Developing Adaptive Interventions for Suicidal College Students Seeking Treatment - SMART

Overview

This pilot study tested the feasibility of utilizing an adaptive intervention strategy for college students who are suicidal when first seeking treatment at a campus clinic. Right now, the typical strategy may rely on a "one size fits all" approach, but in fact suicidal students vary greatly on what and how much they need. This study will pave the way for subsequent larger trials for clinical decision making (trying one approach, and if that doesn't work, another) to be empirically developed and tested in a subsequent large-scale multisite trial with the goal of maximizing resources in overburdened college counseling centers. This pilot study followed by a subsequent large-scale trial could eventually significantly impact service delivery to suicidal college students at college counseling centers.

The primary aim of this proposal was to conduct feasibility research to inform the implementation of a future full-scale SMART (sequential, multiple assignment, randomized trial, Almirall et al., 2012) that will be used to construct adaptive treatment strategies (ATSs) to address suicidality in college students seeking services at college counseling centers (CCCs). ATSs individualize treatment via decision rules that can specify how the type and intensity of an intervention should be sequenced based on variables collected mid-treatment (or at baseline), such as response to treatment. Suicidality is a frequent presenting concern among college students seeking treatment; yet, studies with this population show that some students respond rapidly to treatment whereas others may require considerably more resources. However, at this time, CCCs, which are overburdened and often have to resort to waitlists, have no guidance as to how to sequence different approaches with suicidal students in an empirically-based and cost effective manner. Therefore, empirically validated ATSs are needed in the provision of services to suicidal college students to address the heterogeneity of students with this presentation and the variability in response to interventions. In the present pilot SMART, each participant progressed through two stages of intervention. In the first stage (S1) 62 participants were randomized to one of two brief individual therapy interventions for 4-8 weeks: 1) one that is suicidality-focused (Collaborative Assessment and Management of Suicidality (CAMS; Jobes, 2016) and 2) one that relies on Treatment as Usual (TAU) being provided at a CCC. Responders to either program will discontinue services/ be stepped down and be monitored over time for maintenance. Non-responders to either intervention who remain in treatment (estimated n = 18) were re-randomized to one of two second-stage (S2) higher intensity/dosage intervention options for an additional 4-16 weeks: 1) CAMS (either continued but for a longer period of time or administered for the first time) or 2) Comprehensive Dialectical Behavior Therapy (DBT, Linehan, 1993a, 1993b) which includes individual therapy, skills groups, and phone coaching for the clients and DBT peer consultation for the therapists. This study enrolled moderately to severely suicidal (endorsing thoughts of wanting to die of 2 or above on 0-4 scale) college students in the "emerging adulthood" phase (18-25 years of age) seeking services at a CCC. The aims of this feasibility services research project were to 1) develop and refine a SMART design in a CCC setting; 2) assess the feasibility of conducting a SMART and its embedded ATSs in a CCC setting; 3) obtain estimates of overall response rates to S1 interventions; and 4) explore the utility of incorporating secondary tailoring variables (e.g., level of functioning at pre-treatment, ratio of Wish to Live vs. Wish to Die) in the ATSs in the subsequent larger trial.