Thursday June 30 2011
Paticipants were sent supportive messages for six months
Motivational text messages sent to smokers’ mobile phones can double their chances of giving up tobacco, reported The Guardian.
The story is based on a large UK study that looked at whether a six-month programme of supportive text messages could help smokers quit. It compared a group who received positive messages to another group who were given details of other support programmes. At six months, those receiving texts were twice as likely to quit, with a quitting rate of 10.7% compared to the other group’s rate of 4.9%.
This large, well-designed study took several steps to ensure the accuracy of its results. For example, saliva tests were used to verify how accurately people had reported their non-smoking and, in their analyses, the researchers counted people who dropped out of the study as failing to quit. While the quit rate was relatively low in both groups, the researchers say it was comparable to the number of people who succeed using other forms of help, such as counselling.
As a relatively cheap intervention that could reach large numbers of people, text messaging is potentially cost effective, an issue the authors will address in a forthcoming study. The trial did not directly compare text messaging with other methods for quitting smoking, such as nicotine replacement therapy or behavioural support, so text messaging still needs to be assessed in relation to existing treatments.
Where did the story come from?
The study was carried out by researchers from London School of Hygiene and Tropical medicine, the University of Auckland in New Zealand, and the George Institute at the University of Sydney in Australia. It was funded by the UK Medical Research Council, Cancer Research UK and the Primary Care Research Networks. The study was published in the peer-reviewed medical journal The Lancet.
Generally, the media reported the story accurately. The BBC included comments from independent experts.
What kind of research was this?
This randomised controlled trial (RCT) looked at whether motivational text messages could help people give up smoking. It assessed the effect of the messages on smoking habits at six months. Studies with an RCT design compare a treatment against another treatment, a placebo or no treatment. This study compared the quitting rates in a group who received motivational text messages and another who received occasional non-motivational text messages (the control group). An RCT is considered the best type of study to determine the effectiveness of a treatment.
Many RCTs are double blinded. This means that neither participants nor researchers know to which group the participants were allocated. This resolves the problem that knowing which group they belonged to could affect the results, either consciously or unconsciously. However, the nature of the treatment being tested means that this is not always possible. For example, in this study, it would not have been possible to conceal from participants whether they were receiving the intervention or were in the control group. However, this trial was single blinded as the researchers were not told which participants were in which group.
The researchers say that mobile phone technology has the potential to provide personalised stop-smoking support. Motivational messages and methods to change people’s behaviour, which are used in face-to-face stop-smoking services, can be modified for delivery through mobile phones. Content can also be tailored to address the age, sex and ethnic group of each quitter. Given the widespread ownership of mobile phones, they believe that fully automated stop-smoking support can be delivered to large numbers of people at low cost.
The researcher say that although this type of programme has been shown to increase self-reported smoking abstinence at six weeks, the extent to which these early benefits can be maintained in the longer term needs further investigation. They also point out that some previous studies have not used chemical tests to verify whether participants truly stopped smoking, which might give a better indication of quitting rates.
What did the research involve?
Researchers gathered 11,914 potential participants using advertisements for their trial. To be eligible, participants had to be smokers aged 16 or older who were willing to attempt to quit in the next month and owned a mobile phone. In total, 5,800 volunteers were found to be eligible and entered the trial.
Between 2007 and 2009, they were randomly assigned to either a mobile phone text-messaging stop-smoking programme called txt2stop, or to a control group that received text messages unrelated to quitting. Randomisation was carried out using an independent telephone randomisation system, which also balanced the groups based on their sex, age, educational level and level of nicotine addiction. The system automatically sent the participants motivational or non-motivational texts according to their allocated group. Researchers involved in the trial did not know which participants were allocated to which group, unless the participants told them. All participants were free to take part in other stop-smoking programmes.
