The effectiveness of mass media campaigns in reducing smoking, second-hand smoke exposure and smoking-related disease in England & Wales

Lead Research Organisation: University of Nottingham
Department Name: School of Medicine

Abstract

Tobacco control mass media campaigns involve communication which aims to encourage smokers to stop smoking or avoid smoking in front of others, particularly children. The campaigns may be communicated via television, radio, newspapers and/or billboards. It is likely that they contribute to changes in smokers' behaviour and reductions in smoking, and this in turn helps to reduce the number of people suffering the diseases caused by smoking. There is almost no evidence on how effective mass media campaigns in the UK have been, nor on whether this has been money well spent. The lack of UK evidence enabled the government to stop funding mass media campaigns in April 2010. Very recently, however, the government indicated that it might begin such campaigns again. Research on this topic is therefore very timely and would help ascertain how best to do this.
This project will measure the effectiveness of individual mass media campaigns, and of different types of campaign, in England and Wales over recent years. We will initially identify the characteristics of individual campaigns, such as whether the campaign aimed to encourage smokers to stop or to protect their children from smoke, its emotional and informational content and style, and how it was delivered (eg whether via television, radio or print media and the balance between the three). Then we will consider how much impact each campaign and campaign type has had by looking at a range of indicators of smoking behaviour in the adult population, such as the proportion of people who smoke, the numbers of smokers trying to stop and the proportion who smoke in the home. We will also measure whether the impact of mass media campaigns reduce the health problems caused by smoking. We will do this by measuring, for example, the number of people admitted to hospital with heart attacks or strokes, and the number of children consulting the GP for respiratory infections.
We will use statistical methods that enable us to examine how any changes in these indicators relate to the timing of mass media campaigns. These methods will also allow us to examine whether it is the mass media campaign that is important, or other factors, such as changes in other tobacco policies at the same time, that are responsible. We will look at these changes in smoking behaviour and health in datasets representative of England and Wales as a whole and where available, in data from specific sectors of the population, such as the more socio-economically deprived, to see whether the impact of mass media campaigns is the same in these groups. We will look at the effects of specific regional campaigns by comparing changes in smoking behaviour between regions. We will also look at what changes occurred in smokers behaviour and health at the point when government spend on advertising ceased prior to the election in 2010.
This work will provide essential information to guide future decisions on anti-tobacco mass media spend at national and regional levels.

Technical Summary

This research will evaluate the effectiveness and cost-effectiveness of anti-tobacco mass media campaigns in England and Wales since 2004 using key indicators of smoking behaviour and smoking-related health outcomes in the general population, and in specific population groups. We will use a combination of time series analysis and longitudinal panel study analysis, using data from national datasets.
Our outcomes will include smoking prevalence, rates of smoking cessation medication prescriptions, and GP consultations for children's respiratory disease (The Health Improvement Network primary care data), NHS quitline calls, NHS stop smoking service (SSS) attendees and 4 week quitters, NRT sales, Health Survey for England children's cotinine measurements, and Hospital Episode Statistics on acute MIs.
Autoregressive time series models will be used to estimate population-level change in smoking behaviours and health outcomes in response to specific campaigns, or month-to-month change in advertising reach and spend, specifying the appropriate autoregressive process and transfer function, and modelling underlying trends, seasonality, and impacts of other tobacco control policies. Longitudinal analysis will use International Tobacco Control (ITC) UK data of over 2000 adult smokers and recent ex-smokers to assess individual changes in smoking behaviour and attitudes pre and post campaigns using random effects regression models.
The economic model will link mass-media campaigns (via expenditure and coverage) to intermediate stop-smoking outcomes (calls to quit helplines, SSS attendance), the impact of those stop-smoking outcomes on intermediate health outcomes (AMI and respiratory disease) and the costs avoided following reductions in those health outcomes.
The study will provide evidence as to whether government spending on mass media campaigns is effective and cost-effective, and for prioritising future funding to maximise its impact and reach its target audience.

