom they are encouraged to discuss important aspects of
their development and potential challenges for the therapeutic process in a systematic
and continuous fashion. The primary contact and other staff also implement treatment
principles from ISTDP whenever suitable and indicated in the every day on-goings within
the ward. Table 
1 summarizes the weekly treatment activities of the program.

Table 1. Weekly treatment activities for patients included in the program

Scientific design

The research project is a naturalistic longitudinal study for examining the short-
and long term effectiveness of this time-limited intensive in-patient treatment program
for relieving treatment resistant conditions, mainly anxiety-, depressive-, and personality
disorders. To increase confidence in our findings a naturalistic waitlist control
condition is established. Thus, patients are assessed with self-rated measures prior
to their evaluation session so that development from the time of evaluation until
the onset of treatment can be used as a contrast to effects presumably attributable
to the treatment itself. In the treatment and follow-up phases, observer- and self-rated
assessments are done in the week prior to hospitalization (T1), during the eight week
treatment phase (t01-t08), at termination (T2), in addition to six (T3) and twelve
months (T4) after the end of treatment. Individual psychotherapy sessions are videotaped
for supervision purposes and made amenable to the application of observational coding
systems for rating various process and outcome factors in treatment.

Assessments and assessment schedule

Observer rated measures

Experienced members of staff are specifically trained for assessing patients included
in the treatment program on the relevant observer based measures which are conducted
prior to hospitalization (T1), at termination (T2), and twelve months after termination
(T3). The MINI Neuropsychiatric Interview
28] is used for assessing Axis I diagnoses. The Structured Clinical Interview for the
Diagnostic and Statistical Manual of Mental Disorders, Axis II, Fourth Edition (DSM-IV-R)
(SCID-II)
29] is used for assessing Axis II personality disorders and severity of personality problems.
Finally, the Affect Consciousness Interview (ACI)
30] is used for assessing the functional level of affect integration and affect organization.

The MINI neuropsychiatric interview The Mini International Neuropsychiatric Interview
28] was developed to meet the need for a brief, reliable and valid structured diagnostic
interview for screening a variety of psychiatry disorders and replacing the somewhat
cumbersome Structured Clinical Interview for DSM-IV-R, Axis I (SCID I) and the Composite
International Diagnostic Interview (CIDI) in clinical trials. The MINI contains 120
questions and screens 16 Axis I and one Axis II DSM IV-R disorders for 24 current
and lifetime diagnoses. Like the SCID-P and CIDI, the MINI is organized in diagnostic
sections. Using branching tree logic, the MINI has two to four screening questions
per disorder. Additional symptom questions within each disorder section are asked
only if the screen questions are positively endorsed. Since anxiety and depressive
disorders are a central focus of the program, the presence or absence of such disorders
will be used as specific measures of outcome from the treatment, along with diagnostic
status on other relevant diagnostic categories in the MINI.

The structured clinical interview for the diagnostic and statistical manual of mental
disorders, axis II (SCID-II) The SCID II
29] is a version of the SCID developed for the assessment of DSM-IV-TR Personality Disorders.
The interview covers the eleven DSM-IV Personality Disorders (including Personality
Disorder NOS) and the appendix categories Depressive Personality Disorder and Passive-Aggressive
Personality Disorder. It closely follows the language of the DSM-IV Axis II Personality
Disorders criteria. The scoring is done so that the trait either is absent, sub-threshold,
true, or there is “inadequate information to code”. Traits considered true are then
summed up and diagnoses are indicated when the required number of traits are present
for any given disorder. The SCID-II is administered by trained interviewers and generally
yields decent indications of the personality disorder spectrum.

The Affect Consciousness Interview (ACI) The ACI is a semi-structured interview designed for assessing the consciousness and
integration of discrete affects (Affect Consciousness – AC)
30,31]. Five aspects of the experience of these affects are assessed: Scenes/elicitors,
awareness, tolerance, emotional (nonverbal) expression and conceptual (verbal) expression.
The interviewer asks about the following for each affect: (1) scenes in which the
affect is activated, (2) how the patient becomes aware of and recognizes the affect,
(3) how the affect impacts upon the patient, how the patient copes with the affect,
and what information the patient decodes from the affect activation, (4) to what extent
and how the affect is expressed in nonverbal forms (i.e., emotional expression, a
category that refers to the respondent’s consciousness of own voluntary and involuntary
nonverbal expressions) and finally (5) to what extent and how the affect is expressed
verbally (i.e., conceptual expression). The interviews are administered by trained
interviewers, videotaped, and scored according to criteria specified in the Affect
Consciousness Scales (ACSs)
30]. The ACSs comprise four specified nine-step scales, one for each of the integrative
aspects: Awareness, Tolerance, Emotional Expression and Conceptual Expression. A score
of 1 is the lowest possible, 9 is the highest, and a score of 5 would be considered
normal. On the basis of these indicators, scores on three different levels are calculated:
Overall mean score (Global AC), mean score on each of the four integrating aspects
(e.g., the mean Awareness-score across the nine affects), and mean score on each of
the specific affects (e.g., mean score on awareness, tolerance, emotional expression
and conceptual expression for Tenderness/Care).

