Smoking is the main preventable cause of premature illness and death worldwide. Approximately 20% of deaths in men and 5% in woman over 30 can be attributed to smoking. The World Health Organization (WHO) estimates that 6 million people die from smoking related disease each year, with tobacco being the number one risk factor for non-communicable disease deaths. Indeed, the WHO views reducing tobacco use as a priority with today’s World No Tobacco Day campaigning for a ban on tobacco advertising to reduce uptake of smoking.
But what of those people already addicted to tobacco? Quitting smoking can significantly improve health and wellbeing; unfortunately, many smokers find it difficult to achieve permanent abstinence without support.
There are lots of treatments available for people who want to stop smoking
A number of drug treatments are available to help people to quit smoking. The three most widely used types of treatment are; nicotine replacement therapies (NRT), bupropion (an antidepressant) and varenicline (a nicotine receptor partial agonist), all of which are licensed for use in Europe and the United States. Bupropion and varenicline are available on prescription, whist NRTs such as patches and gums are also available over the counter.
Treatments such as cytisine have also been used in other countries as a potential intervention, but are not yet widely available. Other therapies such as anxiolytics, selective serotonin reuptake inhibitors (SSRI’s), monoamine oxidase inhibitors (MAOI’s), opioid agonists, clonidine, mecamylamine (both originally marketed to lower blood pressure) and a cannabinoid agonist have also been trialed as potential pharmacological treatments.
The efficacy and safety of these treatments for smoking cessation is the focus of a new Cochrane systematic review published today (Cahill et al, 2013). The authors’ primary objectives were to examine how the most widely used pharmacotherapies compare with placebo, and each other, in achieving abstinence. Other less used treatments were also considered where possible. Secondly, comparisons between the benefits and risks of each treatment were conducted.
The authors conducted a meta-analysis on 12 Cochrane reviews between 2008 and 2012, investigating a total of 26 pharmacotherapies. The analysis consisted of 267 randomised control trials (RCTs) including over 101,000 smokers. Eligible trials had to include a comparable placebo condition, other active pharmacological treatments or a combination of both. The primary outcome was smoking cessation, for 6 months or longer from the start of treatment.
To compare the benefits with the risk of harm the authors looked at any adverse or serious adverse (life-threatening) events, which could have been attributed to the treatment. Adverse events included such things as nausea and insomnia. Serious adverse events included the onset of psychological disorders (such as depression and anxiety), suicidal idealization, neurological events (seizures) or cardiovascular problems.
Using odds ratios (see Szumilas, 2010 for an explanation of odds ratios) the authors conducted direct comparisons between NRT, bupropion and varenicline with placebo. They found that:
This review found that NRT, bupropion and varenicline are all effective treatments for smoking cessation
NRT (Nicotine Replacement Therapies)
Taking NRT significantly increased the odds of quitting compared to placebo (OR = 1.84, 95% CI 1.71. to 1.99)
The main side effect of NRT was skin sensitivity and irritation (up to 54% of users). However, this rarely leads to treatment withdrawal. Due to limited data no analyses on serious adverse events was conducted
Taking bupropion significantly increased the odds of quitting compared to placebo (OR = 1.82, 95% CI 1.60 to 2.06) but not NRT (OR = 0.99, 95% CI 0.86 to 1.13)
The most common adverse events reported for bupropion were insomnia (30-40% of patients), dry mouth and nausea. The use of bupropion was associated with seizures occurring in 1:1,500 users, which was lower than previous estimates (1:1,000). There was a marginal, but not significant, increased risk of serious adverse events while taking bupropion (2.5% likelihood) compared to placebo (2.2%)
Taking varenicline significantly increased the odds of quitting compared to placebo (OR = 2.88, 95% CI 2.40 to 3.47) and to both NRT (OR = 1.57, 95% CI 1.29 to 1.91) and bupropion (OR = 1.59, 95% CI 1.29 to 1.96)
Of the available evidence, the most common adverse event caused by varenicline was nausea. Varenicline had no increased risk of serious adverse events compared to placebo. However, the authors note that evidence for the safety of varenicline is still under investigation
More evidence is needed before the promising treatments cytisine and nortriptyline can be recommended for smoking cessation
Secondary analyses demonstrated that different types of NRT administration (patches, gum etc) were generally as effective as each other. Interestingly, a combination of different NRTs is more effective than single formulations and leads to comparable odds of quitting as varenicline.
Whilst there was a smaller evidence base for other pharmacotherapies such as cytisine andnortriptyline (partly because they are not licensed in Europe or the USA) they also increased the chance of quitting compared to placebo. Of the other alternative pharmacotherapies subject to the meta-analyses only clonidine was found to be an effective treatment (although there were notable adverse effects); SSRI’s, MAOI’s, anxiolytics and opioid agonists generally had poor outcomes.
The review of the data on pharmacotherapies for smoking cessation suggests that the most common treatments all increase the odds of smoking cessation.
The authors conclude that:
Efficacy for NRT, bupropion and varenicline is well-established across a strong evidence base.
There was also some evidence that emerging treatments such as cytisine and nortriptyline may also improve odds compared to placebo, but more evidence is needed. Finally, the authors suggest that future research should attempt to determine retrospective safety and efficacy of varenicline and NRT.
The potential dangers of illegal drug use are never far from the media spotlight, and drug use during pregnancy may be associated with particular health problems both for the Mother and baby. Drug use during pregnancy has been linked with a number of negative outcomes, for example, cocaine use has been linked with an increased risk of placental abruption, while opiate use has been linked with neonatal abstinence syndrome and low birth weight. However, much of this research has failed to control for cigarette smoking status, which in itself has been associated with an increased risk of placental abruption, poor fetal growth, and late intrauterine death. Given that the majority of illegal drug users smoke, it is important to determine which maternal and prenatal health issues are specifically associated with illegal drug use.
