Evaluating the impact of opioid substitution treatment on mortality – a CPRD cohort and prison cohort study in UK
Date/Time: 11 May 2017, 13:00-14:00
Venue: Baddiley-Clark Seminar Room
Aim:- We use cohort studies in UK primary care and prisons to examine two hypotheses:- (a) to test the hypothesis that buprenorphine reduces mortality risk compared to methadone in the community; (b) to test whether prison-based opioid substitution treatment (OST) reduces post-release mortality
Methods:– (A) Cohort study of patients prescribed methadone or buprenorphine for opioid substitution treatment (OST) 1998-2014 extracted from the Clinical Practice Research Datalink, comprising 11,033 patients, 26,546 OST episodes and 30,410 person years follow-up. (B) National prospective cohort study of 15,141 incarcerations and 12,260 adult prisoners with OUD recruited from 39 prisons (and released from 123 prisons) in England during 2010-2016 linked to Prison Health, Justice Statistics Analytical Services, Office for National Statistics, and National Drug Treatment Monitoring System. 8,645 exposed to OST on release and 6,496 unexposed.
Results:– (A) The lowest risk period was from 4 weeks on treatment until treatment cessation: all-cause mortality (ACM) and drug related poisoning (DRP) rates of 0.98 and 0.29 per 100 person years respectively. Patients on buprenorphine compared to methadone had lower ACM rate in each treatment period. The fully adjusted IRR was 0.04 (95% 0.015 to 0.17) and 0.17 (95%CI 0.12 to 0.24) for patients on burprenorphine compared to patients on methadone in the first 4 weeks of treatment and first 4 weeks out of treatment respectively. DRP risk was lower during OST for patients prescribed buprenorphine compared to methadone: adjusted IRR 0.08 (95%CI 0.01 to 0.48) and 0.37 (95%CI 0.17 to 0.79) for first four weeks and rest of time on OST respectively. Duration on treatment was shorter on buprenorphine than methadone. There was insufficient evidence that methadone or bupreneorphine reduce drug related poisoning in the population (weighted IRR 1.32 95%CI 0.47-4.57 and 2.15 95%CI 0.45-5.19) for methadone and buprenorphine respectively.
(B) There were 401 deaths during the observation period (160 in first year; 24 in the first month). The mortality risk in the OST exposed group was lower than the unexposed group in the first 4 weeks 0.93 (95%CI 0.4-2.1) per 100 person years (p100py) vs 3.67 (95%CI 2.3-5.8) p100py– unadjusted hazard ratio (HR) 0.25 (95%CI 0.10 to 0.64). There was no evidence of any difference in mortality risk from 4 weeks-4 months and from 4 months-1 year. OST exposed prisoners were more likely to enter community treatment than the non-exposed group (OR 2.47, 95%CI 2.3-2.65). The protective effect of OST exposure was not attenuated after adjustment for demographic or behavioural confounders or for community drug treatment (adjusted HR 0.27, 95% CI 0.11-0.71).
Conclusions:- (A) Buprenorphine lowers all cause and drug related poisoning mortality compared to methadone in the first four weeks of treatment. The combination of short treatment duration and raised mortality risk after leaving treatment may reduce the protective benefit of OST in the population. (B) OST at prison release lowered risk of risk of mortality in the first month by 75% (removing the excess risk of death people with an OUD experience leaving prison compared to rest of time in the community) and increased the likelihood of entering drug treatment in the community.
Bio: - Matt Hickman is a Professor in Public Health and Epidemiology at Bristol, NIHR Senior Investigator, and Honorary Public Health Consultant at Bristol City Council and Public Health England. He is the director of NIHR Health Protection Research Unit on Evaluation of Interventions, and a member and co-investigator of UK CRC Public Health Centre of Excellence (DECIPHer: Centre for the Development and Evaluation of Complex Interventions for Public Health Improvement) and NIHR School of Public Health Research. His research programme focuses on epidemiology and public health consequences of drug misuse – including adolescent substance use, and epidemiology and prevention of HCV and drug related mortality. He is deputy regional editor of Addiction, a member of the Scientific Committee of European Monitoring Centre on Drugs and Drug Addiction and WHO Technical Advisory Group on alcohol and drug epidemiology. MH is co-lead of MRC Addiction Research Clinical Training programme (MARC).