Project 2: Process and outcome

Wittorf et al.
Process and outcome in CBT for positive symptoms in psychotic disorders

Background and Aims
Process-outcome research in psychotherapy represents an empirical strategy for determining which aspects of the therapeutic process are particularly helpful or harmful to patients (Orlinsky et al., 1994). This research links the two domains of process and outcome studies. According to Orlinsky and colleagues, the crucial question is, "what is effectively therapeutic about psychotherapy?" They view the therapeutic contract (treatment model, e.g. rational, goal setting, format), the therapeutic operations (therapist interventions), and the therapeutic bond (therapist's and patient's interpersonal behaviour) as the essential factors of the treatment process. These factors were empirically found to be linked with therapy outcome in many psychiatric disorders. However, regarding psychotherapy in psychotic disorders such findings are still missing up to now. With respect to CBT for positive symptoms in psychotic disorders process-outcome research is not yet a major focus. Studies on this topic (Kuipers et al., 1997; Tarrier et al., 1998; Sensky et al., 2000) focus mainly on effectiveness. On this background, this project will add process-outcome-analyses to the major clinical trial of the network.

The main objective is to determine which therapeutic processes in the CBT condition are relevant for reducing persistent positive symptoms as measured by the positive syndrome of the PANSS nine month after study inclusion (T1). At first, the adherence to the treatment manual will be checked. This more technical aspect is followed by a rating of the patient-therapist-relationship and a rating of the extent of empirical collaboration in applying the CBT. Hypothetically we assume that the most critical aspect of the therapist's competence is the timing and intensity in confronting the patient with objective reality. A stable therapeutic relationship hypothetically can be viewed as an important prerequisite for disputing the patient's perceptual and inferential biases as described by Blackwood and colleagues (2001).

Adherence to Treatment Manuals
To assure the adherence to the manuals all treatment sessions (CBT and ST condition) will be audio taped if patients give consent to do that. Furthermore, therapists will complete a structured session protocol after each therapy session. Four audio tapes of each patient will randomly be selected for analyses: one of the first 3 sessions (early phase), one of the sessions 4-10 (early middle phase), one of the sessions 11-17 (late middle phase), and one of the last 3 sessions (late phase).
The rater will check the audio taped CBT and ST sessions and session protocols with regard to manual adherence. Guided by a checklist it will be assessed whether or not the treatment session followed the CBT or the ST manual with regard to content, the material worked on, and the formal characteristics (e.g. duration of the session). This rating by a checklist will allow for quantifying the adherence.
According to Orlinsky's and colleagues' review (1994) the strongest evidence linking process to outcome concerns the therapeutic bond, i.e. patient-therapist-relationship. The bond as a whole as well as its different aspects (role investment, interactive coordination, communicative contact, and affective attitude) show significant positive associations to therapy outcome. The patient-therapist-relationship will be assessed using a session questionnaire for patients ("Patientenstundenbogen") and for therapists ("Therapeutenstundenbogen", Grawe, 2000) Factor analyses of both instruments conducted in the framework of the German Research Network on Schizophrenia revealed a therapeutic relationship and a problem solving dimension in a sample of 111 first episode patients with schizophrenia participating in CBT for relapse prevention. Both instruments will be applied at the end of each therapy session of the CBT and the ST condition. The main research question is how the self-rated therapeutic relationship interacts with the more "technical" guided discovery (CBT condition) as part of the therapist's competence. Further, the continuous application of the session questionnaires will allow examining the course of the patient-therapist-relationship during the CBT. Hypothetically, the better the patient-therapist-relationship in earlier stages of the therapy the better the outcome will be.

Extent of empirical collaboration and cognitive restructuring
Establishing empirical collaboration between the patient and the therapist as a means for cognitive restructuring is a major aspect of the therapists' competence in CBT. According to the rationale of CBT, we expect that the extent of empirical collaboration correlates with therapeutic outcome (symptom reduction). To measure the extent of empirical collaboration the Cognitive Therapy Scale for Psychosis (CTS-Psy, Haddock et al., 2001) will be applied. The CTS-Psy consists of 10 items (agenda setting, feedback, understanding, interpersonal effectiveness, collaboration, guided discovery, focus on key cognitions, choice of intervention, homework, and quality of intervention) which are rated on a 7-point scale where higher scores indicate better competency. The CTS-Psy demonstrates excellent inter-rater reliability, good validity, and sensitivity to changes (Haddock et al., 2001).
The adequacy of cognitive restructuring (using questions to examine the patient's arbitrary conclusions and considering alternative explanations) depends not only on the quality of this technique, but also on the timing and intensity of this intervention. Hypothetically, the more adequate with respect to the strength of the patient's delusions cognitive restructuring is, the better the outcome (reduction of positive symptoms) will be. The more severe delusions are the less direct confrontation of CBT has to be. In an early stage of CBT questions regarding former times or others' opinions might be appropriate for carefully disputing the patient's cognitive biases. In later stages of CBT the therapist might use a more direct style of confrontation (e.g. verbalizing his/her own doubts with regard to patient's beliefs). The four randomly selected audio tapes will be analysed by means of the CTS-Psy.
The focus of the analyses will be the interrelation of the CBT specific items of the CTS-Psy (e.g. guided discovery, focus on key cognitions) with the therapeutic relationship and the outcome of CBT.