Thursday, February 15, 2007

IN MY VIEW: FDA Tobacco Legislation Would Be Death Knell for Tobacco Harm Reduction

Perhaps the most compelling reason why I view the FDA tobacco legislation being introduced into Congress this week by Senator Edward Kennedy and Representative Henry Waxman as a measure that will harm, rather than protect, the public's health, is that the legislation would essentially put an end to a harm reduction approach in tobacco control. Yesterday, I offered my opinion that the proposed FDA tobacco legislation would put a death knell to harm reduction efforts. Today, I explain exactly why.

Please note that this analysis is based on the text of the legislation that was introduced into the previous two Congresses. To the best of our understanding, the legislation to be introduced in the coming days will be virtually identical to that legislation. Obviously, the analysis which follows depends on the accuracy of that assumption. But all indications are that the bills to be introduced by Senator Kennedy and Representative Waxman will be the same as the bills introduced the past two years.

The Rest of the Story

The critical flaw in the proposed legislation is section 911(g)(1). This is the Modified Risk Product section of the bill.

Based on my analysis, I have concluded that the Modified Risk Product section of the proposed FDA legislation would make it virtually impossible for modified risk products to enter the market, while at the same time, allowing reduced exposure products to essentially be falsely marketed as reduced risk products (thus institutionalizing the very problem that the health organizations have expressed so much concern about).

Here are the specific problems:

1. The legislation lists several criteria for achieving approval of a modified risk product. The most important are the following: "the Secretary shall approve an application for a modified risk tobacco product filed under this section only if the Secretary determines that the applicant has demonstrated that such product, as it is actually used by consumers, will--`(A) significantly reduce harm and the risk of tobacco-related disease to individual tobacco users; and `(B) benefit the health of the population as a whole taking into account both users of tobacco products and persons who do not currently use tobacco products.

Section 911(g)(1)(A), the (A) clause above places an insurmountable obstacle in the path of approval of modified risk products. In order to demonstrate that the product, as actually used by consumers, will significantly reduce the risk of tobacco-related disease to individual users, large-scale, long-term epidemiologic studies are necessary. Even ignoring the requirement under 911(g)(1)(B), the (B) clause above (which itself appears to introduce an insurmountable obstacle), the bill as currently written precludes any harm reduction approach to tobacco control both by making it impossible for such products to meet the conditions for approval and by eliminating any incentive (especially economic) to develop such products. Thus, the bill may have the exact opposite effect that many believe it should have. It protects the existing high-risk products on the market.

It is critical for the public and policy makers to understand the ramifications of section 911(g)(1). In order to market a reduced risk product, a cigarette company would have to demonstrate that the product, as actually used by smokers, would substantially reduce the actual risk of disease among individual tobacco users.

There is only one way to do this. And that is to conduct a long-term epidemiologic study in which one compares the disease risk of the modified product with that of a comparison product over a long time period and among a large population of smokers.

There are all kinds of complications with conducting such a study. First, it would be tremendously expensive. Second, it would take, at a minimum, 10-20 years to follow the smokers long enough to monitor changes in disease. For cancer risks in particular, you would have to follow smokers for about 20 years before you would be able to draw definitive conclusions regarding any reductions in risk.

Third, there are all kinds of research complications that would make it difficult to draw accurate conclusions. The only way to credibly demonstrate a reduction in risk would be to conduct a randomized clinical trial, where smokers were randomized to either smoke conventional cigarettes or the putative reduced risk cigarettes. But conducting a randomized trial of smoking would be unethical; such a study is impossible.

The best that could be hoped for is a natural experiment type of study in which a product is "test-marketed" and that population of smokers (i.e., guinea pigs) are monitored for 10-20 years. But this is a catch-22. How can you test-market the product if you need to obtain FDA approval before you can test-market it? The only way you could do this would be to market the product as a conventional cigarette (not let anyone know that it is a potentially reduced risk product). But to do that, you would destroy the study, because it is your obligation to demonstrate that as actually used by consumers, the product would reduce risk. Smokers might use the product very differently if they believe it is a reduced risk product than if they don't.

In other words, the legislation does not merely make it difficult to market a reduced risk product. It makes it literally impossible.

2. The bill contains a special rule that would allow FDA to approve certain modified risk products that cannot meet the criteria listed under (A) and (B) above. Such products must only claim to be reducing exposure to, or reducing levels of or being free of a particular constituent. Specifically, such products can be approved if "scientific evidence is not available and, using the best available scientific methods, cannot be made available without conducting long-term epidemiological studies for an application to meet the standards set forth in paragraph (1)." In such cases, the major criterion that must be met is as follows: "the scientific evidence that is available without conducting long-term epidemiological studies demonstrates that a measurable and substantial reduction in morbidity or mortality among individual tobacco users is anticipated in subsequent studies."

Thus, Section 901(g)(2)(A) essentially allows products that do not claim to reduce risk but merely claim to reduce exposure to or levels of a particular constituent to be approved by simply showing that a reduction in risk is ANTICIPATED in subsequent studies. This is a very weak standard. As long as there is any promising preliminary evidence, one could argue that reduced risk is expected if it were to be studied (this is a far cry from demonstrating reduced risk, especially given the many uncertainties including the actual use of the product by consumers and unintended side effects or consequences).

The effect of 901(g)(2)(A), then, is essentially to allow the approval of reduced exposure products and to allow these products to be marketed as such, even though there is shoddy scientific evidence of any health benefit. This institutionalizes the very problem (unsubstantiated health claims) that health groups supporting this legislation have expressed so much concern about. The fact that the bill does not allow these products to represent themselves as reducing health risk is largely irrelevant, because consumers are going to perceive them as reducing risk. How else would someone interpret a claim of reduced exposure?

Of note, the bill does require that there be testing of consumer perception which shows that as the product is proposed to be labeled and advertised, it will not mislead consumers into thinking that the product reduces health risk. However, there are no restrictions on who must do the consumer testing and what the conditions or protocols must be. It would be relatively easy for a company to do consumer testing in such a way that they obtain the result they want to show. Anyone familiar with survey research, focus group studies, or other types of formative research knows that the way questions are asked and the specific protocols used can shape consumer responses substantially. Similarly, post-market surveillance protocols are left to the companies.

The proposed FDA legislation is thus the worst of both worlds. On the one hand, it allows companies to gain a government sanction to market products under the guise of reduced risk by merely calling their product reduced exposure (does the term "lights" ring a bell?). On the other hand, it prevents companies from marketing products that may truly be reduced risk products by putting an impossible regulatory burden in the path of such products.

Despite all the rhetoric, the proposed FDA legislation would do nothing to save lives, but would instead ensure that the highest risk products remain firmly entrenched in the marketplace, without any competition from what could be life-saving alternatives. Of course, this lack of effective competition is why Philip Morris adores this legislation.

What it comes down to is protecting the profits of the nation's largest tobacco company at the expense of the public's health. That Philip Morris is lobbying for this legislation makes perfect sense, and it is the company's fiduciary responsibility to do nothing less. That the Campaign for Tobacco-Free Kids and other major anti-smoking groups are joining alongside Philip Morris in promoting this legislation is nothing less than a travesty.

No comments: