Narrative Exposure Therapy for Posttraumatic Stress Disorder associated with repeated interpersonal trauma in patients with Severe Mental Illness: a mixed methods convergent design

Interpersonal trauma exposure and trauma related symptoms are often overlooked
in the treatment of severely mentally ill patients even though prevalence rates
for both are high. As shown in recent research, repeated interpersonal trauma
and Posttraumatic Stress Disorder (PTSD) have a negative influence on the
course of the present severe mental illness (SMI).
In the Netherlands, SMI patients receive Flexible Assertive Community Treatment
(FACT) provided by multidisciplinary community mental health teams. In GGNet,
FACT -teams are trained in screening for traumatic experiences and PTSD
symptoms to improve the treatment of SMI patients with comorbid PTSD. SMI
patients with comorbid PTSD, treated in the FACT teams, are offered evidence
based trauma focused treatment like Eye Movement Desensitization Reprocessing
(EMDR) or Prolonged Exposure (PE), according to the international PTSD
guidelines. Several studies have shown that these exposure based trauma focused
therapies (i.e. EMDR and PE) are effective in SMI patients and well tolerated.
For patients with (comorbid) PTSD associated with repeated interpersonal trauma
there is a large amount of evidence for the effectiveness of Narrative Exposure
Therapy (NET) within various patient groups. NET has not been specifically
studied in SMI patients. Since 2012, our FACT-teams offer NET to SMI patients
with PTSD associated with repeated interpersonal trauma. All NET treatments are
monitored according to the treatment protocol. This trauma specific monitoring
is combined with the routine outcome monitoring (ROM), which is common practice
in mental health care and based on Dutch performance indicators for SMI.
To our knowledge this is the first study which aims to evaluate the outpatient
practice of providing NET to SMI patients with comorbid PTSD associated with
repeated interpersonal trauma and receiving FACT.

The primary aim of this study is to evaluate NET in SMI patients with comorbid
PTSD associated with repeated interpersonal trauma whether:
(a) the PTSD symptoms change following NET and
(b) changes occur in the present SMI symptoms, care needs, quality of life,
and care consumption.

The second aim is to gain insight in patients* experiences with the NET
treatment and to identify influencing factors on treatment results in terms of
symptom changes, care needs, and quality of life.

Quantitative and qualitative methods are combined in this mixed methods
convergent design.
The quantitative method consists of a pretest-posttest design and will be
carried out among adult (21 to 65 years) SMI patients with one of the following
primary diagnoses: schizophrenia spectrum disorder; bipolar disorder, major
depression, or personality disorder and co-morbid PTSD associated with repeated
interpersonal trauma. We will include consenting participants (N=25) receiving
continuous FACT from one mental health center. These participants are
independently living outpatients and referred by their psychiatrist for the NET
according to the guidelines. Participants do not receive involuntary treatment
following Mental Health Law. The qualitative method consists of a Grounded
Theory design with semi-structured in-depth interviews.

Data collection: Quantitative data will be collected from the electronic
patient records on three occasions: one week (T0) prior to NET, one month (T1)
after NET and at six months (T6) follow up. On the first occasion the following
instruments are used for trauma specific measurements: (1) the LEC-5 to verify
the trauma history, (2) the CAPS-5 to verify the existence and severity of
PTSD, (3) the DES to verify the existence and severity of dissociative
symptoms. At T1 and T6 CAPS-5 and DES are administered again. For Routine
Outcome Monitoring (ROM) the following instruments are used and combined at T0,
T1 and T2: (1) the HoNOS to measure psychiatric symptoms, (2) the CAN to
measure care needs, and (3) the MANSA) to measure the experienced quality of
life. At T0 the M.I.N.I. - plus (or the SCID-II in case of a personality
disorder) will be used to verify the current chart diagnosis of the primary
existing SMI. This is in line with the current practice whereby existing
diagnoses are re-assessed every two years. Care consumption in minutes will be
calculated based on the electronic record for the following period: three
months before NET, during NET and during six months follow up.
Qualitative data collection will be performed on one occasion: semi-structured
in-depth interviews are held two months after providing the NET treatment and
four months before follow up measures following a convergent design. All
participants will be interviewed about their experiences with the NET treatment
and its effect on their daily life based on a topic list.

Analysis: quantitative analysis: The results from post-treatment and follow up
measurements will be compared with pretest measures using paired t-tests.
Qualitative analysis: The interviews will be analyzed by the grounded theory
method to identify relevant themes in how patients experience this treatment
and create meaning in daily life functioning. Integrative analysis: Integration
of both quantitative and qualitative results will be focused on the influencing
factors on treatment results in terms of reducing symptoms, care needs, care
consumption, and in improving quality of life.

Discussion: To our knowledge, this is the first evaluation of NET for PTSD in
SMI patients receiving FACT.

The semi-structured interview takes up to 60 minutes and is conducted two
months after NET treatment. At this stage patients are less vulnerable and most
of them function better. The interview is not aimed at the traumatic
experiences but is focused on the treatment experience and the treatment effect
on symptoms and daily life functioning. Moreover, patients are accustomed by
the therapy to talk about their experiences in general.
The risk of worsening of symptoms or suicidality as a result of the interview,
is therefore estimated as low.