A multi-site sample of currently-injecting drug users (IDUs) comprisin
g 344 men and 136 women was recruited in Edinburgh. Sixty-seven per ce
nt of the sample said they had at some time used injecting equipment a
lready used by another person and 25% admitted doing so in the 6 month
s before interview. Whereas women who injected with used equipment obt
ained it predominantly from a sexual partner, for men the source was m
ore often a close friend or someone whose HIV status they were unlikel
y to know. In the 6 months before interview, 40% of men, compared with
20% of women, had more than one heterosexual partner. This difference
was associated with a higher proportion of men with steady partners a
lso having casual partners. Women IDUs were more likely to have regula
r partners who injected (57% vs 26%). Though shaving of injecting equi
pment has already diminished in Edinburgh, further measures are needed
to eliminate it For injectors here, the risk of infection from unprot
ected heterosexual intercourse may now be greater than that from shari
ng injecting equipment, particularly for women. Other methods of encou
raging changes in sexual behaviour need to be investigated and success
ful ones promoted.