Choosing between different alcohol pricing and taxation strategies: a comparative policy appraisal using the Sheffield Alcohol Policy Model

BackgroundPolicies that increase alcohol prices effectively reduce alcohol consumption, one of the top three risk factors for global disease burden. Our aim was to appraise how different alcohol pricing policies balance competing priorities.MethodsWe built an econometric dynamic epidemiological model for England, combining survey and register data on alcohol purchasing, consumption, and 43 harms, and published price elasticities, relative risk, and alcohol attribution. We model five hypothetical taxation options each estimated to give a 10% reduction in average consumption: P1, a uniform 85% increase in existing duty; P2, a 22% sales tax based on product price; P3, a £0·35 per unit volumetric tax; P4, a £0·80 minimum unit price (MUP); P5, a £0·75 MUP with a volumetric tax of £0·30. Outcomes were consumption, annual alcohol-related deaths, hospital admissions, health-care costs, consumer spending, and government revenue. Uncertainty was assessed through sensitivity analyses.FindingsPopulation level health harm-reductions would be highest for increases in the present tax system (P1: deaths ?3026, hospital admissions ?179 000, health-care costs ?£583 million) and for the MUP (P4: ?3081, ?169 000, ?£574 million), and lowest for the sales tax (P2: ?2852, ?168 000, ?£575 million). Harm reductions in high-risk drinkers would be highest for the two MUP options (P4: deaths ?1764, hospital admissions ?84 000, health care ?£218 million; P5: ?1712, ?78 000, ?£205 million), and lowest for sales tax (P2: ?1267, ?64 000, ?£159 million). The overall greatest burden on consumer spending would be from a duty increase under the present system (P1 £3·2 million), and the lowest from a volumetric tax (P3 £2·0 million). The smallest extra annual expenditure for each moderate drinker would be achieved by MUP (P4 £17·80), whereas the largest extra expenditure would be from a tax rise in the present system (P2 £37·20). With the exception of the MUP-only policy, which would have a small negative effect on government revenue from alcohol (P4: 1·3%, ?£121 million), all taxation policies would raise government revenue, by between £1·9 billion and £4·2 billion per year.InterpretationPricing policies can be implemented in ways that balance the priorities of increasing government revenue, maximising harm reductions, and targeting heavy drinkers, while protecting moderate consumers from excessive burden.FundingMedical Research Council and Economic and Social Research Council (grant G1000043).     

Mapping alcohol and tobacco tax interventions for health in the UK: a qualitative framework analysis

BackgroundAlcohol and tobacco use undermine population health, generating substantial costs. Increasing price is an effective means to reduce consumption and tax is a key harm reduction tool. In the UK, alcohol and tobacco tax are managed within fiscal policy, which does not necessarily prioritise health promotion. We aimed to map the objectives and options for alcohol and tobacco tax change in the UK, including the potential for greater coordination to improve health.MethodsWe did five semi-structured interviews with ten participants selected for their expertise in alcohol or tobacco tax policy. Interviews occurred in pairs (ie, one alcohol and one tobacco expert in each interview) to elicit comparison between substances and were supported by a rapid literature review of tax options. Participants were from government, arms-length governmental organisations, and advocacy groups. Informed by a rapid literature review, comparative framework analysis of alcohol and tobacco tax policy objectives, options, and factors pertaining to coordination between tobacco and alcohol was done.FindingsParticipants raised common health objectives (reducing consumption, harm, inequalities) and fiscal objectives (raising revenue, mitigating societal costs). Drawing on options identified in the rapid review, participants discussed common tax options to achieve these objectives: tax rate increases (sudden rises, annual increases, minimum thresholds), changing tax structures (taxing products differently, tax proportional to harm), levies (taxing retailers and manufacturers), and revenue hypothecation (for prevention or treatment of addiction, local services). Participants were positive about policy exchange across alcohol and tobacco and modelling the combined effect of tax changes, but uncertain about formally linking tax policy across substances.InterpretationRaising tax is often considered to improve health by making products less affordable, but a tax regime that raises additional revenue can support prevention and treatment services and mitigate the social and economic costs of consumption. An unresolved issue for policy debate is who should pay this revenue and how revenue would be maintained if the health objective of falling consumption were met. Although more input from fiscal policymakers would deepen findings, our rapid review and interview approach facilitated discussion across alcohol and tobacco, and use of the framework approach ensured consistent analysis.FundingNational Institute of Health Research Public Health Research programme.     

