'Smoking does not relieve stress – but quitting does,' the Daily Mail reports.
The story is based on research that looked at whether giving up – or trying to give up – smoking over a six-month period was associated with any change in anxiety levels.
In this study, researchers looked at data collected from a trial of people who were trying to stop smoking using various nicotine replacement therapies.
Anxiety levels were measured at the beginning of the trial and during follow-up. The current piece of research looked at the data from this trial and found that people who had stopped smoking by the end of the six-month study became less anxious (a nine-point decrease in anxiety score), while those who tried to give up but failed became a little more anxious (a three-point increase in anxiety score).
The changes in anxiety levels were more marked in people with a mental health disorder who reported that they smoked in order to cope with stress.
However, this type of secondary analysis of trial data cannot tell us for certain whether the varying anxiety levels were caused by smoking status or if they were due to other, unmeasured factors. We also don't know how significant these changes would be for each individual in terms of their functioning. However, as acknowledged by the researchers, this type of secondary analysis of trial data cannot tell us for certain whether the varying anxiety levels were caused by smoking status or if they were due to other, unmeasured factors.
Despite these limitations, the findings suggest that managing to quit smoking for good could be good for your mental health as well as your physical health.
Where did the story come from?
The study was carried out by researchers from King's College London, the University of Southampton and the University of Birmingham. It was funded by the Medical Research Council UK and a number of institutions with an interest in cancer prevention, heart disease or promoting stopping smoking.
The study was published in the peer-reviewed British Journal of Psychiatry.
The results of the study were reported fairly accurately in the media.
What kind of research was this?
This was a prospective cohort study that aimed to assess whether successfully giving up smoking or failing to give up had any effect on people's anxiety levels.
The researchers did this by carrying out a secondary analysis of a previous randomised controlled trial (RCT) that had recruited adults who wanted to stop smoking and randomised them to different types of nicotine replacement therapies (NRTs) to help them quit.
Various assessments were performed as part of this trial, including measuring the participants' anxiety at the start of the study and during follow-up. This was the data that the current researchers used for their study.
However, the main aim of the original trial was to look at the effect that individually tailoring NRT to a person’s genetic make-up had upon their likelihood of achieving successful abstinence, rather than looking at the effect of quitting or relapsing upon anxiety.
As such, this secondary analysis of data collected as part of a trial cannot prove cause and effect. It is possible that many other factors, such as changes in employment or relationship status, were responsible for the change in anxiety scores. This limitation is recognised by the authors who acknowledge that there may have been unmeasured variables that could account for the findings, but consider that there is no reason to suspect that any such variables should be differentially associated with abstinence status.
The authors say that the belief that smoking alleviates anxiety is widespread, despite some evidence that smoking may actually cause stress. The belief that smoking relieves stress is a major barrier to smokers giving up and health professionals recommending it to patients.
They say that because of this belief, smokers with psychiatric disorders in particular are less likely than other smokers to be offered advice about giving up. This group, say the researchers, have a life expectancy about 20 years lower than people without such a disorder, a gap that is partly attributable to higher smoking levels.
They also say that while anxiety tends to increase in the first few days of an attempt to stop smoking due to withdrawal from nicotine, it remains unclear what happens to anxiety levels in the longer term once the initial withdrawal phase has ended.
Some research has suggested an association between successfully giving up and reduced stress levels.
What did the research involve?
The researchers used a study originally set up to look at the effects of tailoring nicotine replacement therapy to people's genetic make-up. Participants in the study were recruited from smoking cessation clinics in 29 GP practices in two English cities and were followed for six months. Those smoking at least 10 cigarettes a day who wanted to quit and were aged 18 or over were eligible for inclusion. A total of 633 people agreed to participate.
All participants were prescribed nicotine replacement patches (the dose dependent on how heavily they smoked) and oral NRT (either lozenges or gum – the researchers do not state which).
They attended a total of eight weekly clinic appointments with a research nurse. People started their attempt to give up after the third clinic visit.
