Abstract
INTRODUCTION: Research identifying pathways to heroin use has typically been conducted among urban populations. This study examined heroin initiation following pharmaceutical opioid use in three suburban/exurban Southern California counties. METHODS: Interviewer-administered surveys collected data among 330 participants (65.9 % male; 63.9 % non-Hispanic white) whose initial use of any opioid was a pharmaceutical opioid. Retrospective discrete-time survival analysis identified predictors of heroin initiation, measured as self-reported age of first heroin use. RESULTS: Median age of first pharmaceutical opioid use was 17 years; 50.6 % initially acquired pharmaceutical opioids from an illicit source, 56.7 % first used pharmaceutical opioids for recreational purposes, and 86 % initiated heroin use. Average time from first pharmaceutical opioid use to first heroin use was 8.2 years. Drug/alcohol treatment (adjusted Hazard Ratio [aHR]: 0.67, 95 % CI: 0.50, 0.88) was associated with delayed time to heroin initiation. Obtaining opioids from non-medical sources (aHR: 2.21, 95 % CI: 1.55, 3.14) was associated with accelerated time to heroin initiation. Reporting supply problems with obtaining pharmaceutical opioids (e.g., unable to acquire pharmaceutical opioids) was associated with accelerated time to heroin initiation, but the magnitude of this effect was dependent on one's history of methamphetamine use (p < 0.05). CONCLUSIONS: Time to heroin initiation following pharmaceutical opioid use was accelerated among those reporting supply problems and delayed among those with exposure to substance use treatment. Interventions interrupting supply of opioids might benefit from coordination with evidence-based medication-assisted treatment to minimize the risk of transitioning to heroin use, particularly among those with a long history of non-prescribed pharmaceutical opioid use.