Cigarette brand diversity and price changes during the implementation of plain packaging in the United Kingdom

Background and aim

Plain packaging of cigarettes appeared in the United Kingdom in July 2016 and was ubiquitous by May 2017. The change coincided with another legislative change, raising the minimum pack size from 10 to 20 cigarettes. Laws imposing plain packaging on cigarette packs remove another promotional route from tobacco companies, but the effect of such laws on brand diversity, pricing and sales volume is unknown. This study aimed to (1) describe and quantify changes in brand diversity, price segmentation and sales volumes and (2) estimate the association between the introduction of plain cigarette packaging and cigarette pricing in the United Kingdom.

Design

We used a natural experiment design to assess the impact of plain packaging legislation on brand diversity and cigarette prices. The data comprised a sample of 76% of sales of cigarettes in the UK between March 2013 and June 2017.

Setting

United Kingdom.

Measurements

Cigarette prices, number of brands and products and volumes of sales.

Findings

During the period analysed, there was a slight decrease in the number of cigarette brands. There was also an initial increase observed in the number of cigarette products, due mainly to an increase in the number of products in packs of fewer than 20 cigarettes sold before July 2016, which was then followed by a rapid decrease in the number of products that coincided with the implementation of the new legislation. Cigarette sales volumes during this period did not deviate from the preceding secular trend, but prices rose substantially. Regression results showed that price per cigarette, regardless of pack size, was 5.0 [95% confidence interval (CI) = 4.8–5.3] pence higher in plain than in fully branded packs. For packs of 20 cigarettes, price increases were greater in the lower price quintiles, ranging from 2.6 (95% CI = 2.4–2.7) GBP in the lowest to 0.3 (95% CI = 0.3–0.4) GBP per pack in the highest quintile.

Conclusions

The implementation of standardized packaging legislation in the United Kingdom, which included minimum pack sizes of 20, was associated with significant increases overall in the price of manufactured cigarettes, but no clear deviation in the ongoing downward trend in total volume of cigarette sales.     

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

Aim To compare alcohol purchasing and consumption by ill drinkers in Edinburgh with wider alcohol sales in Scotland. Design Cross‐sectional. Setting Two hospitals in Edinburgh in 2008/09. Participants A total of 377 patients with serious alcohol problems; two‐thirds were in‐patients with medical, surgical or psychiatric problems due to alcohol; one‐third were out‐patients. Measurements Last week's or typical weekly consumption of alcohol: type, brand, units (1 UK unit 8 g ethanol), purchase place and price. Findings Patients consumed mean 197.7 UK units/week. The mean price paid per unit was £0.43 (lowest £0.09/unit) (£1 = 1.6 US$ or 1.2€), which is below the mean unit price, £0.71 paid in Scotland in 2008. Of units consumed, 70.3% were sold at or below £0.40/unit (mid‐range of price models proposed for minimum pricing legislation by the Scottish Government), and 83% at or below £0.50/unit proposed by the Chief Medical Officer of England. The lower the price paid per unit, the more units a patient consumed. A continuous increase in unit price from lower to higher social status, ranked according to the Scottish Index of Multiple Deprivation (based on postcode), was not seen; patients residing in postcodes in the mid‐quintile paid the highest price per unit. Cheapness was quoted commonly as a reason for beverage choice; ciders, especially ‘white’ cider, and vodka were, at off‐sales, cheapest per unit. Stealing alcohol or drinking alcohol substitutes was only very rarely reported. Conclusions Because patients with serious alcohol problems tend to purchase very cheap alcohol, elimination of the cheapest sales by minimum price or other legislation might reduce their consumption. It is unknown whether proposed price legislation in Scotland will encourage patients with serious alcohol problems to start stealing alcohol or drinking substitutes or will reduce the recruitment of new drinkers with serious alcohol problems and produce predicted longer‐term gains in health and social wellbeing.     

Content analysis of tobacco industry data on the illicit tobacco trade in UK newspapers during the standardised packaging debate

BackgroundStandardised tobacco packaging, which commenced in the UK in May, 2016, has been vehemently opposed by transnational tobacco companies. Previous research has concluded that companies invoke the existence of an illicit trade in tobacco to oppose tobacco control policies and exaggerate its true scale. The argument they make is that such policies will further increase this trade. Independent peer-reviewed evidence does not support claims that illicit tobacco increased after the 2012 implementation of standardised packaging in Australia. In the UK, leaked Philip Morris International documents revealed the tobacco company's intention to utilise third-party “media messengers” to make the argument that the policy will exacerbate the illicit tobacco trade. We aimed to explore whether and, if so, how, transnational tobacco companies presented their data on illicit tobacco in UK newspapers during the standardised packaging policy debate and whether Philip Morris implemented its media messengers plan.MethodsArticles about illicit tobacco published in English language UK newspapers from April 1, 2013, to March 31, 2015, from LexisNexis were examined for presence, nature, and timing of tobacco industry data, which included any illicit tobacco monitoring initiative commissioned or undertaken by transnational tobacco companies in the UK or overseas. Search terms included combinations of “illegal”, “illicit”, “smuggling”, “tobacco”, “cigarette”, all four transnational tobacco companies, and known tobacco industry consultants.Findings428 articles about illicit tobacco were identified. 157 articles (37%) cited industry data, of which 90 (57%) referred to undercover test purchases, the remainder citing data from empty pack surveys, seizures, polls, or data of unknown methodology. 98 (62%) of the 157 articles mentioned ex-Scotland Yard Detective Chief Inspector Will O'Reilly, who conducts undercover test purchases for Philip Morris International; in 34 of these articles, this funding was not disclosed. Most articles were published in regional newspapers (77%, 121/157) and the majority mentioned counterfeit tobacco, so-called cheap whites, or both (72%, 113/157). However, the Institute of Trading Standards revealed that between May and November, 2014, over 70% of the illegal tobacco seized was tobacco industry contraband.InterpretationPhilip Morris International's investment in undercover test purchases appears to have enabled the tobacco industry to secure significant press coverage of its data on illicit tobacco. With over half of the articles in this study referring to a consultant working on behalf of the tobacco company, its plan to use third party media messengers as part of its anti-standardised packaging strategy appears to have been mobilised. Industry-funded third party spokesmen, notably former policemen, are now appearing in other countries, suggesting that this is an industry tactic that should be expected in other jurisdictions. Transparency in all media coverage is essential.FundingKAE-R is supported by Cancer Research UK (C27260/A20488). JLH is supported by the UK Centre for Tobacco and Alcohol Studies (MR/KO23195/1). AR is supported by the New Venture Fund.     

Fiscal policy, affordability and cross effects in the demand for tobacco products: the case of Uruguay

Uruguay, a country with a solid tobacco control policy since 2005 shows, contrary to expectations, an insignificant decrease in total tobacco products' sales in the last five years. The hypothesis is that on one side, changes in household income and the income elasticity of the demand for cigarettes were important countervailing factors in the demand of both products. The period 2005-2009 shows a large increase of 36% in household real income in Uruguay due to fast economic recovery after the 2002 crisis. The second factor is the interchangeability of roll your own and cigarettes and the impact on the demand of each product as a reaction to tax and price changes. The tax and price of roll your own tobacco remains substantially lower than that of cigarettes. This fact, and the increased substitution of roll your own for cigarettes seems to be the main reasons for the low impact of the policy of tobacco tax and price increases. This paper then consists of a revision of a 2004 study to estimate separate demands for both main tobacco products and obtain estimates for own price, cross price and income elasticities. Then, a simulation study was performed using the elasticities found and two scenarios of increases in household income: moderate (2.5% per year) and high (5% per year) confirming that countries where income is growing fast and with a potential for substitution towards cheaper products require substantial cigarette tax and price increases for a fiscal tobacco control policy to become effective.     

