Monday, January 22, 2007

Analysis Shows No Increase in Nicotine Yields of Marlboro Cigarettes from 1997-2006; Widespread Media Reports Have Given a Misleading Picture

According to a new report released this Thursday by the Harvard School of Public Health, cigarette companies - including Philip Morris - steadily increased the nicotine yield of their cigarettes - including Marlboro - during the period 1997-2005. The report describes the increase as being a total of 11% over the seven-year period 1998-2005, or an average increase of 1.6% each year during that period.

According to the report, Marlboro brands "showed a significant increase in smoke nicotine yield." During the overall study period of 1997-2005, the report estimates the increase to be 0.019 mg per year. The report concludes that there has been a "statistically significant trend in increased smoke nicotine yield of 0.019 mg per cigarette (1.1%) per year from 1997-2005 as measured by a smoking machine under the MA method." This trend was said to hold for all market categories, and in particular, for the most popular cigarette brand - Marlboro: "The present analysis of the leading U.S. brand family, Marlboro, demonstrates a significant increase in smoke nicotine yield, contradicting the PM USA claims."

Philip Morris has taken issue with the study's conclusions, arguing that there was no increase in the nicotine yield of Marlboro cigarettes between 1997 and 2006 and that there are no consistent trends indicating an increase in these yields over time, although there are random fluctuations over the time period.

The report has received extensive media coverage, with headlines across the country informing the public that there has been a large, significant, and steady increase in nicotine yields over the recent eight-year time period.

For example, a New York Times article reported that: "A Harvard study concluding that cigarette makers have for years deliberately increased nicotine levels in cigarettes to make them more addictive led to renewed calls Thursday for greater federal oversight of the industry."

Major anti-smoking groups also hailed the study and told the public that it shows the need for FDA regulation of tobacco products. The Campaign for Tobacco-Free Kids (TFK) issued a press release stating that: "The Harvard study uses sophisticated methods of analysis and also includes data from 1997 and 2005, demonstrating conclusively that there is a clear upward trend in the levels of nicotine found in Marlboro."

TFK used the study results to call for passage of legislation that would grant the Food and Drug Administration (FDA) authority to regulate tobacco products: "A new study released today by the Harvard School of Public Health shows the critical need for Congress to enact legislation granting the U.S. Food and Drug Administration (FDA) authority over tobacco products. The Harvard study expands on and confirms an August 2006 study released by the Massachusetts Department of Public Health that found that tobacco companies have deliberately increased the levels of nicotine in cigarette smoke since 1998. The FDA legislation would require tobacco companies to disclose to the FDA changes in their products and provide FDA the authority to require them to reduce levels of constituents, like nicotine that make them more harmful or more addictive. ... The fact that the tobacco companies have been able to secretly increase nicotine levels in tobacco smoke occurred only because no federal or state agency currently has regulatory authority over cigarettes or what tobacco companies put in cigarettes. ... The proposed legislation would grant the FDA the authority and resources to stop harmful tobacco company practices that continue to addict children... ."

The American Legacy Foundation, one of the funders of the study, issued a press release stating that: "These data from Harvard University are an expanded study of findings first released in August of last year by the Massachusetts Department of Public Health. Both studies show that nicotine levels in seven brand families -- including Marlboro, Newport and Camel cigarettes -- have increased significantly between 1997 and 2005."

The Rest of the Story

The steady and significant increase in nicotine yields of Marlboro cigarettes over the past eight years or so sounds like a concerning finding. It sounds like Philip Morris is secretly and deceptively increasing the nicotine yields of its leading cigarette - Marlboro - and apparently, lying about it (since they deny the allegations of the report), all in an effort to increase the addictive potential of their cigarettes, and as a result of all of this, harming the public's health in a way that demands passage of the Campaign for Tobacco-Free Kids' FDA tobacco legislation that it has been promoting for the past two years.

At least this is what the major anti-smoking groups want the public to think.

There are, however, a number of major problems with this.

First, and most significant, is the fact that the major allegation of the report - that there has been a steady increase in nicotine yield of Marlboro cigarettes since 1997 and that Philip Morris is wrong to deny that - appears to be inaccurate.

Given the discrepancy between what the newspapers are all saying and what Philip Morris is claiming, I saw only one way for me to resolve the issue for myself: and that is to obtain the data and analyze it for myself.

Using the data on nicotine yields of Marlboro cigarettes provided by Philip Morris to the Massachusetts Department of Public Health (MDPH) for the years 1997 through 2006, I conducted my own analysis of the trends in Marlboro nicotine yields. During this time period, there were 16 Marlboro sub-brands whose nicotine yields were reported to MDPH each year. I assembled a complete panel of data that included the nicotine yields of each of these 16 Marlboro sub-brands for the period 1997-2006.

Before getting to the details of the analysis, take these two pieces of data and see what conclusion you would come to yourself:

Average nicotine yield of all 16 Marlboro sub-brands in 1997: 1.81 mg
Average nicotine yield of all 16 Marlboro sub-brands in 2006: 1.81 mg

To get right down to it, then, the rest of the story is that there was no change in the average nicotine yield of Marlboro cigarettes from 1997 to 2006.

