PT  - JOURNAL ARTICLE
AU  - Sarah Larney
AU  - Matthew Hickman
AU  - David A Fiellin
AU  - Timothy Dobbins
AU  - Suzanne Nielsen
AU  - Nicola R Jones
AU  - Richard P Mattick
AU  - Robert Ali
AU  - Louisa Degenhardt
TI  - Using routinely collected data to understand and predict adverse outcomes in opioid agonist treatment: Protocol for the Opioid Agonist Treatment Safety (OATS) Study
AID  - 10.1136/bmjopen-2018-025204
DP  - 2018 Aug 01
TA  - BMJ Open
PG  - e025204
VI  - 8
IP  - 8
4099  - http://bmjopen.bmj.com/content/8/8/e025204.short
4100  - http://bmjopen.bmj.com/content/8/8/e025204.full
SO  - BMJ Open2018 Aug 01; 8
AB  - Introduction North America is amid an opioid use epidemic. Opioid agonist treatment (OAT) effectively reduces extramedical opioid use and related harms. As with all pharmacological treatments, there are risks associated with OAT, including fatal overdose. There is a need to better understand risk for adverse outcomes during and after OAT, and for innovative approaches to identifying people at greatest risk of adverse outcomes. The Opioid Agonist Treatment and Safety study aims to address these questions so as to inform the expansion of OAT in the USA.Methods and analysis This is a retrospective cohort study using linked, routinely collected health data for all people seeking OAT in New South Wales, Australia, between 2001 and 2017. Linked data include hospitalisation, emergency department presentation, mental health diagnoses, incarceration and mortality. We will use standard regression techniques to model the magnitude and risk factors for adverse outcomes (eg, mortality, unplanned hospitalisation and emergency department presentation, and unplanned treatment cessation) during and after OAT, and machine learning approaches to develop a risk-prediction model.Ethics and dissemination This study has been approved by the Population and Health Services Research Ethics Committee (2018HRE0205). Results will be reported in accordance with the REporting of studies Conducted using Observational Routinely-collected health Data statement.