Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality

Sadie Boniface; Jack W Scannell; Sally Marlow

doi:10.1136/bmjopen-2016-013497

Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality

Boniface, Sadie; Scannell, Jack W; Marlow, Sally 2017-06-06 00:00:00 Open Access Research Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality 1 2 1 Sadie Boniface, Jack W Scannell, Sally Marlow To cite: Boniface S, ABSTRACT Strengths and limitations of this study Scannell JW, Marlow S. Objectives: To assess the evidence for price-based Evidence for the effectiveness alcohol policy interventions to determine whether ▪ This review adds to an emerging literature of sys- of minimum pricing of minimum unit pricing (MUP) is likely to be effective. tematic reviews synthesising findings using the alcohol: a systematic review Design: Systematic review and assessment of studies Bradford Hill criteria for causality in research and assessment using the according to Preferred Reporting Items for Systematic Bradford Hill criteria areas where traditional meta-analyses of rando- Reviews and Meta-Analyses (PRISMA) guidelines, for causality. BMJ Open mised controlled trials are not possible or 2017;7:e013497. against the Bradford Hill criteria for causality. Three appropriate. doi:10.1136/bmjopen-2016- electronic databases were searched from inception to ▪ A range of study designs were included, allowing February 2017. Additional articles were found through for a comprehensive review of a disparate evi- hand searching and grey literature searches. dence base to investigate whether minimum unit Prepublication history and Criteria for selecting studies: We included any pricing of alcohol is likely to reduce alcohol con- additional material is study design that reported on the effect of price- sumption and alcohol-related harm. available. To view please visit based interventions on alcohol consumption or ▪ Studies examining the effects of alcohol taxation the journal (http://dx.doi.org/ alcohol-related morbidity, mortality and wider harms. or changes in alcohol affordability, or studies 10.1136/bmjopen-2016- Studies reporting on the effects of taxation or solely reporting on price elasticity of demand, 013497). affordability and studies that only investigated price were not included. elasticity of demand were beyond the scope of this ▪ Methodological quality of studies was variable. review. Studies with any conflict of interest were Received 15 July 2016 excluded. All studies were appraised for Revised 28 February 2017 methodological quality. INTRODUCTION Accepted 6 March 2017 Results: Of 517 studies assessed, 33 studies were There are many policies and programmes included: 26 peer-reviewed research studies and seven that aim to reduce harms from alcohol. from the grey literature. All nine of the Bradford Hill One of these is minimum alcohol pricing, criteria were met, although different types of study which exists in a number of countries around satisfied different criteria. For example, modelling the world. The most notable example of studies complied with the consistency and specificity this is Canada, where there are government criteria, time series analyses demonstrated the monopolies on alcohol sales and a variety of temporality and experiment criteria, and the analogy types of minimum pricing in operation. For criterion was fulfilled by comparing the findings with example, there is a minimum price per litre the wider literature on taxation and affordability. of a particular drink in British Columbia and Conclusions: Overall, the Bradford Hill criteria for a (higher) minimum price linked to drink causality were satisfied. There was very little evidence 1 3 Department of Psychology that minimum alcohol prices are not associated with type and strength in Saskatchewan. Other and Neuroscience, National consumption or subsequent harms. However the countries with minimum alcohol pricing Addiction Centre, Institute of overall quality of the evidence was variable, a large include Belarus, Kyrgyzstan, the Republic of Psychiatry, King’s College proportion of the evidence base has been produced by Moldova, the Russian Federation and London, London, UK a small number of research teams, and the quantitative School of Social and Ukraine. Minimum alcohol pricing is being uncertainty in many estimates or forecasts is often 5 Political Science, University considered by governments in Ireland and poorly communicated outside the academic literature. of Edinburgh, Edinburgh, UK has also been reviewed in Australia and New Nonetheless, price-based alcohol policy interventions Zealand. Correspondence to such as MUP are likely to reduce alcohol consumption, The situation with regards to minimum Dr Sadie Boniface; alcohol-related morbidity and mortality. [email protected] alcohol pricing in the UK is complex. Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 1 Open Access In England and Wales, there has been a ban on alcohol In light of this ongoing consideration of MUP in the being sold at below cost (the total amount of ‘duty plus UK, in this paper we assess the effectiveness of value added tax (VAT)’) since May 2014; and the ﬁrst minimum alcohol price interventions to reduce alcohol- conviction for selling alcohol below this level took place related harm. Alcohol-related harm costs the National in 2016. Duty plus VAT is equivalent to a 70 cl bottle of Health Service in England £3.5 billion each year and the vodka (37.5% alcohol by volume (ABV)) costing a estimated cost to society is £21 billion per year. The minimum of £8.72, whereas under a minimum price latest annual ﬁgures for England (population of 54 of 50 pence per unit (one UK unit=10 mL or 8 g million) show over 1 million alcohol-related hospital ethanol), this would cost £13.13. In 2012, the UK coali- admissions (2013/2014) and 6500 alcohol-related tion government cited support for minimum unit deaths (2013); and these ﬁgures represent increases pricing (MUP) in its alcohol strategy, and legislation to compared with a decade previously of 115% and 10%, have a minimum price of £0.50 per unit was passed in respectively. Scotland the same year. Following the change to a We systematically review the literature on the effect of Conservative majority government in 2015, it is unclear price interventions or policies such as MUP on alcohol whether there is still central government support for consumption, alcohol-related morbidity and mortality, MUP. In Scotland, the Scotch Whisky Association chal- and wider harms. We use the nine Bradford Hill criteria lenged the 2012 legislation in the Scottish Court of for causality as a framework with the aim of assessing the Session, which referred the case to the Court of Justice likely effectiveness of MUP as a policy to reduce alcohol of the European Union (CJEU) in 2014. In late 2015, consumption and alcohol-related harm. the CJEU referred the case back to the Scottish courts to investigate proportionality (that the same objective cannot be met through increased taxation), which METHODS could have implications for other EU countries consider- A systematic literature search was performed according ing MUP. In late 2016, the Scottish Court of Session to Preferred Reporting Items for Systematic Reviews and ruled that MUP does not contravene EU law; however, Meta-Analyses (PRISMA) guidance (see ﬁgure 1 for the Scotch Whisky Association then appealed to the UK PRISMA ﬂow diagram and online supplementary ﬁle for Supreme Court. excluded studies). Figure 1 PRISMA 2009 flow diagram of studies in this systematic review. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses. 2 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Open Access Identification of studies criteria and agreed which studies provided relevant evi- Three electronic databases were searched for titles or dence for or against each criterion. abstracts containing ‘minimum unit pric$’ OR ‘minimum pric$’ OR ‘ﬂoor pric$’ OR ‘pric$ AND RESULTS policy’ AND alcohol. The databases were PsycINFO The included studies that are published in peer-reviewed (1806 to February Week 2 2017), Embase (1974 to 2017 journals (26 research studies and two systematic reviews) Week 07) and Ovid Medline (1946 to February Week 2 are listed by study type in table 2 with information on 2017). We also searched the websites of ﬁve alcohol study characteristics and methodological quality. Of the charities for publications or reports related to ‘price’, research studies, the methodological quality was rated as and also searched 20 leading UK think tanks for ‘strong’ in 15 studies, ‘moderate’ in 8 studies and ‘weak’ ‘alcohol’ or ‘addiction’. in 3 studies. Both of the systematic reviews were rated Inclusion criteria were: any study design; population- ‘strong’. The seven reports from the grey literature are level studies exploring at least one aspect of the effect of listed in table 3. Five of the seven were rated as of interventions or policies leading to changes in the ‘strong’ methodological quality, with the remaining two minimum price of alcohol, including but not limited to not appropriate to rate using our critical appraisal tools. changes in alcohol sales, consumption, morbidity and mortality; and individual-level studies exploring Consideration against the Bradford Hill criteria minimum alcohol prices and alcohol purchasing, con- for determining causality sumption, morbidity or mortality; written in English. Strength of the association Exclusion criteria were: studies about taxation, afford- In 16/26 studies published in peer-reviewed journals, ability, price elasticity of demand for alcohol and strength of the association between pricing and alcohol general changes in alcohol price not the result of an consumption or alcohol-related harm was evidenced by intervention or policy (there is a large literature on each a summary statistic such as an OR, and by a test of the of these already and reviewing all of these studies was statistical signiﬁcance of the association. As well as the beyond the scope of this review); studies about public statistical signiﬁcance of the summary statistic, the mag- perceptions of MUP; and studies where a conﬂict of nitude of the effect was also considered, as a larger interest was reported in the paper, whether this was in effect size corresponds to a greater population health favour of or against MUP. impact. Studies in Canada found that 10% increases All 33 studies that met the inclusion criteria were in minimum prices were associated with reductions in assessed against the Bradford Hill criteria for causality alcohol consumption of 3.4–8.4%, reductions in and the methodological quality appraised. These alcohol-attributable hospital admissions of 9% and included 26 original research studies and seven studies reductions in alcohol-related mortality of 32%, each of from the grey literature, and in addition two systematic which was statistically signiﬁcant. Cross-sectional reviews pertinent to the analogy criterion were included. 32–36 38–40 studies in the UK, Ireland, Australia and one Of the 26 research studies, there were 9 cross-sectional trial from the USA found statistically signiﬁcant asso- surveys, 8 time series analyses or similar, 7 modelling ciations between cheaper alcohol and heavier drinking. studies, 1 qualitative study and 1 trial. The magnitude of the association varied across these studies, but due to different study measures and out- Analysis of included studies comes, the results are not all comparable. As an indica- Quality of included studies was assessed independently tion, the OR for buying alcohol below a proposed by two reviewers and using validated tools. Owing to the minimum price among heavier drinkers was 1.34 in wide variation in study designs among the included 35 34 Crawford’s study, 1.50 in Cousins’s study and 1.70 in studies, the Effective Public Health Practice Project’s Callinan’s study. There was not any evidence to (EPHPP) tool was used for assessing all quantitative support this criterion from the grey literature. Overall, studies, as recommended by the Cochrane Handbook there is reasonably good support for the strength of the for assessing studies in public health. Qualitative association criterion. studies (n=1) and systematic reviews (n=2) included in this review were not covered by the EPHPP tool and so were assessed using the Critical Appraisal Skills Consistency Programme (CASP) tools speciﬁc to these study designs. This criterion requires looking across all the studies Nine criteria to determine causality were suggested by included in the review to see whether similar conclu- Bradford Hill in an inﬂuential 1965 paper. Increasingly, sions have been drawn. Inverse associations between the Bradford Hill criteria are a standard framework to alcohol pricing and alcohol consumption or harm have assess the impact of interventions where it is not ethical been documented in countries in Europe, North or practical to conduct randomised controlled trials America and Australia, and although most studies are (RCTs). Our interpretation of the Bradford Hill criteria from the last 10 years, there are studies from the 1970s for the purpose of this review is listed in table 1.Two and 1980s as well. There is evidence from different reviewers assessed each study against each of the nine research teams and different types of study including Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 3 Open Access Table 1 Bradford Hill criteria for assessing causation and the definitions used in this review Criterion Bradford Hill criteria (1965) Application in this review 1. Strength of the The strength of a supposed association between an A statistically significant change (p<0.05) in alcohol association intervention and an outcome is determined by the consumption or alcohol-related harms, in the appropriate statistic used