Tuesday 24 January 2017

Are e-cigarettes are creating a brand new generation of cigarette smokers?

The anti-vaping propaganda from California continues to pour forth like the sewage it is. This week the 'gateway' hypothesis was resurrected by Stanton Glantz of UCSF, leading to headlines like this:

E-cigarettes are creating a brand new generation of cigarette smokers, experts warn

The authors of the study were keen to comment on their 'research', as you can imagine...

Senior author Stanton A. Glantz at the University of California said: “E-cigarettes are encouraging – not discouraging – youth to smoke and to consume nicotine, and are expanding the tobacco market.”

Researcher Lauren Dutra added: “We didn’t find any evidence that e-cigarettes are causing youth smoking to decline.”

Central to their argument is the claim that 'adolescent' rates of cigarette smoking have not fallen faster since the emergence of vaping than they did before. They provide this graph as proof:


This graph supposedly shows the proportion of adolescents who have used cigarettes and/or e-cigarettes in the past 30 days. The blue line supposedly represents the pre-vaping trend extrapolated into the future and, according to Glantz and Dutra, demonstrates that:

'The advent of e-cigarettes did not affect declining trends in ... current (use in past 30 days; bottom panel) cigarette use'.

Glantz and Dutra do not define how old these 'adolescents' are, but they cite the National Youth Tobacco Survey as their source. The NYTS data can be downloaded here if you have Microsoft Access or can be viewed individually here: 2004, 2006-09, 2011-12, 2013 and 2014.

Glantz and Dutra base their pre-vaping trend on 2004 to 2009 which is problematic because almost nobody was vaping in 2009, let alone teenagers. Even in 2011, the past month vaping rate in high schools was only 1.5%. The big rise in e-cigarette use occurred between 2012 and 2014 when past month e-cigarette use rose from 2.8% to 13.4%.

That is a relatively minor issue, however. The bigger issue is that Glantz and Dutra's graph doesn't bear much resemblance to the NYTS data. They claim that 'Current smoking decreased from 15.8 per cent in 2004 to 6.4% in 2014' but the only reference Google can find to a smoking rate of 6.4% in 2014 comes from their own study. The NYTS shows a cigarette smoking rate of 9.2% among high school students and a rate of 2.5% among middle school students.

I can't see any way of crunching the numbers to arrive at the figures Glantz and Dutra use, so let's look at what the NYTS actually shows. If we look at the proportion of middle school students who have smoked a cigarette in the past 30 days (ie. current smokers), the rate falls at the kind of steady pace that might be expected if we assume a linear decline (which cannot, in fact, be assumed). The black line shows the 2004-09 trend extrapolated forward, as per Glantz and Dutra.

But if we look at high school students - who are far more likely to be smokers - there is an obvious acceleration in the decline after 2012:

This becomes even more obvious if we abandon the strange vertical axis that Glantz and Dutra use and show the data properly...

All the smoking rates since 2011 are lower than the pre-2009 trend would have predicted and the rate in 2014 is dramatically lower. If Glantz and Dutra 'didn’t find any evidence that e-cigarettes are causing youth smoking to decline', they can't have been looking very hard. Both the National Youth Tobacco Surveys and the Monitoring the Future surveys show an unusually large decline in smoking rates in recent years.

These surveys cannot prove causation but they provide pretty good prima facie evidence that e-cigarettes have indeed caused youth smoking to decline - and they certainly disprove Glantz and Dutra's claim that the 'decline in current smoking ... did not change after the introduction of e-cigarettes'.

If the decline in smoking had continued at the pre-2009 rate, the high school smoking rate would have been 14% in 2014. Instead it was 9.2% - a third lower than Glantz's little model would have predicted.  

'Creating a brand new generation of cigarette smokers' and 'expanding the tobacco market', my arse. Scoundrels.

No comments: