Sunday, 6 September 2009

Smoking in cars study is misleading

I don't think this will be the last we hear of this...

Secondhand smoke in cars worse than in bars

Jurisdictions around the globe have tried to squash secondhand smoke by banning smoking in public places. But only a few have tried to prevent people from lighting up in their cars -- typically only when children are present.

A new study from Johns Hopkins' school of public health takes on the question car smoking -- just how bad is it?

Pretty bad. The amount of secondhand smoke was significantly higher in cars than in bars and restaurants, the paper found.

This comes from the
Baltimore Sun. The study is unpublished (as is so often the case), but will appear in the October issue of Tobacco Control. That the results are already appearing in the mainstream press is a tribute to the anti-smoking lobby's well-drilled PR machine. Expect to see similar headlines around the world over the next few weeks.

The news angle is evident from the article quoted above - secondhand smoke levels are higher in cars than in bars. The political implications are obvious and the researchers make an open appeal to government in their paper:

These high levels of exposure to SHS [secondhand smoke] support the need for education measures and legislation that regulate smoking in motor vehicles when passengers, especially children, are present.

But the headline is extremely misleading. The study gives an average nicotine level in a car of 9.6µg/m3 (micrograms per cubic metre). There are numerous studies that have measured nicotine levels in bars, so we can see whether the car figure really is 'worse than bars'.

Mulcahy (2005) found a level of 35.5µg/m3 in Irish bars.

Nebot (2005) found levels between 19µg/m3 and 122µg/m3 in bars/discos.

Lopez (2008) measured levels of 32.99µg/m3 in discos/pubs.

All of these readings are far higher than the 9.6µg/m3 reported for cars in this new study. So what's going on? A clue is given in the Mulcahy study which reported a 35.5µg/m3 level before the Irish smoking ban, which fell to 5.95µg/m3 afterwards. 

And then the penny drops. The researchers are comparing the 'smoky' cars with smoke-free bars! It is quite unsurprising, then, that nicotine levels in premises where smoking is completely banned by law are lower than in a car where someone is smoking. Do we really need to pay scientists to tell us such things?

The real finding here is that nicotine levels in 'smoky' cars are much lower than in 'smoky' bars. But that wouldn't make for such a eye-catching headline, would it?


Dick Puddlecote said...

Great article, Chris. I'm sure it won't be the only one I ever pinch from you. ;-)

handymanphil said...

Quote:- "A clue is given in the Mulcahy study which reported a 35.5µg/m3 level before the Irish smoking ban, which fell to 5.95µg/m3 afterwards."
I may be a bit thick here but if these were 'smoke free' bars how was any level recorded at all?

Christopher Snowdon said...

Good question, Phil. The answer is that the levels are exceptionally low in both cases. We have modern science to thank for the fact that they can be measured at all. A microgram is one millionth of a gram. 5 or 9 micrograms per square metre is barely above the detectable level. Smoke-free premises have these tiny traces of nicotine from tiny wafts of smoke coming in from outside or from staff or customers who have traces of nicotine in their breath or clothes.

Belinda said...

That would be 'third hand smoke', then?

Michael J. McFadden said...

Belinda, I think it's more likely due to occasional illegal smoking upping the averages.

Chris is certainly right about the measuring instruments. If something is present at a level of 5 micrograms per cubic meter then that means that a standard sitting nonsmoker would inhale about 3 micrograms (resting respiratory rate averages about .6 m^3 per hour if I'm remembering correctly.) which is about 1/300th of a cigarette per hour.

Something I've noticed in reading the car-smoking research and reports about it is that a disproportionate amount of emphasis is always put on plume readings (which is when an actual little plume of smoke from the end of a cigarette brushes across the end of their naughty little "sniffer" machine for a second or so) and on readings with the windows either all rolled up or just cracked a bit - sometimes with the car being stationary. Of course the real measurement of interest would involve at least one and more likely two windows cracked down by at least several inches in moving traffic.

Another game that the Antis will play is comparing RSP 2.5s in smoking and nonsmoking vehicles. While cars produce a lot of pollution they tend NOT to produce a whole lot of RSP 2.5s, so the "pollution" supposedly measured from the cigarettes is really simply a measurement of smoke - obviously higher in a car with smoking than in a car without smoking.

With nicotine it's even more extreme since there's virtually no source of nicotine OTHER than smoking!

- Michael