U.S. Tobacco Taxes: behavioural effects and policy implications

This paper examines U.S. tobacco taxation, the effect of cigarette taxes on smoking and on the health effects of smoking, and equity and efficiency considerations that arise when cigarette excise taxes are used to reduce smoking. Cigarette excise taxes, imposed by the Federal Government, all Stale governments, and nearly 400 cities and counties, add approximately 34 cents per pack to the price of cigarettes. Real cigarette excise tax rates have fallen because tax increases have not kept pace with inflation. Increases in the price of cigarettes decrease smoking, particularly by adolescents. An estimated 100,000 additional persons may live to the age of 65 as a result of doubling the Federal cigarette tax in 1983. Because cigarette taxes are regressive and are borne primarily by smokers, inequities may arise when they are used to reduce smoking. Success in achieving a tobacco-free society will require that tobacco taxes be replaced with alternative sources of revenue.     

Australian smokers' and recent quitters' responses to the increasing price of cigarettes in the context of a tobacco tax increase

Aims To track smokers' responses to the increasing price of cigarettes after a tax increase, and assess socio‐demographic differences in responses. Design The Cancer Institute NSW's Tobacco Tracking Survey (CITTS) is a continuous tracking telephone survey. Weekly data were collected between May and September 2010. Settings New South Wales, Australia. Participants A total of 834 smokers and 163 recent quitters (quit in last 12 months). Measurements Responses to the price increase included smoking‐related changes (tried to quit, cut down) and product‐related changes (changed to lower priced brands, started using loose tobacco, bought in bulk). Recent quitters were asked how much the increasing price of cigarettes influenced them to quit. Findings Overall, 47.5% of smokers made smoking‐related changes and 11.4% made product‐related changes without making smoking‐related changes. Multinomial logistic regressions showed that younger smokers (versus older) were more likely to make product‐related changes and smoking‐related changes in comparison to no changes. Low‐ or moderate‐income smokers (versus high‐income) were more likely to make smoking‐related changes compared to no changes. Highly addicted smokers (versus low addicted) were more likely to make product‐related changes and less likely to make smoking‐related changes. The proportion of smokers making only product‐related changes decreased with time, while smoking‐related changes increased. Recent quitters who quit after the tax increase (versus before) were more likely to report that price influenced them. Conclusions The effect of increasing cigarette prices on smoking does not appear to be mitigated by using cheaper cigarette products or sources. These results support the use of higher cigarette prices to encourage smoking cessation.     

Exploring the case for investing in early years in Wales: a broad evidence review

BackgroundLife-expectancy and healthy life-expectancy depend on the socioeconomic and wider determinants of health. At birth there is an 18·9 year difference in healthy life expectancy between the least and most deprived areas of Wales. Multidisciplinary public health policy requires that prevention interventions are increasingly assessed in terms of the financial return on investment. The aim of this study was to explore the return on investment of a broad range of early years' interventions, where possible applied to the context of Wales.MethodsWe constructed a pragmatic framework to explore international and UK evidence on the relative return on investment of: devoting public sector resources to programmes and practices that support babies and young children (up to age 7 years), and their families; promoting skills development through education; and fostering positive environments that protect health assets relevant to the early years. This review was not done systematically because of the breadth of interventions considered. Rather, we used snowballing to search for literature starting with centres of excellence in the USA and UK, and recent key policy reports. We restricted our search to studies published in English during the past 20 years.FindingsWe identified estimates of return on investment, mainly from the USA and UK, from investment in the child and whole family, education and skills, and a child's environment and wider community. Methods of calculating return on investment were not standardised so comparisons across programmes were problematic. We found evidence that for every £1 spent on contraceptive services between £11 and £14 in savings could be returned annually to the National Health Service (NHS) in Wales. We estimate that increasing the proportion of women who exclusively breastfeed at 4 months (presently 9% in Wales) to that recorded at birth (57%) could save the NHS in Wales £1·5 million annually in treating common early childhood diseases. Our finding that every £1 invested in adventure playgrounds could provide a return on investment of £1·32 shows the importance of physical activity and play.InterpretationThere are efficiency and equity arguments for investing in early years. Investment in early years should be considered in the same way as investment in the wider economic development across the Welsh economy.FundingPublic Health Wales.     

