Treatment integrity, therapeutic alliance and outcome : an evaluation of the relationship in cognitive behaviour therapy and befriending for psychosis
Outcome studies of CBT for psychosis have shown positive results from
cognitive behaviour therapy and supportive treatments. It has been suggested
that this results partly from the common, non-specific aspects of therapy.
Although this is possible, in many studies, treatment integrity has not been
adequately examined and it may be that the cross-contamination of treatments
might account for some results obtained. This research examines treatment
differences and relationships with outcome, for treatment integrity and working
alliance, in early sessions of CBT and befriending for psychosis.
This study used outcome and therapy data from sixty dyads involved in a
randomised controlled trial of CBT and befriending. Observer ratings of
adherence, competence and alliance were made from early treatment sessions.
Treatment differences in adherence, competence and alliance were examined and
their relationship with global and negative symptom change was explored.
CBT was adherent to a cognitive model of psychosis and was characterised by
competent administration of techniques. Befriending did not include specific
cognitive therapy techniques. There were significant differences between CBT
and befriending on measures of competence and adherence. Working alliance
was high in both interventions with comparable levels of therapeutic `bond'
between treatments. Significant treatment differences were observed in `task'and `goal' rated alliance with higher levels observed in the CBT group. Higher
levels of goal rated alliance, insight oriented techniques and competently
administered specific and non-specific CBT techniques differentiated those
individuals who improved and those who did not.
Befriending and CBT were procedurally different interventions, with comparable
levels of therapeutic bond. Higher task and goal alliance scores suggest that
patients invested more in the CBT intervention. Factors that facilitate
improvement may change over the course of therapy but in early therapy, aspects
of working alliance and adherent and competent administration of techniques
appear to be implicated in improvement. Further research in this area is required