to measure the protective expected direction. The exact magnitude of the effect of an intervention (eg, relative risk or OR). association was assessed on a study by study This is the most important factor determining basis causation 2. Consistency Has it been repeatedly observed by different Whether different studies conducted in different persons, in different places, circumstances and locations, in different populations, by different times? investigators and at different times have reported similar findings 3. Specificity Specificity is present when the intervention is If pricing was the only reason that alcohol exclusive to the outcome and when the outcome consumption or alcohol-related harm could have has no other known cause or associated risk fallen, this adds to the argument for causality. factors; cautions that this criterion should not be However, if a price intervention was one of a overemphasised and that if specificity is not number of alcohol policy interventions, then this apparent, this does not preclude causation criterion is not satisfied 4. Temporality Refers to temporal relationship of association The pricing intervention studied must have taken between exposure and disease outcome; to infer place before a change in alcohol consumption or causality, exposure must precede outcome harm was observed 5. Dose– If the association is one in which a dose–response If interventions leading to a larger increase in prices response curve or biological gradient can be observed, this had a greater effect on alcohol consumption and adds to the case for causality alcohol-related harm than interventions where the price change was small, or if studies demonstrate that different minimum prices have differing effects, in the expected direction 6. Plausibility A likely biological mechanism linking the Studies that found an association between price intervention to the observed findings helps to and population-level alcohol consumption and that explain causality; plausibility depends on biological heavier drinkers tend to purchase the cheapest knowledge of the day alcohol could demonstrate plausibility 7. Coherence When the evidence from different disciplines Describes whether studies conducted in different sources ‘hangs well together’ and does not conflict settings or disciplines had complementary findings. with other generally known facts, this criterion is Will not be demonstrated by a single study in met isolation but rather the evidence base as a whole 8. Experiment Experimental evidence from laboratory studies or In addition to laboratory studies and RCTs, natural RCTs could potentially provide strongest support for experiments with before-and-after measures could causation also show the effectiveness of minimum unit pricing This criterion often provides the strongest support in a ‘real-world’ setting for causation and describes whether there is empirical evidence for the association 9. Analogy Causality is supported by analogy if there are Other areas of relevance include whether higher similar associations or causal relationships in other taxation on alcohol is associated with reduced areas of relevance, weakest form of evidence of alcohol consumption and alcohol-related harm, and causality may require drawing on additional literature outside of the main systematic review RCTs, randomised controlled trials. cross-sectional studies, time series analyses and econo- Shefﬁeld Alcohol Policy Model is one such model and 25–29 47 45 51 metric modelling studies. Support for the consistency has been applied in England, Scotland and criterion is very strong. Canada and provides very strong support for the speci- ﬁcity criterion. Further support is provided by other dif- 49 50 30 31 Specificity ferent modelling studies in the UK and Australia The speciﬁcity criterion relates to whether changes in and a (non-randomised) trial in the USA. Thus, alcohol consumption or harm could be attributed to any- support for the speciﬁcity criteria is very strong. thing other than the price intervention. Many studies included have statistically adjusted for confounding Temporality factors; however, the best support for the speciﬁcity cri- It is important that pricing interventions take place terion comes from the econometric modelling studies before changes to alcohol consumption and harm to because there is no risk of residual confounding. The attribute causality. Strong support for this criterion 4 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Open Access Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 5 Table 2 Studies published in peer-reviewed journals included in Bradford Hill criteria assessment Study characteristics Study assessment First author Conflict and year Population or Pricing intervention Peer of Quality Bradford Hill Study type published Country Study design participants studied Outcomes studied reviewed interest rating criteria met Natural Bhattacharya Russia Time series Populations of 77 Substantial increases in Mortality Yes Not Strong SA, CON, TE, experiments 2013 analysis of Russian oblasts administratively set alcohol stated PL, CO, EX, and time panel data set (provinces), 1970– prices 1985–1988, along series 2000 with six other antialcohol analyses measures Herttua Finland Time series General population Modelled 1% increase in the Alcohol-related Yes None Strong SA (not 2015 analysis using population average minimum price of all mortality universal registry alcoholic beverages based findings— on actual price increases subgroup only), adjusted for inflation using CON (counter Consumer Price Index findings) TE, PL, CO, EX Stockwell Canada Cross-sectional Population of British Actual minimum price Alcohol Yes None Strong SA, CON, TE, 2012 versus time Columbia increased over a 20-year consumption DR, CO, EX series analysis period. Study modelled a (measured by sales) of ecological 10% increase in the average data minimum price of all alcoholic beverages adjusted by monthly Consumer Price Index Stockwell Canada Cross-sectional Population of Actual minimum price Alcohol Yes Not Strong SA, CON, TE, 2012 versus time Saskatchewan increased over a 7-year consumption stated DR, CO, EX series analysis period. Study modelled a (measured by sales) of ecological 10% increase in the average data minimum price of all alcoholic beverages adjusted by monthly Consumer Price Index Stockwell Canada Cross-sectional Populations of 89 Actual minimum price Alcohol-attributable Yes Not Strong SA, CON, TE, 2013 versus time geographic areas in increased over a 20-year hospital admissions stated DR, PL, CO, series analysis British Columbia period. Study modelled 10% EX of ecological increase in the average data minimum price of all alcoholic beverages adjusted by monthly Consumer Price Index Treisman Russia Secondary Population of Russia Price liberalisation of vodka Mortality Yes Not Strong SA, CON, TE, 2010 analysis of in early 1990s—in 1993, real stated PL, CO, EX historical data price of vodka was around with focus on 25% of that in 1990 price changes 1990–1994 Continued Open Access 6 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Table 2 Continued Study characteristics Study assessment First author Conflict and year Population or Pricing intervention Peer of Quality Bradford Hill Study type published Country Study design participants studied Outcomes studied reviewed interest rating criteria met Wald 1984 Poland Analysis of Population of Poland Poor harvest led to high Alcohol Yes Not Weak CON, TE, PL, routine data prices, rationing and illegal consumption and stated CO, EX 1970–1981 sales alcohol-related hospital admissions Zhao 2013 Canada Cross-sectional Populations of 16 Actual minimum price Acute, chronic and Yes None Strong SA, CON, TE, versus time health service increased over a 20-year wholly DR, PL, CO, series analysis delivery areas in period. Study modelled 10% alcohol-attributable EX of ecological British Columbia, increase in the average mortality data Canada minimum price of all alcoholic beverages adjusted by monthly Consumer Price Index. Also looked at outlet density Modelling Brennan England Modelling study The UK national MUP of £0.40, £0.45 and Alcohol Yes None Strong CON, SP, DR, studies 2014 using SAPM surveys of general £0.50. Ban on below cost consumption, PL, CO population selling consumer spending, (subgroups of 47 health harms, moderate, harmful, QALYs hazardous) Holmes England Modelling study The UK national MUP of 45p Alcohol Yes None Strong CON, SP, PL, 2014 using SAPM surveys of general consumption, CO population consumer spending, (subgroups of 47 health harms, moderate, harmful, QALYs hazardous) Meier 2009 The UK Modelling study The UK national Ten pricing policy options, Alcohol Yes None Strong CON, SP, DR, using SAPM surveys of general including different levels of consumption, PL, CO population MUP (of 33 analysed) consumer spending, (subgroups of 47 health harms, moderate, harmful, crime, employment hazardous) Meier 2016 England Modelling study The UK national MUP of £0.50 compared with Alcohol Yes None Strong CON, SP, PL, using SAPM surveys of general three alcohol taxation consumption in CO population interventions different income and (subgroups of socioeconomic moderate, increasing groups risk, heavy) Purshouse England Modelling study The UK national 18 different pricing policies Alcohol Yes None Strong CON, SP, DR, 2010 using SAPM surveys of general (including MUP) consumption, PL, CO population consumer spending, (subgroups of 47 health harms, moderate, harmful, QALYs hazardous) Continued Open Access Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 7 Table 2 Continued Study characteristics Study assessment First author Conflict and year Population or Pricing intervention Peer of Quality Bradford Hill Study type published Country Study design participants studied Outcomes studied reviewed interest rating criteria met Sharma Australia Counterfactual Representative MUP of A$2 Alcohol purchasing Yes None Strong CON, SP, PL, 2016 analysis sample of and consumption CO households (n=884) completing 12-month Homescan shopping survey Vandenberg Australia Modelling study Representative MUP of A$1 compared with Alcohol purchasing Yes None Strong CON, SP, PL, 2016 sample of a specific alcohol tax and consumption CO households (n=885) completing Homescan shopping survey Cross- Black 2011 Scotland Cross-sectional 377 hospital patients The UK alcohol units Alcohol Yes None Moderate SA CON, DR, sectional survey with serious alcohol purchased below proposed consumption PL, CO studies problems MUP of £0.40p/£0.50p Callinan Australia Cross-sectional Drinkers 18+ Australian standard drinks Alcohol Yes Not Moderate SA, CON, DR, 2015 survey participating in purchased below proposed consumption stated PL, CO Australian minimum prices of A$0.80/A International Alcohol $1.00/A$1.25 Control study (n=1681) Cousins Ireland Cross-sectional 3187 adults in 2013 Alcohol units purchased AUDIT-C score Yes None Strong SA, CON, PL, 2016 survey National Alcohol below proposed minimum CO Diary Survey price of €1.00 Crawford England Cross-sectional 515 members of the The UK alcohol units AUDIT score Yes None Moderate SA, CON, PL, 2012 survey public purchased below proposed CO MUP of £0.50 Falkner New Cross-sectional 115 adults New Zealand standard Alcohol Yes No Moderate SA, CON, PL, 2015 Zealand survey undergoing alcohol drinks purchased below consumption CO detoxification proposed minimum prices of NZ$1.00/NZ$1.10/NZ$1.20 Forsyth Scotland Cross-sectional Shopkeepers of 144 MUP of £0.50 Products affected Yes None Weak CON, PL 2014 survey off licences in and hospital (weakly), CO Glasgow admissions Ludbrook The UK Cross-sectional Expenditure and Purchasers of alcohol < Income of Yes Not Moderate SA, CON, PL, 2012 survey Food Survey data £0.45 per unit purchasers of cheap stated CO from 20062008 alcohol (n=18 624) Sharma Australia Cross-sectional Representative MUP of A$1 and taxation Alcohol Yes None Moderate SA, CON, DR, 2014 survey sample of consumption PL, CO households (n=885) (measured by completing shopping projected sales) survey Continued Open Access 8 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Table 2 Continued Study characteristics Study assessment First author Conflict and year Population or Pricing intervention Peer of Quality Bradford Hill Study type published Country Study design participants studied Outcomes studied reviewed interest rating criteria met Sheron The UK Cross-sectional Adult patients in a The UK alcohol units Alcohol Yes Not Moderate SA, CON, DR, 2014 survey liver unit of a hospital purchased below £0.50 consumption stated PL, CO (n=204) Intervention Babor 1978 The USA Trial (not 34 male volunteers in ‘Happy hour’ with a reduction Alcohol Yes Not Weak SA, CON, SP, studies randomised) live-in research in set price of alcohol for one consumption stated TE, CO, EX facility group of participants Qualitative Seaman Scotland Qualitative 130 participants aged Hypothetical minimum price Alcohol Yes None Moderate CON, CO studies 2013 study 16–30 increases consumption and substitution with other substances Systematic Wagenaar Worldwide Systematic Studies tended to Alcohol price and taxation Alcohol Yes None Strong AN reviews 2009 review and cover general interventions studied consumption meta-analysis population together (measured by alcohol sales or self-reported consumption) Wagenaar Worldwide Systematic Studies tended to Alcohol price and taxation Alcohol-related Yes Not Strong AN 2010 review and cover general interventions studied morbidity (disease, stated meta-analysis population together injury, suicide, traffic crashes, sexually transmitted diseases, other drug use, crime and misbehaviour) and mortality Abbreviations for the Bradford Hill criteria: AN, analogy; CO, coherence; AUDIT, Alcohol Use Disorders Identification