Planned Impact

The research will be of immediate benefit to policy makers, at local, national and international levels. Our results will provide an essential UK-specific evidence base on the effectiveness and cost-effectiveness of recent national and regional campaigns on a comprehensive set of short and long term smoking behaviour and health outcomes. Policy makers at local and national levels will gain information to evaluate the impact of previous spending on anti-tobacco campaigns, and to help to prioritise future spending in terms of medium (television, radio etc) and type of campaign to optimise efficiency and effectiveness. The findings will be used by organisations such as ASH and UKCTCS to engage policy makers, nationally and internationally, with evidence for the impact of campaigns, and if appropriate of the potential public health consequences of decisions made on mass media spending, including the recent halt in spending in England & Wales. It is expected that the results will therefore have a considerable influence on future decisions regarding anti-smoking mass media spend in this country and elsewhere. We have developed this work in collaboration with ASH, who are actively involved in tobacco control policy development in the UK and are ideally placed to ensure that the results of this work are taken into consideration in future policy development.
The research will also be of immediate benefit to several national charities, such as Cancer Research UK & the British Heart Foundation, which have also historically funded anti-tobacco mass media campaigns, by demonstrating whether such campaigns are effective in reducing the heath consequences of active and passive cigarette smoking and, if so, how their impact can be maximised.
The proposed project also has the potential to be of benefit to the commercial sector, notably advertising and pharmaceutical companies. Although we will not directly examine the effectiveness of NRT advertising in this project (as that is the one of the focuses of our funded TAG work), we will be examining, for example, factors that influence NRT sales, which will be valuable to pharmaceutical companies. Cuts in government funding may place greater reliance on pharmaceutical company funded campaigns in the future, and this study which looks at the important characteristics of government funded campaigns is likely also to provide valuable information for the content and style of pharmaceutical company advertising.
Commercial, public and third sector bodies involved in health behaviour change more generally may also benefit from this research.
Finally, and most importantly, the study will be of benefit to smokers, and those exposed to second-hand cigarette smoke. In the UK the prevalence of smoking in 2007 was about 21%, equivalent to 10 million people. One in two of these smokers will die prematurely as a result of smoking - half of these in middle age. Smoking cessation reduces the risk of several cancers, as well as other smoking-related diseases such as chronic obstructive pulmonary disease and coronary heart disease, with some of these health benefits happening quite quickly, others taking a number of years. Increasing rates of smoking cessation is therefore crucial to reducing morbidity and premature death. Moreover, second-hand smoke exposure has been estimated to be responsible for more than 10,000 premature deaths per year in the UK. Smoking also has significant cost implications for society; the cost to the NHS of treating diseases caused by smoking is approximately £2.7 billion a year. Smoking also results in lost productivity caused by smoking breaks and increased absenteeism amongst smokers due to ill-health. By establishing the most effective use of mass media to increase smoking cessation and reduce second-hand-smoke exposure, this research has huge potential to contribute to the nation's health and wealth.

Publications

10 25 50
 
Description ASH UKCTAS mass media briefing
Geographic Reach National 
Policy Influence Type Implementation circular/rapid advice/letter to e.g. Ministry of Health
 
Description delivery of training on mass media impacts to researchers and policy makers
Geographic Reach National 
Policy Influence Type Influenced training of practitioners or researchers
Impact better training for tobacco control and other public health professionals attending UKCTAS training to ensure evidence based practice
URL http://www.ukctas.ac.uk/ukctas/what-we-do/education-and-training/masters.aspx
 
Description policy briefing with marketing lead for Public Health England
Geographic Reach National 
Policy Influence Type Implementation circular/rapid advice/letter to e.g. Ministry of Health
Impact Our findings are in line with the PHE approach (i.e. a mixture of hard hitting health warnings and positive messaging) and PHE should and are continuing with an approach that corresponds with our findings to maximise effectiveness of subsequent campaigns.
 
Description Mass media for public health messages
Amount £207,068 (GBP)
Funding ID 13/163/17 
Organisation National Institute for Health Research 
Sector Public
Country United Kingdom
Start 10/2015 
End 03/2017
 
Description Scottish Government Chief Scientist Office
Amount £216,452 (GBP)
Funding ID CZH/4/978 
Organisation Chief Scientist Office 
Sector Public
Country United Kingdom
Start 11/2013 
End 10/2015
 
Title econometric time series methods applied in public health 
Description we have applied and described econometric time series models in a relatively novel context, as they have been little used in public health. 
Type Of Material Data analysis technique 
Provided To Others? No  
Impact time series models provide a stronger evidence base for the effect of associations with time related exposures than the more usual pre and post evaluations, and have enabled appropriate modelling of confounders, and non-linear associations. 
 
