In September 2009, preliminary results were reported from the British Randomised Injectable Opiate Treatment Trial (RIOTT) on supervised heroin assisted treatment for patients who are dependent on heroin but do not respond to conventional methadone substitution treatment.1 One hundred and twenty seven patients were randomised to six months of supervised injectable heroin, supervised injectable methadone, or optimised oral methadone. Improvement was seen in the primary outcome measure of reduced use or abstinence from “street” heroin in all treatment groups, but the greatest effect was seen in people who received supervised injectable heroin—three quarters remained abstinent from illegal heroin, and number of days with criminal …