Test; CON, consistency; DR, dose–response; EX, experiment; MUP, Minimum Unit Pricing; PL, plausibility; SA, strength of the association; SAPM, Sheffield Alcohol Policy Model; SP, specificity; TE, temporality; QALYs, Quality Adjusted Life Years. Open Access Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 9 Table 3 Studies published in the grey literature included in Bradford Hill criteria assessment Study characteristics Study assessment Author and Bradford year Population or Pricing intervention Peer Conflict Quality Hill criteria published Country Study design participants studied Outcomes studied reviewed of interest rating met Angus 2016 Scotland Modelling Scottish general MUP of 30p, 40p, 50p, Alcohol consumption, Not stated None Strong CON, SP, study using population survey 60p and 70p, compared consumer spending, DR, PL, SAPM (subgroups of with taxation interventions exchequer and retail CO moderate, harmful, revenue, 47 health hazardous) harms Booth 2008 Worldwide Review of Studies tended to Various minimum unit Alcohol consumption Yes None Strong AN reviews and cover general prices and taxation and various measures systematic population interventions of alcohol harm review Brennan England Modelling Adults in England General price increases. Alcohol consumption, Not stated None Strong CON, SP, 2008 study using MUP of £0.20, £0.25, consumer spending, DR, PL, SAPM £0.30, £0.35, £0.40, sales duty and VAT, CO £0.45, £0.50, £0.60 and 47 health harms, £0.70. Restrictions on crime and employment off-trade price promotions. Hill McManus Canada Modelling Adults in two MUP of C$1.50 Alcohol consumption, No None Strong CON, SP, 2012 study using Canadian consumer spending, PL, CO SAPM provinces (Ontario hospital admissions, and British mortality, crime Columbia) Institute for Great Economic Shopping data MUP of £0.45 Alcohol consumption Not stated Not stated Not CON, SP, Fiscal Studies Britain modelling from 25 248 British possible CO 2010 study using households to rate market research data Institute for Great Economic Population of Great MUP of £0.45 and Alcohol consumption Not stated Not stated Not CON, SP, Fiscal Studies Britain analysis Britain increased alcohol taxation possible CO 2013 to rate Meng 2010 Scotland Modelling Adults in Scotland MUP of £0.20, £0.25, Alcohol consumption, Not stated None Strong CON, SP, study using £0.30, £0.35, £0.40, consumer spending, DR, PL, SAPM £0.45, £0.50, £0.60 and 47 health harms, CO £0.70. Restrictions on crime, employment off-trade price promotions. AN, analogy; CO, coherence; CON, consistency; DR, dose–response; EX, experiment; PL, plausibility; SA, strength of the association; SAPM, Sheffield Alcohol Policy Model; SP, specificity; TE, temporality. Open Access 32 40 comes from research following the introduction of MUP drinking the cheapest alcohol. The modelling in Canada, where minimum price increases preceded studies which use survey data in turn suggest heavier 23 26 reductions in alcohol consumption, alcohol-attribut- drinkers will be most affected by MUP. Overall, the evi- able hospital admissions and alcohol-related mortal- dence base provides strong support for this criterion. ity. Studies where price changes preceded the expected changes in alcohol consumption or harm have Experiment 19 22 23 also been conducted in Russia, Poland and We have not identiﬁed any RCTs of minimum pricing or Finland. Overall, there is very strong support for the price-based interventions to reduce alcohol consump- temporality criterion. tion. There is a small (and not randomised) trial from the 1970s which found participants living in controlled Dose–response/biological gradient conditions and offered a daily ‘happy hour’ discount This criterion is supported if different price levels have drank signiﬁcantly more alcohol than those who were been found to have differing effects on consumption or not offered the discount. There is, however, substantial harm. Many of the studies using the Shefﬁeld Alcohol evidence in support of the experiment criterion from Policy Model explore the impact of a range of potential time series analyses or natural experiments, for example, 25 27 29 45 51 MUP options, and these consistently suggest where minimum pricing was introduced in Canada that the higher the MUP the greater the reductions in and where prices ﬂuctuated in the late 1980s and early 19 22 alcohol consumption or alcohol-related harms. The 1990s in Russia, and to a lesser extent in Finland, Canadian studies of minimum pricing lend further where minimum price increases were associated with support for this criterion because the analysis presents reduced mortality only among men with a basic educa- the effect on consumption or harm of a modelled 1% tion. These studies provide tentative support for the increase in price, meaning dose response can be experiment criterion. inferred. Dose response is supported to a lesser extent by evidence from cross-sectional studies that Analogy heavier drinkers are more likely to pay less than a pro- To address the analogy criterion, areas related to 32–34 39 40 posed MUP. Overall, there is strong support for minimum alcohol pricing must be considered. There is the dose–response criterion, although the relationship is 52 evidence from literature on the affordability of alcohol difﬁcult to quantify. that consumption and harm are very responsive to the affordability of alcohol. Large systematic reviews have Plausibility 53 investigated the price elasticity of demand for alcohol, This criterion refers to whether there is evidence that and have found that higher alcohol pricing and taxation alcohol price can be used as an economic mechanism to (considered together) are associated with reductions in inﬂuence consumption at a population level, and alcohol consumption, alcohol-related morbidity and whether heavy drinkers tend to purchase cheaper 43 44 46 mortality. Overall, the support for the analogy cri- alcohol. There is evidence from 21/26 research studies terion is very strong, although Bradford Hill describes and 4/7 studies in the grey literature that the price of this as the weakest evidence for causality. alcohol is inversely related to alcohol-related morbidity, hospital admissions or mortality. Moreover, there is also evidence from numerous cross-sectional studies in the DISCUSSION 32–36 38–40 UK, Ireland and Australia and one trial from We assessed 26 research studies and two systematic the USA that heavier drinking was signiﬁcantly asso- reviews, plus a further seven studies from the grey ciated with purchasing alcohol below speciﬁed prices, literature in this review of the evidence for priced-based further suggesting that economic mechanisms such as interventions—such as MUP—to reduce alcohol con- minimum pricing would particularly affect the heaviest sumption and alcohol-related harm. All nine of the drinkers. This provides strong support for the plausibility Bradford Hill criteria for causality were met and the vast criterion. majority of studies offered support for price-based alcohol policy interventions. However, the evidence for Coherence two of the criteria, although present, was not as strong This criterion refers to whether studies from different as it was for the other criteria. These criteria were disciplines have had complementary ﬁndings and strength of the association (criterion 1) and experiment whether these ﬁtor ‘hang’ well together. It is different (criterion 8), and according to Bradford Hill, these are to consistency, which is more concerned with reproduci- the two criteria that can provide the strongest evidence bility of ﬁndings. The ﬁndings of the majority of studies for causality. Therefore, although all of the criteria were supported the coherence criterion in that they suggest supported, we conclude that it is highly probable, but 2 3 21 24 that real-world MUP or minimum price not deﬁnite, that introducing MUP for alcohol would 19 20 23 increases led to reductions in alcohol consump- reduce alcohol consumption and alcohol-related harms. tion and alcohol-related harm and cross-sectional It is also of note that different types of study tended to surveys ﬁnd that it is the heavier drinkers that are satisfy different Bradford Hill criteria, and that different 10 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Open Access study designs also produced evidence of the effective- rated as strong or moderate. However, it was not possible ness of minimum pricing in relation to different out- to appraise two of the studies from the grey literature comes. This is summarised in ﬁgure 2. This underlines using this tool, and there were some challenges assessing the importance of including a variety of study designs in the econometric modelling studies against this frame- this review. work. However, overall we think that our quality Strengths of this study are that this is the ﬁrst to have appraisal across the different studies is broadly compar- systematically reviewed the literature relevant speciﬁcally able. It should also be noted that although a number of to alcohol minimum pricing policies. We had broad studies were rated as ‘strong’, this is in relation to their inclusion criteria with regards to study design, price respective study designs and does not reﬂect the pos- intervention and outcome measure, allowing for a com- ition of the study type in the hierarchy of evidence prehensive review of the evidence base. Application of framework. the Bradford Hill criteria as part of a narrative systematic This is the ﬁrst systematic review that has addressed literature review is a useful and emergent technique for the effectiveness of minimum alcohol price interven- identifying causality: a PubMed search for systematic tions such as MUP using the Bradford Hill criteria. It reviews with ‘Bradford Hill’ mentioned in the title or was beyond the scope of this review to study the impact abstract yielded 28 results, 90% of which were published of generalised increases in alcohol prices (as opposed to in the last 5 years. The limitations of this systematic minimum prices). However, where such studies have review relate mainly to the broad range of studies been carried out, a minimum price or ﬂoor price has included. It was not possible to conduct any kind of been recommended, for example, in Gruenewald’s 2006 meta-analysis and therefore we do not present a pooled study in Sweden which found that the lowest quality (the estimate for the likely effect of MUP on certain out- cheapest) alcohol has the highest price elasticity. comes. The exact effect of any MUP would be inﬂu- Previous systematic reviews of alcohol price and con- 43 44 enced by a range of factors, including: the minimum sumption and alcohol-related harm have tended to price level chosen, how broadly it is applied, how consider the effect of price increases and increased tax- strongly it is enforced and contextual factors such as ation together. These reviews found signiﬁcant effects affordability (in the UK, alcohol was 54% more afford- on consumption and morbidity and mortality. Although able in 2014 than it was in 1980 ), other governmental price regulation and taxation are closely related policy 55 45 regulations and the price-level pre-MUP. Occasionally, options, evidence from surveys and modelling studies minimum pricing has been implemented as part of a suggests that the effects of each are different, although range of measures, and these studies were considered it is known that the majority of tax increases are passed 43 56 alongside studies where MUP was implemented in isola- on as increased prices for consumers. It was beyond tion. This emphasises the importance of the speciﬁcity the scope of this review to discuss whether MUP is criterion. regressive in detail, but as it only affects the prices of the There were also challenges with the quality appraisal. cheapest drinks, which are usually consumed by the The EPHPP quality assessment tool was used to assess heaviest drinkers, MUP is likely to narrow health 28 31 quantitative studies and the majority of studies were inequalities. A recent rapid evidence review Figure 2 This model shows that different study types tended to produce evidence of effectiveness of minimum pricing in relation to different outcomes. Studies cited in the figure are key examples of the literature in that area and do not represent an exhaustive list. Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 11 Open Access Provenance and peer review Not commissioned; externally peer reviewed. published in The Lancet examined alcohol control pol- icies in England and recommended a combination of Data sharing statement No additional data are available. MUP and tax increases to reduce alcohol harm and Open Access This is an Open Access article distributed in accordance with increase government revenue, rather than either in iso- the terms of the Creative Commons Attribution (CC BY 4.0) license, which lation. It is also important to highlight that a consider- permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http:// able proportion of included studies were produced by a creativecommons.org/licenses/by/4.0/ small number of research teams. Also, with regards to the econometric modelling studies, uncertainty in esti- mates or forecasts is often poorly communicated outside REFERENCES of the academic literature. The overall risk of bias in the 1. Anderson P, Chisholm D, Fuhr DC. Effectiveness and included studies was minimised by excluding studies cost-effectiveness of policies and programmes to reduce the harm caused by alcohol. Lancet 2009;373:2234–46. with a conﬂict of interest (either for or against MUP). It 2. Stockwell T, Auld MC, Zhao J, et al. Does minimum pricing reduce was not possible to assess publication bias using an ana- alcohol consumption? The experience of a Canadian province. Addiction 2012;107:912–20. lytical technique such as a funnel plot due to the narra- 3. Stockwell T, Zhao J, Giesbrecht N, et al. The raising of minimum tive nature of the review; however, we anticipate that by alcohol prices in Saskatchewan, Canada: impacts on consumption including grey literature in this review, we have mitigated and implications for public health. Am J Public Health 2012;102: e103–10. publication bias as far as reasonably possible. 