Guidelines in Whose Interest? - Smoking Cessation

Physicians are frequently exhorted to follow clinical practice guidelines. Guidelines are now being given "teeth" by the currently fashionable "Pay for Performance" (P4P) movement. Yet, as we have discussed, guidelines may be written by people with financial ties to organizations that have vested interests in products or services that the guidelines might recommend.

The latest example of this comes from an article in the Wall Street Journal. The Journal reported that US government sponsored guidelines about smoking cessation (helping people quit cigarettes) were written by people with financial ties to companies that make drugs intended to aid smoking cessation, and that the guidelines recommended these drugs as first line therapy.

Michael Fiore is in charge of revising federal guidelines on how to get smokers to quit. He also runs an academic research center funded in part by drug companies that make quit-smoking aids, and he personally has received tens of thousands of dollars in speaking and consulting fees from those companies.

Conflict of interest? No, says Dr. Fiore, who has consistently declared that doctors ought to use stop-smoking medicine. He says his opinion -- reflected in current federal guidelines -- is based on scientific evidence from hundreds of studies.

The Public Health Service, part of the Department of Health and Human Services, issued guidelines in 2000 calling for smokers to use nicotine patches, gums and other pharmaceutical aids to quit, with a few exceptions such as pregnant women. Dr. Fiore, a University of Wisconsin professor of medicine, headed the 18-member panel that created those guidelines. He and at least eight others on it had ties to the makers of stop-smoking products.

The panel is now working on a revision of the guidelines, scheduled for completion early next year. Dr. Fiore, an internist, is again chairman. He says this time only seven of 26 members have industry ties.

As the federal government weighs the data in making new recommendations, many of its advisers are receiving money from companies with a stake in the outcome. Dr. Fiore holds a chair at Wisconsin that is funded by GlaxoSmithKline. He directs a tobacco research center that received nearly $1 million in funding from makers of quit-smoking medicine in 2004 and $400,000 in 2005. Between 1999 and 2004, Dr. Fiore personally pocketed $10,000 to $40,000 a year from the quitting-aid industry for honorariums and consulting work. He says he stopped such work in 2005.

However, the evidence that supports using drugs to help people quit smoking may not be so clear. First, as the Wall Street Journal article pointed out, new observational studies have failed to show that using drugs leads to higher quit rates in the long run. These studies, however, can be confounded by differences among people who choose to quit "cold turkey" versus those who choose to use drugs.

Second, the evidence from controlled trials of drugs versus placebos may not be unequivocal. For example, I reviewed a prominent article that reported a trial that compared placebo with the nicotine patch, sustained-release buproprion, and both the patch and buproprion. [Jorenby DE, Leischow SJ, Nides MA. A controlled trial of sustained-release buproprion, a nicotine patch, or bot for smoking cessation. N Engl J Med 1999; 340: 685-691.]

The article had two major problems. First, it excluded patients with any serious medical problems, who were on any psychoactive drug, or who had substance abuse problems. Thus the study's results may not apply to many of the patients who most need to stop smoking.

Second, many patients enrolled in the study dropped out before completion of follow-up. Fully 19.8% were completely lost to follow-up. We do not know whether the patients who dropped out quit cigarettes in the long run, and whether the quit rates in such patients randomized to different treatments may have varied. Thus, given that the magnitude of the drop-out rate was similar to the magnitude of the difference in the quit rates among the treatment groups, it is possible that were we to learn whether the patients who dropped out quit, this data could completely reverse the results.

For example, the one-year quit rates for the patients who stayed in the study were 15.6% for the placebo group, 16.4 for the nicotine patch, 30.3 for buproprion, and 35.5 for both buproprion and the patch. Suppose that all the patients on placebo who dropped out quit long term, and that none who were on buproprion and the patch who dropped out quit. Then the quit rates for those two groups would have been 64.4% versus 35.5%, a complete reversal compared to the published results. Such a high drop out rate appears to be a fatal flaw in this study.

Finally, the Wall Street Journal article noted that many experts in smoking cessation have failed to reveal their financial ties to companies that make smoking cessation drugs when speaking or writing. For example,

Those who advocate medication sometimes fail to disclose that they have financial ties to companies. In an article on Voice of America's Web site last year, Jack Henningfield, identified only as a smoking-cessation expert, urged smokers to 'go to the consumer-friendly Web site that I like, which is www.quit.com.'

Dr. Henningfield is a principal of Pinney Associates, a consulting firm whose largest client is GlaxoSmithKline, operator of the quit.com site. Other articles citing Dr. Henningfield's views on smoking have identified him as a professor at Johns Hopkins School of Medicine without mentioning the GlaxoSmithKline connection. Dr. Henningfield, who holds a doctorate in psychology, is an adjunct professor at Johns Hopkins. He says only 10% of his income comes from Hopkins.

Dr. Henningfield says he always tells journalists about his financial ties to industry. But in an interview with The Wall Street Journal last summer, Dr. Henningfield promoted the use of stop-smoking medicine without volunteering any information about those ties. He says he thought GlaxoSmithKline's public-relations firm had already provided the information.

In at least two medical-journal articles that Dr. Fiore wrote or co-wrote promoting the use of stop-smoking medicine, no mention was made of his financial ties to the makers of those treatments. Dr. Fiore says the editors of those journals may have ignored his disclosure or he may have failed to provide it. If the latter, 'I am sorry about that,' he says, adding that those are two of more than 150 medical-journal articles he has published.

In summary, some authors of the guidelines for smoking cessation had financial ties to companies that make drugs used to help smokers quit. The guidelines favored first-line use of these drugs, yet the evidence supporting such use may not be crystal clear. Smoking cessation experts have failed to disclose such financial ties when speaking or writing about relevant issue.

As I have concluded before, just because a set of recommendations is called a guideline, or is written under the auspices of some august academic or scientific organization does not guarantee that its recommendations are based on the best evidence, or are not influenced by vested interests.And yet guidelines remain a buzz word for those who say they want to control health care costs, and improve quality and access. And giving guidelines teeth, otherwise known as "pay for performance," (P4P) is now the most fashionable solution to the cost, quality, and access problems. (See post here.)We need to remain very skeptical about how guidelines are developed, and whose vested interests they may serve, and very careful about giving guidelines teeth through P4P.