The Office of National Statistics has just published its annual survey of smoking across the UK, suggesting a substantial drop in smoking in Scotland and in the other nations. This contrasts with a largely static estimate from the Scottish Health Survey. So what is going on?
It is generally accepted that smoking is the biggest cause of ill health in Scotland and that the costs of smoking create a huge drag on many of the most vulnerable household budgets, but also that smoking is in ongoing decline and not the force it once was in the land.
Yet we remain uncertain about exactly how many people in Scotland smoke.
The official figure, as used by the Scottish Government, comes from the annual Scottish Health Survey, from which the latest suggestion was that 21% of adults smoke and that this rate has not gone down over the last few years. Does this mean that we are failing, at a time when other parts of the UK are reporting an ongoing decline?
Well, no, because the data showing smoking reducing elsewhere comes from the UK-wide survey from the Office of National Statistics. This survey also suggests that the smoking rate in Scotland is declining steadily and has increasingly diverged from the Health Survey results in recent years. So which of these stories are we to believe?
The only clue we have as to why these surveys might give different results comes from the different ways they process the raw survey data. We know that the Scottish Health Survey makes a greater correction to allow for the fact that people in more disadvantaged groups are more likely to smoke but less likely to take part in government surveys. This would explain a higher estimate for the smoking rate, but not the divergence between the two surveys we have seen in recent years.
Otherwise we need to look for different factors operating in Scotland and in our neighbours (and it’s hard to think of anything that could cause such a striking divergence over the last few years) or fall back on the normal statistical variations you get in any such survey samples.
Fortunately help is at hand, in the form of two other Scottish Government surveys, the Scottish Household Survey and the Scottish Crime and Justice Survey. Together these three surveys deliver the Scottish Survey Core Questions and, crucially, they all ask about smoking status and so provide a substantially larger sample than just the Scottish Health Survey. So plotting the Scottish Survey Core Questions result should give a more reliable estimate of what surveys in Scotland are suggesting for the smoking rate. And here it is….
Taking the combined Scottish Survey Core Questions result removes the worrying divergence and suggests an ongoing regular decline in the smoking rate, at a similar rate to that in the ONS survey but with the slightly higher estimate that could be expected from the enhanced inequalities correction.
Which is good enough to convince me that while the exact smoking rate in Scotland is difficult to determine, because of the challenge of correcting a survey sample to reflect the whole population, the overall smoking rate is continuing to decline.
That leaves us with the real question of whether the rate is declining fast enough to achieve the Scottish Government target of a 5% smoking rate by 2034 (the “Tobacco-free Generation”).
And the answer to that? Not quite fast enough.