Intended for healthcare professionals


Helping pregnant smokers to quit

BMJ 2014; 348 doi: (Published 11 March 2014) Cite this as: BMJ 2014;348:g1808
  1. Leonie S Brose, lecturer in addictions
  1. 1UK Centre for Tobacco and Alcohol Studies, Institute of Psychiatry, King’s College London
  1. leonie.brose{at}

Behavioural support works, evidence on drug treatment still patchy

Although most smokers manage to quit during pregnancy, a proportion does not. In England, 26% of women smoke in the year before their pregnancy and 12% smoke through to delivery.1 The rate is similar in other high income countries, whereas in low and middle income countries, smoking rates are more variable and seem to be increasing among young women.2 In addition to the countless negative consequences for the smoker’s own mental and physical health, smoking in pregnancy is linked to a wide range of poor health outcomes for the child.3 Thus there is an urgent need to help pregnant smokers who find it difficult to quit.

Nicotine replacement therapy (NRT) is clearly effective in smokers in general,4 but a recent review of randomised controlled trials found no good evidence in pregnant women.5 In the linked paper, Berlin and colleagues (doi:10.1136/bmj.g1622) report a welcome randomised placebo controlled trial that found no benefit of transdermal nicotine patches in pregnant smokers.6 Recruitment seems to have been slow, but with over 400 participants this is the second largest trial of NRT in pregnancy to date. Compared with previous trials, Berlin and colleagues’ study had several advantages. In an attempt to match participants’ nicotine intake to that achieved by smoking, a substantial proportion of the participants received higher doses of NRT than in previous trials. Adherence to treatment was high; an area that has been problematic for studies of NRT in pregnancy. Additionally, the patches were administered from quit date to delivery, providing a longer duration of treatment than in previous trials, and participants were advised to continue treatment even if they restarted smoking. Despite this careful design, the patches did not seem to increase smoking abstinence from quit date to delivery, an outcome stringently assessed by self reported abstinence confirmed by levels of carbon monoxide in expired air. Important limitations of the new trial were that treatment could start only in the second trimester, only participants who could not quit unaided could be included, and there was a gap of two weeks after the first session before the next session of behavioural support and adjustment of NRT. This time delay is crucial as it covers the period when smokers are generally at most risk of relapse, borne out by figures in this study. This delay perhaps explains why both groups had low success rates.

Hard evidence on the effectiveness and efficacy of drug treatments for pregnant smokers is lacking. Whereas other smokers can choose from a range of effective options, such as bupropion, varenicline, and cytisine, these agents have not been evaluated in pregnant women and cannot be recommended for this group. Of all the available NRT products and delivery modes, Berlin and colleagues chose transdermal patches. In non-pregnant smokers, combining patches with faster acting forms of NRT works substantially better than using patches alone.4 5 6 7 These combinations are more flexible, help target cravings better, and are commonly used in practice. To date, no trials have studied combination treatments in pregnant smokers; however, data from clinical practice strongly suggest a benefit of combined NRT over single forms.8

Fortunately, and in contrast with disappointing results and a slim evidence base for drug treatment, good evidence shows that psychosocial interventions help pregnant smokers to quit. Effective, evidence-based strategies include individual counselling, financial incentives, peer support, and feedback on fetal health or the by-products of tobacco smoking (for example, measuring the levels of carbon monoxide in expired air).9 Counselling and feedback work best when delivered in combination and tailored to individual women. Although financial incentives had the largest effect sizes in a recent systematic review,9 this particular result should be interpreted with caution as it was based on few trials with small samples. Most of the included studies evaluating psychosocial interventions were from high income countries, leaving a major gap in the evidence about how best to support pregnant women trying to quit smoking in low and middle income countries.

Further research has analysed trial and treatment protocols of these efficacious interventions and identified the active ingredients.10 The behavioural support offered to women in Berlin and colleagues’ trial included some of these, but such support was not based on a comprehensive manual and important elements are likely to have been omitted. Again, we know that psychosocial support is critical in the first few weeks of a quit attempt,11 which was when the participants in the present study did not receive any.

So how does this affect practitioners supporting pregnant smokers? Healthcare practitioners must routinely intervene with all smokers. There is no need to ask whether a smoker wants to stop smoking before offering advice and help.12 13 This advice should be followed by a referral to specialist smoking cessation services or instigation of other intensive and evidence based behavioural support. Given the limitations of the new trial and the potential benefit of combination therapy, it may be too early to abandon the option of NRT entirely. Pregnant smokers who fail to quit are exposing their fetuses not only to nicotine but also to a multitude of other potentially teratogenic substances in tobacco smoke. If effective, nicotine substitution using established NRT or other delivery systems (potentially including electronic cigarettes) would be a far safer alternative. However, a much greater effort is still needed to identify, test, and deliver more effective treatments for pregnant smokers who struggle to quit.


Cite this as: BMJ 2014;348:g1808


  • Research, doi:10.1136/bmj.g1622
  • I thank Ann McNeill and my colleagues for their comments on a draft of this editorial.

  • Competing interests: I have read and understood the BMJ Group policy on declaration of interests and declare the following interests: None.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


Log in

Log in through your institution


* For online subscription