The annual cost of blood transfusions in the United Kingdom

This study estimated the annual cost of blood transfusions in the UK during 1994/1995. The analysis was based on published data, information derived from interviews with relevant NHS personnel and a purpose-designed structured questionnaire of blood donors. The cost to the UKs blood transfusion services of providing blood and blood products for transfusion was £165.5 million in 1994/1995. During this period, 2.75 million conventional donations of whole blood and 144 000 apheresis donations of platelets and plasma were collected: 2.58 million units of red blood cells were issued, resulting in ≈ 866 000 red blood cell transfusions; 334 000 units of fresh frozen plasma and 1.16 million units of platelets were issued, resulting in ≈ 17 000 and 188 000 isolated plasma and platelet transfusions, respectively. Hospital resource use attributable to providing blood transfusions during 1994/1995 cost the NHS £52.6 million. In total, blood transfusions cost the NHS £218.2 million during 1994/1995. Of this, red blood cell transfusions accounted for 76% of the annual cost, isolated platelet transfusions 16%, isolated plasma transfusions 1% and other products 7%. Donors incurred direct costs of £3.1 million and indirect costs of £11.2 million were accrued due to lost productivity. Additionally, blood donors gave up 2.5 million hours of their leisure time donating blood.     

Effects of beverage alcohol price and tax levels on drinking: a meta-analysis of 1003 estimates from 112 studies

Aims We conducted a systematic review of studies examining relationships between measures of beverage alcohol tax or price levels and alcohol sales or self‐reported drinking. A total of 112 studies of alcohol tax or price effects were found, containing1003 estimates of the tax/price–consumption relationship. Design Studies included analyses of alternative outcome measures, varying subgroups of the population, several statistical models, and using different units of analysis. Multiple estimates were coded from each study, along with numerous study characteristics. Using reported estimates, standard errors, t‐ratios, sample sizes and other statistics, we calculated the partial correlation for the relationship between alcohol price or tax and sales or drinking measures for each major model or subgroup reported within each study. Random‐effects models were used to combine studies for inverse variance weighted overall estimates of the magnitude and significance of the relationship between alcohol tax/price and drinking. Findings Simple means of reported elasticities are ?0.46 for beer, ?0.69 for wine and ?0.80 for spirits. Meta‐analytical results document the highly significant relationships (P < 0.001) between alcohol tax or price measures and indices of sales or consumption of alcohol (aggregate‐level r = ?0.17 for beer, ?0.30 for wine, ?0.29 for spirits and ?0.44 for total alcohol). Price/tax also affects heavy drinking significantly (mean reported elasticity = ?0.28, individual‐level r = ?0.01, P < 0.01), but the magnitude of effect is smaller than effects on overall drinking. Conclusions A large literature establishes that beverage alcohol prices and taxes are related inversely to drinking. Effects are large compared to other prevention policies and programs. Public policies that raise prices of alcohol are an effective means to reduce drinking.     

Economic policies for tobacco control in developing countries

Raising tobacco taxes can have an income distributional impact on the population. Since lower socio-economic groups usually smoke more, they also contribute more to total cigarette tax collection. Thus, those who can afford it least contribute the most in terms of tobacco taxes. This means that tobacco taxes are regressive. However, tobacco tax increases are likely to be progressive, decreasing the relative tax incidence on the poor, vis-à-vis the rich. This is based on the premise that the poor are likely to be more sensitive to price changes, and would thus reduce their cigarette consumption by a greater percentage than the rich in response to an excise tax-induced increase in cigarette prices. Recent empirical studies confirm this hypothesis by demonstrating that the price responsiveness of cigarette demand increases with income. Research in China confirmed that reducing cigarette expenditures could release household resources for spending on food, housing, and other goods that improve living standards. Therefore, in the long run, tobacco control measures will reduce social inequality.     

