Imperial College London


Faculty of MedicineSchool of Public Health

Professor of Mathematical Epidemiology



+44 (0)20 7594 3263mc.boily




LG26Norfolk PlaceSt Mary's Campus






BibTex format

author = {Tavitian-Exley, I and MAHEU-GIROUX, M and PLATT, L and HEIMER, R and UUSKÜLA, A and LEVINA, O and VICKERMAN, P and Boily, M},
doi = {10.1016/j.drugpo.2017.11.010},
journal = {International Journal of Drug Policy},
pages = {96--105},
title = {Differences in risk behaviours and HIV status between primaryamphetamines and opioid injectors in Estonia and Russia},
url = {},
volume = {53},
year = {2018}

RIS format (EndNote, RefMan)

AB - Background and objectivePeople who inject drugs (PWID) account for over half of new HIV infections in Eastern Europe and central Asia, where opioids continue to be the dominant illicit drugs injected. Stimulants including amphetamines (ATS) have been associated with HIV infection risk in several settings. We sought to examine whether primary ATS injection was associated with greater HIV risk, compared to opioid injection in two European locales with significant HIV epidemics.MethodsPWID in Kohtla-Järve and St. Petersburg were recruited using respondent-driven sampling in 2012–2013. Survey data on demographic characteristics, service use, injecting and sexual risk behaviours and HIV-status (and HCV in Kohtla-Järve) were compared between primary opioid and ATS injectors using logistic regression models.ResultsOf 591 injectors recruited in Kohtla-Järve and 811 in St. Petersburg, 195 (33%) and 27 (4%) primarily injected ATS in each city. In both cities, ATS injectors were younger than opioid injectors, initiated injection later, injected less frequently and were more likely to have been paid for sex. In both cities, PWID had high levels of multiple sex partners. In Kohtla-Järve, ATS-injectors had lower odds of back-loading and greater odds of polydrug use than opioid-injectors. In St. Petersburg, where over half of PWID reported unsafe sharing practices, ATS-injectors were less likely to report these practices. ATS-injection was negatively associated with being HIV positive in Kohtla-Järve (aOR=0.6; 95%CI: 0.5–0.8) and St. Petersburg (aOR=0.3; 95%CI: 0.1–0.7). ATS-injection was negatively associated with HCV-reactivity in Kohtla-Järve (aOR=0.5; 95%CI: 0.3–0.6).ConclusionsIn both locations, primary ATS injection was associated with lower injecting risk behaviours, lower odds of HIV and being paid for sex compared to opioid injection. Interventions targeting the characteristics and needs of ATS injectors are need
AU - Tavitian-Exley,I
AU - Boily,M
DO - 10.1016/j.drugpo.2017.11.010
EP - 105
PY - 2018///
SN - 0955-3959
SP - 96
TI - Differences in risk behaviours and HIV status between primaryamphetamines and opioid injectors in Estonia and Russia
T2 - International Journal of Drug Policy
UR -
UR -
VL - 53
ER -