THE STEP TRIAL: A SEQUENTIAL MULTIPLE ASSIGNMENT RANDOMISED TRIAL (SMART) OF INTERVENTIONS FOR PATIENTS AT ULTRA-HIGH RISK OF PSYCHOSIS - STUDY RATIONALE, DESIGN AND BASELINE DATA

Abstract Background Although approximately twenty randomised controlled trials have now been conducted with young people identified as being at high clinical risk of psychotic disorder, it remains unclear what the optimal type and sequence of treatments are for this clinical population. There has also been increased focus on clinical outcomes other than transition to psychotic disorder, such as psychosocial functioning, persistent attenuated psychotic symptoms and non-psychotic disorders. At Orygen, we are currently conducting a trial of a sequence of interventions consisting of two psychosocial therapies (support and problem solving [SPS] and cognitive-behavioural case management [CBCM]) and antidepressant medication. The primary outcome of the study is functional outcome after 6 months. This presentation will outline the background, rationale, design, recruitment and retention data and preliminary baseline results. Methods STEP is a sequential multiple assignment randomised trial (SMART) of treatments for young people (12–25 year olds) who meet ultra high risk for psychosis (UHR) criteria. Participants were recruited from primary (headspace) and secondary/tertiary (Orygen Youth Health) mental health services in Melbourne, Australia. The trial consists of three steps: Step 1: SPS (1.5 months); Step 2: SPS vs Cognitive Behavioural Case Management (4.5 months); Step 3: Cognitive Behavioural Case Management + Antidepressant Medication vs Cognitive Behavioural Case Management + Placebo (6 months). Patients who do not respond by the end of each step graduate to the next step in treatment. Responders are randomised to SPS or monitoring. Treatment response is based a combination of reduced attenuated psychotic symptoms, rated using the Comprehensive Assessment of At-Risk Mental States (CAARMS), and functional improvement (Social and Occupational Functioning Assessment Scale [SOFAS]) at the end of the treatment step. A ‘fast fail’ option is built into Step 3, whereby patients who deteriorate or have not responded 3 months into Step 3 are offered a choice of continuing existing treatment or commencing omega-3 fatty acids or low-dose antipsychotic medication. The intervention is for 12 months, with follow up at 18 and 24 months. A pilot study using the same design is currently being conducted at The University of California Davis. Results Recruitment has recently completed, with 342 patients recruited over a 2.4 year period, representing the largest UHR treatment study conducted to date. Preliminary results indicate an 8% response rate to Step 1 and a 23% response rate to Step 2. Discontinuation rates are 15% (step 1), 43% (step 2), 32% (step 3), primarily due to participants being lost to follow up or not wanting to start medication. The current transition to psychosis rate is 10.2%. Baseline clinical data are currently being analysed and will be presented at the conference. Discussion Preliminary results indicate high non-response rates following SPS and moderate non-response rates following extended SPS or CBCM, possibly partly due to the stringent definition of response, which required substantial and persistent improvement in both attenuated psychotic symptoms and functioning. Discontinuation rates are low to moderate, reflecting the complexity and severity of this clinical population. The recruitment and retention data show that it is possible to conduct large-scale and complex stepped care trials with this high risk population in a primary mental health care setting (headspace services). Outcomes will inform the most effective type and sequence of treatments for improving psychosocial functioning, symptoms and reducing risk of developing psychotic disorder in this group, as well as identify predictors of treatment response.