Systematic analysis of funding awarded for viral hepatitis‐related research to institutions in the United Kingdom, 1997–2010

Viral hepatitis is responsible for great health, social and economic burden both globally and in the UK. This study aimed to assess the research funding awarded to UK institutions for viral hepatitis research and the relationship of funded research to clinical and public health burden of viral hepatitis. Databases and websites were systematically searched for information on infectious disease research studies funded for the period 1997–2010. Studies specifically related to viral hepatitis research were identified and categorized in terms of funding by pathogen, disease and by a research and development value chain describing the type of science. The overall data set included 6165 studies (total investment £2.6 billion) of which £76.9 million (3.0%) was directed towards viral hepatitis across 323 studies (5.2%). By pathogen, there were four studies specifically investigating hepatitis A (£3.8 million), 69 studies for hepatitis B (21.4%) with total investment of £14.7 million (19.1%) and 236 (73.1%) hepatitis C studies (£62.7 million, 81.5%). There were 4 studies investigating hepatitis G, and none specifying hepatitis D or E. By associated area, viral hepatitis and therapeutics research received £17.0 million, vaccinology £3.1 million and diagnostics £2.9 million. Preclinical research received £50.3 million (65.4%) across 173 studies, whilst implementation and operational research received £19.4 million (25.3%) across 128 studies. The UK is engaged in much hepatology research, but there are areas where the burden is great and may require greater focus, such as hepatitis E, development of a vaccine for hepatitis C, and further research into hepatitis‐associated cancers. Private sector data, and funding information from other countries, would also be useful in priority setting.     

Modelling the long-term epidemiological and cost impact of a multicomponent lifestyle intervention and a sugar-sweetened beverage tax in the European Union: results from the EConDA project

BackgroundInformation on the efficacy and costs associated with targeting obesity can inform implementation of cost-effective policies. Within the Economics of Chronic Diseases (EConDA) project, we aimed to quantify the health and economic impact of two approaches to reduce chronic disease in eight European countries through to 2050: a multicomponent lifestyle intervention (MCLI) and a sugar-sweetened beverage tax.MethodsBody-mass index (BMI), disease epidemiology, and cost data for the diseases modelled (type 2 diabetes, coronary heart disease, stroke, and hypertension) were obtained from the literature for Bulgaria, Finland, Greece, Lithuania, the Netherlands, Poland, Portugal, and the UK. When country-specific data were not available, other countries were used as a proxy after adjustment for between-country differences. Mean change in BMI after annual MCLI to reduce energy intake and improve physical activity through behaviour change was obtained from the literature and was estimated after a 20% excise tax on sugar-sweetened beverages. A microsimulation model created country-representative synthetic populations. We determined simulated health outcomes in a baseline scenario using projections of BMI trends for each country up to 2050 and in two scenarios applying the interventions' respective BMI reduction to the BMI trends in the population. Simulated individuals developed diseases on the basis of their BMI and outputs provided the incidence and costs associated with the different scenarios.FindingsThe prevalence of obesity and incidence of BMI-related disease was forecast to increase in all countries to 2050. MCLI had the largest effect on type 2 diabetes, ranging from 147 cases per 100 000 avoided (SD 17) in Bulgaria to 1003 (21) in Portugal by 2050. The highest costs avoided were for stroke (up to €1·15 million, SD €0·008 million) in Portugal. Although tax on sugar-sweetened beverages also had the largest effect on diabetes (from 5 per 100 000 cases avoided [SD 8] in the Netherlands to 87 [13] in the UK), the costs avoided were lower than for MCLI.InterpretationThese results highlight the variation in impact depending on the type of intervention aimed at the prevention of chronic diseases. An integrated approach to reducing obesity is required, so that a battery of approaches, varying in uptake and targeted population group, could be useful to obtain a greater effect on BMI reduction.FundingEuropean Commission Health Programme and the Executive Agency for Health and Consumers (project number 2012 12 13).     

