Small graph of survival curves from Doll et al's 1994 BMJ publication on the Doctor's study

Is cutting down a ‘waste of time’? 3

Newspapers today are reporting on a new study from Scottish researchers on smoking reduction. Today’s Sun says ‘Cutting down on cigs is a waste of time’. But only last month headlines said ‘Smokers urged to ‘cut down’ rather than quit’ (from The Telegraph). What’s going on?

First, let’s take a look at what the new Scottish study from Carole Hart and colleagues found. They examined two separate groups (cohorts) of men and women from central Scotland, just over 5,000 smokers in total, who were middle-aged in the 1970s. These cohort members were questioned twice, once in the (roughly) early-to-mid 1970s, and once again in the mid-to-late 1970s. At these screenings, information on their smoking status, and a wide range of other health data was gathered.

These people were then followed-up till 2010, and their dates and causes of death were obtained. The resulting mortality rates (basically, ‘death rates’) were compared, adjusting for other potentially relevant factors. They found that, while study participants who reported stopping smoking completely at their second screening in the late 70s had much better mortality outcomes than those who continued smoking, no such consistent pattern was seen for ‘reducers’.

In one sub-group (heavier smokers who reported reduction) in the smaller of the two cohorts, the researchers did find quite large improvements in mortality outcomes, but this wasn’t seen consistently in rest of the study data. The researchers contrast their findings with a similar recent study from Israel that found more consistent evidence of improved mortality outcomes in reducers.

Maybe this all seems a bit strange – it has been established for a long time that heavier smoking results in worse health outcomes than light smoking, e.g. this graph from the well-known British Doctors Study:

Graph showing survival curve with varying cigarette consumption from 1994 Doll & Hill study.

From Richard Doll et al. (1994), courtesy of the BMJ.

This is something that we continue to see in large, well-designed modern studies. So if smoking more seems to be consistently worse, why doesn’t smoking less seem to be consistently better?

To answer this we need to consider two things going on independently: one is how much reduction in number of cigarettes smoked actually benefits health (and what factors might work against it); and the second is how well we are able to detect improvement in health through smoking reduction – even if it were to exist – in these kind of studies.

For the latter point, these studies – as with any research – do have limitations. Smoking behaviour changes over time, and only measuring smoking reduction between two points in the 1970s does not allow very accurate assessment of how many classed here as ‘reducers’ actually maintained their reduction (or how many ‘maintainers’ went on to subsequently reduce their consumption, increase it, or to quit completely).

These changes could have the effect of underestimating – or maybe overestimating (because what exactly could be going on here is quite difficult to predict) any benefits of reduction. Research that better records changes in cigarette smoking at more frequent points in time would be useful to improve our knowledge here.

There are also reasons to believe that some kinds of reduction in the number of cigarettes smoked do not translate directly into the reductions in death and disease we might expect – chief among these is probably compensatory smoking (pdf). Compensatory smoking reflects the tendency among smokers to adjust their style of cigarette consumption (by changing things like the frequency and intensity of puffs) to achieve their preferred dosage of nicotine. So if the number of cigarettes smoked changes, but a smoker still wants to get the same level of nicotine, there is a tendency to change smoking style in a way that maintains their nicotine levels but that also might not decrease their exposure to harmful cigarette smoke as much as expected. (Nicotine itself, in the doses smokers are exposed to, is thought to be relatively safe – it is the other by-products of burning tobacco that principally cause the harm.)

What does all this tell us? In theory it would be logical to expect some reduction in disease to accompany reduced cigarette consumption. In practice, this has not been the consistent finding of studies, likely in part due to methodological difficulties and in part due to modified individual smoking behaviour such as compensatory smoking. This does not mean reduction never has a health benefit for any given individual (it would be hard to imagine a situation where a reduction from 40 cigarettes per day to 2 per day had completely zero health benefits), rather that the benefits of reduction, in a very general sense, on death rates across the whole population are not very clear.

Fortunately, there are other benefits of reduction. Recent guidance by the English National Institute for Health and Care Excellence (NICE) on tobacco harm reduction describes that those who reduce are more likely to make quit attempts, and are more likely to quit smoked tobacco completely compared to those who do not reduce. NICE also notes that evidence shows reducers who use ‘clean’ sources of nicotine – like medicinal nicotine replacement therapy – are more likely to reduce, and also more likely to quit. It is important to note that today’s study measured smoking reduction in an era before clean nicotine sources became widely available.

Clean nicotine sources might also have a role in reducing compensatory smoking in smoking reduction, as providing controlled dosage of nicotine in a less harmful form could reduce the issue of smokers adjusting their cigarette smoking habits to compensate for smoking a lower number of cigarettes. More research is needed on this, however a review of studies found that smokers using controlled-dosage nicotine products like nicotine gum and inhalators, at the same time as they continued to smoke cigarettes, maintained similar nicotine levels, but reduced their exhaled carbon monoxide (a measure of exposure to tobacco smoke).

So, in terms of health, reduction is neither a complete ‘waste of time’ as today’s Sun suggests nor should it be universally recommended over quitting completely, as implied by the Telegraph headline of last month. As the NICE guidance makes clear, whether reduction could be beneficial for any one person depends on a range of factors, including just how willing or able the individual is to give up smoking entirely – issues we can return to in a blog at a later date.



  1. Sean Semple (Mon 08/07/2013) wrote:
    This is an excellent and very clear analysis of the conflicting media reports and the original scientific study that generated these headlines. It might also be worth noting that ‘reducers’ may also experience lower risk of ill-health (morbidity) and so have greater ‘healthy life years’ when compared to non-reducers. There are also likely to experience other indirect benefits that impact on the quality of their life. From a purely practical perspective, ‘reducers’ will have more disposable income and there is a fairly strong association between disposable income and health and well-being.

  2. Interesting to note that this article promotes licensed NRTs despite their utterly abysmal 13 month quit rate, and completely ignores the enfant terrible of the electronic cigarette. Why ignore a very real solution to this (highly profitable) problem?

  3. Hi Joanne – this blog was looking at one particular aspect of reducing harm from smoking (cigarette reduction), and as such deliberately didn’t really go into issues surrounding e-cigarettes. (The vast majority of the studies underlying the NICE recommendations are based on traditional medicinal nicotine products – though of course, I’d expect the last few paragraphs to apply to any product that delivered nicotine effectively and was acceptably safe).

    Re: ‘abysmal’ quit rates for traditional approaches, this is discussed a little bit in another entry in this blog, if you are interested in that.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s