CT Scans for Lung Cancer Screening—Who Should Decide?


What's riskier - a CT scan or a federal bureaucracy?

The Washington Post has a front page article today about the "debate" over using CT scans to detect the early stages of lung cancer. As the Post notes a recent study…

…found that screening certain heavy smokers and ex-smokers could slash their chances of dying from lung cancer. The finding was hailed as one of the most important advances in decades toward reducing the toll from the nation's leading cancer killer…. the scans could slash the death rate by a stunning 20 percent compared with old-fashioned chest X-rays.

So who might benefit? 

More than 222,000 Americans receive lung cancer diagnoses each year, and more than 157,000 die from the disease — more than from cancers of the breast, colon and prostate combined. Lung cancer has remained notoriously difficult to treat, in part because it is often diagnosed too late.

There are about 100 million current and former smokers in the United States, all of whom are at increased risk. Many more might be prone to the disease because of family history or exposure to substances such as radon and asbestos.

CT scans create three-dimensional images of the lungs, instead of the two-dimensional perspectives captured by chest X-rays. Scans are more likely to spot small tumors, boosting the chances of survival.

So what's the debate all about? Costs and side effects. First, the side effects. Almost any medical diagnostic test can turn up false positives which result in further interventions which themselves pose risks to patients. In this case, the Post observes:

The scans produced false alarms in about 40 percent of cases in the study. While screening saved 88 lives among the trial participants, 16 patients died from apparent complications from follow-up procedures, including six who did not have cancer.

Let's just say that that is not an insignificant rate of false positives and deleterious side effects. Of course, smart people in the government are now mulling over whether government health care programs should pay for the scans which also affects the decisions made by many private insurers. However, the Post does report the right answer: 

"We've always said it's a personal decision an individual needs to make," said Claudia Henschke, a professor of radiology at Mount Sinai Medical Center in New York.

Well, yes. Individuals worried about their risks should certainly be allowed to pay for the scans out-of-pocket. In addition, if we had a health insurance market that was not hamstrung by regulations and mandates, competitive private policies might well offer to cover such scans. Of course, people dependent on government health insurance will have to live (or die) by the decisions of those smart folks at the Centers for Medicare and Medicaid Services. 

Disclosure: After smoking for about ten years , I quit my 3-pack a day habit 25 years ago. I might want to have a scan sometime in the future. 

NEXT: Reason Morning Links: Pillbilly Crackdown, Defunding the IDF, Ron Paul's Media Blackout, Rick Perry's Crony Capitalism

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. I don’t believe you smoked 3 packs a day for 10 years. It takes a while to get up to 3 packs a day. And to have quickly gotten up to that point and suddenly quit so quickly…yea right.

    1. All it takes is the ability to smoke at your desk and long hourse of work or study.

    2. tjones: In the good ole days, I would smoke my first cigarette before I got out of bed. One could smoke as much as one wanted at one’s desk and in classrooms. Cigarette machines used to dispense a pack of matches (20) to go with your pack of cigs. I estimate that I would have about 10 matches leftover after finishing each pack of cigs.

      In other news, my genotype screening tests suggest that I carry gene variants that put me at higher risk of lung cancer from smoking.

      1. I smoked up to a pack a day in high school (ah, those were the days) and into college. I quit in college (can’t even remember why) by . . . quitting. Basically, I just stopped buying cigarettes. I don’t recall any particular hardship.

        But 60 cigs a day – youch. Any trouble quitting, Ron?

        1. RCD: Smoking was a sublime bad habit and yes, it took a while to quit. I had smoking dreams for many years afterwards.

  2. Can’t let smokers get these CT scans, they’re the one of only two revenue positive demographics. Black men can’t carry everyone else by themselves. They need smokers.

    1. Black men can’t carry everyone else by themselves.

      I dunno, they look pretty strong.

  3. How about we let the fuckers who decided to smoke suck on it, while non-smokers get to go the front of the line?

    1. I don’t think that there is any shortage of CT scanners in the US. How about we let people pay out of pocket for any services they want that are available?

      And why should smoking be treated differently from any other risky activity? If you ski, you are more likely to break your leg or dislocate your shoulder. How about the non-skiers go to the front of the line for x-rays, casts and surgery?

  4. The scans produced false alarms in about 40 percent of cases in the study.

    Was that “40% of positives were wrong”?

    This sounds like another scam to terrorize the rubes into paying for expensive yet unreliable “magic”.

    “The doctor can’t speak with you right now, he’s on the phone with the Jaguar dealer.”