Impact of a levy on sales of sugar-sweetened beverages within a national chain of restaurants: interrupted time-series analysis

BackgroundConsumption of sugar-sweetened beverages (SSBs) is associated with type 2 diabetes, weight gain, and dental caries. Taxes and levies on these beverages have been proposed as a way to reduce their consumption. However, only limited evidence exists for the effectiveness of taxes and levies where implemented. We aimed to assess the impact of a levy of £0·10 per drink on sales of SSBs delivered nationally in the UK within a restaurant setting using electronic point of sale data.MethodsPoint of sale data for non-alcoholic drinks were obtained from all eligible restaurants (n=37) in one restaurant chain between June 23, 2014, and Feb 28, 2016. We conducted an interrupted time-series analysis using linear mixed-models with restaurant as a random effect to assess whether there was a step-change in sales per customer of levy-eligible SSBs at 12 weeks and 6 months after implementation. We assessed effects on non-levied drinks (other soft drinks, juices, bottled waters, and diet cola) to explore substitution effects. Analyses also adjusted for seasonality and clustering. We stratified analyses by sales area (London, the North of the UK, the South of the UK) to assess regional effects.FindingsOverall, compared with the preintervention period, sales per customer of levy-eligible SSBs declined by 11·7% (p=0·002) at 12 weeks and 9·8% (p=0·004) at 6 months. For non-levy drinks, sales per customer of fruit juice (main menu) increased at 12 weeks (+3·3%, p=0·051) and 6 months (+19·7%, p<0·0001). At 6 months, there was a decrease in sales per customer of diet cola (−7·6%, p=0·002) and bottled waters (−6·7%, p=0·009), whereas sales of other non-levied soft drinks increased (+14·6%, p<0·001). We found marked regional variations, with declines in sales of levy-eligible SSBs in London (−14·2%, p<0·001) and the South (−13·8%, p<0·001), but no change in the North.InterpretationTo our knowledge, this is the first study to assess the impact on sales of a levy on SSBs delivered at scale in a national chain of restaurants. We found evidence of substantial declines in sales of levy-eligible SSBs, with the greatest effects seen in London and the South. A levy on SSBs within restaurant settings has the potential to reduce sales and therefore consumption of these beverages.FundingNational Institute for Health Research.     

Impact of Maryland’s 2011 alcohol sales tax increase on alcoholic beverage sales

ABSTRACT

Background: Increasing alcohol taxes has proven effective in reducing alcohol consumption, but the effects of alcohol sales taxes on sales of specific alcoholic beverages have received little research attention. Data on sales are generally less subject to reporting biases than self-reported patterns of alcohol consumption. Objectives: We aimed to assess the effects of Maryland’s July 1, 2011 three percentage point increase in the alcohol sales tax (6–9%) on beverage-specific and total alcohol sales. Methods: Using county-level data on Maryland’s monthly alcohol sales in gallons for 2010–2012, by beverage type, multilevel mixed effects multiple linear regression models estimated the effects of the tax increase on alcohol sales. We controlled for seasonality, county characteristics, and national unemployment rates in the main analyses. Results: In the 18 months after the tax increase, average per capita sales of spirits were 5.1% lower (p < 0.001), beer sales were 3.2% lower (p < 0.001), and wine sales were 2.5% lower (p < 0.01) relative to what would have been expected from sales trends in the 18 months prior to the tax increase. Overall, the alcohol sales tax increase was associated with a 3.8% decline in total alcohol sold relative to what would have been expected based on sales in the prior 18 months (p < 0.001). Conclusion: The findings suggest that increased alcohol sales taxes may be as effective as excise taxes in reducing alcohol consumption and related problems. Sales taxes also have the added advantages of rising with inflation and taxing the highest priced beverages most heavily.     

Lessons from comparing narrative synthesis and meta-analysis in a systematic review

