Pregnancy and child

No conclusive evidence that e-cigs tempt teens to smoke

"Young people who try e-cigarettes are much more likely to start smoking, scientists have concluded," The Daily Telegraph reports. Though the conclusion, such as it is, is based on just 16 teenagers.

The study relies on results from two questionnaires, sent a year apart to about 700 young people in the US on whether they had ever smoked e-cigarettes or tobacco. 

Just 16 of these youngsters had tried e-cigarettes at the start of the study, six of them had tried a cigarette by the next year and five thought they might in the future.

Importantly, the research did not ask people how often they had used e-cigarettes or smoked tobacco, so we have no idea whether they were "addicted" to nicotine. 

This study leaves many unanswered questions, such as why young people tried e-cigarettes or tobacco.

Ultimately, this is a very small number to base such sweeping conclusions on. 

Where did the story come from?

The researchers are from the University of Pittsburgh School of Medicine, Dartmouth University and the University of Oregon in the US. 

The study was published in the peer-reviewed medical journal JAMA Pediatrics on an open-access basis, so it is free to read online.

It was funded by the National Cancer Institute and National Center for Advancing Translational Sciences.

The Telegraph and the Mail Online overplayed the fears about e-cigarettes being a gateway for teens into tobacco use. 

The Telegraph wrongly stated that 68% of those who had tried e-cigarettes went on to smoke tobacco – the true figure was 37.5%. The Mail Online reported the percentages correctly, but did not say that these results were based on just 16 young people who had tried e-cigarettes.

Also, the reporting of the study could give the impression that the findings represented a consensus opinion, which is certainly not the case. The study has come in for harsh criticism from independent experts in public health.

For example, Professor Robert West, professor of health psychology at UCL, is quoted as saying: "This kind of propaganda by major medical journals brings public health science into disrepute and is grist to the mill of apologists for the tobacco industry who accuse us of 'junk science'."

What kind of research was this?

This was a longitudinal cohort study, which means the researchers followed a group of people over time to see what happened to them. These studies are good at finding links between things, but cannot show that one thing causes another.

What did the research involve?

Researchers reviewed the results of questionnaires sent to people aged 16 to 26, which asked them about whether they had ever smoked tobacco cigarettes (defined as just one puff) or ever tried e-cigarettes, and about their attitudes to smoking. They followed up with another questionnaire one year later, and asked them the same questions.

They then used statistical analysis to see whether people who said they had tried e-cigarettes, but did not smoke and would not accept a cigarette if offered one, had tried smoking tobacco or changed their attitudes to it.

Of the people surveyed, 728 said they had never smoked and would not accept a cigarette if offered one. Only 507 of these people responded to the survey again a year later, so the researchers used statistical techniques to estimate the likely responses of some of the people who dropped out, based on responses from people in similar circumstances. This gave them a total of 694 people to base the survey on.

The researchers looked to see whether some other factors were also associated with the chances of someone trying a tobacco cigarette during the year. These included people's age, whether their parents smoked, whether their friends smoked, and how likely they were to try risky things.

What were the basic results?

Only 16 of the 694 people in the study had ever tried e-cigarettes at the start of the study. Of those, six (38%) tried a tobacco cigarette during the year of the study. Another five (31%) said they might try a tobacco cigarette if they were offered one, but had not done so yet.

Young people who had not tried an e-cigarette at the start of the study were less likely to say they had tried a tobacco cigarette at the end of it. The study found 65 of the 678 (10%) who had not tried an e-cigarette went on to try tobacco, and 63 (9%) said they might try a tobacco cigarette if offered one.

After adjusting their figures to take account of other factors, the researchers calculated that people were eight times as likely to try tobacco in the following year if they had tried an e-cigarette (adjusted odds ratio (AOR) 8.3, 95% confidence interval (CI) 1.2 to 58.6).

Looking at other factors that were linked to the chances of smoking tobacco, the study found young people who said they were open to trying risky things were more than twice as likely to try tobacco (AOR 2.6, 95% CI 1.3 to 5.2), and those who had more friends who smoked were almost twice as likely to try tobacco (AOR 1.8, 95% CI 1.2 to 2.9).

Not surprisingly, people who had tried e-cigarettes were more likely than those who had not tried them to say they were open to trying new or risky things.

How did the researchers interpret the results?

The researchers said their findings showed that e-cigarettes might make young people more likely to try smoking tobacco. They said: "Because e-cigarettes deliver nicotine more slowly than traditional cigarettes, they may serve as a 'nicotine starter' allowing a new user to advance to cigarette smoking," as they get used to the effects. 

They also say that using e-cigarettes might lead people to get used to the habit of smoking. They say the results of their study "support regulations to limit sales and decrease the appeal of e-cigarettes to adolescents and young adults".

Conclusion

On the face of it, this study seems to support the idea that young people progress to smoking tobacco via e-cigarettes. However, there are many limitations, which means we cannot draw such a conclusion from the study findings.

The first serious limitation is that only 16 of the 694 young people in the study had actually tried e-cigarettes. With numbers so small, we cannot be sure the results are reliable. There is a high chance that another group of 16 young people who had tried e-cigarettes might have given different answers.

Also, this type of study can never prove that one thing (in this case trying e-cigarettes) causes another (trying tobacco cigarettes). Young people try lots of things while they grow up, and some people are more likely than others to take risks. It is perhaps not surprising that those who try e-cigarettes are also more likely to try tobacco.

The language in the study could be misleading. For example, it describes people who had ever tried an e-cigarette as "e-cigarette users" and people who have taken even one puff of a cigarette as "smokers". Teenagers may try something once and then never try it again. 

It also talks about young people's "progression to smoking", which you might think means the numbers who had actually started to smoke. However, the definition of progression includes those who moved from saying they would definitely not accept a cigarette if offered one, to saying that they were not likely to accept one but could not rule it out completely. 

This may be why The Daily Telegraph got its figures wrong – it combined the young people who had tried smoking with the young people who had not ruled it out completely.

It is important to know whether e-cigarettes encourage people to start smoking tobacco. Tobacco is much more harmful than e-cigarettes, because of the toxins created when tobacco is burned. A recent evidence review carried out by Public Health England stated that e-cigarettes were probably 95% safer than smoking tobacco.

This study does not add much to our knowledge about whether e-cigarettes encourage young people to start smoking tobacco. We would need to see bigger, more detailed studies over time that look at how often people use e-cigarettes and tobacco, to get closer to answering that question.


NHS Attribution