4. World Health Organization. European status report on alcohol and Overall, the ﬁndings of this review lend strong support health 2014. Pricing policies. World Health Organization, 2014. http:// www.euro.who.int/__data/assets/pdf_file/0003/244902/ for policies such as MUP in reducing alcohol consump- Pricing-policies.pdf tion and alcohol-related harm, with all nine of the 5. Department of Health (Ireland). The Public Health (Alcohol) Bill Bradford Hill criteria met, and little by way of counter 2015. 2015 (cited 17 February 2017). http://health.gov.ie/blog/ publications/public-health-alcohol-bill-2015/ ﬁndings. As it is unlikely to be feasible to conduct RCTs 6. The Guardian. Minimum alcohol price not in the public interest, says of MUP, the decision whether or not to introduce MUP health agency. The Guardian. 1 May 2014 (cited 17 February 2017). https://www.theguardian.com/world/2014/may/01/minimum-alcohol- will not be informed by a systematic review and price-not-in-the-public-interest-says-health-agency meta-analysis of RCTs, and therefore, this synthesis of evi- 7. New Zealand Ministry of Justice. Alcohol minimum pricing policies. dence according to the Bradford Hill criteria is of value. (cited 17 February 2017). https://justice.govt.nz/justice-sector-policy/ key-initiatives/sale-and-supply-of-alcohol/alcohol-minimum-pricing- Unanswered questions about the effectiveness of MUP report/ remain; for example, this review has highlighted that 8. Home Office. Guidance on banning the sale of alcohol below the cost of duty plus VAT. For suppliers of alcohol and enforcement support was moderate or tentative for two of the Bradford authorities in England and Wales. 2015 (cited 3 June 2016). Hill criteria (‘strength of the association’ and ‘experi- Available from: https://www.gov.uk/government/uploads/system/ uploads/attachment_data/file/415522/HO_Guidance_on_BBCS.pdf ment’, respectively). There may be opportunities to 9. Off Licence News. First conviction for selling alcohol “below cost” explore this in countries such as Scotland if MUP is imple- sees retailer fined. 2016 (cited 3 June 2016). http://www. mented. If Scotland were to implement MUP, then it offlicencenews.co.uk/news/fullstory.php/aid/15750/First_conviction_ for_selling_alcohol__below_cost__sees_retailer_fined.html would be possible to evaluate the validity of the Shefﬁeld 10. Home Office. Duty plus VAT permitted price calculator (2015). 2016. Alcohol Policy Model studies conducted using Scottish https://www.gov.uk/government/publications/banning-the-sale-of- alcohol-below-the-cost-of-duty-plus-vat-march-2015 data. It would also be possible to conduct a longitudinal 11. HM Government. The Government’s Alcohol Strategy. 2012 (cited 3 study to evaluate the effectiveness of MUP in reducing June 2016). https://www.gov.uk/government/uploads/system/ alcohol consumption and alcohol-related morbidity and uploads/attachment_data/file/224075/alcohol-strategy.pdf 12. Scottish Government. Minimum unit pricing. (cited 17 February mortality. The ﬁndings of this natural experiment would 2017). http://www.gov.scot/Topics/Health/Services/Alcohol/ have relevance elsewhere within and outside the UK. minimum-pricing 13. Court of Justice of the European Union. Press Release 155/15: the Scottish legislation introducing a minimum price per unit of alcohol is Twitter Follow Sadie Boniface @sadieboniface contrary to EU law if less restrictive tax measures can be introduced. 2015 (cited 17 February 2017). http://curia.europa.eu/jcms/upload/ docs/application/pdf/2015-12/cp150155en.pdf Acknowledgements We would like to thank Rebecca McDonald for advice on 14. Judiciary of Scotland. Scotch Whisky Association and others v Lord using the Bradford Hill criteria in a systematic review, Dr James Nicholls for Advocate and Advocate General for Scotland. (cited 11 July 2016). advice on study interpretation and Dr Daniel Stahl for statistical advice on http://www.scotland-judiciary.org.uk/9/1672/Scotch-Whisky- Association-and-others-v-Lord-Advocate-and-Advocate-General- some of the included studies. for-Scotland Contributors SM conceived the idea. SB conducted the initial search. SM and 15. BBC News. Bid to appeal against minimum alcohol pricing to SM contributed to independently reviewing abstracts, hand-searching Supreme Court. 2016 Nov 18 (cited 17 February 2017) http://www. bbc.co.uk/news/uk-scotland-scotland-politics-38026073 reference lists, completing data extraction and conducting quality appraisal. 16. Lifestyles Statistics Team, Health and Social Care Information All authors contributed to the analysis and interpretation of the results and Centre. Statistics on alcohol. England, 2015. 2015 (cited 3 June contributed to writing the manuscript. SB is guarantor. 2016). http://www.hscic.gov.uk/catalogue/PUB17712/ alc-eng-2015-rep.pdf Funding Open access for this article was funded by King’s College London 17. Cochrane Handbook. 21.4 Assessment of study quality and risk of Open Scholarship Fund. bias. (cited 2 June 2016). http://handbook.cochrane.org/chapter_21/ 21_4_assessment_of_study_quality_and_risk_of_bias.htm Competing interests SB and SM work at King’s College London, which as an 18. Hill AB. The environment and disease: association or causation? institution is listed as a member of the Alcohol Health Alliance. SM has Proc R Soc Med 1965;58:295–300. received funding indirectly from UKCTAS, which as an institution is also listed 19. Bhattacharya J, Gathmann C, Miller G. The Gorbachev Anti-Alcohol as a member of the Alcohol Health Alliance. None of the authors have any Campaign and Russia’s Mortality Crisis. Am Econ J Appl Econ relationship with the Alcohol Health Alliance. 2013;5:232–60. 12 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Open Access 20. Herttua K, Mäkelä P, Martikainen P. Minimum prices for alcohol and 41. Babor TF, Mendelson JH, Greenberg I, et al. Experimental analysis educational disparities in alcohol-related mortality. Epidemiology of the ‘happy hour’: effects of purchase price on alcohol 2015;26:337–43. consumption. Psychopharmacology (Berl) 1978;58:35–41. 21. Stockwell T, Zhao J, Martin G, et al. Minimum alcohol prices and 42. Seaman P, Edgar F, Ikegwuonu T. The role of alcohol price in young outlet densities in British Columbia, Canada: estimated impacts on adult drinking cultures in Scotland. Drugs Educ Prev Policy alcohol-attributable hospital admissions. Am J Public Health 2013;20:278–85. 2013;103:2014–20. 43. Wagenaar AC, Salois MJ, Komro KA. Effects of beverage alcohol 22. Treisman D. Death and prices. Econ Transit 2010;18:281–331. price and tax levels on drinking: a meta-analysis of 1003 estimates 23. Wald I, Moskalewicz J. Alcohol policy in a crisis situation. from 112 studies. Addiction 2009;104:179–90. Br J Addict 1984;79:331–5. 44. Wagenaar AC, Tobler AL, Komro KA. Effects of alcohol tax and 24. Zhao J, Stockwell T, Martin G, et al. The relationship between price policies on morbidity and mortality: a systematic review. minimum alcohol prices, outlet densities and alcohol-attributable Am J Public Health 2010;100:2270–8. deaths in British Columbia, 2002–09. Addiction 2013;108: 45. Angus C, Holmes J, Pryce R, et al. Model-based appraisal of the 1059–69. comparative impact of minimum unit pricing and taxation policies in 25. Brennan A, Meng Y, Holmes J, et al. Potential benefits of minimum Scotland an adaptation of the Sheffield Alcohol Policy Model version unit pricing for alcohol versus a ban on below cost selling in England 3. ScHARR, University of Sheffield, 2016 April (cited 7 April 2016). 2014: modelling study. BMJ 2014;349:g5452. https://www.shef.ac.uk/polopoly_fs/1.565373!/file/Scotland_report_ 26. Holmes J, Meng Y, Meier PS, et al. Effects of minimum unit pricing 2016.pdf for alcohol on different income and socioeconomic groups: 46. Booth A, Brennan A, Meier PS, et al. Independent review of the a modelling study. Lancet 2014;383:1655–64. effects of alcohol pricing and promotion: part A—systematic reviews. 27. Meier PS, Purshouse R, Brennan A. Policy options for alcohol price Project report for the Department of Health September 2008. regulation: the importance of modelling population heterogeneity. ScHARR University of Sheffield, 2008. Addiction 2010;105:383–93. 47. Brennan A, Purshouse R, Taylor K, et al. Independent review of the 28. Meier PS, Holmes J, Angus C, et al. Estimated effects of different effects of alcohol pricing and promotion: part B. Modelling the alcohol taxation and price policies on health inequalities: potential impact of pricing and promotion policies for alcohol in a mathematical modelling study. PLOS Med 2016;13:e1001963. England: results from the Sheffield Alcohol Policy Model version 29. Purshouse RC, Meier PS, Brennan A, et al. Estimated effect of 2008(1-1). ScHARR University of Sheffield, 2008. https://www.shef. alcohol pricing policies on health and health economic outcomes in ac.uk/polopoly_fs/1.95621!/file/PartB.pdf England: an epidemiological model. Lancet 2010;375:1355–64. 48. Hill McManus D, Brennan A, Stockwell T, et al. Model-based 30. Sharma A, Etilé F, Sinha K. The effect of introducing a minimum appraisal of alcohol minimum pricing in Ontario and British price on the distribution of alcohol purchase: a counterfactual Columbia: a Canadian adaptation of the Sheffield Alcohol Policy analysis. Health Econ 2016;25:1182–200. Model version 2. 2012. http://www.uvic.ca/research/centres/carbc/ 31. Vandenberg B, Sharma A. Are alcohol taxation and pricing policies assets/docs/report-model-based-appraisal.pdf regressive? Product-level effects of a specific tax and a minimum 49. Institute for Fiscal Studies. The impact of Introducing a minimum unit price for alcohol. Alcohol Alcohol 2016;51:493–502. price on alcohol in Britain. IFS Briefing Note BN109. 2010. https:// 32. Black H, Gill J, Chick J. The price of a drink: levels of consumption www.ifs.org.uk/publications/5287 (accessed 27 March 2017) and price paid per unit of alcohol by Edinburgh’s ill drinkers with a 50. Institute for Fiscal Studies. Price-based measures to reduce alcohol comparison to wider alcohol sales in Scotland. Addiction consumption. IFS Briefing Note BN138. 2013. https://www.ifs.org.uk/ 2011;106:729–36. publications/6644 (accessed 27 March 2017) 33. Callinan S, Room R, Livingston M, et al. Who purchases low-cost 51. Meng Y, Purshouse R, Brennan A, et al. Model-based appraisal of alcohol in Australia? Alcohol Alcohol 2015;50:647–53. alcohol minimum pricing and off-licensed trade discount bans in 34. Cousins G, Mongan D, Barry J, et al. Potential impact of minimum Scotland using the Sheffield Alcohol Policy Model (v.2): second unit pricing for alcohol in Ireland: evidence from the national alcohol update based on newly available data. 2010. http://www.shef.ac.uk/ diary survey. Alcohol Alcohol 2016;51:734–40. polopoly_fs/1.96510!/file/scotlandupdate.pdf 35. Crawford MJ, Parry AMH, Weston ARW, et al. Relationship 52. Rabinovich L, Brutscher PB, de Vries H, et al. The affordability of between price paid for off-trade alcohol, alcohol consumption and alcoholic beverages in the European Union. 2012. http://ec.europa. income in England: a cross-sectional survey. Alcohol Alcohol eu/health/ph_determinants/life_style/alcohol/documents/alcohol_ 2012;47:738–42. rand_en.pdf (accessed 27 March 2017) 36. Falkner C, Christie G, Zhou L. The effect of alcohol price on 53. Gallet CA. The demand for alcohol: a meta-analysis of elasticities. dependent drinkers’ alcohol consumption. N Z Med J Rochester, NY: Social Science Research Network, 2007 May (cited 2015;128:9–17. 6 January 2016). Report No.: ID 985689. http://papers.ssrn.com/ 37. Forsyth AJM, Ellaway A, Davidson N. How might the alcohol abstract=985689. minimum unit pricing (MUP) impact upon local off-sales shops and 54. Gruenewald PJ, Ponicki WR, Holder HD, et al. Alcohol prices, the communities which they serve? Alcohol Alcohol beverage quality, and the demand for alcohol: quality substitutions 2014;49:96–102. and price elasticities. Alcohol Clin Exp Res 2006;30:96–105. 38. Ludbrook A, Petrie D, McKenzie L, et al. Tackling alcohol misuse: 55. Ally AK, Meng Y, Chakraborty R, et al. Alcohol tax pass-through purchasing patterns affected by minimum pricing for alcohol. Appl across the product and price range: do retailers treat cheap alcohol Health Econ Health Policy 2012;10:51–63. differently? Addiction 2014;109:1994–2002. 39. Sharma A, Vandenberg B, Hollingsworth B. Minimum pricing of 56. Rabinovich L, Hunt P, Staetsky L, et al. Further study on the alcohol versus volumetric taxation: which policy will reduce heavy affordability of alcoholic beverages in the EU [Internet]. 2012 (cited consumption without adversely affecting light and moderate 17 February 2017). http://www.rand.org/pubs/technical_reports/ consumers? PLoS ONE 2014;9:e80936. TR1203.html 40. Sheron N, Chilcott F, Matthews L, et al. Impact of minimum price per 57. Burton R, Henn C, Lavoie D, et al. A rapid evidence review of the unit of alcohol on patients with liver disease in the UK. Clin Med effectiveness and cost-effectiveness of alcohol control policies: an 2014;14:396–403. English perspective. Lancet 2016. Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 13 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMJ Open British Medical Journal http://www.deepdyve.com/lp/british-medical-journal/evidence-for-the-effectiveness-of-minimum-pricing-of-alcohol-a-FEy5QgWp0R