    1. Yes, that means the scan uncovered something, which in radiologist-speak gets reported as “opacity of uncertain significance, may represent artifact but cannot exclude early malignancy. Clinical correlation and follow-up recommended.” In other words, your doctor gets a report that says there might be a cancer there, which means we will certainly recommend follow-up procedures which might include a biopsy of the suspicious area. You can’t fault the physician for ordering those follow-up tests, as if we miss an early cancer because we didn’t order further investigation in today’s legal environment, we might as well just start writing the checks to the malpractice attorney. That’s the problem with radiologic scans. The “abnormality” may not actually exist, but there’s no way to know for sure. There is no magic scan. That’s why this technique is fraught with peril.

      1. CP: With regard to CT scan cancer risk, according to ScienceDaily:

        “We found that while most patients accrue small cumulative cancer risks, 7 percent of the patients in our study had enough recurrent CT imaging to raise their estimated cancer risk by 1 percent or more above baseline levels,” said Aaron Sodickson, M.D., Ph.D., assistant director of Emergency Radiology at Brigham and Women’s Hospital and researcher at the Center for Evidence-Based Imaging in Boston.

        One percent above baseline? I may be misinterpreting, but isn’t baseline for men a 44 percent lifetime risk of cancer and for women a 37 percent lifetime risk?

        1. It has been established that the risks of actually causing cancer through repeated CT scans is relatively small, but it has not been studied specifically in a high risk group for cancer such as smokers, at least not that I have seen. The studies also have not had sufficient time (CT scanning is a relatively new technology) to truly estimate the risks. At best, the risk increase given is a guess based on currently available data.

        2. “One percent above baseline? I may be misinterpreting, but isn’t baseline for men a 44 percent lifetime risk of cancer and for women a 37 percent lifetime risk?”

          Your lifetime risk of developing cancer is approximately 50% for men and one in three for women. For lung cancer specifically (for all patients), the risk is 8% in men and 6% for women, according to the American Cancer Society. In people with a history of smoking, the risk is between 10-15%, depending on what source is consulted. Increasing risk from 10% to 11% or 15% to 16% is a significant increase, particularly with the risk of death or debilitating complications from false positive prompted procedures, not to mention contrast exposure. The fact that the lives saved by the scans are at this point largely conjecture makes the risk/reward calculation highly difficult at this point until more research is available.

          1. CP: Why not give patients the relative risk information and let them decide which risks they want to take?

            1. Absolutely. If someone wants to get a CT scan every year and is aware of the risks, they should be able to do it. The evidence, however, is not strong that it is of enough benefit to justify it being covered by private (or especially public) insurance programs.

        3. Perhaps a better way to assess the radiation risk from X-ray procedures is to note that there is approximately 1 excess cancer per 1000 millirem per 10,000 exposures. (These units are no longer used but grad school was a long time ago and I’m not gonna do the conversion now.) A thoracic or abdominal x-ray with barium will give a whole body equivalent dose of about 250-300 mrem, a cranial ct scan about 1100 mrem. So assuming a thoracic ct scan delivers a dose of about 500 millirem, you can expect 1 extra cancer for every 20,000 exposures.

          Putting this in a personal context it means that you could get a CT scan every day for more than 50 years before developing a cancer (with some dose rate provisos that I won’t go into). This doesn’t seem to bad but in an epidemiological context with, say, 1 million exposures, you would expect 50 cancers. That sounds much worse than 50 years cancer free. So CT scans are not free and some people will die but more will live. You choose.

          1. Maybe they’ll live, maybe not. You haven’t accounted for the risks of the subsequent procedures to false positive tests (of which there were quite a few in the cited study), morbidity from IV contrast dye exposure, and the lack of definite evidence that earlier diagnosis would save the lives of a substantial number of patients. As I said in an earlier post, lots of variables at play here.

  5. Having reviewed the study in question, the conclusions the Post article draws are somewhat dubious. The scans produce a large number of false positives (leading to biopsies and other tests, all of which are costly and several of which killed patients in this study). The increased risk of cancer from repeated CT scans is already well documented and is of particular interest in this population, as they are already prone to developing malignancy. The scans could reduce the death rate by enabling earlier detection, but its also possible that might not make any statistically significant difference, given the increased risk of cancer from the scans, the deaths from procedures resulting from false positives, and so on. The matter needs long term study if any meaningful conclusions are to be drawn. In the meantime, people should certainly be able to get a CT scan of their lungs any time they desire, provided they understand the risks of radiation exposure and the potential complications of contrast use…provided they can pay for them. Asking the insurance market (i.e. the rest of us) to bear the cost of what essentially is an elective scan, particulary since the cancer it is designed to detect is largely self-inflicted through smoking, is ridiculous.

    One of the most frustrating things about being a physician is patients showing up demanding some treatment because “a new study” was reported by the news media. Many of the studies published are shoddy at best, garbage at worst. Many of them are ultimately completely discredited. Unfortunately, that doesn’t help the patients who took them at face value and may have actually harmed themselves.

    1. How small is the set of people helped that accidental surgical death and the CT cancer risk of going in once a year can obviate them?

