Helping patients in hospital to quit smoking

This issue reports a study that nicely encapsulates the problems facing hospitals who want to encourage and help their patients to stop smoking (p 87).¹ The study focuses on inpatients with cardiac disorders who want to stop smoking and examines whether a brief intervention (averaging 34 minutes) delivered on the ward by nurses would help them stop. It finds no effect at six weeks and at follow up after one year. This mirrors results with other patient groups. Most notably two large randomised controlled trials, one in the United Kingdom and one in Denmark, have found that a brief intervention by midwives failed to improve pregnant smokers' chances of stopping.²,³ All three studies also found that busy staff had considerable difficulty finding time to undertake the counselling. How can one square this with claims that brief advice from healthcare professionals leads patients to stop smoking and that behavioural support increases the chances of success of attempts to quit?⁴

Although it seems confusing, in fact there is a pattern. As regards brief advice from healthcare professionals, the key point is that this advice is delivered opportunistically to all smokers, takes just a few minutes and is aimed not at helping smokers to stop but at triggering an attempt to quit.⁴ What happens in the attempt will then depend on the smokers and the treatment they subsequently get. It is apparent from studies such as that of Hajek et al¹ that the treatment package needs to be commensurate with the severity of the dependence it seeks to overcome.

As regards that treatment package, a Cochrane review clearly shows that behavioural support and counselling increases a smoker's chances of achieving lasting abstinence.⁵ However, close analysis of the evidence indicates that the behavioural support must be reasonably extensive and delivered by someone employed and trained for the task rather than someone trying to fit it into other duties. For example, there are four trials in the Cochrane review involving hospital patients receiving counselling totalling 60 minutes or more with follow up for at least six months and biochemical confirmation of abstinence. These studies yield an odds ratio of 1.97 (95% confidence interval 1.33 to 2.90, n=944) in favour of the intervention. There are two studies in this population with similarly rigorous methods involving counselling of around 30 minutes or less. For these the intervention was not effective (odds ratio 1.23, 0.76 to 2.13, n=931). Analysis of studies contained in two more recent reviews yield similar conclusions.⁶,⁷

Similarly, a close inspection of the Cochrane systematic review of controlled trials on counselling pregnant smokers shows that where counselling was delivered by staff employed for the purpose, preterm abstinence rates were considerably higher in the intervention groups (odds ratio 2.04, 1.69 to 2.47, n=3775).⁸ In contrast, studies in the Cochrane review plus the two more recent randomised trials show no effect of counselling delivered by midwives fitting it into their other duties (odds ratio 1.22, 0.89 to 1.66, n=2118).

Hajek et al's treatment package did not incorporate any pharmacotherapy (nicotine replacement or bupropion), presumably because of concerns over safety in this group.¹ However, McRobbie and others have recently reported the results of a randomised trial which finds that bupropion improved abstinence rates over six months in people with cardiovascular disease (H McRobbie et al, 3rd European society for research in nicotine and tobacco conference, Paris, 2001). Nicotine replacement has been evaluated in such patients as well and appears to be well tolerated, although efficacy has not yet been clearly established.⁹

Optimal treatment for the life threatening disorder known as tobacco dependence is now accepted to consist of structured behavioural support involving multiple sessions combined with nicotine replacement or bupropion.⁴ Therefore it would no longer seem acceptable for hospitals to attempt to meet the smokers' needs with brief interventions undertaken by staff who have only scant training and expertise and many other calls on their time.

Fortunately, at present the NHS has a specialist smoking service organised at the health authority level, and in theory all that hospital staff need do is to refer their patients to that service. In practice, hospitals will need to work closely with the established smoking cessation services in their locality to set up comprehensive referral systems. Unfortunately, the future of this fledgling service is uncertain because the UK government has so far failed to make a decision on funding the service for the forthcoming year.¹⁰,¹¹

Papers p 87