Loading next page...

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher: British Medical Journal
Copyright: Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/
eISSN: 2044-6055
DOI: 10.1136/bmjopen-2016-013497
Publisher site: See Article on Publisher Site

Abstract

Open Access Research Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality 1 2 1 Sadie Boniface, Jack W Scannell, Sally Marlow To cite: Boniface S, ABSTRACT Strengths and limitations of this study Scannell JW, Marlow S. Objectives: To assess the evidence for price-based Evidence for the effectiveness alcohol policy interventions to determine whether ▪ This review adds to an emerging literature of sys- of minimum pricing of minimum unit pricing (MUP) is likely to be effective. tematic reviews synthesising findings using the alcohol: a systematic review Design: Systematic review and assessment of studies Bradford Hill criteria for causality in research and assessment using the according to Preferred Reporting Items for Systematic Bradford Hill criteria areas where traditional meta-analyses of rando- Reviews and Meta-Analyses (PRISMA) guidelines, for causality. BMJ Open mised controlled trials are not possible or 2017;7:e013497. against the Bradford Hill criteria for causality. Three appropriate. doi:10.1136/bmjopen-2016- electronic databases were searched from inception to ▪ A range of study designs were included, allowing February 2017. Additional articles were found through for a comprehensive review of a disparate evi- hand searching and grey literature searches. dence base to investigate whether minimum unit Prepublication history and Criteria for selecting studies: We included any pricing of alcohol is likely to reduce alcohol con- additional material is study design that reported on the effect of price- sumption and alcohol-related harm. available. To view please visit based interventions on alcohol consumption or ▪ Studies examining the effects of alcohol taxation the journal (http://dx.doi.org/ alcohol-related morbidity, mortality and wider harms. or changes in alcohol affordability, or studies 10.1136/bmjopen-2016- Studies reporting on the effects of taxation or solely reporting on price elasticity of demand, 013497). affordability and studies that only investigated price were not included. elasticity of demand were beyond the scope of this ▪ Methodological quality of studies was variable. review. Studies with any conflict of interest were Received 15 July 2016 excluded. All studies were appraised for Revised 28 February 2017 methodological quality. INTRODUCTION Accepted 6 March 2017 Results: Of 517 studies assessed, 33 studies were There are many policies and programmes included: 26 peer-reviewed research studies and seven that aim to reduce harms from alcohol. from the grey literature. All nine of the Bradford Hill One of these is minimum alcohol pricing, criteria were met, although different types of study which exists in a number of countries around satisfied different criteria. For example, modelling the world. The most notable example of studies complied with the consistency and specificity this is Canada, where there are government criteria, time series analyses demonstrated the monopolies on alcohol sales and a variety of temporality and experiment criteria, and the analogy types of minimum pricing in operation. For criterion was fulfilled by comparing the findings with example, there is a minimum price per litre the wider literature on taxation and affordability. of a particular drink in British Columbia and Conclusions: Overall, the Bradford Hill criteria for a (higher) minimum price linked to drink causality were satisfied. There was very little evidence 1 3 Department of Psychology that minimum alcohol prices are not associated with type and strength in Saskatchewan. Other and Neuroscience, National consumption or subsequent harms. However the countries with minimum alcohol pricing Addiction Centre, Institute of overall quality of the evidence was variable, a large include Belarus, Kyrgyzstan, the Republic of Psychiatry, King’s College proportion of the evidence base has been produced by Moldova, the Russian Federation and London, London, UK a small number of research teams, and the quantitative School of Social and Ukraine. Minimum alcohol pricing is being uncertainty in many estimates or forecasts is often 5 Political Science, University considered by governments in Ireland and poorly communicated outside the academic literature. of Edinburgh, Edinburgh, UK has also been reviewed in Australia and New Nonetheless, price-based alcohol policy interventions Zealand. Correspondence to such as MUP are likely to reduce alcohol consumption, The situation with regards to minimum Dr Sadie Boniface; alcohol-related morbidity and mortality. [email protected] alcohol pricing in the UK is complex. Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 1 Open Access In England and Wales, there has been a ban on alcohol In light of this ongoing consideration of MUP in the being sold at below cost (the total amount of ‘duty plus UK, in this paper we assess the effectiveness of value added tax (VAT)’) since May 2014; and the ﬁrst minimum alcohol price interventions to reduce alcohol- conviction for selling alcohol below this level took place related harm. Alcohol-related harm costs the National in 2016. Duty plus VAT is equivalent to a 70 cl bottle of Health Service in England £3.5 billion each year and the vodka (37.5% alcohol by volume (ABV)) costing a estimated cost to society is £21 billion per year. The minimum of £8.72, whereas under a minimum price latest annual ﬁgures for England (population of 54 of 50 pence per unit (one UK unit=10 mL or 8 g million) show over 1 million alcohol-related hospital ethanol), this would cost £13.13. In 2012, the UK coali- admissions (2013/2014) and 6500 alcohol-related tion government cited support for minimum unit deaths (2013); and these ﬁgures represent increases pricing (MUP) in its alcohol strategy, and legislation to compared with a decade previously of 115% and 10%, have a minimum price of £0.50 per unit was passed in respectively. Scotland the same year. Following the change to a We systematically review the literature on the effect of Conservative majority government in 2015, it is unclear price interventions or policies such as MUP on alcohol whether there is still central government support for consumption, alcohol-related morbidity and mortality, MUP. In Scotland, the Scotch Whisky Association chal- and wider harms. We use the nine Bradford Hill criteria lenged the 2012 legislation in the Scottish Court of for causality as a framework with the aim of assessing the Session, which referred the case to the Court of Justice likely effectiveness of MUP as a policy to reduce alcohol of the European Union (CJEU) in 2014. In late 2015, consumption and alcohol-related harm. the CJEU referred the case back to the Scottish courts to investigate proportionality (that the same objective cannot be met through increased taxation), which METHODS could have implications for other EU countries consider- A systematic literature search was performed according ing MUP. In late 2016, the Scottish Court of Session to Preferred Reporting Items for Systematic Reviews and ruled that MUP does not contravene EU law; however, Meta-Analyses (PRISMA) guidance (see ﬁgure 1 for the Scotch Whisky Association then appealed to the UK PRISMA ﬂow diagram and online supplementary ﬁle for Supreme Court. excluded studies). Figure 1 PRISMA 2009 flow diagram of studies in this systematic review. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses. 2 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Open Access Identification of studies criteria and agreed which studies provided relevant evi- Three electronic databases were searched for titles or dence for or against each criterion. abstracts containing ‘minimum unit pric$’ OR ‘minimum pric$’ OR ‘ﬂoor pric$’ OR ‘pric$ AND RESULTS policy’ AND alcohol. The databases were PsycINFO The included studies that are published in peer-reviewed (1806 to February Week 2 2017), Embase (1974 to 2017 journals (26 research studies and two systematic reviews) Week 07) and Ovid Medline (1946 to February Week 2 are listed by study type in table 2 with information on 2017). We also searched the websites of ﬁve alcohol study characteristics and methodological quality. Of the charities for publications or reports related to ‘price’, research studies, the methodological quality was rated as and also searched 20 leading UK think tanks for ‘strong’ in 15 studies, ‘moderate’ in 8 studies and ‘weak’ ‘alcohol’ or ‘addiction’. in 3 studies. Both of the systematic reviews were rated Inclusion criteria were: any study design; population- ‘strong’. The seven reports from the grey literature are level studies exploring at least one aspect of the effect of listed in table 3. Five of the seven were rated as of interventions or policies leading to changes in the ‘strong’ methodological quality, with the remaining two minimum price of alcohol, including but not limited to not appropriate to rate using our critical appraisal tools. changes in alcohol sales, consumption, morbidity and mortality; and individual-level studies exploring Consideration against the Bradford Hill criteria minimum alcohol prices and alcohol purchasing, con- for determining causality sumption, morbidity or mortality; written in English. Strength of the association Exclusion criteria were: studies about taxation, afford- In 16/26 studies published in peer-reviewed journals, ability, price elasticity of demand for alcohol and strength of the association between pricing and alcohol general changes in alcohol price not the result of an consumption or alcohol-related harm was evidenced by intervention or policy (there is a large literature on each a summary statistic such as an OR, and by a test of the of these already and reviewing all of these studies was statistical signiﬁcance of the association. As well as the beyond the scope of this review); studies about public statistical signiﬁcance of the summary statistic, the mag- perceptions of MUP; and studies where a conﬂict of nitude of the effect was also considered, as a larger interest was reported in the paper, whether this was in effect size corresponds to a greater population health favour of or against MUP. impact. Studies in Canada found that 10% increases All 33 studies that met the inclusion criteria were in minimum prices were associated with reductions in assessed against the Bradford Hill criteria for causality alcohol consumption of 3.4–8.4%, reductions in and the methodological quality appraised. These alcohol-attributable hospital admissions of 9% and included 26 original research studies and seven studies reductions in alcohol-related mortality of 32%, each of from the grey literature, and in addition two systematic which was statistically signiﬁcant. Cross-sectional reviews pertinent to the analogy criterion were included. 32–36 38–40 studies in the UK, Ireland, Australia and one Of the 26 research studies, there were 9 cross-sectional trial from the USA found statistically signiﬁcant asso- surveys, 8 time series analyses or similar, 7 modelling ciations between cheaper alcohol and heavier drinking. studies, 1 qualitative study and 1 trial. The magnitude of the association varied across these studies, but due to different study measures and out- Analysis of included studies comes, the results are not all comparable. As an indica- Quality of included studies was assessed independently tion, the OR for buying alcohol below a proposed by two reviewers and using validated tools. Owing to the minimum price among heavier drinkers was 1.34 in wide variation in study designs among the included 35 34 Crawford’s study, 1.50 in Cousins’s study and 1.70 in studies, the Effective Public Health Practice Project’s Callinan’s study. There was not any evidence to (EPHPP) tool was used for assessing all quantitative support this criterion from the grey literature. Overall, studies, as recommended by the Cochrane Handbook there is reasonably good support for the strength of the for assessing studies in public health. Qualitative association criterion. studies (n=1) and systematic reviews (n=2) included in this review were not covered by the EPHPP tool and so were assessed using the Critical Appraisal Skills Consistency Programme (CASP) tools speciﬁc to these study designs. This criterion requires looking across all the studies Nine criteria to determine causality were suggested by included in the review to see whether similar conclu- Bradford Hill in an inﬂuential 1965 paper. Increasingly, sions have been drawn. Inverse associations between the Bradford Hill criteria are a standard framework to alcohol pricing and alcohol consumption or harm have assess the impact of interventions where it is not ethical been documented in countries in Europe, North or practical to conduct randomised controlled trials America and Australia, and although most studies are (RCTs). Our interpretation of the Bradford Hill criteria from the last 10 years, there are studies from the 1970s for the purpose of this review is listed in table 1.Two and 1980s as well. There is evidence from different reviewers assessed each study against each of the nine research teams and different types of study including Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 3 Open Access Table 1 Bradford Hill criteria for assessing causation and the definitions used in this review Criterion Bradford Hill criteria (1965) Application in this review 1. Strength of the The strength of a supposed association between an A statistically significant change (p<0.05) in alcohol association intervention and an outcome is determined by the consumption or alcohol-related harms, in the appropriate statistic used to measure the protective expected direction. The exact magnitude of the effect of an intervention (eg, relative risk or OR). association was assessed on a study by study This is the most important factor determining basis causation 2. Consistency Has it been repeatedly observed by different Whether different studies conducted in different persons, in different places, circumstances and locations, in different populations, by different times? investigators and at different times have reported similar findings 3. Specificity Specificity is present when the intervention is If pricing was the only reason that alcohol exclusive to the outcome and when the outcome consumption or alcohol-related harm could have has no other known cause or associated risk fallen, this adds to the argument for causality. factors; cautions that this criterion should not be However, if a price intervention was one of a overemphasised and that if specificity is not number of alcohol policy interventions, then this apparent, this does not preclude causation criterion is not satisfied 4. Temporality Refers to temporal relationship of association The pricing intervention studied must have taken between exposure and disease outcome; to infer place before a change in alcohol consumption or causality, exposure must precede outcome harm was observed 5. Dose– If the association is one in which a dose–response If interventions leading to a larger increase in prices response curve or biological gradient can be observed, this had a greater effect on alcohol consumption and adds to the case for causality alcohol-related harm than interventions where the price change was small, or if studies demonstrate that different minimum prices have differing effects, in the expected direction 6. Plausibility A likely biological mechanism linking the Studies that found an association between price intervention to the observed findings helps to and population-level alcohol consumption and that explain causality; plausibility depends on biological heavier drinkers tend to purchase the cheapest knowledge of the day alcohol could demonstrate plausibility 7. Coherence When the evidence from different disciplines Describes whether studies conducted in different sources ‘hangs well together’ and does not conflict settings or disciplines had complementary findings. with other generally known facts, this criterion is Will not be demonstrated by a single study in met isolation but rather the evidence base as a whole 8. Experiment Experimental evidence from laboratory studies or In addition to laboratory studies and RCTs, natural RCTs could potentially provide strongest support for experiments with before-and-after measures could causation also show the effectiveness of minimum unit pricing This criterion often provides the strongest support in a ‘real-world’ setting for causation and describes whether there is empirical evidence for the association 9. Analogy Causality is supported by analogy if there are Other areas of relevance include whether higher similar associations or causal relationships in other taxation on alcohol is associated with reduced areas of relevance, weakest form of evidence of alcohol consumption and alcohol-related harm, and causality may require drawing on additional literature outside of the main systematic review RCTs, randomised controlled trials. cross-sectional studies, time series analyses and econo- Shefﬁeld Alcohol Policy Model is one such model and 25–29 47 45 51 metric modelling studies. Support for the consistency has been applied in England, Scotland and criterion is very strong. Canada and provides very strong support for the speci- ﬁcity criterion. Further support is provided by other dif- 49 50 30 31 Specificity ferent modelling studies in the UK and Australia The speciﬁcity criterion relates to whether changes in and a (non-randomised) trial in the USA. Thus, alcohol consumption or harm could be attributed to any- support for the speciﬁcity criteria is very strong. thing other than the price intervention. Many studies included have statistically adjusted for confounding Temporality factors; however, the best support for the speciﬁcity cri- It is important that pricing interventions take place terion comes from the econometric modelling studies before changes to alcohol consumption and harm to because there is no risk of residual confounding. The attribute causality. Strong support for this criterion 4 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Open Access Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 