Tuesday, January 24, 2012

Chief Scientist of RTI Attacks Alpert et al. Study But Fails to Disclose Conflict of Interest; Which Side of the Story Would You Believe?

Two weeks ago, I reported the results of a new study by Alpert et al. which showed that nicotine replacement medications (NRT) are not effective in helping smokers to quit long-term. Based on this finding, the article argued that public expenditures for NRT provision to smokers is a waste of resources.

Calling this a landmark study, I wrote: "This study provides important empirical evidence that over the long-term, population-based use of NRT in real-life situations - outside of the clinical trial setting where there is extensive counseling and other support provided - is not an effective treatment to aid smoking cessation. It does not appear to be any better than smokers attempting to quit unaided. This does not mean that in individual patients, NRT is not effective and should not be considered. But it does mean that on a population basis, as a matter of public health policy, the role of NRT has been way over-rated. I agree that with limited funding available, using those funds to provide NRT is of limited value. Instead, limited public funding for tobacco control should be used for proven programs such as smoking prevention media campaigns, which have been shown to reduce youth smoking by as much as 50%."

In a response to the Alpert et al. article, Dr. Douglas Kamerow - chief scientist at RTI International and associated editor at BMJ - published a commentary in BMJ last week in which he dismisses the findings of the Alpert et al. study based primarily on what he states are a very low sample size and a high likelihood of recall bias. He also suggests that the results of "hundreds" of randomized clinical trials are being nullified by a "single" cohort study.

From what I can tell from the commentary page, no conflicts of interest are noted.

The Rest of the Story

The rest of the story is that unlike the original Alpert et al. study, in which none of the investigators had any conflicts of interests (including the fact that none has received funding for their work from pharmaceutical companies), this commentary is written by an individual with a quite profound conflict of interest:

His company - RTI International - receives or has received funding from a host of pharmaceutical companies, many of which produce and market the very products about which he is opining in this article. Yet nowhere in the article is this conflict revealed.

Among the pharmaceutical companies listed as clients of RTI are:
  1. Abbott Laboratories;
  2. AstraZeneca;
  3. Bristol-Myers Squibb;
  4. Eli Lilly and Company;
  5. The Johnson & Johnson Family of Companies;
  6. Merck & Co., Inc.;
  7. Novartis;
  8. Novo Nordisk;
  9. Pfizer;
  10. Roche;
  11. Sanofi-Aventis;
  12. Takeda Pharmaceuticals UK; and
  13. Tioga Pharmaceuticals.
That is no fewer than 13 pharmaceutical companies which are clients of RTI, several of which produce and/or market nicotine replacement medications which are the precise topic of the commentary.

How BMJ let this through without a disclosure of this severe conflict of interest is beyond me, but how the author failed to voluntarily disclose these conflicts is equally mystifying. The fact that these severe conflicts of interest are hidden creates the appearance that indeed, someone is trying to hide something. Perhaps the commentary would have less credibility if it were known that the author's company relies upon contracts from pharmaceutical companies that have a huge financial stake in the profitability of NRT, and therefore, in the conclusions of this very commentary.

This is not only a scientific, but an economic battle. The results of the scientific debate over the efficacy of NRT has billions of dollars worth of financial implications for the pharmaceutical clients of RTI International. Doesn't it seem that a disclosure of this conflict of interest would have been appropriate and necessary in order to provide the reader with information needed to judge the scientific objectivity of the commentary?

There are a number of statements made in the commentary which I view as scientifically flawed. First, the author asserts that there is only one cohort study and that those who are calling for policy change are dismissing hundreds of clinical trials on the basis of a single cohort study. This is a misrepresentation of the scientific literature. There are many population-based studies which call into question the effectiveness of NRT when used in real-life situations, outside of the clinical trial setting.

Second, the author laments the fact that many NRT users did not continue to use the NRT for the recommended eight weeks, which lowers the sample size. Yet that very piece of information is itself evidence that these smokers found NRT ineffective. Why did most of them fail to use NRT for the full eight weeks? Probably because they resumed smoking. That is, the NRT failed. The very reason why the sample size was so low is that NRT is quite ineffective and many smokers using NRT relapsed.

Finally, the author asserts that the majority of cold turkey quitters are highly motivated to quit, much more so than the subjects in NRT clinical trials. This is most likely not true, as most clinical trials involve the recruitment of smokers who are willing to try to quit smoking, are treated with intensive counseling, multiple follow-up visits, and frequent supportive phone calls, all in the context of what is essentially an unblinded study. The chances of quitting in real-life settings, in my opinion, are much lower than in these clinical trials. Thus, the fact that quitting rates in unaided attempts appear higher than with the use of NRT is, I believe, strong evidence that NRT's effectiveness has been grossly exaggerated based on the clinical trial results.

Unfortunately, because of the undisclosed and hidden conflict of interest, it creates the appearance that financial interests, rather than scientific ones, may be influencing the scientific points made in this article, albeit subconsciously. I have no idea whether this occurred here or not. But that's precisely the problem with undisclosed conflicts of interest. They open the door to speculation as to the potential role of investigator bias.

The rest of the story is that this commentary was written by an author with severe conflicts of interest by virtue of the fact that his company relies upon a large number of pharmaceutical companies as clients, many of which produce the very drugs about which the commentary opines. This severe conflict of interest, unfortunately, is not disclosed in the article, creating at least the appearance that there was an attempt to hide it. This calls into question the scientific objectivity of the commentary, and creates at least an appearance that this lack of objectivity could be in part a result of the conflict of interest.

One thing is for sure. The commentary does a great job of protecting the financial interests of many of RTI International's corporate clients.

No comments: