Article Text


Do “Light” cigarettes undermine cessation?
  1. John R Hughes
  1. University of Vermont, Burlington, Vermont, USA
  1. John R Hughes, MD, University of Vermont, Department of Psychiatry, 38 Fletcher Place, Burlington, VT 05401-1419,{at}

Statistics from

Despite significant tobacco control efforts, in the USA the average smoker tries to stop only once every 2.5 years.1 In Europe, it is even less frequent.2 One hypothesised factor to explain this reluctance to try to quit is that tobacco industry promotion of “Light” cigarettes as a reasonable (and of course easier) alternative to quitting has undermined cessation resolve.3-5 The articles in this issue are some of the best, if not the best, tests of whether this hypothesis is true and, if so, what one could do about it. This editorial summarises both this new evidence and prior evidence for and against this hypothesis.

Correlational studies

If Light cigarettes undermine cessation, one would expect Light smokers to be less likely to have quit by the time of a survey than regular smokers. Analyses of the 1986 Adult Use of Tobacco Survey data did find this pattern.3 6 Also, one would expect that Light users to report less interest or desire to quit; however, here all four studies found either similar or more interest in quitting among Light/Ultra Light users than among regular users.7-9 Retrospective studies have also found 23–55% of Light/Ultra Light smokers report switching as a step toward, not away from, cessation.6 10-12

In terms of prospective studies, when 8000 smokers in military boot camp were followed for one year, those who switched to low tar cigarettes were not less likely to go on to cessation.7Also, when a cohort of 1000 smokers from the Community Intervention Trial for smoking cessation (COMMIT) were followed, there was no difference in quit rates between smokers who did and did not switch to low tar cigarettes.13

Interpreting these retrospective, cross sectional, and prospective studies is difficult because smokers self select into groups. On the one hand, Light smokers are the more educated, more health conscious smokers6; thus, they would be expected to have higher than normal quit rates. Finding they have similar rates of cessation to regular cigarette users could actually be seen as evidence that Lights undermine cessation. On the other hand, Light smokers appear to be smokers who are more nicotine dependent7 8 and who switch to Light cigarettes because they think their chances of quitting are very small. If this were the case, then one would expect lower cessation rates in Light users, not because Lights undermined cessation, but rather because Light users were poor cessation candidates from the beginning.

Experimental studies

Luckily, several experimental studies not subject to these selection biases are available. A 1998 study reported 25–32% of low tar smokers said they would be more likely to quit if they knew that low tar cigarettes were actually not any safer.10 This finding prompted four experiments to determine if this is actually true.4 12 14-16 One of these was a non-randomised comparison (n = 1000) that found twice as many smokers in Massachusetts who saw a Department of Health advertisement against Light cigarettes quit smoking than smokers who did not see the ad.14 Furthermore, during the period the ads ran, Massachusetts had a higher quit rate than the rest of the USA. The other three were the randomised experiment described in the review article by Kozlowski and colleagues4 6 16 and the two experiments of Shiffman and associates12 15 published in this supplement. In these studies, ads correcting false beliefs about Lights increased self reported interest in quitting, likelihood of quitting or interest in obtaining treatment.

These four studies are important because they are both direct, experimental validations of the undermining hypothesis as well as demonstrations that counter-marketing may be an effective intervention. Experimental studies eliminate many of the biases of prospective and retrospective studies and typically are believed to be more valid. In addition, the consistency of the findings across the replication tests and the similarity of their findings across different methods, measures, and designs is a strong argument for convergent validity.

These experimental studies could be criticised because their effect sizes were small and their outcomes were self reported intention to quit, not actual quit attempts or cessation. But given the brevity of the intervention, small effects would be anticipated. Thus, a large trial with a more intensive and longer lasting counter-marketing intervention and with actual quit behaviour as an outcome is clearly indicated. However, I think the value of this trial will be, not as a test of replication (as I do believe it will replicate), but rather as a feasibility/demonstration project. Positive results in such a trial should prompt inclusion of anti-Lights ads into all tobacco control media campaigns.17 18


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.