5 Table 2 Studies published in peer-reviewed journals included in Bradford Hill criteria assessment Study characteristics Study assessment First author Conflict and year Population or Pricing intervention Peer of Quality Bradford Hill Study type published Country Study design participants studied Outcomes studied reviewed interest rating criteria met Natural Bhattacharya Russia Time series Populations of 77 Substantial increases in Mortality Yes Not Strong SA, CON, TE, experiments 2013 analysis of Russian oblasts administratively set alcohol stated PL, CO, EX, and time panel data set (provinces), 1970– prices 1985–1988, along series 2000 with six other antialcohol analyses measures Herttua Finland Time series General population Modelled 1% increase in the Alcohol-related Yes None Strong SA (not 2015 analysis using population average minimum price of all mortality universal registry alcoholic beverages based findings— on actual price increases subgroup only), adjusted for inflation using CON (counter Consumer Price Index findings) TE, PL, CO, EX Stockwell Canada Cross-sectional Population of British Actual minimum price Alcohol Yes None Strong SA, CON, TE, 2012 versus time Columbia increased over a 20-year consumption DR, CO, EX series analysis period. Study modelled a (measured by sales) of ecological 10% increase in the average data minimum price of all alcoholic beverages adjusted by monthly Consumer Price Index Stockwell Canada Cross-sectional Population of Actual minimum price Alcohol Yes Not Strong SA, CON, TE, 2012 versus time Saskatchewan increased over a 7-year consumption stated DR, CO, EX series analysis period. Study modelled a (measured by sales) of ecological 10% increase in the average data minimum price of all alcoholic beverages adjusted by monthly Consumer Price Index Stockwell Canada Cross-sectional Populations of 89 Actual minimum price Alcohol-attributable Yes Not Strong SA, CON, TE, 2013 versus time geographic areas in increased over a 20-year hospital admissions stated DR, PL, CO, series analysis British Columbia period. Study modelled 10% EX of ecological increase in the average data minimum price of all alcoholic beverages adjusted by monthly Consumer Price Index Treisman Russia Secondary Population of Russia Price liberalisation of vodka Mortality Yes Not Strong SA, CON, TE, 2010 analysis of in early 1990s—in 1993, real stated PL, CO, EX historical data price of vodka was around with focus on 25% of that in 1990 price changes 1990–1994 Continued Open Access 6 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Table 2 Continued Study characteristics Study assessment First author Conflict and year Population or Pricing intervention Peer of Quality Bradford Hill Study type published Country Study design participants studied Outcomes studied reviewed interest rating criteria met Wald 1984 Poland Analysis of Population of Poland Poor harvest led to high Alcohol Yes Not Weak CON, TE, PL, routine data prices, rationing and illegal consumption and stated CO, EX 1970–1981 sales alcohol-related hospital admissions Zhao 2013 Canada Cross-sectional Populations of 16 Actual minimum price Acute, chronic and Yes None Strong SA, CON, TE, versus time health service increased over a 20-year wholly DR, PL, CO, series analysis delivery areas in period. Study modelled 10% alcohol-attributable EX of ecological British Columbia, increase in the average mortality data Canada minimum price of all alcoholic beverages adjusted by monthly Consumer Price Index. Also looked at outlet density Modelling Brennan England Modelling study The UK national MUP of £0.40, £0.45 and Alcohol Yes None Strong CON, SP, DR, studies 2014 using SAPM surveys of general £0.50. Ban on below cost consumption, PL, CO population selling consumer spending, (subgroups of 47 health harms, moderate, harmful, QALYs hazardous) Holmes England Modelling study The UK national MUP of 45p Alcohol Yes None Strong CON, SP, PL, 2014 using SAPM surveys of general consumption, CO population consumer spending, (subgroups of 47 health harms, moderate, harmful, QALYs hazardous) Meier 2009 The UK Modelling study The UK national Ten pricing policy options, Alcohol Yes None Strong CON, SP, DR, using SAPM surveys of general including different levels of consumption, PL, CO population MUP (of 33 analysed) consumer spending, (subgroups of 47 health harms, moderate, harmful, crime, employment hazardous) Meier 2016 England Modelling study The UK national MUP of £0.50 compared with Alcohol Yes None Strong CON, SP, PL, using SAPM surveys of general three alcohol taxation consumption in CO population interventions different income and (subgroups of socioeconomic moderate, increasing groups risk, heavy) Purshouse England Modelling study The UK national 18 different pricing policies Alcohol Yes None Strong CON, SP, DR, 2010 using SAPM surveys of general (including MUP) consumption, PL, CO population consumer spending, (subgroups of 47 health harms, moderate, harmful, QALYs hazardous) Continued Open Access Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 7 Table 2 Continued Study characteristics Study assessment First author Conflict and year Population or Pricing intervention Peer of Quality Bradford Hill Study type published Country Study design participants studied Outcomes studied reviewed interest rating criteria met Sharma Australia Counterfactual Representative MUP of A$2 Alcohol purchasing Yes None Strong CON, SP, PL, 2016 analysis sample of and consumption CO households (n=884) completing 12-month Homescan shopping survey Vandenberg Australia Modelling study Representative MUP of A$1 compared with Alcohol purchasing Yes None Strong CON, SP, PL, 2016 sample of a specific alcohol tax and consumption CO households (n=885) completing Homescan shopping survey Cross- Black 2011 Scotland Cross-sectional 377 hospital patients The UK alcohol units Alcohol Yes None Moderate SA CON, DR, sectional survey with serious alcohol purchased below proposed consumption PL, CO studies problems MUP of £0.40p/£0.50p Callinan Australia Cross-sectional Drinkers 18+ Australian standard drinks Alcohol Yes Not Moderate SA, CON, DR, 2015 survey participating in purchased below proposed consumption stated PL, CO Australian minimum prices of A$0.80/A International Alcohol $1.00/A$1.25 Control study (n=1681) Cousins Ireland Cross-sectional 3187 adults in 2013 Alcohol units purchased AUDIT-C score Yes None Strong SA, CON, PL, 2016 survey National Alcohol below proposed minimum CO Diary Survey price of €1.00 Crawford England Cross-sectional 515 members of the The UK alcohol units AUDIT score Yes None Moderate SA, CON, PL, 2012 survey public purchased below proposed CO MUP of £0.50 Falkner New Cross-sectional 115 adults New Zealand standard Alcohol Yes No Moderate SA, CON, PL, 2015 Zealand survey undergoing alcohol drinks purchased below consumption CO detoxification proposed minimum prices of NZ$1.00/NZ$1.10/NZ$1.20 Forsyth Scotland Cross-sectional Shopkeepers of 144 MUP of £0.50 Products affected Yes None Weak CON, PL 2014 survey off licences in and hospital (weakly), CO Glasgow admissions Ludbrook The UK Cross-sectional Expenditure and Purchasers of alcohol < Income of Yes Not Moderate SA, CON, PL, 2012 survey Food Survey data £0.45 per unit purchasers of cheap stated CO from 20062008 alcohol (n=18 624) Sharma Australia Cross-sectional Representative MUP of A$1 and taxation Alcohol Yes None Moderate SA, CON, DR, 2014 survey sample of consumption PL, CO households (n=885) (measured by completing shopping projected sales) survey Continued Open Access 8 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Table 2 Continued Study characteristics Study assessment First author Conflict and year Population or Pricing intervention Peer of Quality Bradford Hill Study type published Country Study design participants studied Outcomes studied reviewed interest rating criteria met Sheron The UK Cross-sectional Adult patients in a The UK alcohol units Alcohol Yes Not Moderate SA, CON, DR, 2014 survey liver unit of a hospital purchased below £0.50 consumption stated PL, CO (n=204) Intervention Babor 1978 The USA Trial (not 34 male volunteers in ‘Happy hour’ with a reduction Alcohol Yes Not Weak SA, CON, SP, studies randomised) live-in research in set price of alcohol for one consumption stated TE, CO, EX facility group of participants Qualitative Seaman Scotland Qualitative 130 participants aged Hypothetical minimum price Alcohol Yes None Moderate CON, CO studies 2013 study 16–30 increases consumption and substitution with other substances Systematic Wagenaar Worldwide Systematic Studies tended to Alcohol price and taxation Alcohol Yes None Strong AN reviews 2009 review and cover general interventions studied consumption meta-analysis population together (measured by alcohol sales or self-reported consumption) Wagenaar Worldwide Systematic Studies tended to Alcohol price and taxation Alcohol-related Yes Not Strong AN 2010 review and cover general interventions studied morbidity (disease, stated meta-analysis population together injury, suicide, traffic crashes, sexually transmitted diseases, other drug use, crime and misbehaviour) and mortality Abbreviations for the Bradford Hill criteria: AN, analogy; CO, coherence; AUDIT, Alcohol Use Disorders Identification Test; CON, consistency; DR, dose–response; EX, experiment; MUP, Minimum Unit Pricing; PL, plausibility; SA, strength of the association; SAPM, Sheffield Alcohol Policy Model; SP, specificity; TE, temporality; QALYs, Quality Adjusted Life Years. Open Access Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 9 Table 3 Studies published in the grey literature included in Bradford Hill criteria assessment Study characteristics Study assessment Author and Bradford year Population or Pricing intervention Peer Conflict Quality Hill criteria published Country Study design participants studied Outcomes studied reviewed of interest rating met Angus 2016 Scotland Modelling Scottish general MUP of 30p, 40p, 50p, Alcohol consumption, Not stated None Strong CON, SP, study using population survey 60p and 70p, compared consumer spending, DR, PL, SAPM (subgroups of with taxation interventions exchequer and retail CO moderate, harmful, revenue, 47 health hazardous) harms Booth 2008 Worldwide Review of Studies tended to Various minimum unit Alcohol consumption Yes None Strong AN reviews and cover general prices and taxation and various measures systematic population interventions of alcohol harm review Brennan England Modelling Adults in England General price increases. Alcohol consumption, Not stated None Strong CON, SP, 2008 study using MUP of £0.20, £0.25, consumer spending, DR, PL, SAPM £0.30, £0.35, £0.40, sales duty and VAT, CO £0.45, £0.50, £0.60 and 47 health harms, £0.70. Restrictions on crime and employment off-trade price promotions. Hill McManus Canada Modelling Adults in two MUP of C$1.50 Alcohol consumption, No None Strong CON, SP, 2012 study using Canadian consumer spending, PL, CO SAPM provinces (Ontario hospital admissions, and British mortality, crime Columbia) Institute for Great Economic Shopping data MUP of £0.45 Alcohol consumption Not stated Not stated Not CON, SP, Fiscal Studies Britain modelling from 25 248 British possible CO 2010 study using households to rate market research data Institute for Great Economic Population of Great MUP of £0.45 and Alcohol consumption Not stated Not stated Not CON, SP, Fiscal Studies Britain analysis Britain increased alcohol taxation possible CO 2013 to rate Meng 2010 Scotland Modelling Adults in Scotland MUP of £0.20, £0.25, Alcohol consumption, Not stated None Strong CON, SP, study using £0.30, £0.35, £0.40, consumer spending, DR, PL, SAPM £0.45, £0.50, £0.60 and 47 health harms, CO £0.70. Restrictions on crime, employment off-trade price promotions. AN, analogy; CO, coherence; CON, consistency; DR, dose–response; EX, experiment; PL, plausibility; SA, strength of the association; SAPM, Sheffield Alcohol Policy Model; SP, specificity; TE, temporality. Open Access 32 40 comes from research following the introduction of MUP drinking the cheapest alcohol. The modelling in Canada, where minimum price increases preceded studies which use survey data in turn suggest heavier 23 26 reductions in alcohol consumption, alcohol-attribut- drinkers will be most affected by MUP. Overall, the evi- able hospital admissions and alcohol-related mortal- dence base provides strong support for this criterion. ity. Studies where price changes preceded the expected changes in alcohol consumption or harm have Experiment 19 22 23 also been conducted in Russia, Poland and We have not identiﬁed any RCTs of minimum pricing or Finland. Overall, there is very strong support for the price-based interventions to reduce alcohol consump- temporality criterion. tion. There is a small (and not randomised) trial from the 1970s which found participants living in controlled Dose–response/biological gradient conditions and offered a daily ‘happy hour’ discount This criterion is supported if different price levels have drank signiﬁcantly more alcohol than those who were been found to have differing effects on consumption or not offered the discount. There is, however, substantial harm. Many of the studies using the Shefﬁeld Alcohol evidence in support of the experiment criterion from Policy Model explore the impact of a range of potential time series analyses or natural experiments, for example, 25 27 29 45 51 MUP options, and these consistently suggest where minimum pricing was introduced in Canada that the higher the MUP the greater the reductions in and where prices ﬂuctuated in the late 1980s and early 19 22 alcohol consumption or alcohol-related harms. The 1990s in Russia, and to a lesser extent in Finland, Canadian studies of minimum pricing lend further where minimum price increases were associated with support for this criterion because the analysis presents reduced mortality only among men with a basic educa- the effect on consumption or harm of a modelled 1% tion. These studies provide tentative support for the increase in price, meaning dose response can be experiment criterion. inferred. Dose response is supported to a lesser extent by evidence from cross-sectional studies that Analogy heavier drinkers are more likely to pay less than a pro- To address the analogy criterion, areas related to 32–34 39 40 posed MUP. Overall, there is strong support for minimum alcohol pricing must be considered. There is the dose–response criterion, although the relationship is 52 evidence from literature on the affordability of alcohol difﬁcult to quantify. that consumption and harm are very responsive to the affordability of alcohol. Large systematic reviews have Plausibility 53 investigated the price elasticity of demand for alcohol, This criterion refers to whether there is evidence that and have found that higher alcohol pricing and taxation alcohol price can be used as an economic mechanism to (considered together) are associated with reductions in inﬂuence consumption at a population level, and alcohol consumption, alcohol-related morbidity and whether heavy drinkers tend to purchase cheaper 43 44 46 mortality. Overall, the support for the analogy cri- alcohol. There is evidence from 21/26 research studies terion is very strong, although Bradford Hill describes and 4/7 studies in the grey literature that the price of this as the weakest evidence for causality. alcohol is inversely related to alcohol-related morbidity, hospital admissions or mortality. Moreover, there is also evidence from numerous cross-sectional studies in the DISCUSSION 32–36 38–40 UK, Ireland and Australia and one trial from We assessed 26 research studies and two systematic the USA that heavier drinking was signiﬁcantly asso- reviews, plus a further seven studies from the grey ciated with purchasing alcohol below speciﬁed prices, literature in this review of the evidence for priced-based further suggesting that economic mechanisms such as interventions—such as MUP—to reduce alcohol con- minimum pricing would particularly affect the heaviest sumption and alcohol-related harm. All nine of the drinkers. This provides strong support for the plausibility Bradford Hill criteria for causality were met and the vast criterion. majority of studies offered support for price-based alcohol policy interventions. However, the evidence for Coherence two of the criteria, although present, was not as strong This criterion refers to whether studies from different as it was for the other criteria. These criteria were disciplines have had complementary ﬁndings and strength of the association (criterion 1) and experiment whether these ﬁtor ‘hang’ well together. It is different (criterion 8), and according to Bradford Hill, these are to consistency, which is more concerned with reproduci- the two criteria that can provide the strongest evidence bility of ﬁndings. The ﬁndings of the majority of studies for causality. Therefore, although all of the criteria were supported the coherence criterion in that they suggest supported, we conclude that it is highly probable, but 2 3 21 24 that real-world MUP or minimum price not deﬁnite, that introducing MUP for alcohol would 19 20 23 increases led to reductions in alcohol consump- reduce alcohol consumption and alcohol-related harms. tion and alcohol-related harm and cross-sectional It is also of note that different types of study tended to surveys ﬁnd that it is the heavier drinkers that are satisfy different Bradford Hill criteria, and that different 10 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Open Access study designs also produced evidence of the effective- rated as strong or moderate. However, it was not possible ness of minimum pricing in relation to different out- to appraise two of the studies from the grey literature comes. This is summarised in ﬁgure 2. This underlines using this tool, and there were some challenges assessing the importance of including a variety of study designs in the econometric modelling studies against this frame- this review. work. However, overall we think that our quality Strengths of this study are that this is the ﬁrst to have appraisal across the different studies is broadly compar- systematically reviewed the literature relevant speciﬁcally able. It should also be noted that although a number of to alcohol minimum pricing policies. We had broad studies were