You might have heard the recent news reports about fancy new CT scans for lung cancer. It was widely reported that these scans reduce the rate of lung cancer death among heavy smokers. There are lots of problems with these reports, but a problem of missing numbers particularly irks me. I'll focus here today on the most important missing number, and the very odd view of health implied by this blunder.
The reports clearly state that the smokers who were screened with the three-dimensional computed tomography (CT) were somewhat less likely to die from lung cancer. To be precise, there were 20 percent fewer deaths from lung cancer, a percentage which, by the way, is very hard for most people to evaluate.
What's missing? The much more important number is the total percentage of screened participants who died (compared to the control group), regardless of why they died. Epidemiologists call this all-cause mortality. That is, did the CT increase longevity or just reduce death from lung cancer?
Some people were saved from lung cancer death with the screening, but they soon died (at the same age) from stomach cancer or liver failure instead. Other people had strokes and heart attacks, and they too suffered and died at the same age that they would have succumbed to lung cancer.
Digging deeper, I found some suggestion that deaths from all causes declined slightly (about 7 percent) in the lung screening study. But we don't yet know whether this is good or bad. If the overall death rate decreased due to the discovery in the CT scans of life-threatening conditions, then that's great. But if deaths from say stroke and liver failure are also lower in those who had their lungs screened, then something is fishy about the results. It may be that those in the screened group differ in other ways from the comparison group. Perhaps something other than their lung scanning is really protecting their health.
The purpose of cancer screening is to promote health and long life. So why don't we hear more about whether those screened stayed healthier and lived longer overall? Why the narrow focus on one cause of death? Well, the study is being done by experts on lung cancer and funded by those interested in treating lung cancer. So, if fewer people die from lung cancer, they consider it a great success. Strokes and livers are someone else's bailiwick.
But there's an additional problem, that's harder to recognize. Another reason for the narrow focus on one disease at a time is a constricted and misleading view of health. The view is that health is the absence of disease--in this case the absence (or cure) of lung cancer. And the solution therefore is to screen for disease after disease, and voilà you're healthy! This is great in those few cases (like pap smears for cervical cancer) where it involves a low-cost, low-risk test for a life-threatening condition, but it turns out to be a terrible way to think about staying healthy.
In my own research, we always try to take into account length of life and the different aspects of being healthy and productive. Improving the treatment for people with lung cancer is obviously important, but the missing numbers in cancer screening reports are often the ones that really deserve our attention.
Copyright © 2010 Howard S. Friedman, all rights reserved.