Description Collaboration between University of Nottingham and ASH 
Organisation Action on Smoking and Health (ASH)
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution We used our Nottingham Tobacco Control Database to carry out the preliminary descriptive analysis of mass media campaigns (measured in terms of Television Rating Points) and markers of smoking cessation activity. We have also provided help with interpretation of statistical and economic tobacco control research findings to support ASH in the targetting of its campaigns.
Collaborator Contribution ASH are a campaigning public health charity working to reduce the harm caused by smoking. ASH has provided initial funding to extract some of the tobacco campaign data and to enable us to use our Nottingham Tobacco Control Database (developed as part of our NPRI funded work) to do some preliminary descriptive analysis of the association between anti-smoking mass media campaigns and smoking cessation activity. ASH presented some of this preliminary work in the All Party Parliamentary Group on Smoking and Health "inquiry into the effectiveness and cost-effectiveness of tobacco control" paper submitted to the Spending Review in 2010 as evidence for the likely effectiveness of mass media campaigns, thus using our work to attempt to drive national tobacco control policy and spend.
Impact Some of the preliminary work on the effectivness of anti-smoking mass media campaigns was included in the All Party Parliamentary Group on Smoking and Health "Inquiry into the effectiveness and cost-effectiveness of tobacco control" paper submitted to the Spending Review 2010. http://www.ash.org.uk/files/documents/ASH_743.pdf. ASH was a collaborator on two subsequent applications for funding to extend this work; a successful application to CRUK and a further application for NPRI funding. This is a multi-disciplinary collaboration involving academic researchers in tobacco control and epidemiology and a campaigning public health charity.
Start Year 2010
 
Description Collaboration between University of Nottingham and UCL 
Organisation University College London
Department UCL Cancer Institute
Country United Kingdom 
Sector Academic/University 
PI Contribution Data from the Smoking Toolkit has been included in the Nottingham Tobacco Control database, and we are collaborating in work arising from the Smoking Toolkit. We have contributed to debate with RW on the relative strengths and weakness of the various datasets available for monitoring smoking behaviour and the most appropriate methods for analysis. We collaborated with RW in our preliminary work on the impact of anti-smoking mass media campaigns, and are continuing to use toolkit data to explore impacts of mass media campaigns on quit attempts.
Collaborator Contribution Access to, and on-going consultation on, the Smoking Toolkit, a monthly series of panel studies of smokers in England (from Professor Robert West, UCL).
Impact The Smoking Toolkit has more detailed information on individuals' smoking behaviour and reasons for their behaviour than is available from other data sources, and is therefore a valuable tool in our evaluation of tobacco policy and included in our developed Nottingham Tobacco Control database. This collaboration is multi-disciplinary involving expertise in statistics, epidemiology and health policy in Nottingham and health behaviour at UCL.
Start Year 2008
 
Description Evaluation of the impact of tobacco control mass media campaigns on quitting behaviour, smoking prevalence and smoking-related health outcomes. 
Organisation University of Stirling
Country United Kingdom 
Sector Academic/University 
PI Contribution I am a co-investigator on this project, which is investigating the impact of tobacco control mass media campaigns in Scotland. I was able to provide expertise on the evaluation of tobacco control media campaigns based on my previous experience in this field.
Collaborator Contribution The University of Stirling was the lead applicant.
Impact Peer-reviewed publication: Haghpanahan et al. The Impact of TV Mass Media Campaigns on Calls to a National Quitline and the Use of Prescribed Nicotine Replacement Therapy: A Structural Vector Autoregression Analysis. Addiction. 2017.
Start Year 2013
 