People in the group receiving the text-messaging programme were asked to set a quit date within two weeks of being assigned to the group. They received five text messages a day for the first five weeks, and then three a week for the next 26 weeks. These included messages to encourage motivation and behaviour change, encouraging them to persevere with their attempts to quit, such as an introductory message on their chosen quit date: “This is it! ... TODAY is the start of being QUIT forever, you can do it!” The text support group could also send a request for a “lapse text” if they smoked, which read: “Don’t feel bad or guilty if you’ve slipped ... Slip-ups can be a normal part of the quitting process.”
By texting the word “crave”, quitters could receive instant messages to distract and support them and they could also text each other for support.
The programme was also personalised based on information gathered at the start of the study, sending a selection of texts to address each smoker’s concerns related to quitting, such as potential weight gain. In total, participants had access to a core programme of 186 messages and a further database of 713 personalised messages. Participants in the intervention group were also given top-up vouchers towards phone credit.
Control group participants received short, simple, fortnightly messages relating to the importance of trial participation but not encouraging them to quit. All participants were also offered relevant helpline numbers.
Researchers gathered data from participants on whether they had quit smoking, defined as no more than five cigarettes smoked in the past week at four weeks, and no more than five cigarettes smoked at six months of follow-up. This was done through a website or by telephone. The researchers verified self-reported quitting at six months with a postal saliva test that tested for levels of cotinine, a substance produced in the body as a result of nicotine breakdown, which can be used to distinguish smokers from non-smokers. Those who said they had given up but whose test showed they were still smokers were counted as smokers in the analysis.
The researchers used validated statistical methods to look at the potential effect of the text-messaging programme. They looked at the effects of the intervention by age (split into people over or under 35 years old), level of nicotine addiction, employment status, receipt of a mobile phone top-up voucher and use of other stop-smoking treatments.
What were the basic results?
The randomisation process allocated 2,915 smokers to the txt2stop intervention and 2,885 to the control group. Eight people were excluded because they were randomised more than once. Results were available for 5,524 (95%) participants.
Researchers found that at six months, 10.7% of those on the txt2stop programme had stopped smoking, compared to 4.9% of those in the control group, with results verified by saliva tests. This equated to a more-than-double rate of quitting for those in the intervention group (relative risk [RR] 2.20, 95% confidence interval [CI] 1.80 to 2.68). The absolute difference was 5.8%, meaning about six extra people in every hundred people offered the motivational texts would stay off cigarettes at six months compared to those not receiving the texts.
Similar results were obtained when participants that were “lost to follow-up” were treated as smokers.
The intervention had similar results in both younger and older smokers and in all socioeconomic groups. Biochemical results showed that more than a quarter of participants who self-reported quitting were smokers. There was no evidence of any difference between the two groups in the use of existing smoking support services.
How did the researchers interpret the results?
The researchers say that the txt2stop programme significantly improved stop-smoking rates at six months and should be considered for inclusion in smoking cessation services.
This large study had several strengths, such as using reliable methods for randomisation and ensuring that staff and researchers had no foreknowledge of how treatment was allocated. Results were calculated for both the 95% of participants who completed the study and the full set of participants, counting any participants who were not followed up as smokers. Reports of quitting smoking were verified using saliva tests, which helps increase the accuracy of estimated quitting rates.
However, as with any study, it had some limitations:
- Occasionally, participant told study staff which group they were in, the knowledge of which could have influenced the results.
- It was not possible to mask participants to the treatment they were allocated to. The authors say that, in particular, this could have reduced motivation in the non-treatment group as they would have known they were not getting the intervention. The researchers say that they tried to minimise this by offering contact details for existing stop-smoking services.
- The biochemical tests used are not 100% accurate.
- The non-text group were not asked to nominate a date to stop smoking. If they had, it may have given them a specific goal to aim for, which might have affected their chances of quitting successfully.
Overall, this large, well-conducted trial suggests that text message programmes offering support for stopping smoking could be a valuable addition to smoking cessation services. The researchers plan to publish details of how the content of the test messages might fit into current theories of how to change behaviour, as well as further economic analysis of the cost of the intervention. These are both important questions as the treatment will need to be assessed in relation to existing stop-smoking services.
Analysis by Bazian.