Factors associated with treatment retention and outcomes in a healthy lifestyles intervention for people with psychotic disorders: client-related factors and early therapeutic alliance

Scope: Research consistently demonstrates treatment dropout can be problematic. The ramifications of dropout for people with comorbid psychotic disorders and physical health vulnerabilities extend to the wider community, fiscally and otherwise. This thesis reviews and expands on research examining client and therapeutic alliance factors potentially associated with retention. Consideration is also given to differing treatment modalities. A critical literature review and manuscript detailing a secondary-analysis of data collected during a randomised controlled trial is provided. Purpose: The current study aimed to improve understanding of factors influencing treatment retention and longer-term outcomes in people with psychotic disorders involved in a healthy lifestyles intervention. This intervention compared a telephone versus face-to-face delivered intervention. Methodology: This study examined a subsample of 178 smokers with a psychotic disorder from a larger study. Recruitment for the original study was through self-referral, health service providers and research registers. Inclusion criteria included a completed participant baseline alliance rating (Agnew Relationship Measure); only then permitting inclusion of the therapist ratings (of the relevant participant-therapist dyad) in the analysis. Data were examined from baseline, mid-intervention and 12 months post-recruitment. Primary outcomes were treatment retention (dichotomised as low or high), and selected symptom and health behaviour measures. Preliminary and major analyses included descriptive statistics, univariate and multivariate analyses. Results: The average number of treatment sessions attended was 3.88 sessions (low retention group) and 14.59 sessions (high retention group). Participants with high treatment retention tended to be older and had higher rates of schizophrenia diagnosis. Participants from the telephone-delivered condition had significantly higher treatment retention rates (80.2% vs. 54.3%). Early therapeutic alliance did not predict the duration participants remained in treatment. Aspects of participant early alliance ratings (Confidence subscale) were predictive of greater improvement in depression at 12 months, and the early alliance therapist ratings (Bond subscale) predicted significant improvement in psychiatric symptomatology. However, some elements within therapist ratings (higher Confidence subscale scores) predicted less improvement in psychiatric symptomatology at 12 months. Early alliance was not predictive of change in health behaviours over time. Some study site effects were also identified, together with significant ‘early alliance by intervention condition’ interactions; for example, participants with low perceived therapeutic confidence or openness at baseline appear to have benefited preferentially from the telephone-based intervention. Conclusions: Contrary to expectations, early therapeutic alliance did not predict retention in treatment in a healthy lifestyles intervention for smokers with a psychotic disorder. Elements of therapeutic alliance as rated by both client and therapist did appear to be predictive of some clinical outcomes. A telephone-delivered treatment achieved greater client retention in comparison to a face-to-face intervention. Implications: Retention of people with psychotic disorders in a healthy lifestyles intervention could be facilitated by monitoring patterns of early therapeutic alliance and maintaining flexibility in the treatment modality adopted.