Monday, June 06, 2005

New Study Promotes CT Screening as Smoking Cessation Tool

A study in the April 15 issue of Cancer reports that abnormal CT scans obtained as part of a three-year study of the utility of screening for lung cancer among smokers served as a motivational tool that increased smoking abstinence among these patients. Among smokers receiving annual chest CT scans for the early detection of lung cancer, 42% who had three successive abnormal scans quit smoking, compared to 28% with two abnormal scans, 24% with one abnormal scan, and 20% with no abnormal scans. Receiving a recommendation the previous year for follow-up due to an abnormal CT scan finding was significantly associated with smoking cessation.

The authors conclude not only that increased positive findings on chest CT screening increase smoking cessation, but that overall, the use of this type of screening enhanced smoking cessation rates, as “abstinence rates for those receiving 1, 2, or 3 positive screenings are much greater than the expected annual abstinence rates of 5-7% found in the general population.”

The paper’s overall conclusion is that: “Better understanding of the predictors of smoking cessation, including multiple lung CT scans, may inform the debate on the use of screenings for early-stage lung carcinoma to increase smoking abstinence… .” The authors note that: “The cost of annual lung CT scan screenings will have to be weighed against the potential benefits of increased smoking abstinence rates in high-risk populations of smokers.”

The Rest of the Story

On the surface, this may appear to be convincing evidence that supports the use of chest CT scanning of smokers as a motivational tool to encourage smoking cessation. However, closer consideration of the study reveals a number of very concerning issues.

First, at the most basic level, this study does not provide evidence that annual CT scanning of smokers, in itself, enhances smoking cessation. The paper compares the rate of smoking abstinence among study participants (which was about 20% at three year follow-up for smokers with normal CT findings) with rates among the general population of smokers (which are only 5-7%). However, this is not a relevant comparison. A relevant comparison would be with smoking abstinence rates among smokers who are highly motivated to quit, so motivated that they enroll in a lung cancer screening study.

In other words, the very fact that all subjects in this study were motivated enough to enroll in an intensive, long-term study to specifically screen for lung cancer makes them a special population, and comparing cessation rates among this group to population cessation rates makes little sense. So the absolute finding of “high” rates of smoking cessation in this study is not particularly meaningful.

Second, the study finding that being told that one has a suspicious lung nodule that needs follow-up motivates people to quit smoking is actually somewhat disturbing. Why? Because the overwhelming majority of these nodules are not, in fact, malignant. According to the most recent published data from this Mayo Clinic study, 92% of abnormal chest CT findings in this study are benign. What this means is that it is largely false positive results that are motivating quitting behavior. Smokers are quitting, therefore, based on a largely false belief that they may have lung cancer.

In and of itself, this is not necessarily a bad thing; however, to recommend the use of CT screening to promote smoking cessation is essentially promoting the use of instilling false fears in patients in order to motivate a change in a health behavior. That I have serious problems with.

Third, and most importantly, this study seems to suggest that a finding of increased cessation among smokers who are screened via annual chest CT supports the use of this strategy. But I don’t find that to be relevant to the consideration of using chest CT as a medical screening tool. If chest CT scanning is to be justified, it is going to have to be justified entirely on the basis of being an effective screening tool. If it is not justified on the basis of being an appropriate screening technique, then the benefits that may accrue to patients in terms of smoking cessation certainly do not justify the widespread use of this screening strategy.

The paper speaks about a “debate on the use of screenings for early-stage lung carcinoma to increase smoking abstinence.” I find that disturbing, because I don’t think there should be any such debate. We should absolutely not be using CT screenings as a strategy to increase smoking cessation. That is simply not an appropriate use of a medical screening procedure.

I have several serious ethical concerns with the approach to the evaluation of chest CT screening suggested in this paper. Is it ethical to promote a strategy for smoking cessation that relies largely upon instilling false and inappropriate fears among patients to motivate them to change their behavior? Is it ethical, in the first place, to use a medical screening procedure (that is designed for the early detection of disease) as a health behavior change strategy?

