Feasibility and Acceptability of a Suicide Prevention Protocol in the Context of an International Clinical Trial for Mental Health and Substance Abuse

Most behavioral health clinical trials exclude patients at risk of suicide, thus limiting data on the effectiveness of treatment with this population. Given the association between substance abuse and suicidal behavior, inclusion of suicidal patients is particularly relevant in clinical trials for treatment of alcohol and drug abuse. The aims of this study were to (1) describe the suicide risk prevention protocol used in an international multi-site NIDA-funded clinical trial for Latino immigrants with co-occurring mental health and substance abuse problems; and (2) provide data on the feasibility and acceptability of the protocol. Method: The suicide risk prevention protocol used in three study sites was described. Sociodemographic and clinical characteristics of the patients in the “high suicide risk” group were compared to all other patients in the screening sample, using a chi-square test to identify significant differences between groups. The same measures, as well as indicators of trial participation, were compared within the subsample that participated in the clinical trial. Results: Participants (n=2284) who identified as Latino immigrants ages 18 and up were screened in clinics and community agencies with the AC-OK Screen for Co-occurring Disorders. Participants who endorsed past suicidal ideation, plans, or attempts then received the Paykel suicide questionnaire. Affirmative responses to either of two Paykel questions (presence of active suicidal ideation or plans in the past 30 days; presence of suicide attempt in the past 30 days) resulted in activation of the suicide prevention protocol. 20.3% of patients (n=464) screened positive to one or two of the AC-OK items. 4.3% (n=98) screened positive to one or two Paykel questions (“high suicide risk” group), and 38.8% of these (n=38) later entered the clinical trial. There were no differences in age, gender, site, or education between the high suicide risk group and other clinical trial participants. The high risk group had higher mental health and substance abuse symptom scores at baseline (p<0.01) and were more likely to endorse suicide risk items in follow-up surveys (23.7% compared to 2.6%). There was no difference between groups in number of assessment sessions completed after baseline (retention in trial) or in the number of clinical sessions attended in the treatment group (adherence to intervention). Discussion: This study highlights the feasibility of implementing a consistent suicide risk prevention protocol in a multi-site clinical trial serving a predominantly low-income, medically underserved population. Not only were a large proportion of screened participants who received a suicide risk intervention later successfully recruited to the clinical trial, but their retention rates and intervention adherence rates were similar to other trial participants (despite their higher level of clinical symptoms). The majority of patients in the suicide risk group did not require further suicide risk assessment. Furthermore, the availability of the initial suicide prevention protocol to a large screening sample provides the opportunity to extend suicide prevention activities to a larger population than those that go on to participate in clinical trials.