Or to say it another way: the average nicotine yield of Marlboro cigarettes in 2006 was exactly the same as in 1997, nine years earlier (at least to two decimal points).

Even if one examines the average nicotine yield of Marlboro cigarettes in 2005, there still was very little change from 1997 to 2005: a slight increase from 1.81 mg to 1.85 mg. That represents an increase of just 2.2% over an eight-year period, which (if one assumes a steady linear increase) corresponds to an annual increase of just 0.3% per year, or 0.005 mg per year. This is about one-fourth of the increase that is described for Marlboro in the report.

If you examine the data in more detail, you can see that the assumption that there has been an increasing linear trend in Marlboro nicotine yields over the past 8 or 9 years appears to be unclear. Instead, it appears that there was a small increase from 1997 to 2003, and then a small decrease from 2003 to 2006 (the average nicotine yields by year, starting in 1997, are: 1.81 mg, 1.74, 1.83, 1.85, 1.84, 1.89, 1.93, 1.87, 1.85, 1.81).

If you model the Marlboro nicotine yield trend as a steady, linear increase, you do get a pattern of an overall increasing trend, with an increase of 0.008 mg per year. However, one can more accurately model the data as showing two lines: first, a line with increasing from 1997 to 2003; and second, a line with decreasing slope from 2003 to 2006. If one does this, there appears to be a 0.024 mg per year increase in nicotine yield between 1997 and 2003 and a 0.038 mg per year decrease between 2003 and 2006.

While it is not clear whether these trends reflect an actual intention to increase nicotine yields of Marlboro sub-brands from 1997 to 2003 and then decrease the yields in the past three years or whether this is random variation over time, what is clear is that there is no steady increase in the nicotine yields of Marlboro cigarettes as a group.

If one examines trends by Marlboro sub-brand, there are a total of only 4 sub-brands (out of the 16) which had higher nicotine yields in 2006 than 1997 (to one decimal point). There are 5 sub-brands which had lower nicotine yields in 2006 than 1997 (to one decimal point). The other 7 sub-brands had essentially no change between 1997 and 2006.

My analysis applies only to Marlboro cigarettes and thus I am not commenting on the conclusions regarding the trends in nicotine yields of other cigarette brands. But it does appear clear that with regard to Marlboro, there has certainly not been a steady, significant, and substantial increase in nicotine yields over the past 9 years.

Moreover, examining the data for one other Philip Morris brand - Basic - in a similar manner, one finds a very similar result: there was an increase in nicotine yields from 1998 to 2004, but a decrease from 2004 to 2006, such that over the entire period 1998 to 2006, the average nicotine yield actually fell slightly.

It appears that the Campaign for Tobacco-Free Kids was wrong in stating that the data demonstrate "conclusively that there is a clear upward trend in the levels of nicotine found in Marlboro." Going from 1.81 mg to 1.81 mg is hardly a clear upward trend.

The American Legacy Foundation was also apparently wrong in telling the public that nicotine levels in Marlboro "have increased significantly between 1997 and 2005."

Anti-smoking groups appear to have been too hasty in jumping upon the conclusions of the report and disseminating the findings to the media before all the conclusions were carefully evaluated and scrutinized, especially using the most recent data (from 2006) that are publicly and easily available.

In its haste to use these findings to support the need for FDA tobacco legislation, the Campaign for Tobacco-Free Kids has told us that it is the increase in nicotine levels that presents the critical need to pass this legislation. Now that it is clear that nicotine yields in Marlboro cigarettes are not on the rise, but are in fact decreasing, will the Campaign now agree that the legislation is no longer needed?

Beyond the fact that the data do not demonstrate an increase in nicotine yields for Marlboro, there are a number of other problems with the conclusions that anti-smoking groups are drawing and disseminating to the public regarding this issue.

For one, even if there had been an increase in nicotine yields in (Marlboro) cigarettes, this does not necessarily mean that the public's health is harmed. It is well documented that smokers compensate in response to changes in nicotine yields to maintain exposure to a relatively constant nicotine dose. This is why "light" cigarettes are not safer products. While the nicotine levels are lower, smokers compensate by simply smoking more; this negates the potential benefits of reduced nicotine and tar levels.

In a similar way, smokers might be expected to compensate by smoking slightly less if nicotine yields increase. This could actually have a marginally positive health benefit if it reduces overall cigarette consumption.

In fact, the report acknowledges this important point: "The increase in smoke nicotine yield does not necessarily signify any change in exposure within the population of smokers, particularly as human smoking behavior is compensatory and will adjust for differences in smoke yield."

If the increase in smoke nicotine yield (assuming it were a real and sustained increase over time) doesn't necessarily indicate any change in exposure among smokers, then what exactly is the significance of these findings? How do the findings indicate any net public health detriment from the changing nicotine yields?

