On behalf of science and practice – using psychotherapy research to address clinical practice concerns
Date issued
Authors
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
License
Abstract
The many efforts to seek common ground for psychotherapy research and practice
could not yet fix the gap that exists between the two stakeholders of mental health services
as it has been the case for decades. The mutual integration of research and practice is an
essential ingredient in effective service delivery of psychological treatments. In this
connection, practice-oriented studies that are conducted in naturalistic settings provide a
context in which both parties can learn from and complement each other. With the aim of
contributing to the improvement of psychological practice in terms of patient outcomes,
professional well-being, and training, two research topics that are relevant to clinical practice
concerns were investigated in this dissertation: the use of data-driven clinical support tools
and occupational stress among psychotherapists.
Study 1 investigated whether basic outcome monitoring in outpatient psychotherapy
(i.e., assessment intervals of 5 to 15 sessions) can be used for personalized outcome
prediction. This is significant because outcome prediction and monitoring have been proven
useful to prevent stagnation, deterioration, or premature dropout from psychological
treatment. However, session-by-session evaluation is rare in most clinical settings in the
context of limited time and resources. In this study, individual treatment progress and
dropout risk were predicted using modern statistical machine learning techniques such as the
nearest neighbor method and least absolute shrinkage and selection operator regression. For
the prediction of individual treatment progress, each patient’s nearest neighbors were
selected based on variables that have been identified as significant predictors of symptom
change (i.e., baseline distress, intrinsic treatment motivation, previous inpatient treatment,
and number of suicide attempts in the past). Lower intrinsic treatment motivation, a lack of
university entrance qualification, higher baseline impairment, previous inpatient treatment,
and diagnosed personality or eating disorder accounted for higher dropout probability.
Addressing potential obstacles in routine symptom assessment, this study points to the
practical significance of evidence-based research. The findings demonstrate that innovative
outcome prediction is not limited to elaborated assessment and provide a reasonable
approach for successfully predicting individual patient outcomes as long as session-bysession
assessment is not a valid standard.
While patient symptoms and outcomes have always been the focus of investigation,
therapist well-being is not sufficiently studied in psychotherapy research. High prevalences
of work-related stress, that can lead to reduced professional competence and associated risks for their patients, make this topic an ethical priority in the workplace. Therefore, Study 2
investigated how potential work stressors (i.e., patient distress, working alliance, and
treatment outcome) relate to occupational stress among psychotherapists. The results show
that treatments with high-distress patients, poor working alliance, and non-remission are
associated with higher levels of occupational stress. Furthermore, occupational stress was
significantly negatively related to perceived learning opportunities and patient treatment
satisfaction.
When interpreting the results, particular attention was given to psychotherapy
training as the potential of clinical support tools is developed in clinical training regarding
their utility and implementation and because trainees experience more insecurity and stress
at work compared to more experienced therapists. The conclusions of this dissertation
highlight the importance of research-supported psychological treatments (e.g., the use of
clinical support tools as a useful supplement to evaluation based on therapist judgment).
Furthermore, approaches for the prevention of occupational stress among therapists are
proposed. Both, the integration of data-driven clinical support tools into psychological
practice and the risks of occupational stress burnout should be targeted in the educational
context.