Year: 2016
Type: article
Abstract: Fischbacher et al. 1 invite a clearer ‘theory of change’ explanation for why those at highest risk of opioid-related death (ORD) might be expected to benefit most from naloxone's availability to peers 2. The period soon after prison release is recognized internationally as high risk for ORDs 3-8. As Scotland's National Naloxone Programme (NNP) took hold, and before optimal availability 8, recency of prison release would prompt peers to administer any available naloxone to the ex-prisoner—be it naloxone-on-release (NOR) or community-prescribed take-home naloxone (THN), supportive evidence being: one-third of NOR was administered to the ex-prisoner but only one-eighth of THN to the prescribee 2. We agree 2 that the relative reduction in ORDs (and therefore effectiveness) would be expected to be higher for the primary outcome of prison release ORDs than for the secondary outcome of prison release and/or hospital-discharge ORDs. Indeed, the secondary outcome was defined to be powerful at 3 years in respect of only 20%, rather than 30%, effectiveness 8. Prior to autumn 2015, we are not aware that Scottish hospitals had a protocol for issuing THN on hospital discharge to opioid-dependent patients. NHS Greater Glasgow and Clyde approved such a guideline in September 2015 9. As Scottish hospitals have not acted on injectors’ high risk soon after hospital discharge 10, 11, it is unsurprising that those with a history of injection drug use had not taken on board that, for them, the ORD risk in the 4 weeks after hospital discharge is high, roughly half that in the 4 weeks after prison release. Fischbacher et al. suggest that only the primary outcome provided evidence of effectiveness, but we showed 22% effectiveness for Scotland's secondary outcome (95% confidence interval = 7–33%), consistent with previous expectations 2, 8. Cost-effectiveness was evaluated as if only prison-release ORDs were prevented because sufficient to justify NNP's prescription costs 2. There is indeed no significant interaction by gender or age-group in effect size for the primary outcome 2, interactions being more often in the eye of the beholder than evidenced. Specificity need not relate exposure to a single outcome. Hill 12 recognized smoking as a major cause of lung cancer and as one cause of cardiovascular mortality; nickel refining pre-1923 caused sinus and lung cancers. Formal evaluation of Scotland's NNP was science-led and statistically powerful by design 8: not a naive before/after evaluation. Previous epidemiological assumptions on effect size and probable numbers of ORDs 8 were met within the 3-year horizon that the Scottish Government had set initially 2. Further testing at 5 years is powerful to discern 20% effectiveness for just hospital-discharge ORDs. Our acknowledgements and supplementary material 2 recognized the immense range of practical experience and oversight that the National Naloxone Advisory Group contributed, the National Naloxone Co-ordinator's inspirational role, the work of peer-trainers and, in quantifying injectors’ access to naloxone, Scotland's Needle Exchange Surveillance Initiative 8, 13, with information to follow from the Scottish Prisoner Surveys in 2014 and 2015 conducted by the Scottish Prison Service and on engagement by general practitioners. Our paper is certainly not the last word on Scotland's NNP, and we look forward to engaging in future debate and meta-analytical studies on this key public health policy. S.M.B. is a co-principal investigator for the MRC-funded N-ALIVE pilot Trial which, by unanimous decision of its Trial Steering and Data Monitoring Committee, ceased randomization on 8 December 2014 and offered naloxone-on-release to already randomized, not-yet-released prisoners who had been assigned to the control group. All four authors serve on Scotland's National Naloxone Advisory Group, which C.H. chairs, but write in a personal capacity. S.M.B. was, on two occasions, funded, invited speaker at the Academic and Health Policy Conference on Correctional Health. S.M.B. holds GSK shares.
Source: Addiction
DOI: 10.1111/add.13391
Language: English
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2340767251