Jim Lewsey

Aim: This simulation study is to assess the utility of physician's prescribing preference (PPP) as an instrumental variable for moderate and smaller sample sizes. Materials & methods: We designed a simulation study to imitate a comparative effectiveness research under different sample sizes. We compare the performance of instrumental variable (IV) and non-IV approaches using two-stage least squares (2SLS) and ordinary least squares (OLS) methods, respectively. Further, we test the performance of different forms of proxies for PPP as an IV. Results: The percent bias of 2SLS is around approximately 20%, while the percent bias of OLS is close to 60%. The sample size is not associated with the level of bias for the PPP IV approach. Conclusion: Irrespective of sample size, the PPP IV approach leads to less biased estimates of treatment effectiveness than OLS adjusting for known confounding only. Particularly for smaller sample sizes, we recommend constructing PPP from long prescribing histories to improve statistical power.

Background and aims On 1 May 2018, Scotland implemented Minimum Unit Pricing (MUP) of £0.50 per unit of alcohol with the aim to lower alcohol consumption and related harms, and reduce health inequalities. We measured the impact of MUP on the most likely categories of road traffic accidents (RTAs) to be affected by drink‐driving episodes (fatal and nighttime) up to 20 months after the policy implementation. Further, we checked whether any association varied by level of socio‐economic deprivation. Methods An interrupted time series design was used to evaluate the impact of MUP on fatal and nighttime RTAs in Scotland and any effect modification across socio‐economic deprivation groups. RTAs in England and Wales (E&W) were used as a comparator. Covariates representing severe weather events, bank holidays, seasonal and underlying trends were adjusted for. Results In Scotland, MUP implementation was associated with 40.5% (95% confidence interval: 15.5%, 65.4%) and 11.4% (−1.1%, 24.0%) increases in fatal and nighttime RTAs, respectively. There was no evidence of differential impacts of MUP by level of socio‐economic deprivation. While we found a substantial increase in fatal RTAs associated with MUP, null effects observed in nighttime RTAs and high uncertainty in sensitivity analyses suggest caution be applied before attributing causation to this association. Conclusion There is no evidence of an association between the introduction of minimum unit pricing for alcohol in Scotland and a reduction in fatal and nighttime road traffic accidents, these being outcome measure categories that are proxies of outcomes that directly relate alcohol consumption to road traffic accidents.