“…Mortality studies in cohorts of problem drug users or injectors are arguably a better approach still (see Wahren et al, 1997;Versino et al, 2000;Langendam et al, 2001;Bargagli et al, 2001;Risser et al, 2001;Gossop et al, 2002;Hickman et al, 2003;Galai et al, 2003;Copeland et al, 2004;Bargagli et al, 2005;Bloor et al, 2008) because: a) they document not only drugsrelated but also HIV-related and other-cause mortality (see Ross, 2002 andBloor et al, 2008), b) the covariate influences of sex and current age on the risk of drugs-related death can be estimated for individuals whose drugs history, including treatment referral (Buster et al, 2002 andFugelstad et al, 2007), route of administration of primary drug and initiation age, was documented in a standardized manner at enrolment to the cohort (Hickman et al, 2003 andCopeland et al, 2004), and c) with shared protocol for enrolment, comparably-estimated risks can be compared across cohorts. , for example, applied evidence-synthesis techniques to published data from Scotland to deduce that the relative risk of drugs-related death per 100 injectors could be 1.8 times as high for males (95% CI: 1.3 to 2.3) and two to six times higher for older injectors.…”