Self rated measures

A number of self-rated questionnaires are completed prior to the evaluation session,
prior to treatment onset, and throughout the treatment- and follow-up phases, reflecting
various aspects of patient functioning and therapeutic process. One questionnaire
assesses basic demographic information, along with information on prior treatment,
educational level, current occupational status, the use of medications, and current
financial situation. As a principle measure of level of psychopathology and psychological
functioning, as well as a process measure reflecting weekly improvement/deterioration
the Outcome Questionnaire- 45 (OQ-45) is used
32]. The OQ-45 is first administered before the evaluation session, in the week before
onset of treatment, and then before every single individual psychotherapy session
throughout the treatment phase, as well as six and twelve months after termination.
In addition, the Symptom Checklist 90-Revised
33] (SCL-90-R), and the Inventory of Interpersonal Problems (IIP-64)
34] are used prior to evaluation, before onset of treatment, in weeks two and five of
treatment, at termination, and six and twelve months after termination. The Working
Alliance Inventory, patient version (WAI-P)
35] is used as a measure of conscious working alliance and is administered in week two,
five, and at termination.

Outcome questionnaire- 45 (OQ-45) The OQ-45 is a symptom and distress inventory developed by Lambert et al.
32]. It has been found to be useful for examining the effectiveness of psychotherapy
over time
36]. The instrument is designed to assess “patient functioning” and scores are used to
track changes in symptomatology on a session-by-session basis. The OQ-45 consists
of 45 items tapping various aspects of psychological distress each associated with
a 5-point Likert scale. Responses refer to the last seven days and range from “never”
to “almost always”. Once responses to nine negatively worded items have been reverse-coded,
sum scores are calculated with higher scores representing increasing levels of psychopathology.
Overall scores are calculated along with scores on each of three subscales, assessing
Symptom Distress (SD), Social-Role functioning (SR) and Interpersonal Relationships
(IR) respectively.

Symptom checklist-90-revised (SCL-90-R) The SCL-90-R is a widely used and comprehensive symptom inventory that measures symptom
distress on nine dimensions and three global indexes
33]. Intensity of 90 symptoms during the last seven days is rated on a five point Likert
scale ranging from not at all (0) to very much (4). The Global Severity Index (GSI),
the average score across all 90 items, is regarded a highly nuanced and valuable indicator
of overall current level of distress
37]. The primary symptom dimensions that are assessed are somatization, obsessive-compulsive,
interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid
ideation, psychoticism, and a category of “additional items” assessing other aspects
of patient symptoms such as sleep disturbances, guilt/self-blame, and eating/appetite
problems. The depression-, anxiety-, and phobic anxiety- subscales will be used as
specific outcome measures of these, to the program, central outcome dimensions.

Inventory of interpersonal problems (IIP-64) General and specific interpersonal problems are assessed using the 64 item IIP-circumplex
version
34]. The IIP-64 consists of two types of items. The first 39 items begin with the phrase:
“It is hard for me to…” The remaining 25 items represent “Things that you do too much.”
Each item is rated on a five point Likert scale ranging from not at all (0) to very
much (4). The general or elevation factor of the IIP-64 has been consistently linked
to both symptom severity and negative affectivity
38]. The second (Agency) and third (Communion) factors, yielding the IIP-64 circumplex
structure, generally show good construct validity in terms of fit with a quasi-circumplex
model, along with distinct convergent-discriminant correlation patterns with different
forms of personality pathology, supporting the notion that the IIP-64 model adequately
represents its theoretically alleged distinctions in interpersonal functioning
31,39]. The overall score of the IIP-64 is used as an indicator of general interpersonal
problems. The IIP-64 also yields eight octant sum-scores indicating specific problems
with being PA?=?Domineering/Controlling, BC?=?Vindictive/Self-Centered, DE?=?Cold/Distant,
FG?=?Socially Inhibited, HI?=?Non-Assertive, JK?=?Overly Accommodating, LM?=?Self-Sacrificing
and NO?=?Intrusive/Needy.