If anything, the way this issue is being presented to the public by anti-smoking groups could actually harm efforts to protect the public's health. By focusing on increases in nicotine yields as if they are necessarily harmful to public health, anti-smoking groups are implying that decreases in nicotine yields would be a good thing. But the truth is that reduced nicotine yields could be harmful to public health because they would likely increase cigarette consumption (due to compensation), leading to increased tar delivery and higher rates of lung and other cancers as well as chronic lung disease.

Another problem with the way the information is being presented to the public by anti-smoking groups is the implication that the tobacco companies are doing something wrong if they are increasing nicotine yields (assuming, again, then this were true, which is no longer clear - it's clearly not true for Marlboro and Basic). This implies that if the companies were decreasing nicotine yields, they would be doing something good. But because of compensation, as discussed above, a decline in nicotine yields could be devastating to the public's health. It is just not accurate to present the issue in a way that equates nicotine yields with harm to the public's health.

Yet another problem I have with the reporting of this story by anti-smoking groups is the implication that tobacco companies should not be controlling the levels of nicotine in their cigarettes and that there is something illegal or wrong with them doing so.

There is nothing illegal about the nicotine in cigarettes, nor is there anything wrong with manipulating nicotine yields. What is wrong, if anything, is the fact that cigarettes are made to be an addictive product by the failure to remove the nicotine. Whether the average nicotine yields in cigarettes are 1.9 mg or 1.7 mg, cigarettes are still addictive. If we're going to allow the companies to market an addictive product, I frankly don't see any difference between a product that delivers an average of 1.9 mg of nicotine versus 1.7 mg. And if we're going to allow nicotine in cigarettes, then what is wrong with cigarette companies controlling the amount of nicotine in the product.

The companies are truly in a catch-22 situation. If they increase the nicotine levels, anti-smoking advocates lambast them for trying to addict people. If they decrease the nicotine levels, then we lambast them for misleading people about the public health benefit of the altered product and for causing cigarette consumption to increase. I guess the only way companies could avoid being lambasted is to maintain nicotine levels at exactly the same level - but then they'd be attacked for maintaining levels of addiction in the population.

The point is that if anti-smoking advocates are disturbed by the addictiveness of cigarettes, then there's only one thing that can be done - and that's to require the elimination of the nicotine. Short of that, there's nothing that can be done. At least not anything beneficial. Requiring reductions in nicotine levels would be the worst thing we could do, because cigarette consumption would rise due to compensation, causing increased tar delivery and increased disease and death.

This brings us to our final point. The Campaign for Tobacco-Free Kids is, in my view, full of crap in proclaiming how the supposedly increasing nicotine yields demand FDA tobacco legislation to be enacted. The truth is that the legislation that TFK has promoted during the last two legislative sessions would have specifically precluded the FDA from eliminating the nicotine in cigarettes. Any effort to make cigarettes a non-addictive product by removing the nicotine would have certainly been tied up by lawsuits claiming that this provision in the legislation precludes any such action (which I believe it does).

If the supposedly rising nicotine yields in cigarettes should cause TFK to do anything, it would be to force TFK to withdraw its support for the current form of the legislation, which specifically does not allow FDA to eliminate the nicotine from cigarettes.

I'm not necessarily calling for the elimination of nicotine from cigarettes here. What I am saying is that if anti-smoking groups are going to complain about the nicotine levels and lambast the tobacco companies for manipulating these levels, then they should be calling for the elimination of the nicotine. That's the only way to address the problem. If they are not willing to do that, then they should shut up already. Their support for legislation that precludes FDA from addressing the problem is appalling given the public statements that these groups are making about how awful it is that cigarettes are addictive and that cigarette companies control nicotine levels and that they have been (supposedly) increasing nicotine levels steadily over the past eight years.

Finally, TFK's assertion that tobacco companies have been able to "secretly" increase nicotine levels is absurd. Those (supposed) increases were headlines in the New York Times. They were reported publicly. How secret is that?

This story has a number of important lessons for the public. For one, it shows how careful one has to be in blindly accepting what is reported to you by the media. It shows the importance for careful scrutiny, review, and analysis of information before it is necessarily accepted as being accurate. It also demonstrates how biased the anti-smoking movement and its groups are: how they jump without haste into attacking tobacco companies for wrongdoing before there is an adequate basis upon which to do so.

The rest of the story is that at least for the leading cigarette brand in the United States, there has not been a steady increase in nicotine yields over the past 8 or 9 years as widely reported by anti-smoking groups to the media. As 1.81 mg is not higher than 1.81 mg, the conclusions that were widely disseminated by anti-smoking groups to the media concerning increases in nicotine yields for Marlboro cigarettes were incorrect.

Unfortunately for the anti-smoking movement, this is going to be an embarrassment, because it represents an instance in which Philip Morris was clearly correct, and the anti-smoking groups were wrong.

Sometimes when you are too hasty in attacking the enemy, it comes back to haunt you. My calls for being more careful in drawing our scientific and policy conclusions and attacking others in the tobacco control movement ought to be considered more seriously. If they are not, this is exactly the kind of unfortunate outcome that can occur.

No comments: