In this guest post, Rob Pryce, a Research Associate at the Sheffield Alcohol Research Group based in the University of Sheffield’s School of Health and Related Research (ScHARR), looks at recent research on the effect of the smoking ban on alcohol consumption.
The smoking ban came in to force in Scotland in March 2006, followed by Wales (April 2007), Northern Ireland (April 2007) and then England (July 2007). It prohibited smoking in public places including workplaces, pubs, bars and restaurants. Smoking bans are now commonplace throughout the world, and research has examined their effect on smoking and associated health. However, the smoking ban may also have had unintended consequences on other health behaviours, including alcohol consumption.
My recent research looks at how the introduction of the smoking ban changed alcohol expenditure, with the expectation that it would have a different effect both for on-premise (pubs, bars and restaurants) and off-premise (shops), and for smokers and non-smokers.
Existing literature suggested that smokers might reduce expenditure in the on-premise sector, perhaps because they can drink and smoke at the same time if they do so at home instead of in a pub. There is also the potential that non-smokers may visit on-premises more frequently, and thus increase expenditure, because the venues become smoke-free and thus more enjoyable.
The research used two methods: a simple approach which used all of the data from 2001 to 2014, and a reduced dataset which just used data until England’s smoking ban was introduced. The latter allows a control group to be identified, to suggest what would have happened had a smoking ban not been introduced. The dataset used (the Living Costs and Food Survey) is useful for this research because it records alcohol expenditure split by on-premise and off-premise, as well as whether the household purchased tobacco in the two week expenditure diary.
The results show that smoking households reduce their alcohol expenditure in the on-premise by around 15%, and do not increase off-premise expenditure. Meanwhile, non-smoking households do not significantly increase their spending on alcohol, either in the on-premise or off-premise. This means that on-premise venues will probably have seen a fall in revenues, and it is highly likely that the reductions in revenue are concentrated in certain venue types and geographical areas. It also means that alcohol consumption overall was affected by the smoking ban.
The main policy messages is that the smoking ban may have had unintended consequences on on-premise venues as smokers reduced the amount they spent on drinking out. Future work could look at the differential impact of the smoking ban on expenditure in different venues or settings. There may also be underlying differences in the context of the drinking occasions arising from the introduction of a smoking ban which could be examined with richer data.
The findings also show that tobacco policy can have an effect on alcohol consumption and that the two behaviours should not be viewed in isolation. The smoking ban likely reduced alcohol consumption in smokers as they spent less in the on-trade and there was no change in off-premise expenditure.
This means that the ban has secondary outcomes on alcohol consumption amongst a group where health risks such as cancers of the head and neck might be particularly high due to the dual consumption of alcohol and tobacco. The work also highlights the potential secondary effects of minimum unit pricing, which might lead to a change in smoking behaviour even though not the aim of the policy, where rates of cancer may fall by even more if people reduce their smoking following an increase in the price of alcohol.
Future work should bear in mind that there may be spillover effects with tobacco and alcohol, and as such headline tobacco and alcohol policies have a wide range of potential unintended consequences which may have positive implications for public health.
Recent Comments