The working alliance inventory (WAI) Patient rated working alliance is assessed by the Working Alliance Inventory (WAI)

31] short version with 12 items rated on a seven point scale
40]. The WAI covers three aspects of working alliance: therapeutic bond, task, and goal.
The goals subscale assesses the extent to which patient and therapist agree on the
goals that are the target of the intervention. The tasks subscale assesses the extent
to which patient and therapist agree on the in-counseling behaviors and ideas that
form the substance of the counseling process. The bond subscale assesses the extent
to which patient and therapist share mutual trust, acceptance, and confidence in the
process at hand. Table 
2 summarizes the assessment instruments and assessment schedule.

Table 2. Overview of assessment instruments and assessment schedule in the treatment and follow-up
phases

Sample size and data analytic strategies

Sample size

The research project is designed to include a minimum of 250 patients; however, all
patients treated at the unit are invited to participate, so the sample size is expected
to increase further over time.

Statistical power

In research where demonstrating the effectiveness of treatment is an objective, statistical
power will always be an important factor. Exact estimation of statistical power is
only possible for specific analytic designs aimed at delimited research questions
and necessarily based on conditions that are not exhaustively known when a study is
initiated. However, in general a sample of 250 patients will yield a statistical power
of .80 for the detection of small to medium effect sizes for the overall effect in
one-way analyses of variance. This means that the probability of detecting significant
differences between points of measurement is 80% for an effect size (Cohens’s d) greater
than 0.3 -0.5. This estimate will also apply to change scores between to points of
measurement if we assume that the average correlation between time-points is greater
than .50. In pair-wise comparison of subgroups with n?=?80 the probability of detecting
small to medium effects similarly will be about .80.

Data analytic strategies

The data from the project lends itself to various research topics. Most centrally,
the data can contribute valuable knowledge on the effectiveness of short-term and
time-limited residential treatment for treatment resistant psychiatric conditions.
For answering this research question using continuous outcome variables (e.g. OQ-45,
IIP-64, AC) multilevel (or hierarchical linear) modeling will be applied using the
linear mixed models module in the SPSS/PASW, version 18.0. The application of multilevel
modeling for the analysis of longitudinal data, in this case repeated measurements
on the same individuals, is strongly recommended in the literature
41,42]. For longitudinal data, measurements are nested within individuals, so that measurements
represent units at the first level and individuals represent units at the second.
It is usually proposed that as requirements for longitudinal analyses, all variables
must be collected at three or more measurement waves, that a continuous outcome changes
systematically over time, and that a meaningful unit for time is included
42]. Each of these requirements is met by the design of the present project. Generally,
based on experiences with previous patients going through the program we expect low
attrition rates. Almost all patients are expected to complete the treatment program
and deliver data for the complete treatment phase. Some attrition is expectable for
the follow-phase, but here to, based on previous experience, we expect that as much
as 80 – 90% will deliver follow-up data at least on one occasion.

Using Cohen’s
1] formula, effect sizes can be calculated as the growth-curve estimated difference
between end-status and baseline divided by the pooled standard deviation across all
measurement points of the relevant outcome variable. For dichotomous outcome variables
(e.g. presence or absence of Axis I and II disorders) treatment effects will be tested
with simpler survival analysis and cross tabulation.

Treatment response for individual patients on self-rated instruments is defined in
terms of clinically significant change (CSC) according to Jacobson and Truax
3]. Thus, we will identify the number of patients that are reliably improved and have
moved from the dysfunctional to the functional range (recovered), the number of patients
reliably improved but not recovered, the number of patients reliably worsened, and
the number of patients unchanged on central outcome measures.

The data also lend themselves to a number of investigations into the conceptual and
empirical relationships between various measures of psychological health and functioning,
and to tests of the psychometric properties of different measures included in the
study, e.g. on the relationship between symptom patterns and interpersonal problem
types, on the validity of the subscales of the OQ-45 and the SCL-90-R and so on and
so forth. For such investigations traditional exploratory and semi-confirmatory factor
analyses, multiple linear regression models, and correlation analyses will be performed.
Finally, data may prove to be a valuable source of information on the predictors of
change and treatment response when treating treatment resistant patients. For answering
research questions related to this topic, individual growth curves will be utilized
and relevant predictors entered into the multilevel models of change.

Research ethical considerations

Participation in the research part of the program is voluntary. Participants are informed
that they can withdraw their consent to participate at any time without disclosing
reasons for their cancellation. There are no negative consequences of such withdrawal
for the treatment delivered. All participants sign a written and informed consent.
The protocol for the study was evaluated by the Regional Committee for Medical and
Health Research Ethics in Eastern Norway and a letter of exemption was issued classifying
the study as a quality control project approving dissemination of results.