Monday, June 29, 2015

IN MY VIEW: Anti-Tobacco Advocates are OK With Letting Smokers Die at the Current Rate

In what to me is a shocking admission, two anti-tobacco advocates have argued that there is no need for e-cigarettes - a new and promising smoking cessation tool - because current progress in reducing smoking is sufficient and more smokers each year are trying to quit.

In other words, what they are saying is that they are content with letting smokers die at the current rate. There is no need to expand the smoking cessation alternatives available to smokers, they argue. This means that they are OK with letting smokers die at the current rate and not pursuing a wider range of alternative cessation strategies. They are apparently content with the status quo; that is, the current anti-smoking interventions and cessation treatments (and the current rate of decline in cigarette smoking).

This is shocking to me because it runs counter to not only public health, but to any concept of humanity. How can we possibly be content with letting smokers die at the current rate? How could we possibly not want to accelerate the rate of decline in smoking by making a new smoking cessation tool available to smokers? How can we possibly be content with the current slow rate of decline in smoking prevalence? How can we simply accept the more than 400,000 deaths from smoking each year, and not want to drastically reduce that by pursuing a new, alternative, and potentially much more effective tool for smoking cessation?

This argument is presented in a press release appearing on Stan Glantz's blog and is based on a new article by Kulik and Glantz which appears in the journal Tobacco Control.

According to the press release, Dr. Glantz argues as follows:

"Smokeless tobacco and, more recently, e-cigarettes have been promoted as a harm reduction strategy for smokers who are “unable or unwilling to quit.” The strategy, embraced by both industry and some public health advocates, is based on the assumption that as smoking declines overall, only those who cannot quit will remain.  A new study by researchers at UC San Francisco has found just the opposite.  

The researchers analyzed survey data spanning 18 years in the United States and six years in the European Union. They found that, contrary to the prevailing assumptions, as the fraction of the population that smoked declined, the remaining smokers actually smoked less and were more likely to quit than to stick with it. The authors said their findings challenge the need to promote new forms of nicotine delivery, such as e-cigarettes, since the smoking population continues to quit smoking as a result of proven policies and interventions."


According to Glantz:

"“The fact that the smoking population is softening has important implications for public health policy. These results suggest that current tobacco control policies have been leading to softening of the smoking population without the need to promote new recreational nicotine products like e-cigarettes.”"

Later in the press release, Dr. Kulik echoes this argument:

"We show that there is no real need to distribute e-cigarettes as part of a tobacco policy package because the smoking population is softening."

The Rest of the Story

Whether the smoking population is softening or not is irrelevant to the question of whether we should pursue new and potentially more effective alternative treatments or strategies for smoking cessation. The reason to promote the search for these more effective treatments or alternative strategies is that we are not content with watching more than 400,000 people die from smoking each year and we want to decrease the number of smoking-related deaths as much and as quickly as humanly possible. Smoking prevalence is declining, but it is declining quite slowly. The rapid rates of decline in smoking prevalence that were observed in past decades have plateaued and started to drop off. This should lead us to be open to alternative cessation strategies, not to close the door on them and be content with the current rate of progress.

The authors of this article set up a straw man by arguing that the reason for the promotion of electronic cigarettes is that the smoking population is hardening. Then, if they can show that the smoking population is actually softening, they can argue that e-cigarettes are not needed. However, this is a straw man argument because the reason why advocates like myself are promoting electronic cigarettes has nothing to do with whether the smoking population is hardening or softening.

Even if it were true that the smoking population is softening (and I'll show below that it is not), I would still be promoting electronic cigarettes because there are so many smokers (the overwhelming majority) who fail to quit with existing therapies. We cannot be content to just let these smokers die. Instead, we need to find more effective or at least equally effective alternative strategies that they can use, rather than have them give up on quitting. Electronic cigarettes are just such an alternative and they have helped thousands of smokers to quit who could not and would not have quit using existing treatments.

This reason for promoting electronic cigarettes as an alternative cessation strategy that should be made available to smokers has nothing to do with whether the smoking population is hardening or softening. Even if it were shown convincingly that there was a huge amount of softening of the smoking population, I would never use that as an argument that the status quo is acceptable and that we should be content with the current slow decline in smoking progress and smoking-related disease, disability, and death.

The Rest of the Rest of the Story

Although the potential value of electronic cigarettes is not conditioned on whether the smoking population is hardening or softening, this article misinterprets its own data to incorrectly conclude that the smoking population is softening.

