BACKGROUND: This cross-sectional study estimated the prevalence of contraceptive methods and investigated whether abortion rates influence contraceptive behaviour among injection drug-using (IDU) women in St Petersburg, Russia. METHODOLOGY: A self-administered questionnaire of behaviour in the last 3 months was applied to a convenient sample of IDU women. RESULTS: Of 80 sexually active participants, 67% had had an abortion. No participant reported using hormonal contraceptives or intrauterine devices (IUDs). The only valid method of contraception used was condoms, which was reported by half of the participants. Consistent condom use was reported by 22% of participants and was no more likely among those who had an abortion. Condom use was significantly associated with having multiple or casual sex partners [prevalence ratio (PR) 1.75, 95% (confidence interval) CI 1.11-2.78, p = 0.01], having an IDU sex partner (PR 0.55, 95% CI 0.36-0.85, p = 0.029) and with a negative attitude toward condoms (PR 0.53, 95% CI 0.33-0.84, p = 0.01). Abortions were less likely among those who had multiple or casual sex partners (PR 0.69, 95% CI 0.49-0.97, p = 0.03). CONCLUSIONS: Despite the high prevalence of abortions among IDU women, none reported the use of hormonal contraception or IUDs. Having had an abortion was not associated with greater likelihood of using condoms. Participants mostly used condoms with casual or multiple sex partners, suggesting that condoms were used mainly to prevent HIV/sexually transmitted infection transmission and not to prevent pregnancy. Programmes to prevent unwanted pregnancies and reduce abortion-related health risks among this understudied vulnerable group are needed.