Hypothesis:
Personalized CBT will achieve a comparable treatment effect to the standardized CBT
condition, i.e. a stronger reduction in the outcome measures (intercept and slope).
Participants:
Recruitment takes place at selected university outpatient clinics throughout Germany.
Patients will be recruited via the waiting list of the university's outpatient clinic, in
cooperation with other currently running studies at the same university recruiting
chronic pain patients, and via various media (e.g. newspaper articles) and doctors'
offices. Study therapists will be recruited in the university's outpatient clinic as
well. Inclusion criteria for patients are at least 18 years of age, having access to a
smartphone, and the main diagnosis of chronic pain. The diagnosis will be checked using
the brief version of the Diagnostic Interview for Mental Disorders (Mini-DIPS). For
patients recruited via the waiting list, screening for suitability will take place during
the first consultation at the university psychotherapy training center's outpatient
clinic. Suitable participants will be informed about the study and referred if they agree
to be contacted. Furthermore, patients that had to be excluded from other currently
running studies will be referred if they agreed to be contacted as well.
Procedure & Measures:
Counterbalanced Repeated Measures Design: At this level, the primary outcome measure is
pain disability index (PDI) and the WHO Disability Assessment Schedule 2.0 (WHODAS 2.0).
Therapeutic alliance (Helping Alliance Questionnaire (HAQ), Working Alliance Inventory
Short Revised (WAI-R); both therapist and patient version) and side effects (Negative
Effects Questionnaire, NEQ) as well as expectations (Patient Questionnaire on Therapy
Expectation and Evaluation, PATHEV) are collected as secondary outcome variables. These
are collected before and after the diagnostic phase (A), after both interventions (C1,
C2) and at follow-up (3 months). The HAQ, WAI-R, and NEQ constitute exceptions in this
context. As these instruments pertain to the therapeutic relationship or the overall
therapy process, they are not administered prior to the diagnostic phase, since no
therapeutic contact has yet taken place. In addition, the therapist's case concept and a
Perceived Causal Network (PECAN) of the therapist are collected after the diagnostic
phase. Additional outcome measures are collected before the diagnostic phase (A), before
the 3th baseline (B3) and at follow-up. As additional outcome measures the Depression
Anxiety Stress Scale (DASS-21), the German Pain Solutions Questionnaire (PaSol), the
Patient Global Impression of Change (PGIC) and Pain Self-Efficacy Questionnaire will be
assed.
SCED: The participants begin with the standard diagnostic phase of routine clinical care
(phase B, 5 sessions) with psychoeducation and the development of therapy goals. After a
randomized baseline (phase A1, 1-3 weeks), the intervention phase (phase C) begins.
Participants are randomly assigned to one of two groups. Group 1 begins with personalized
CBT followed by standardized CBT whereas Group 2 begins with standardized CBT followed by
personalized CBT. A second baseline takes place in both groups after the first
intervention before the beginning of the second intervention (A2, randomized 1-3 weeks).
After the two different therapeutic phases, another baseline (A3, 2 weeks) and afterwards
an EMA phase of 3 weeks will be completed. In addition, there are two booster sessions
with the therapist one and three months after the last therapy session. During the EMA
phases, data will be collected 6 times per day. In all other phases, the questionnaires
are asked 3 times a week.
Analysis:
To evaluate group differences, a multilevel analysis (MLM) is calculated to take the
nested data structure into account. The effects within the individual participant are
calculated at level 1 and across the participants at level 2. To determine the required
sample size, we performed a data simulation assuming a normal distribution with two
predictors in the MLM: Treatment and order of treatment. The simulation revealed that a
sample size of 59 participants is required, with a power of 0.80 aimed for to detect
small effect sizes. Based on the dropout rate observed in a previous pilot study, a
sample size of N = 75 is planned.
The Bayes Factor and visual analysis are used to continuously evaluate the intervention
effect at the individual level. In the visual analysis, we look at level, trend,
variability, immediacy, overlap, and consistency.