At the first clinic visit participants' anxiety levels were measured using the short form of an established standardised questionnaire, with scores ranging from 20 to 80. They were also asked about their motives for smoking, with three possible responses – whether they smoked "mainly for pleasure", "mainly to cope" or "about equal". They were also asked to report their current medical history, including their psychiatric history and their use of medication. Where responses to these questions were unclear, researchers checked medical records.
Researchers also collected information on other factors such as age, gender, ethnicity, socioeconomic status and nicotine dependence.
Six months after enrolment the participants were contacted by telephone or by post. They completed follow-up questionnaires on both their current smoking status and anxiety levels. Those who reported they were still not smoking were asked to send saliva samples by post, which were analysed for the presence of cotinine (a chemical related to nicotine that can act as a marker for whether someone has been smoking recently).
The researchers analysed whether smoking status at six months was associated with a change in people's anxiety levels. They also looked at whether the diagnosis of a psychiatric disorder had any effect on this association. They adjusted their results to take account of potential confounders such as age and gender.
What were the basic results?
Of the 633 original participants 491 (77.6%) completed the study. At six months, 68 (14%) were not smoking. Fifteen per cent of those who abstained and 23% of continuing smokers had a diagnosed psychiatric disorder.
The researchers found that overall, at six months participants who had relapsed experienced a three-point increase in anxiety levels measured at the start of the study, while those who were still not smoking experienced a nine-point reduction in anxiety levels.
They say this represents a points difference of 11.8 (95% confidence interval 7.7-16.0) in anxiety score at six months after stopping smoking between people who relapsed and people who attained abstinence.
The increase in anxiety in those who relapsed was largest for those with a current diagnosis of a psychiatric disorder and whose main reason for smoking was to cope with stress. The reduction in anxiety on those who successfully abstained after six months was also larger for these groups.
How did the researchers interpret the results?
The researchers conclude that people who manage to give up smoking experience a marked reduction in anxiety, whereas those who fail to quit experience a modest increase in the long term.
They say that the data contradicts the assumption that smoking is a stress reliever, although it suggests that failing to quit may generate anxiety.
This study found that, overall, quitting smoking at six months was associated with a moderate reduction in anxiety levels, while failing to quit was associated with a small increase in anxiety.
However, the researchers did this by carrying out a secondary analysis of a trial that had recruited adults who wanted to quit smoking and randomised them to different nicotine replacement doses to help them quit.
As part of this trial various assessments were made, including measuring the participants' anxiety at the start of the study and during follow-up.
The current research used this data. However, it is likely that the main aim of the original trial was to look at the effects of different NRT types upon successful abstinence, rather than looking at the effect of quitting or relapsing upon anxiety. As such, this secondary analysis of data collected as part of a trial has a number of limitations:
- Most importantly, it is uncertain if the varying anxiety levels were caused by smoking status. Although researchers adjusted their results for possible confounders such as age and gender, it is possible that other unmeasured factors – such as a change in employment or relationship status – affected the results. This limitation is recognised by the authors who acknowledge that there may have been unmeasured variables that could account for the findings, but consider that there is no reason to suspect that any such variables should be differentially associated with abstinence status.
- We also don't know how significant the changes in scores would have been and whether they would have made any changes to the individual's daily life and functioning, which would depend to a large extent on how severe the person's anxiety was to begin with. Of note, only 14 people in the whole trial were reported to have had a diagnosed anxiety disorder at the start of the study (three of whom managed to quit at six months).
- Further to this, as the researchers point out, it is possible that as most diagnoses of psychiatric disorder were self-reported and not all were verified by medical records, some could have been incorrect.
- Only a small proportion of people (68) abstained from smoking at six months, which may reduce the power of this study to reliably detect differences between quitters and non-quitters.
- The study was also short-term and it remains uncertain whether there is any link between long-term quitting and changes in anxiety levels.
Nevertheless, the results of this study are of interest, suggesting that quitting smoking may have benefits for mental as well as physical health. The study's authors conclude with the recommendation that clinicians should encourage people with a psychiatric disorder who use smoking as a coping mechanism to try to quit.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 2 January 2013
Daily Mail, 2 January 2013
Links to the science
British Journal of Psychiatry. Published online January 2 2013