Earlier this year, the Acta Obstetricia et Gynecologica Scandinavica published a retrospective cohort study, aimed at determining specific maternal and prenatal health issues associated with illegal drug use while controlling for cigarette smoking. They found that illegal drug use was associated with several serious health implications beyond that of smoking cigarettes.
The study compared 561 drug using women with 4,463 who had reported smoking cigarettes during pregnancy
Black and colleagues accessed the Aberdeen Maternity and Neonatal databank to identify women who had smoked during pregnancy, and contacted specialist ante-natal clinics in the same area of Scotland to identify women who had taken illegal drugs during pregnancy. 561 women were identified as illegal drugs users while pregnant. Of this group, 459 (82%) were prescribed methadone, 433 (77%) used heroin, 387 (69%) used benzodiazepines, and 190 (34%) used cocaine. The majority (85.6%) used more than one illegal drug and smoked cigarettes (96%). This drug using group was compared to 4,463 women who had reported smoking cigarettes during pregnancy.
Maternal baseline data included age, body mass index, and parity (childbirth history).
Primary outcome measures were:
Low birth weight (birthweight less than 2,500g) & Standardised birthweight score (SBS takes into account gestation, fetal sex, and maternal parity)
Neonatal unit admission
Perinatal death (stillbirth and fetal loss after 20 weeks gestation, and neonatal death up to 7 days of life)
Other outcome measures were:
Gestational hypertension (diastolic pressure >90 mm Hg on two occasions at least four hours apart, or a single reading of >110 mm Hg; from 20 weeks gestation onwards in a previously normotensive woman)
Pre-eclampsia (gestational hypertension plus ≥1 episode of proteinuria of 0.3 g/24 hrs)
Induction of labour (artificial rupture of membranes or use of prostaglandins to initiate labour)
Oxytocin augmentation (intrapartum oxytocin use after spontaneous onset of labour)
Manual removal of placenta
Postpartum haemorrhage (estimated blood loss >500ml)
Babies exposed to illegal drugs were more likely to have a low birthweight and be admitted to a neonatal unit
In terms of baseline maternal differences, illegal drugs users were less likely to be aged under 20 or over 35 years, or to be primigravid (pregnant for the first time). In addition, illegal drug users were 50% more likely to be underweight than women who smoked cigarettes only.
The authors calculated adjusted odds ratio (AOR) with 95% confidence intervals (CI) to identify associations between illegal drug use during pregnancy and maternal and neonatal outcomes, and to determine whether these associations were significantly different from those between cigarette smoking and outcomes.
Babies who were exposed to illegal drugs were more likely to
Have a low birthweight [AOR 1.9, CI 1.4-2.6]
Be admitted to the neonatal unit [AOR 16.1, CI 12.9-20.1]
Women who had used illegal drugs during pregnancy were more likely to have
Antepartum haemorrhage [AOR 1.6, CI 1.2-2.1]
Deep vein thrombosis [AOR 21.1, CI 8.8-50.8]
A preterm delivery at any gestation [AOR 1.6, CI 1.3-2.1].
However, women who had used illegal drugs during pregnancy were less likely to develop gestational hypertension than those who smoked cigarettes [AOR 0.3, CI 0.2-0.4].
The authors conclude that there is an:
Increased risk of antepartum haemorrhage, preterm delivery, low birth weight infants and admission of infants to the neonatal unit in pregnancies affected by illegal drug use, over and above that which can be explained by the effects of smoking cigarettes. This study also identified a lower prevalence of gestational hypertension in women using illegal drugs.
The authors suggest that the reduced risk of gestational hypertension in the women using illegal drugs may be due to lower BMI and the fact that many illegal drugs have a direct hypotensive effect.
The authors also suggest that although smoking is associated with a low BMI, the finding that women using illegal drugs still had a significantly lower BMI:
May be a reflection of these women investing more time and energy on satisfying their addiction than ensuring an adequate dietary intake.
Strengths and limitations
Future prospective studies should gather more information about substance use and confounding factors
It is important to correctly identifying which behaviours are associated with specific health problems, and to help patients improve their own health as well as that of their children. This is a fairly large cohort study that has identified health problems which are specifically associated with illegal drug use during pregnancy.
However, given that this is a retrospective study, there are limitations. For example, it is possible that some of the ‘control’ cigarette smokers had also taken illegal drugs during pregnancy but were not known to the specialist ante-natal clinics. In addition, smoking behaviours (e.g. number of cigarettes smoked per day) were not known; it is possible that the illegal drug users also smoked significantly more cigarettes. Although the study suggested significant effects of BMI, these must be taken with caution as 24% of this data was missing for the illegal drug users.
To determine whether these findings are valid, future research needs to conduct a prospective study in which a greater level of information can be obtained in terms of drug, alcohol, and smoking use during pregnancy while taking into account potentially confounding factors such as social deprivation and diet.
Acta Obstet Gynecol Scand. 2013 Jan;92(1):47-52. doi: 10.1111/j.1600-0412.2012.01519.x. Outcomes of pregnancy in women using illegal drugs and in women who smoke cigarettes. Black M, Bhattacharya S, Fairley T, Campbell DM, Shetty A. [PubMed abstract]
Kuczkowski. The effects of drug abuse on pregnancy. Current Opinion in Obstetrics & Gynecology 2007 Dec;19(6):578-85. [PubMed abstract]