      1. That’s the problem. Nobody knows because no long term trial has been done. I don’t know how many people might be harmed as a result of recurring CT scans versus lives saved by early detection. There are a lot of confounding variables here and careful study is needed before I could in good conscience recommend this to a patient. In any event, the complete lack of long term study is a big reason why insurance should not be paying for these scans. Even if the scans prove effective at reliably detecting early cancer with no risk, does a six month (or however long) head start really improve long term survival? Nobody knows yet.

        1. CP and BL: See my post above with regard to CT scan risk.

  6. Why would I care if heavy smokers live or die?

    1. Because your tax money is paying for their long, drawn-out deaths from lung cancer.

      1. Lung cancer is actually pretty cheap. It kills quick and early, and is probably a net savings for the public fisc, since it saves SocSec and extended Medicare benefits.

        1. Perversely, the most patriotic thing you could do is work all your life, smoke like a chimney, and drop dead at 64.

          1. Perversely, the most patriotic thing you could do is work all your life, smoke like a chimney, and drop dead at 64.

            In a welfare based economy, absolutely. You want very small windows of non-productivity.

            Get ’em working early, and dying as quickly after retirement as possible.

        2. “”It kills quick and early,””

          So it does. I have a friend that was diagnosed with lung cancer the end of June, died in July.

      2. So it is in our interest for them to go undiagnosed as long as possible – preferably until they drop dead.

    2. Maybe you know some of them? Maybe because nice, decent people care when other people die? You are certainly not obliged to care, but there are many reasons why you might.

      1. Nope.

  7. 40 percent false positive rate? What’s the matter, was the cancer detecting coin booked that day?

  8. It would seem to me that for the moment a 40% false positive rate is unacceptably high given the consequences of such a false positive. I would hope there’s something being developed that could bring that down.

  9. Yes, a 40% false-positive rate for this type of study is very high. Keeping in mind that working up any positive requires the physician to aspirate tissue out of the lung, which carries a high risk of collapsing it, I’d want more certainty.

    Plus, consider the source. Many of them do great work, but radiologists are some of the most recklessly interventionist doctors out there.

    1. Radiologists are some of the most exposed doctors out the for malpractice suits. They are ridiculously risk averse for that reason.

      Risk aversion = high false positives.

  10. Whatever. It’s been almost a year since I quit smoking – two packs a day (plus). I’m not too concerned about the C – at almost 50, I’m gonna die of something fairly soon anyway. So no scans for me – I’ve had enough from all the other shit that’s gone wrong through the years that I’m surprised I don’t glow at night.

    But I suuuuuuuure do feel a whole lot better when I wake up in the morning NOT hacking a lung out, which is why I quit in the first place. Cause I LIKED smoking. I just like breathing better. Quality of life – improved immediately. Therefore, I highly recommend that my friends who smoke quit, but it’s ultimately your choice…it’s a free country (kind of, for a while longer).

    That is all.

  11. How much more carcinogenic are CT scans compared to a chest X ray? Does a chest X ray ever give false postives? At what rate?

    1. Imhotep: Excellent question. Reuters reports:

      With chest X-rays, there was a 9 percent chance of a false-positive after one test, and a 15 percent chance after two, the researchers report in the Annals of Internal Medicine.

      1. This is true, but the false positive chest x-ray is typically followed by a CT scan, which will resolve many of the false positives.

    2. xrays are far less carcinogenic (on the order of hundreds to thousands of times less I think).
      X-rays have much smaller false positive rates, but also much higher false negatives (not a sensitive screen).
      In short: less radiation, less visualization

  12. Why not use MRI instead?

    1. MRI still costs more than CT scans, for one thing.

    2. Because the image is of your lungs, and you have to breathe. MRI scans take a long time, on the order of half and hour to an hour, whereas CT scans can be done in seconds. As a result, you can hold your breathe through a CT scan for a high-resolution image, but you can’t for an MRI.

  13. Courtesy New York State Health Dept.


  14. Sounds like a pretty good plan to me dude.


  15. Suggest there are too many false positives from mammograms, and the lobby for a Ribbon Bully Disease will go absolutely nuts.

    But smokers are one of the groups it’s acceptable to be bigoted against. It’s OK for obituaries to say somebody was a heavy smoker and died of lung cancer, but could you imagine an obituary saying somebody died of AIDS and liked engaging in promiscuous unprotected anal sex with gay men?

  16. New study shows a 15% rate of pneumothorax in folks getting lung nodules biopsied ? 6% requiring a chest tube.
    Multiply times the 95% false positive rate of 25,000 scans (or 50,000,000 if it’s rolled out?) Number needed to screen of 320 in a high-high-risk population, which number will go way higher when lower-risk people are screened as would seem likely.
    So, I’m not recommending my ex-heavy smoking family member get a chest CT. At least not till things are more clear. Would you?

Please to post comments

Comments are closed.