The association between psychosocial and structural-level stressors and HIV injection drug risk behavior among Malaysian fishermen: A cross-sectional study
AIMS To identify the prevalence and correlates of midazolam injection among injecting drug users in Thailand. DESIGN Serial cross-sectional mixed-methods study. SETTING Bangkok, Thailand. PARTICIPANTS A total of 435 adults who had injected drug(s) in the past 6 months were recruited through peer outreach and word of mouth in Bangkok in 2011. MEASUREMENTS Multivariable logistic regression was used to analyze self-reported data collected through an interviewer-administered survey in July-October 2011 (n = 435). Additionally, qualitative data were collected in June-July 2009 (n = 10) and analyzed to explore the health problems attributable to midazolam injection. FINDINGS Among 435 survey participants, the prevalence of daily midazolam injection in the past 6 months was 37.0% (95% confidence interval: 32-42). While 75.8% of the daily midazolam injectors identified heroin as their drug of choice, 91.8% of these individuals reported substituting heroin with midazolam when heroin was unavailable. In multivariable analyses, daily midazolam injection was associated with femoral vein injection [adjusted odds ratio (AOR) = 4.48], polydrug use (AOR = 4.94), daily heroin injection (AOR = 3.25), binge use (AOR = 1.75) and perceived increase in policing pressure (AOR = 1.95) (all P < 0.05). Qualitative accounts indicated severe health problems associated with midazolam injection, including nerve and vascular injuries. CONCLUSION Midazolam misuse is increasing among injecting drug users in Bangkok and appears to be used primarily as a substitute for heroin. Midazolam injection was associated independently with various risk factors for injection-related complications. Given the many deleterious effects of midazolam injection, novel interventions for midazolam injectors are needed urgently.