Aim: To estimate associations between recent licit and illicit substance use and subsequent suicide attempt among people who inject drugs (PWID).
Design: Secondary analysis of longitudinal data from a prospective cohort study of PWID followed bi-annually between 2004 and 2011.
Setting: Montréal, Canada.
Participants: Seven hundred and ninety-seven PWID who reported injection drug use in the previous 6 months, contributing to a total of 4460 study visits. The median number of visits per participant was five (interquartile range: 3-8).
Measurements: An interviewer-administered questionnaire eliciting information on socio-demographic factors, detailed information on substance use patterns and related behaviours, mental health markers and suicide attempt. The primary exposure variables examined were past-month use of alcohol [heavy (>= 60 drinks); moderate (one to 59 drinks); none], sedative-hypnotics, cannabis, cocaine, amphetamine and opioids [regular (>= 4 days); occasional (1-3 days); none]. The outcome was a binary measure of suicide attempt assessed in reference to the previous 6 months.
Findings: In multivariate analyses, a positive association was found among licit substances between heavy alcohol consumption [adjusted odds ratio (AOR) = 2.05; 95% confidence interval (CI) = 1.12-3.75], regular use of sedative-hypnotics (AOR = 1.89; 95% CI = 1.21-2.95) and subsequent attempted suicide. Among illicit substances, occasional use of cannabis (AOR = 1.84; 95% CI = 1.09-3.13) had a positive association with subsequent suicide attempt. No statistically significant association was found for the remaining substances.
Conclusion: Among people who inject drugs, use of alcohol, sedative-hypnotics and cannabis, but not cocaine, amphetamine or opioids, appears to be associated with an increased likelihood of later attempted suicide.