Description University of Stirling, HS research 
Organisation University of Stirling
Department Complex Interventions in Public Health
Country United Kingdom 
Sector Academic/University 
PI Contribution We worked with this group to develop and successfully attain funding from the Scottish government (Chief Scientist Office) for a parallel study to characterise and evaluate mass media campaigns in Scotland.
Collaborator Contribution The funded work in Scotland will enable us to draw direct comparisons of the effectiveness of mass media campaigns in England and Scotland.
Impact Successful funding bid to the Scottish Government (CSO funded grant entitled "Evaluation of the impact of tobacco control mass media campaigns on quitting behaviour, smoking prevalence and smoking-related health outcomes".) This collaboration involves public health academics, statisticians and econometricians (Nottingham and Stirling)
Start Year 2013
 
Description institute for social marketing, university of stirling 
Organisation University of Stirling
Country United Kingdom 
Sector Academic/University 
PI Contribution we were joint applicants on a funded NIHR project to carry out a review of reviews of impact of mass media in changing health behaviours and we are currently carrying out this work jointly
Collaborator Contribution these partners included the PI for this project
Impact no outputs yet, work in progress. yes it is mult-disciplinary involving statisticians, behavioural scientists, systematic review teams, social marketing experts etc . This work is intended eventually to provide guidelines for use of mass media to change behaviour.
Start Year 2015
 
Description imperial MSc teaching 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact we have delivered teaching on tobacco control mass media campaigns on the MSc in Cardiology (Smoking cessation module) at Imperial College, London, each year.
Year(s) Of Engagement Activity 2014,2015,2016
 
Description policy briefings 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact policy briefings summarising our findings, their interpretation and implications for future campaigns have been discussed on 3 occasions with the lead for tobacco control media campaings in Public Health England, resulting in joint understanding of the effectiveness of previous campaigns, discussions on the content of future campaigns, and in particular, discussions on how campaigns might be structured to better enable their evaluation.

discussions on how future campaigns might be timed and varied to better enable their evaluation are continuing.
Year(s) Of Engagement Activity 2013,2014
 
Description smokers panel 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Public/other audiences
Results and Impact The research plans and progress were shared with panels of smokers at the start (the UK Centre for Tobacco Control Studies smokers panel in Bath) and near the end of the project (the UK Centre for Tobacco and Alcohol Studies smokers panel in Nottingham) to guage the interest and insights from smokers on the direction of this research and our findings. This activity helped particularly in the interpretation of our findings with respect to the different impacts of different campaigns, in demonstrating the importance of this research to smokers, smokers being keen for campaigns to protect children from second hand smoke and to better inform them about the support available for them, and in raising some interesting new avenues for future campaigns. A subset of the Smokers Panel also took part in the research in the intervening period, giving a smokers perspective on the emotions evoked by different campaigns for comparison with those of researchers.

the presentation in 2012 generated a lot of interest and discussion and led to the further presentation made to the new UK Centre for Tobacco and Alcohol Studies in 2014. The emphasis from smokers, and from our findings, of the need for both positive and negative campaigns have been shared with Public Health England.
Year(s) Of Engagement Activity 2012,2014
URL http://www.ukctas.ac.uk/ukctas/events/event-items/ukctas-smokers-panel.aspx
 
Description symposium on mass media campaigns (UKNSCC) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Health professionals
Results and Impact shared platform with marketing policy makers from Public Health England, improving understanding of research and practitioner perspectives, sparking communication about working better together to design and evaluate effective mass media campaigns

further discussions which are continuing with Public Health England on how future campaigns could be designed to aid their evaluation and thus improve understanding on their effectivneness.
Year(s) Of Engagement Activity 2014
URL http://www.uknscc.org/uknscc2014_presentation_306.php
 
Description webpage 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact We havd created a webpage describing the study aims, objectives and results (including links to published papers), linked to the UK Centre for Tobacco and Alcohol Studies website. We will continue to update this site to provide access to our findings and their policy interpretations to enable policy makers and the public to see the impacts of current mass media spend.

providing this information by this, and other means, has enabled discussions with policy makers regarding future campaigns, and the possibility of designing campaigns to aid future evaluation.
Year(s) Of Engagement Activity 2013,2014
URL http://www.ukctas.ac.uk/ukctas/what-we-do/featuredprojects/mass-media.aspx