BackgroundNarrative synthesis approaches have been faulted for vote-counting (in which effectiveness is determined by tallying direction and statistical significance of study results) and lack of transparency. Yet narrative syntheses can also highlight trends and explanations for findings in a way that statistical meta-analysis cannot. Debate about the absence of meta-analysis in situations where it could have been done continues in the methodological literature. Using a systematic review of the effects of plain packaging of tobacco products, we compared narrative syntheses and new meta-analyses. We chose this review because its findings, which informed UK public health policy, have been misrepresented by the tobacco industry.MethodsWe revisited a published systematic review on plain packaging of tobacco products. We compared the findings from the narrative synthesis originally used in the review that accounted for study quality with the findings from a new multilevel meta-analysis that included all available effect sizes. We compared them in terms of their inclusiveness, approach to heterogeneity, and overall findings.FindingsAlthough 21 studies (n=27 166) were included in the narrative synthesis, we included seven studies in our new meta-analysis (56 effect sizes, n=5365). The narrative synthesis found that the 21 studies were “highly consistent” in the decreased attractiveness of plain packaging compared with branded packaging, primarily on the basis of direction and significance of effects as reported in the 21 studies. The pooled effect size for the seven studies in the meta-analysis demonstrated that plain packs were less attractive (Cohen's d=–0·59, 95% CI −0·71 to −0·47). Whereas the narrative synthesis highlighted key generalities across subgroups and study designs, the meta-analysis highlighted possible effect modification by tobacco brand.InterpretationThis study adds to existing debates by showing the contribution and implications of each synthesis method, especially with respect to opportunities for exploring heterogeneity. The findings have implications for the credibility of different approaches. Even when narrative synthesis is preferred, a meta-analysis and forest plot can help generate additional hypotheses, test and confirm hypotheses, and understand heterogeneity. In this particular case, use of multilevel meta-analysis helped to make clear aspects of statistical heterogeneity that narrative synthesis alone might not have detected.FundingMedical Research Council.     

Youth attitudes towards tobacco control: a preliminary assessment

The attitudes of Ontario youth toward the sale and price of cigarettes, making smoking against the law, and tobacco company truthfulness were assessed in 2001 and compared to adult attitudes in 2000 and youth attitudes in 2003. Youth were more supportive of restricting cigarette sales and raising prices than adults, and more likely to agree that the government should make smoking against the law, but they were less distrustful of tobacco companies. In 2003, youth were more supportive of sales restrictions and making smoking illegal, and more distrustful of tobacco companies, than in 2001. More comprehensive assessments and continued monitoring of youth attitudes are needed.     

The Demand for Tobacco Products in Finland

Four time-series models are introduced for analysing the demand for addictive substances. The models are applied to estimate the demand for tobacco products in Finland. Separate models were found appropriate for different products. The price of cigarettes is the most important single determinant of the demand for tobacco products. The demand for cigarettes is twice as sensitive to falling prices (elasticity – 0.94) than to rising prices (elasticity – 0.49). The demand for cigarettes is also responsive to changes in real income. The most important factor influencing the demand for pipe tobacco is the price of cigarettes. The demand for cigars is adequately explained by changes in its price. The 1977 Tobacco Act and the anti-smoking publicity preceding it appear to have reduced the cigarette demand permanently by 7%. The results imply that taxation would seem a powerful instrument for achieving the objectives of restricting consumption of tobacco products and raising government revenue. Yet, taxation together with extensive anti-smoking publicity would have a more advantageous effect than either of them used in isolation.     

Alcohol prices, beverage quality, and the demand for alcohol: quality substitutions and price elasticities

BACKGROUND: Although the published literature on alcohol beverage taxes, prices, sales, and related problems treats alcoholic beverages as a simple good, alcohol is a complex good composed of different beverage types (i.e., beer, wine, and spirits) and quality brands (e.g., high-, medium-, and low-quality beers). As a complex good, consumers may make substitutions between purchases of different beverage types and brands in response to price increases. For this reason, the availability of a broad range of beverage prices provides opportunities for consumers to mitigate the effects of average price increases through quality substitutions; a change in beverage choice in response to price increases to maintain consumption. METHODS: Using Swedish price and sales data provided by Systembolaget for the years 1984 through 1994, this study assessed the relationships between alcohol beverage prices, beverage quality, and alcohol sales. The study examined price effects on alcohol consumption using seemingly unrelated regression equations to model the impacts of price increases within 9 empirically defined quality classes across beverage types. The models enabled statistical assessments of both own-price and cross-price effects between types and classes. RESULTS: The results of these analyses showed that consumers respond to price increases by altering their total consumption and by varying their brand choices. Significant reductions in sales were observed in response to price increases, but these effects were mitigated by significant substitutions between quality classes. CONCLUSIONS: The findings suggest that the net impacts of purposeful price policy to reduce consumption will depend on how such policies affect the range of prices across beverage brands.