Background Injecting drug use (IDU) is an important risk for viral hepatitis transmission. Detailed, transparent estimates of the scale of the problem at regional and global levels have never been made. We report national, regional and global prevalence and population size estimates for hepatitis C (HCV) and hepatitis B (HBV) among people who inject drugs. Methods Systematic search of peer-reviewed (Medline/Embase/PsycINFO) and grey literature databases, conference abstracts and online resources, with a widely distributed call for additional data. From 4386 peer-reviewed and 1019 grey literature sources, 1125 were reviewed in full. Studies were extracted to a customised database and graded according their methods. Serological reports of HCV antibodies/anti-HCV, HBV antibodies/anti-HBc, and/or HBV surface antigen/HBsAg among IDUs samples with n>40 participants, <100% HIV-positive, and sampling frames that did not exclude participants on the basis of age or sex were included. Using endorsed decision rules, prevalence estimates were calculated with anti-HCV and anti-HBV as proxies for exposure and HBsAg for current infection. These were combined with IDU population sizes to estimate the number of HBV and HCV positive IDUs. Findings Eligible reports of anti-HCV among IDUs were located for 77 countries. Prevalence was 60–80% in 26 countries and >80% in 12. We estimate worldwide about 10.0 million (range 6.0–15.2M) IDUs might be anti-HCV positive. China, (1.6M), the USA (1.5M) and the Russian Federation (1.3M) had by far the largest such populations. HBsAg reports were found for 59 countries, ranging from 5–10% in 21 countries and over 10% in 10. Worldwide, 6.4 million IDU might be anti-HBc positive (2.3–9.7M), and 1.2 million (0.3–2.7M) HBsAg positive. Interpretation The prevalence of anti-HCV among IDUs is far greater than HIV. Viral hepatitis clearly poses a challenge to public health. Variation in the coverage and quality of existing research creates uncertainty around estimates. Better and more complete data and reporting are required to estimate the scale of the problem, to inform efforts to prevent and treat HCV and HBV among IDUs.

No abstract

Mortality among opioid-dependent users varies across countries and populations. Treatment is clearly protective against mortality even in non-randomized observational studies. Study characteristics predict mortality levels; these should be taken into account in future studies.

No abstract

No abstract

No abstract

Opioid dependence is a substantial contributor to the global disease burden; its contribution to premature mortality (relative to prevalence) varies geographically, with North America, eastern Europe and southern sub-Saharan Africa most strongly affected.

No abstract