rated as ‘strong’, this is in relation to their inclusion criteria with regards to study design, price respective study designs and does not reﬂect the pos- intervention and outcome measure, allowing for a com- ition of the study type in the hierarchy of evidence prehensive review of the evidence base. Application of framework. the Bradford Hill criteria as part of a narrative systematic This is the ﬁrst systematic review that has addressed literature review is a useful and emergent technique for the effectiveness of minimum alcohol price interven- identifying causality: a PubMed search for systematic tions such as MUP using the Bradford Hill criteria. It reviews with ‘Bradford Hill’ mentioned in the title or was beyond the scope of this review to study the impact abstract yielded 28 results, 90% of which were published of generalised increases in alcohol prices (as opposed to in the last 5 years. The limitations of this systematic minimum prices). However, where such studies have review relate mainly to the broad range of studies been carried out, a minimum price or ﬂoor price has included. It was not possible to conduct any kind of been recommended, for example, in Gruenewald’s 2006 meta-analysis and therefore we do not present a pooled study in Sweden which found that the lowest quality (the estimate for the likely effect of MUP on certain out- cheapest) alcohol has the highest price elasticity. comes. The exact effect of any MUP would be inﬂu- Previous systematic reviews of alcohol price and con- 43 44 enced by a range of factors, including: the minimum sumption and alcohol-related harm have tended to price level chosen, how broadly it is applied, how consider the effect of price increases and increased tax- strongly it is enforced and contextual factors such as ation together. These reviews found signiﬁcant effects affordability (in the UK, alcohol was 54% more afford- on consumption and morbidity and mortality. Although able in 2014 than it was in 1980 ), other governmental price regulation and taxation are closely related policy 55 45 regulations and the price-level pre-MUP. Occasionally, options, evidence from surveys and modelling studies minimum pricing has been implemented as part of a suggests that the effects of each are different, although range of measures, and these studies were considered it is known that the majority of tax increases are passed 43 56 alongside studies where MUP was implemented in isola- on as increased prices for consumers. It was beyond tion. This emphasises the importance of the speciﬁcity the scope of this review to discuss whether MUP is criterion. regressive in detail, but as it only affects the prices of the There were also challenges with the quality appraisal. cheapest drinks, which are usually consumed by the The EPHPP quality assessment tool was used to assess heaviest drinkers, MUP is likely to narrow health 28 31 quantitative studies and the majority of studies were inequalities. A recent rapid evidence review Figure 2 This model shows that different study types tended to produce evidence of effectiveness of minimum pricing in relation to different outcomes. Studies cited in the figure are key examples of the literature in that area and do not represent an exhaustive list. Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 11 Open Access Provenance and peer review Not commissioned; externally peer reviewed. published in The Lancet examined alcohol control pol- icies in England and recommended a combination of Data sharing statement No additional data are available. MUP and tax increases to reduce alcohol harm and Open Access This is an Open Access article distributed in accordance with increase government revenue, rather than either in iso- the terms of the Creative Commons Attribution (CC BY 4.0) license, which lation. It is also important to highlight that a consider- permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http:// able proportion of included studies were produced by a creativecommons.org/licenses/by/4.0/ small number of research teams. Also, with regards to the econometric modelling studies, uncertainty in esti- mates or forecasts is often poorly communicated outside REFERENCES of the academic literature. The overall risk of bias in the 1. Anderson P, Chisholm D, Fuhr DC. Effectiveness and included studies was minimised by excluding studies cost-effectiveness of policies and programmes to reduce the harm caused by alcohol. Lancet 2009;373:2234–46. with a conﬂict of interest (either for or against MUP). It 2. Stockwell T, Auld MC, Zhao J, et al. Does minimum pricing reduce was not possible to assess publication bias using an ana- alcohol consumption? The experience of a Canadian province. Addiction 2012;107:912–20. lytical technique such as a funnel plot due to the narra- 3. Stockwell T, Zhao J, Giesbrecht N, et al. The raising of minimum tive nature of the review; however, we anticipate that by alcohol prices in Saskatchewan, Canada: impacts on consumption including grey literature in this review, we have mitigated and implications for public health. Am J Public Health 2012;102: e103–10. publication bias as far as reasonably possible. 4. World Health Organization. European status report on alcohol and Overall, the ﬁndings of this review lend strong support health 2014. Pricing policies. World Health Organization, 2014. http:// www.euro.who.int/__data/assets/pdf_file/0003/244902/ for policies such as MUP in reducing alcohol consump- Pricing-policies.pdf tion and alcohol-related harm, with all nine of the 5. Department of Health (Ireland). The Public Health (Alcohol) Bill Bradford Hill criteria met, and little by way of counter 2015. 2015 (cited 17 February 2017). http://health.gov.ie/blog/ publications/public-health-alcohol-bill-2015/ ﬁndings. As it is unlikely to be feasible to conduct RCTs 6. The Guardian. Minimum alcohol price not in the public interest, says of MUP, the decision whether or not to introduce MUP health agency. The Guardian. 1 May 2014 (cited 17 February 2017). https://www.theguardian.com/world/2014/may/01/minimum-alcohol- will not be informed by a systematic review and price-not-in-the-public-interest-says-health-agency meta-analysis of RCTs, and therefore, this synthesis of evi- 7. New Zealand Ministry of Justice. Alcohol minimum pricing policies. dence according to the Bradford Hill criteria is of value. (cited 17 February 2017). https://justice.govt.nz/justice-sector-policy/ key-initiatives/sale-and-supply-of-alcohol/alcohol-minimum-pricing- Unanswered questions about the effectiveness of MUP report/ remain; for example, this review has highlighted that 8. Home Office. Guidance on banning the sale of alcohol below the cost of duty plus VAT. For suppliers of alcohol and enforcement support was moderate or tentative for two of the Bradford authorities in England and Wales. 2015 (cited 3 June 2016). Hill criteria (‘strength of the association’ and ‘experi- Available from: https://www.gov.uk/government/uploads/system/ uploads/attachment_data/file/415522/HO_Guidance_on_BBCS.pdf ment’, respectively). There may be opportunities to 9. Off Licence News. First conviction for selling alcohol “below cost” explore this in countries such as Scotland if MUP is imple- sees retailer fined. 2016 (cited 3 June 2016). http://www. mented. If Scotland were to implement MUP, then it offlicencenews.co.uk/news/fullstory.php/aid/15750/First_conviction_ for_selling_alcohol__below_cost__sees_retailer_fined.html would be possible to evaluate the validity of the Shefﬁeld 10. Home Office. Duty plus VAT permitted price calculator (2015). 2016. Alcohol Policy Model studies conducted using Scottish https://www.gov.uk/government/publications/banning-the-sale-of- alcohol-below-the-cost-of-duty-plus-vat-march-2015 data. It would also be possible to conduct a longitudinal 11. HM Government. The Government’s Alcohol Strategy. 2012 (cited 3 study to evaluate the effectiveness of MUP in reducing June 2016). https://www.gov.uk/government/uploads/system/ alcohol consumption and alcohol-related morbidity and uploads/attachment_data/file/224075/alcohol-strategy.pdf 12. Scottish Government. Minimum unit pricing. (cited 17 February mortality. The ﬁndings of this natural experiment would 2017). http://www.gov.scot/Topics/Health/Services/Alcohol/ have relevance elsewhere within and outside the UK. minimum-pricing 13. Court of Justice of the European Union. Press Release 155/15: the Scottish legislation introducing a minimum price per unit of alcohol is Twitter Follow Sadie Boniface @sadieboniface contrary to EU law if less restrictive tax measures can be introduced. 2015 (cited 17 February 2017). http://curia.europa.eu/jcms/upload/ docs/application/pdf/2015-12/cp150155en.pdf Acknowledgements We would like to thank Rebecca McDonald for advice on 14. Judiciary of Scotland. Scotch Whisky Association and others v Lord using the Bradford Hill criteria in a systematic review, Dr James Nicholls for Advocate and Advocate General for Scotland. (cited 11 July 2016). advice on study interpretation and Dr Daniel Stahl for statistical advice on http://www.scotland-judiciary.org.uk/9/1672/Scotch-Whisky- Association-and-others-v-Lord-Advocate-and-Advocate-General- some of the included studies. for-Scotland Contributors SM conceived the idea. SB conducted the initial search. SM and 15. BBC News. Bid to appeal against minimum alcohol pricing to SM contributed to independently reviewing abstracts, hand-searching Supreme Court. 2016 Nov 18 (cited 17 February 2017) http://www. bbc.co.uk/news/uk-scotland-scotland-politics-38026073 reference lists, completing data extraction and conducting quality appraisal. 16. Lifestyles Statistics Team, Health and Social Care Information All authors contributed to the analysis and interpretation of the results and Centre. Statistics on alcohol. England, 2015. 2015 (cited 3 June contributed to writing the manuscript. SB is guarantor. 2016). http://www.hscic.gov.uk/catalogue/PUB17712/ alc-eng-2015-rep.pdf Funding Open access for this article was funded by King’s College London 17. Cochrane Handbook. 21.4 Assessment of study quality and risk of Open Scholarship Fund. bias. (cited 2 June 2016). http://handbook.cochrane.org/chapter_21/ 21_4_assessment_of_study_quality_and_risk_of_bias.htm Competing interests SB and SM work at King’s College London, which as an 18. Hill AB. The environment and disease: association or causation? institution is listed as a member of the Alcohol Health Alliance. SM has Proc R Soc Med 1965;58:295–300. received funding indirectly from UKCTAS, which as an institution is also listed 19. Bhattacharya J, Gathmann C, Miller G. The Gorbachev Anti-Alcohol as a member of the Alcohol Health Alliance. None of the authors have any Campaign and Russia’s Mortality Crisis. Am Econ J Appl Econ relationship with the Alcohol Health Alliance. 2013;5:232–60. 12 Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 Open Access 20. Herttua K, Mäkelä P, Martikainen P. Minimum prices for alcohol and 41. Babor TF, Mendelson JH, Greenberg I, et al. Experimental analysis educational disparities in alcohol-related mortality. Epidemiology of the ‘happy hour’: effects of purchase price on alcohol 2015;26:337–43. consumption. Psychopharmacology (Berl) 1978;58:35–41. 21. Stockwell T, Zhao J, Martin G, et al. Minimum alcohol prices and 42. Seaman P, Edgar F, Ikegwuonu T. The role of alcohol price in young outlet densities in British Columbia, Canada: estimated impacts on adult drinking cultures in Scotland. Drugs Educ Prev Policy alcohol-attributable hospital admissions. Am J Public Health 2013;20:278–85. 2013;103:2014–20. 43. Wagenaar AC, Salois MJ, Komro KA. Effects of beverage alcohol 22. Treisman D. Death and prices. Econ Transit 2010;18:281–331. price and tax levels on drinking: a meta-analysis of 1003 estimates 23. Wald I, Moskalewicz J. Alcohol policy in a crisis situation. from 112 studies. Addiction 2009;104:179–90. Br J Addict 1984;79:331–5. 44. Wagenaar AC, Tobler AL, Komro KA. Effects of alcohol tax and 24. Zhao J, Stockwell T, Martin G, et al. The relationship between price policies on morbidity and mortality: a systematic review. minimum alcohol prices, outlet densities and alcohol-attributable Am J Public Health 2010;100:2270–8. deaths in British Columbia, 2002–09. Addiction 2013;108: 45. Angus C, Holmes J, Pryce R, et al. Model-based appraisal of the 1059–69. comparative impact of minimum unit pricing and taxation policies in 25. Brennan A, Meng Y, Holmes J, et al. Potential benefits of minimum Scotland an adaptation of the Sheffield Alcohol Policy Model version unit pricing for alcohol versus a ban on below cost selling in England 3. ScHARR, University of Sheffield, 2016 April (cited 7 April 2016). 2014: modelling study. BMJ 2014;349:g5452. https://www.shef.ac.uk/polopoly_fs/1.565373!/file/Scotland_report_ 26. Holmes J, Meng Y, Meier PS, et al. Effects of minimum unit pricing 2016.pdf for alcohol on different income and socioeconomic groups: 46. Booth A, Brennan A, Meier PS, et al. Independent review of the a modelling study. Lancet 2014;383:1655–64. effects of alcohol pricing and promotion: part A—systematic reviews. 27. Meier PS, Purshouse R, Brennan A. Policy options for alcohol price Project report for the Department of Health September 2008. regulation: the importance of modelling population heterogeneity. ScHARR University of Sheffield, 2008. Addiction 2010;105:383–93. 47. Brennan A, Purshouse R, Taylor K, et al. Independent review of the 28. Meier PS, Holmes J, Angus C, et al. Estimated effects of different effects of alcohol pricing and promotion: part B. Modelling the alcohol taxation and price policies on health inequalities: potential impact of pricing and promotion policies for alcohol in a mathematical modelling study. PLOS Med 2016;13:e1001963. England: results from the Sheffield Alcohol Policy Model version 29. Purshouse RC, Meier PS, Brennan A, et al. Estimated effect of 2008(1-1). ScHARR University of Sheffield, 2008. https://www.shef. alcohol pricing policies on health and health economic outcomes in ac.uk/polopoly_fs/1.95621!/file/PartB.pdf England: an epidemiological model. Lancet 2010;375:1355–64. 48. Hill McManus D, Brennan A, Stockwell T, et al. Model-based 30. Sharma A, Etilé F, Sinha K. The effect of introducing a minimum appraisal of alcohol minimum pricing in Ontario and British price on the distribution of alcohol purchase: a counterfactual Columbia: a Canadian adaptation of the Sheffield Alcohol Policy analysis. Health Econ 2016;25:1182–200. Model version 2. 2012. http://www.uvic.ca/research/centres/carbc/ 31. Vandenberg B, Sharma A. Are alcohol taxation and pricing policies assets/docs/report-model-based-appraisal.pdf regressive? Product-level effects of a specific tax and a minimum 49. Institute for Fiscal Studies. The impact of Introducing a minimum unit price for alcohol. Alcohol Alcohol 2016;51:493–502. price on alcohol in Britain. IFS Briefing Note BN109. 2010. https:// 32. Black H, Gill J, Chick J. The price of a drink: levels of consumption www.ifs.org.uk/publications/5287 (accessed 27 March 2017) and price paid per unit of alcohol by Edinburgh’s ill drinkers with a 50. Institute for Fiscal Studies. Price-based measures to reduce alcohol comparison to wider alcohol sales in Scotland. Addiction consumption. IFS Briefing Note BN138. 2013. https://www.ifs.org.uk/ 2011;106:729–36. publications/6644 (accessed 27 March 2017) 33. Callinan S, Room R, Livingston M, et al. Who purchases low-cost 51. Meng Y, Purshouse R, Brennan A, et al. Model-based appraisal of alcohol in Australia? Alcohol Alcohol 2015;50:647–53. alcohol minimum pricing and off-licensed trade discount bans in 34. Cousins G, Mongan D, Barry J, et al. Potential impact of minimum Scotland using the Sheffield Alcohol Policy Model (v.2): second unit pricing for alcohol in Ireland: evidence from the national alcohol update based on newly available data. 2010. http://www.shef.ac.uk/ diary survey. Alcohol Alcohol 2016;51:734–40. polopoly_fs/1.96510!/file/scotlandupdate.pdf 35. Crawford MJ, Parry AMH, Weston ARW, et al. Relationship 52. Rabinovich L, Brutscher PB, de Vries H, et al. The affordability of between price paid for off-trade alcohol, alcohol consumption and alcoholic beverages in the European Union. 2012. http://ec.europa. income in England: a cross-sectional survey. Alcohol Alcohol eu/health/ph_determinants/life_style/alcohol/documents/alcohol_ 2012;47:738–42. rand_en.pdf (accessed 27 March 2017) 36. Falkner C, Christie G, Zhou L. The effect of alcohol price on 53. Gallet CA. The demand for alcohol: a meta-analysis of elasticities. dependent drinkers’ alcohol consumption. N Z Med J Rochester, NY: Social Science Research Network, 2007 May (cited 2015;128:9–17. 6 January 2016). Report No.: ID 985689. http://papers.ssrn.com/ 37. Forsyth AJM, Ellaway A, Davidson N. How might the alcohol abstract=985689. minimum unit pricing (MUP) impact upon local off-sales shops and 54. Gruenewald PJ, Ponicki WR, Holder HD, et al. Alcohol prices, the communities which they serve? Alcohol Alcohol beverage quality, and the demand for alcohol: quality substitutions 2014;49:96–102. and price elasticities. Alcohol Clin Exp Res 2006;30:96–105. 38. Ludbrook A, Petrie D, McKenzie L, et al. Tackling alcohol misuse: 55. Ally AK, Meng Y, Chakraborty R, et al. Alcohol tax pass-through purchasing patterns affected by minimum pricing for alcohol. Appl across the product and price range: do retailers treat cheap alcohol Health Econ Health Policy 2012;10:51–63. differently? Addiction 2014;109:1994–2002. 39. Sharma A, Vandenberg B, Hollingsworth B. Minimum pricing of 56. Rabinovich L, Hunt P, Staetsky L, et al. Further study on the alcohol versus volumetric taxation: which policy will reduce heavy affordability of alcoholic beverages in the EU [Internet]. 2012 (cited consumption without adversely affecting light and moderate 17 February 2017). http://www.rand.org/pubs/technical_reports/ consumers? PLoS ONE 2014;9:e80936. TR1203.html 40. Sheron N, Chilcott F, Matthews L, et al. Impact of minimum price per 57. Burton R, Henn C, Lavoie D, et al. A rapid evidence review of the unit of alcohol on patients with liver disease in the UK. Clin Med effectiveness and cost-effectiveness of alcohol control policies: an 2014;14:396–403. English perspective. Lancet 2016. Boniface S, et al. BMJ Open 2017;7:e013497. doi:10.1136/bmjopen-2016-013497 13