I can imagine, for example, that giving all 20-year-olds a colonoscopy might be an extremely effective strategy to change health behaviors (such as fiber intake and fat intake) that may be risk factors for colon cancer. Giving 18-year-old smokers a bronchoscopy would almost certainly induce a large number of them to quit. But these strategies are not ethical, because they are simply not justified as screening procedures.

My point here is not that chest CT scans will not eventually turn out to be justified as a lung cancer screening tool (although see my previous posts [post1 post2] about why there is not evidence at this time to justify CT screening for lung cancer). Instead, my point is that whether CT scanning increases smoking cessation is not relevant to the consideration of this strategy as a medical screening procedure. If it is justified, then it must stand on its own as a screening strategy, not on the basis of any effect on health behavior change.

So I disagree with the paper’s contention that “The cost of annual lung CT scan screenings will have to be weighed against the potential benefits of increased smoking abstinence rates in high-risk populations of smokers.” That is not an appropriate consideration. Even if the costs are outweighed by potential benefits of smoking cessation, the use of annual lung CT scans is not justified, appropriate, or ethical if it is not determined to be an effective and appropriate medical screening technique.

The rest of the story suggests that the appeal of advanced, state-of-the-art, high-technology solutions to society’s problems may be clouding the judgment and decision-making framework in the medical profession, and if left unchecked, could eventually result in the inappropriate and potentially unethical promotion of a costly, intensive, and fear-producing medical procedure.


Cervantes said...

I definitely agree that promoting a technique such as CT scanning to motivate people to quit smoking is quite ridiculous. But here's an anecdote that may be suggestive, I'm not sure of what. My father was a lifelong smoker until he was about 55, when a chest x-ray found a lesion which his doctor clearly thought was cancerous. It turned out to be scar tissue (too long a story to explain), but that week of waiting in terror to find out was enough for him. He quit cold turkey, and for good.

I can't think of any way to create such experiences for people, or the equivalent, unfortunately. Oh well.

Michael Siegel said...

Thanks Cervantes for the comment and the anecdote. The experience that your father had is not at all atypical. In fact, it is observations like that which led investigators to suspect that CT screening may have an effect on smoking cessation.

I should emphasize that I don't think there is any problem investigating this hypothesis and if it is true, then that is a great thing and could have a tremendous public health benefit. My only point is that the screening technique needs to be justified on its face, not based on effects on health behavior change.

Incidentally, the experience you relate differs from what is being reported in this study because your father had a suspicious lesion detected on X-ray. In his case, CT scanning was being used for a diagnosis, rather than as an initial screening test.

Cervantes said...

A major problem, of course, is that if people get a clean scan, they may say, "Great! Now I have permission to keep smoking!"

Fred Grannis MD said...

You are all wet on this commentary Mike.

Unintended side effects are an important issue in medicine. For example, this email I got today shows an unintended ill effect.

"Sildenafil use, sexual risk behavior and risk for sexually transmitted diseases, including HIV infection
Men who have sex with other men and use sildenafil are more likely to practice unsafe sex and to use illicit drugs. This is a potentially important public health problem. The American Journal of Medicine."

Unintended beneficial side effects are also very valid topics of medical publication. For example, Linda Ferry's publication of facilitation of smoking cessation as an unintended side effect of the anti-depressant bupropion has resulted in major clinical benefit in smoking cessation.

The article that you trash is one of three or four that have made this same observation. People who participate in smoking cessation have improved chances of smoking cessation. It is certainly important to point out that this data is retrospective, and that there may be biases that should be addressed in future prospective studies, but you step far over that line in unreasonable critique.

With respect to anxiety induced by screening, I am pretty sick of hearing epidemiologists harp on how bad this problem is. As a clinician who treats lung cancer, as well as an ex-smoker who is at risk of lung cancer, I think that providing information to individuals at risk that gives them some anxiety that they will get a disease unless they make a change in their lifestyle is the bedrock of all medical education and preventive medicine.

On the scale of life's misfortunes, anxiety about having lung cancer measures far down the scale from pain, dyspnea, neurological symptoms etc encountered by those 156,000 Americans who will actually die of lung cancer in the U.S this year.

Lung cancer screening will save the lives of tens of thousands of Americans each year in the very near future. If it also helps tens of thousands of smokers to quit, then so much the better for all involved.

Fred Grannis MD