to the presentation (Powerpoint or viewer needed)

Research and practitioner perspectives of tobacco control mass media campaigns in England Dr Tessa Langley UK Centre for Tobacco and Alcohol Studies, University of Nottingham Matthew Walmsley, Public Health England

Research perspectives of tobacco control mass media campaigns in England Dr Tessa Langley UK Centre for Tobacco and Alcohol Studies University of Nottingham

Outline • Rationale for MMC • Intervention evidence • Recent research on MMC in England • Ongoing/future research

The rationale for mass media campaigns • Well-defined behaviourally focussed messages • Potential for widespread and repeated exposure • Incidental exposure • Low cost per head

International evidence • Durkin et al. 2012 – review of cessation

campaigns in adults - Can promote quitting and reduce adult smoking - Small effect sizes but significant populationlevel effects - NHE messages perform best - Rapid decay of campaign effects - Average of 12 exposures per head per quarter needed to reduce adult smoking prevalence

Recent research on mass media campaigns in England: Context • Very little UK evidence • Freeze on public health campaigns April 2010 • Campaign re-introduced September 2011 • Smokefree as 1 of 4 social marketing programmes for public health • Budget: £16m 2012-13 (£38m 2009-10)

 2 year MRC funded project

MRC (NPRI) project Aim: To evaluate the impact of UK anti-tobacco mass media campaigns carried out since 2004 on a comprehensive set of key indicators of adult smoking behaviours Completed/ongoing studies • Characterisation of recent campaigns in England in terms of aims, informational and emotional content and style • Effects of mass media campaigns on population-level indicators of smoking and quitting behaviour: calls to the NHS smoking helpline, use of SSS, consumption, and prevalence • Impact of campaigns on smoking behaviour in the home • Cost-effectiveness of tobacco control mass media campaigns

Jargon Television ratings (TVRs): The % of a particular audience that has seen an advertisement e.g. 100 TVRs = each person has viewed ad once, or 50% have viewed twice e.g. 1200 TVRs = each person has viewed ad 12 times, 25% have seen ad 48 times. Gross rating points (GRPs): Sum of TVRs for individual adverts (but often used interchangeably) e.g. Sum of TVRs for all tobacco control adverts

Are MMC in England maximally effective? Evidence suggests that – 400 GRPs per month are needed to reduce smoking prevalence by 0.3% – Sustained behaviour change requires sustained campaign exposure due to the short-lived effects of campaigns – Adverts with high emotional content and testimonial adverts are most effective at increasing quit rates

 Study to characterise publically-funded tobacco control campaigns in England (2004-2010) - in line with recommendations?

Intensity 25

% of months

20

• 1 in 5 months had no MMC • 2 in 5 had 400+ GRPs

23 19

20

20

15

13

10 5

0

3 0

0-200

200-400

400-600

600-800

3

800-1000 1000-1200

Total GRPs Total TVRs Jan 04-Mar 10: 24507

Langley et al. Addiction 2013.

Informational content • From Apr 2008-Mar 2010 over 60% of adverts advertised the SSS • A quarter contained information about negative consequences

61.4

45.9

% of TVRs

40.1 26.3

Negative consequences of smoking Benefits of quitting using particular methods Jan 04-Mar 10 Langley et al. Addiction 2013.

Apr 08-Mar 10

Style Other (e.g. Graphical aid) 11%

Testimonial 17%

Acted 72%

% TVRs for each style Langley et al. Addiction 2013.

Characterisation of campaigns: Conclusions Based on existing recommendations – Only a small proportion of tobacco control advertisements utilised most effective strategies negative health effects messages and testimonials – Intensity of campaigns was lower than international recommendations

However, subsequent research has shown that MMC in England have been effective – both positive and negative campaigns

Evidence on campaigns in England Langley et al. Addiction 2014

• In 2010 government ceased spending on national public health mass media campaigns in England for 18 months • Interrupted time series analysis to quantify the impact of the campaign freeze on a range of measures of quitting behaviour • Quitline, quit support pack requests, Smokefree website hits, NHS Stop Smoking Services

Immediate drops in • Quitline calls – 65% • Literature requests – 98% • Web hits – 34%

No change in use of SSS (intensive support)

Evidence on campaigns in England cont. Sims et al. Addiction 2014 • Analysis of monthly cross-sectional surveys to estimate effect of campaigns 2002-2010 on prevalence and consumption • Adjusted for other tobacco control policies, cigarette costliness and individual characteristics • 400 point increase in tobacco control GRPs associated with a significant 1.80% reduction in average consumption in following month • Campaigns accounted 11.2% of the total decline in consumption over the period 2002-2009 • 400 point increase in GRPs was significantly associated with 3% lower odds of smoking two months later • Campaigns accounted for 13.5% of decline in prevalence over this period

Our findings suggest that overall, national tobacco control MMC influence smoking behaviour

But which type of tobacco control mass media campaign is most effective?

Recall of campaign types Richardson et al. BMC Public Health. 2014. • Data on recall of televised campaigns from ITC UK Survey, 2005-09 • Merged with GRP data – Campaigns categorised as “positive” or “negative” according to emotional content • Negative campaigns: For every additional 1,000 GRPs in the six months prior to survey, 41% increase in likelihood of recall (OR = 1.41, 95% CI: 1.24–1.61) • Positive campaigns: no significant effect

Quitline – by campaign type

Note: Plots generated using ggplot2 package

Preliminary results Prevalence and consumption • Increased exposure to both positive and negative campaigns associated with lower odds of smoking

• Increase in exposure to negative emotive campaigns associated with decrease in average cigarette consumption • No effect of positive campaigns on consumption

Preliminary results Quit attempts – Increased exposure to both positive and negative campaigns is associated with an increase in the odds of participants reporting having made a quit attempt within the last three months Smokefree homes – Aggregated ads have no effect on smokefree home prevalence – SHS ads increase odds of smokefree home

Are MMC cost-effective? • MMC are expensive – are costs justified by benefits? • Atusingwize et al. Systematic review of economic evaluations (All international studies, under review) found that – evidence on the cost-effectiveness of tobacco control mass media campaigns is limited (10 studies) – Methods are of acceptable quality, but studies highly heterogeneous – All suggest that TC MMC offer good value for money, compared with no campaign

Conclusions • Spending cuts have increased the need for high quality evidence • Increasing UK evidence base

• Evidence suggest TCC do influence quitting behaviour and reduce smoking – Both positive and negative emotive campaigns

• More research on what works and optimal level of exposure is warranted

Ongoing & future research • Scottish study (Stirling) • Co-ordination re. evaluation of MMC with PHE

Acknowledgements Funders National Prevention Research Initiative Action on Smoking and Health

Colleagues Nottingham: Sarah Lewis, Lisa Szatkowski Bath: Michelle Sims, Anna Gilmore, Ruth Salway KCL: Ann McNeill