Journal

BMJ Open – British Medical Journal

Published: Jun 6, 2017

Keywords: alcoholpolicyminimum unit pricingPUBLIC HEALTHBradford Hill

References

Effectiveness and cost-effectiveness of policies and programmes to reduce the harm caused by alcohol

Anderson, P; Chisholm, D; Fuhr, DC
Does minimum pricing reduce alcohol consumption? The experience of a Canadian province

Stockwell, T; Auld, MC; Zhao, J
The raising of minimum alcohol prices in Saskatchewan, Canada: impacts on consumption and implications for public health

Stockwell, T; Zhao, J; Giesbrecht, N
The environment and disease: association or causation?

Hill, AB
The Gorbachev Anti-Alcohol Campaign and Russia's Mortality Crisis

Bhattacharya, J; Gathmann, C; Miller, G
Minimum prices for alcohol and educational disparities in alcohol-related mortality

Herttua, K; Mäkelä, P; Martikainen, P
Minimum alcohol prices and outlet densities in British Columbia, Canada: estimated impacts on alcohol-attributable hospital admissions

Stockwell, T; Zhao, J; Martin, G
Death and prices

Treisman, D
Alcohol policy in a crisis situation

Wald, I; Moskalewicz, J
The relationship between minimum alcohol prices, outlet densities and alcohol-attributable deaths in British Columbia, 2002–09

Zhao, J; Stockwell, T; Martin, G
Potential benefits of minimum unit pricing for alcohol versus a ban on below cost selling in England 2014: modelling study

Brennan, A; Meng, Y; Holmes, J
Effects of minimum unit pricing for alcohol on different income and socioeconomic groups: a modelling study

Holmes, J; Meng, Y; Meier, PS
Policy options for alcohol price regulation: the importance of modelling population heterogeneity

Meier, PS; Purshouse, R; Brennan, A
Estimated effects of different alcohol taxation and price policies on health inequalities: a mathematical modelling study

Meier, PS; Holmes, J; Angus, C
Estimated effect of alcohol pricing policies on health and health economic outcomes in England: an epidemiological model

Purshouse, RC; Meier, PS; Brennan, A
The effect of introducing a minimum price on the distribution of alcohol purchase: a counterfactual analysis

Sharma, A; Etilé, F; Sinha, K
Are alcohol taxation and pricing policies regressive? Product-level effects of a specific tax and a minimum unit price for alcohol

Vandenberg, B; Sharma, A
The price of a drink: levels of consumption and price paid per unit of alcohol by Edinburgh's ill drinkers with a comparison to wider alcohol sales in Scotland

Black, H; Gill, J; Chick, J
Who purchases low-cost alcohol in Australia?

Callinan, S; Room, R; Livingston, M
Potential impact of minimum unit pricing for alcohol in Ireland: evidence from the national alcohol diary survey

Cousins, G; Mongan, D; Barry, J
Relationship between price paid for off-trade alcohol, alcohol consumption and income in England: a cross-sectional survey

Crawford, MJ; Parry, AMH; Weston, ARW
The effect of alcohol price on dependent drinkers’ alcohol consumption

Falkner, C; Christie, G; Zhou, L
How might the alcohol minimum unit pricing (MUP) impact upon local off-sales shops and the communities which they serve?

Forsyth, AJM; Ellaway, A; Davidson, N
Tackling alcohol misuse: purchasing patterns affected by minimum pricing for alcohol

Ludbrook, A; Petrie, D; McKenzie, L
Minimum pricing of alcohol versus volumetric taxation: which policy will reduce heavy consumption without adversely affecting light and moderate consumers?

Sharma, A; Vandenberg, B; Hollingsworth, B
Impact of minimum price per unit of alcohol on patients with liver disease in the UK

Sheron, N; Chilcott, F; Matthews, L
Experimental analysis of the ‘happy hour’: effects of purchase price on alcohol consumption

Babor, TF; Mendelson, JH; Greenberg, I
The role of alcohol price in young adult drinking cultures in Scotland

Seaman, P; Edgar, F; Ikegwuonu, T
Effects of beverage alcohol price and tax levels on drinking: a meta-analysis of 1003 estimates from 112 studies

Wagenaar, AC; Salois, MJ; Komro, KA
Effects of alcohol tax and price policies on morbidity and mortality: a systematic review

Wagenaar, AC; Tobler, AL; Komro, KA
Independent review of the effects of alcohol pricing and promotion: part A—systematic reviews

Booth, A; Brennan, A; Meier, PS
Alcohol prices, beverage quality, and the demand for alcohol: quality substitutions and price elasticities

Gruenewald, PJ; Ponicki, WR; Holder, HD
Alcohol tax pass-through across the product and price range: do retailers treat cheap alcohol differently?

Ally, AK; Meng, Y; Chakraborty, R
A rapid evidence review of the effectiveness and cost-effectiveness of alcohol control policies: an English perspective

Burton, R; Henn, C; Lavoie, D

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality

Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality

Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality

References

Abstract

Journal

Recommended Articles

References

Our policy towards the use of cookies