Ken Wassum, Associate Director, Clinical Development & Support, Tobacco:

Nicotine replacement therapy (NRT) has been used for decades to help smokers quit. This week’s newspapers and online sources were awash with “breaking news” that nicotine replacement therapy does not work. This study* came out of the Harvard school of public health, and frankly its findings should not be worth the hype.

I will take this a step further by saying that NRT options work better than ever. Tobacco treatment professionals now better understand how to recommend combinations of these medications as well as match them with highly effective behavioral strategies that improve quitting success greatly.

Here is what the Harvard researchers did. Between 2001 and 2006 they interviewed a total of 787 individuals who had recently quit smoking. They conducted 3 waves of interviews over the course of 5 years. Results showed that those who used NRT (nicotine patches or nicotine gum) and/or professional assistance were more likely to relapse back to smoking as those who quit without NRT or professional help. I do no dispute their findings, but I do question their interpretation of the results.

We have known for years that their findings are largely true, but it does not mean that NRT is not effective. It certainly does not mean that professional help is not effective. What it means is that those who seek treatment (counseling and/or use of quit medications) are typically more nicotine dependent than those who do not seek treatment and as a result have less success in quitting. Those who are less dependent typically feel they can quit on their own.  As Saul Shiffman pointed out in his 2008 publication** surveys of the type done by the Harvard researchers are subject to recall bias. He states, “past quit attempts are easily forgotten, particularly as many are undertaken spontaneously, and many are short-lived. Treatment seeking itself likely makes quit attempts more memorable because smokers must undertake unusual actions, like seeking counseling or purchasing and using medications. Thus, treatment failures may be remembered and unaided failures forgotten, leading to biased estimates of effectiveness.”

Many experts in the field of tobacco dependence treatment feel the Harvard study had some limitations. The most obvious was the self-selection bias, where individuals are not randomized to a certain treatment, but rather select a treatment or quitting approach based on their assessment of their own needs. I am not saying that randomized controlled trials (RCTs) for cessation medications do not have some problems – they do. In my opinion they frequently exclude too many individuals who are representative of real world smokers. But randomization goes a long way to remove the element of bias in study results.

It is sad that many reputable news agencies ran the article without doing due diligence by researching the findings. The effect is that many people who very much need to quit smoking and would benefit hugely from professional treatment will not seek help because they now may believe that these medications do not work. In fact, they do. 

In my opinion, rather than saying that medication treatments are not effective or that they are not a good use of scarce public health expenditures, what we need is more research showing us how to better implement what we already know is effective. And one of these effective treatment elements is NRT.

*Alpert, H.  A prospective cohort study challenging the effectiveness of population-based medical intervention for smoking cessation.  Tobacco Control 2012.
**Shiffman, S. Use of Smoking-Cessation Treatments in the United States. Am J Prev Med 2008.