Causes of death in a cohort treated for opioid dependence between 1985 and 2005 Louisa Degenhardt 1,2 , Sarah Larney 1,3 , Deborah Randall 4 , Lucy Burns 1 & Wayne Hall 5,6 National Drug and Alcohol Research Centre, University of New South Wales, Sydney, Australia, 1 Melbourne School of Population and Global Health, University of Melbourne, Melbourne, Australia, 2 Alpert Medical School, Brown University, Providence, RI, USA, 3 Centre for Health Research, School of Medicine, University of Western Sydney, Penrith, Australia, 4 University of Queensland Centre for Clinical Research, University of Queensland, Herston, Australia 5 and National Addiction Centre, Kings College, London, UK 6 ABSTRACT Aims To examine changes in causes of death in a cohort treated for opioid dependence, across time and age; quantify years of potential life lost (YPLL); and identify avoidable causes of death. Design People in New South Wales (NSW) who registered for opioid substitution therapy (OST), 1985–2005, were linked to a register of all deaths in Australia. Setting NSW, Australia. Measurements Crude mortality rates (CMRs), age–sex-standardized mortality rates (ASSRs) and standardized mortality ratios (SMRs) across time, sex and age. Years of potential life lost (YPLL) were calculated with reference to Australian life tables and by calculating years lost before the age of 65 years. Findings There were 43 789 people in the cohort, with 412 216 person-years of follow-up. The proportion of the cohort aged 40+ years increased from 1% in 1985 to 39% in 2005. Accidental opioid overdoses, suicides, transport accidents and violent deaths declined with age; deaths from cardiovascular disease, liver disease and cancer increased. Among men, 89% of deaths were potentially avoidable; among women, 86% of deaths were avoidable. There were an estimated 160555 YPLL in the cohort, an average of 44 YPLL per decedent and an average of 29 YPLL before age 65 years. Conclusions Among a cohort of opioid-dependent people in New South Wales, 1985–2005, almost nine in 10 deaths in the cohort were avoidable. There is huge scope to improve mortality among opioid-dependent people. Keywords Ageing, heroin, injecting drug use, mortality, opioid dependence. Correspondence to: Louisa Degenhardt, National Drug and Alcohol Research Centre, University of New South Wales, Sydney 2052, Australia. E-mail: l.degenhardt@unsw.edu.au Submitted 28 November 2012; initial review completed 25 February 2013; final version accepted 13 August 2013 INTRODUCTION Elevated mortality rates and excess mortality have been well described in opioid-dependent populations in devel- oped countries. Dependent opioid users in these countries may be 15 times more likely to die than their non-opioid- using peers [1]. Drug overdose is the most common cause of death and chronic diseases, suicide and injury also contribute to excess mortality [1–3]. In recent decades changes have been observed in the contribution of different causes of death to premature mortality in opioid users. For example, AIDS-related deaths in Italy and Spain increased throughout the 1980s and 1990s, and then decreased after the introduc- tion of highly active antiretroviral therapy [2,4–6]. Recent studies have reported an increased contribution of liver disease [7,8] and cancer [9] to mortality in Australian cohorts. Few studies, however, have described the full range of causes of death in this population and how they have varied over time and with age [1,10]. This is a particularly pertinent issue given the ageing of opioid-dependent populations in Australia and western Europe [11,12]. A second important public health consideration is to what extent deaths among opioid-dependent people are avoidable, given current knowledge about effective pre- ventive and health care [13]. Analysis of ‘avoidable mortality’ and years of potential life lost may identify potential targets for intervention [10]. Avoidable mortal- ity includes conditions in which current health care may reduce case fatality, as well as conditions for which there are effective preventive interventions [13]. RESEARCH REPORT doi:10.1111/add.12337 © 2013 Society for the Study of Addiction Addiction, 109, 90–99