A recent article in the Wall Street Journal reported on a study that concluded that nicotine replacement therapy (NRT) doesn't positively impact smoking cessation. Not surprisingly, this report has raised a number of questions for insurers, consumers, and the FDA.
The study, A prospective cohort study challenging the effectiveness of population-based medical intervention for smoking cessation, conducted by the Harvard School of Public Health and the University of Massachusetts, examined a sample of 787 adult smokers who had recently quit smoking. The study's objective was to "examine the effectiveness of NRT... to better inform healthcare coverage decisions."
Participants were surveyed over three time periods: 2001-2002, 2003-2004, and 2005-2006. In the initial interview, they were asked whether they had used a nicotine replacement therapy in the form of the nicotine patch, nicotine gum, nicotine inhaler, or nasal spray to help them quit. They also were asked if they had joined a quit-smoking program or received help from a doctor, counselor, or other professional. During the subsequent interviews, they were asked if they had started smoking again.
One of the most significant conclusions, and the source of much of the controversy surrounding the report was the relapse rates between the first, second and third interviews reported by the participants were no different between individuals who used NRT and those who did not. The lead author of the study, Hillel Alpert, was quoted in the Harvard Gazette as saying "This study shows that using NRT is no more effective in helping people stop smoking cigarettes in the long term than trying to quit on one's own."
Since the study only examined relapse rates, this claim is not supported by the evidence put forth in the study. Effectiveness is measured in more than one dimension -- relapse rates are only one piece. As numerous, larger studies around the globe have indicated (according to a review from the Cochrane Collaboration of 123 different controlled studies), NRT is effective among smokers who are motivated to quit, regardless of setting. The quit rate among smokers who use NRT is 1.5 to 2 times higher than smokers who quit cold turkey.
So, if we were to compare two equal sized populations of smokers trying to quit smoking -- one that used NRT and one that did not -- the population of people who successfully quit using NRT will be larger than the population of people who successfully quit without it. Even if both populations relapse at equal rates in the years to come, nearly twice as many people will have successfully quit smoking for good using NRT.
Furthermore, even if the ex-smokers relapse, there are health benefits to quitting for even a short time period. According to the US Surgeon General's Report, 3 months after quitting, circulation and lung function improve, after 9 months, coughing and shortness of breath decrease, and after 1 year, the risk of coronary artery disease is cut in half (not to mention the social benefits of not constantly being engulfed in a cloud of smoke!).
So, reimbursing NRT may still be cost effective, even if some smokers relapse. Another study in the UK actually examined the cost effectiveness of NRT and found that NRT was a low-cost treatment when measured per life-year saved. And this study examined the cost of nicotine patches 15 years ago; since then, the patent on nicotine patches has expired, so they are available at an even lower cost today.
Alpert, according to the Harvard Gazette's coverage, argues that even though clinical trials have found NRT to be effective, the new findings demonstrate the importance of empirical studies regarding effectiveness when used in the general population. But since the study only includes participants who have already quit smoking, it doesn't provide sufficient evidence to conclude the value of NRT is questionable. There are confounding variables that the study does not take into account -- for example, the number of people who would have failed to quit in the first place had they not used NRT.
To really understand the effectiveness of NRT in a real-world setting, researchers would have to start with a population of smokers who are trying to quit, not a population of smokers who have already quit, and follow them over several years. And since so few people who want to quit smoking are successful, the population studied would have to be very large. Only then can the study measure NRT effectiveness in a real-world setting.
So what does all of this mean for insurers? Not much. The Harvard study examined such a narrow indicator that the value of NRT can't really be argued one way or another based on the evidence provided. Because the study only looked at relapse rates, it does not sufficiently undermine the numerous other studies that have indicated that NRT does improve quit rates and can improve health outcomes.
For consumers, the water is even murkier. The study underscores the point that NRT is not a long-term treatment; it's meant to be taken for 6-12 weeks to help with initial symptoms of withdrawal. It doesn't make the urge to smoke go away forever. Still, based on other studies -- like those included in the Cochrane Collaboration's review -- it can be a valuable tool for people who are trying to quit.
Alpert's point about the importance of real-world evidence in evaluating different treatment options is a good one, though, and as long-term, observational studies become more prevalent, there are clear implications for the FDA. Medical device and pharmaceutical manufacturers will increasingly look to real-world, post-market evidence to find new information about existing products, including information to support new indications for existing products. These studies will be increasingly complex, as they examine many dimensions of similar products and protocols. The FDA will need to learn to process this information quickly, and should re-examine its approach to ensuring product safety.
Currently, the FDA usually makes demands for clinical evidence based on large-scale clinical trials to gain product approval. But if it were to adapt a more flexible approach, looking at more diverse kinds of studies, it could revolutionize the medical products industry as we know it.
Manufacturers could design clinical studies that rely on data reflecting a real-world evidence approach and modeling, rather than merely relying on the current "gold standard" -- traditional, large-scale, randomized, placebo-controlled clinical trials (RCTs) with their associated inclusion/exclusion issues. Instead of waiting years for an actual clinical outcome from RCTs, manufacturers could use a "surrogate endpoint" (like quit rates, in the case of NRT) to establish an event that can reasonably predict improved clinical outcomes. Additional confirmatory studies would provide longitudinal evidence that the treatment either does or does not provide clinical benefits for specific sets of patients, allowing the FDA to react accordingly.
New products would reach the market faster and enable many patients the opportunity to try a new treatment sooner rather than later. This approach, in turn, could reduce development costs, and medical products companies could focus their attention on innovating new life-saving therapies.
To reach this goal, the FDA will need to shift its focus from an almost exclusive emphasis on pre-market gatekeeping to post-approval oversight. There are clearly inefficiencies in the agency's ability to bring new products to market, and it does need to do a superior job in its post-approval oversight role. What's really needed is an overhaul of both the agency's regulatory process and its focus.