The evidence presented by the article and used to argue that the smoking population is softening is that, according to the article, as smoking prevalence decreases, quit attempts and quit ratio increase. However, the data involve a panel of 50 states with annual measurements over time. Each data point represents the smoking prevalence, quit attempts, and quit ratio in a given state in a given year. So a good deal of the variation in quit attempts and quit ratios is due to differences in smoking prevalence between states, not to differences in smoking prevalence over time.

The study did attempt to make some inferences about the degree to which there were differences in the relationship between smoking prevalence and quit attempts or quit ratios over time by including two time-dependent variables. However, neither of these variables were significant in the models. Thus, there was no change in the relationship between smoking prevalence and either quit attempts or quit ratios over time. In other words, the observed relationship could be explained simply by the fact that states with a lower smoking prevalence tend to have smokers with higher quit attempts and a higher quit ratio. The analysis conducted in the paper is not answering the correct question. It is largely answering the question of whether states with higher smoking prevalence have higher smokers with more quit attempts. In fact, it could be that the reason the smoking prevalence is lower in these states is precisely because the smokers in that state are more likely to try to quit and are more likely to quit successfully.

To examine the question of hardening vs. softening, what really needs to be examined is the relationship not between smoking prevalence and quit ratios and attempts across states, but the relationship between quit attempts and quit ratios over time. What we really want to know is what percentage of quit attempts made each year are successful and how that changes over time across the United States. A state-by-state analysis is of less interest than a national analysis. But most importantly, what we need to know is what proportion of smokers made quit attempts and what proportion succeeded in quitting each year, and then how those proportions changed over time. This is what will tell us whether the percentage of successful quit attempts has been increasing or decreasing over time.

This article confirms that the proportion of smokers making quit attempts in the United States has increased over time. However, trends in smoking prevalence during the same time period confirm that the decline in prevalence has declined over the past decade and a half. And since the number of quit attempts has gone way up, this means that the proportion of quit attempts that are successful has dropped. In other words, smokers are finding it more difficult, not easier, to quit. This actually suggests that the smoking population is hardening.

Dick Puddlecote has provided an excellent explanation of this phenomenon and does a fine job of showing why the conclusions of this study are inconsistent with the very data that it presents. He summarizes his argument as follows:

"Glantz has ... proven that more people are embarking on quit attempts. However, since there is no steepening decline in prevalence from real life data - in fact it is the opposite - this can only mean one thing; that a smaller percentage of smokers are successful with their quit attempts than in the past and making people quit is therefore getting harder."

Interestingly, data from the UK (the Smoking Toolkit Study) quite ironically show that while the proportion of successful quit attempts was stable prior to the dramatic rise in e-cigarette use, this proportion has increased as e-cigarette use has proliferated in the UK. Moreover, the advent of e-cigarettes was associated with large increases in the number of quit attempts being made. Furthermore, the most common aid now used in quit attempts is e-cigarettes.

To be clear, whether the smoking population is hardening or softening is irrelevant to the issue of whether e-cigarettes should be promoted or not. If e-cigarettes are a bona fide smoking cessation aid for many smokers who could not or would not quit using existing therapies, then e-cigarettes should be promoted as an alternative for this sub-population of smokers (which is, in fact, the vast majority - about 90% to 97%). Even if smokers were making ever more quit attempts and becoming more successful, that is no reason to deny the large proportion of smokers who are not successful quitting using traditional drugs the opportunity to try e-cigarettes as an alternative approach.

The rest of the story is that in this new article, the study sets up an irrelevant straw man argument, misconstrues the reasons why e-cigarettes are being promoted, misinterprets the data analysis, and draws an invalid conclusion about the "irrelevant" research question which it set out to answer. But the worst part of the story is that the article, or at least the statements accompanying the article, essentially argue that we should not pursue e-cigarettes because we should be content with the status quo, as smokers are making more quit attempts each year.

But that status quo is actually the death of more than 400,000 smokers each year. That status quo is a decreasing rate of decline in smoking prevalence. How any anti-tobacco activist could be content with the status quo is not clear to me. I honestly believe that it is Dr. Glantz's extreme hatred for e-cigarettes that has led him to express this inane (and rather inhumane) argument.

Current projections suggest that the U.S. is not going to reach the Healthy People 2020 goal of reducing adult smoking prevalence to 12% in the next five years. Why Glantz and Kulik seem to be content with that failure is not clear to me. I would argue that we should not accept the slow rate of progress and we should not be content to rely on current interventions. We need to expand these interventions, not sit on them. Most importantly, we need to pursue alternative smoking cessation treatments for the vast majority of smokers who are unsuccessful using current approaches. Why are Glantz and Kulik throwing these smokers under the bus?

No comments: