Thursday, January 13, 2011

4 reasons to not be screened for lung cancer

NOTE: A revised and updated version of this post, including my take on the July 29, 2013 U.S. Preventive Services Task Force draft recommendations for lung cancer screening, is available here.


Lung cancer is the leading cause of cancer death in the U.S. and claimed more than 150,000 lives last year. Since most lung cancers are triggered by tobacco use, the best way for family doctors like me to prevent lung cancer in 2011 is to counsel patients not to smoke, or if they already smoke, to quit. Unfortunately, tobacco-damaged lungs often don't heal completely, and ex-smokers continue to be at higher risk for lung and other cancers than never-smokers are.

That's why doctors used to routinely advise that heavy smokers get chest X-rays in hopes of catching cancers at more treatable stages. However, several studies have since found no difference in lung cancer death rates among smokers who got annual chest X-rays versus those who did not. That's likely because the disease is too far advanced to cure by the time it's visible on a chest X-ray.

Computed tomography (CT scans) of the chest, on the other hand, are much more powerful and can detect even smaller lung cancers than X-rays can. Does that mean those cancers will be more curable? To answer this question, several years ago researchers at the National Cancer Institute randomly assigned current and former heavy smokers between the ages of 55 and 74 to annual screening with CT scans or chest X-rays. Last November, they stopped the study early due to evidence showing that patients assigned to receive CT scans had a 20 percent lower risk of dying from lung cancer than those who got chest X-rays. Still, many cancer experts cautioned that these were preliminary results, and that until all of the data from the study are analyzed and published in a peer-reviewed medical journal, patients should not rush to their doctors to request CT scans.

But for the sake of argument, let's say that in the coming months the study's initial results are confirmed, and that screening heavy smokers or ex-smokers for lung cancer with CT scans does, in fact, reduce lung cancer deaths. And let's say that you're worried about your risk, so you visit your family doctor to discuss whether it makes sense to undergo this test. Because most health professionals believe cancer screening saves lives, they may unintentionally minimize, or neglect to discuss, a test's downsides. Screening tests have risks just like any other medical procedure, and it's important for your doctor to thoroughly review those risks with you. Here are 4 good reasons to consider not getting a chest CT scan:

1. The risk of developing cancer from the CT scan itself isn't trivial. A recent analysis published in the Archives of Internal Medicine found that a single chest CT scan exposed patients to the radiation equivalent of more than 100 chest X-rays, and that at age 60, an estimated 1 in 1000 women or 1 in 2000 men would eventually develop cancer from that single scan. (Participants in the lung cancer screening study actually underwent three consecutive annual CT scans.)

2. False alarms are extremely common. In the NCI's lung cancer screening study, researchers found that 1 in 3 patients had at least one false-positive result after undergoing two CT scans. Of those patients, 1 in 14 needed an invasive lung biopsy to be sure they were cancer-free.

3. Even if screening catches lung cancer early, there's no guarantee your prognosis will be better. This is due to "overdiagnosis," or the unnecessary diagnosis of a condition (typically cancer) that will never cause symptoms in a patient's lifetime, either because it's so slow-growing or the patient dies from some other cause. An estimated 1 in 3 breast cancers detected by mammograms is overdiagnosed, and a 2007 study published in the journal Radiology suggested that the proportion of lung cancers overdiagnosed by CT scans could be as high or higher, especially in women. But because there's no way of knowing at the time of diagnosis if a lung cancer will be fatal, inevitably many patients will be needlessly subjected to the side effects of treatment.

4. Finally, it's highly likely that a CT scan for lung cancer will find some other abnormality that will require further investigation. You might think this is a good thing, but studies show that most of these abnormalities turn out to be false alarms, too. In fact, in 2008 the federally-supported U.S. Preventive Services Task Force decided against endorsing CT screening for colorectal cancer due to concerns that investigating all the abnormalities that CT scans turn up could outweigh the cancer-prevention benefits. You may still decide it's worth it for you or a loved one to get a CT scan for the relatively small (about 1 in 300) chance it will prevent death from lung cancer. But before you do, be sure you have all the facts.


The above post was first published on my Healthcare Headaches blog at


  1. Great post with important information. The more we read about the false positives and risks of testing, the better chance that a patient will be informed enough to make their best decisions. Thanks, Dr Lin.

  2. Congratulations Dr Kenneth for your excellent reasoning.
    What is your mammogram recommendation for a low risk 50 yo woman?
    Dr Salomon

  3. Are you kidding me? The title of your blog post alone is incredibly irresponsible. For those people who are in high risk populations, a CT scan can save their lives. And with low-dose CT scanners, the radiation controversy should be put to rest very soon.

    I am all for common sense, but knowing the statistics and the all-too-many people are initially misdiagnosed, your overdiagnosis argument holds little water.

    I would love for you to have a good discussion with founder of the Bonnie J. Addario Lung Cancer Foundation to change your mind and bring balance to the points in this post.

    Trust me, if it were you or one of your family members, you would have the CT done.

    Nicolle Foland

  4. Overdiagnosis of cancer is a real and well-documented phenomenon, backed by many scientific studies. The problem is, on an individual level it's impossible to know if one's life is actually "saved" by cancer screening, since there's no way to compare the outcome with what would have happened in the absence of the test. Obviously, anyone who has a cancer detected via a screening test (be it a chest CT scan, mammogram, PSA test, etc.) who survives treatment for the cnacer will believe that his or her life was "saved." But the statistics from the National Lung Screening Study say otherwise; as I wrote, the number of people needed to screen to prevent one lung cancer death was 300. Which is why I don't agree with your view that this post is "irresponsible"; rather, it provides a more balanced view of the benefits and harms that is missing in far too many screening discussions.

  5. I can appreciate your parsing of words, however, I do not agree that Lung Cancer is over-diagnosed; rather that it is mis- and under-diagnosed precisely because there is no early detection method. Until research is funded to find an early detection method, a CT is the best chance we have to find it earlier, and therefore improving chances for survival. If I could have my best friend, who was diagnosed at stage IV in April of 2009, on this earth for five more years, while you might not call that a life saved, I would call it the most precious gift I could receive.

  6. I agree with Nicolle. My Mother passed away from lung cancer November 1, 2010 after exposure to radon in her home...not from smoking.

  7. So, let me put it this don't believe lung cancer mortality rates would decrease if there were an early detection screening protocol? Early detection screening, which lung cancer STILL does NOT have, can certainly save lives...and/or definitely prolong them. My mom was diagnosed on a fluke. No, she didn't smoke! Luckily, she is still here. Had she been diagnosed during a routine screening test, perhaps the cancer would have been caught much earlier and she wouldn't be terminal now. I WILL have a CT done myself sooner, rather than later, to potentially "save" my own life. As a doctor, I would think your #1 priority would be to try to save lives by doing more than preaching about smoking. What are you doing to help the 80% of new lung cancer cases that are diagnosed in people who have never smoked or quit decades ago??? Lung cancer cannot be stopped by smoking cessation alone. According to the Lung Cancer Alliance, lung cancer in never smokers is the 6th leading cause of cancer deaths. Smokers or not, early detection and legislation to make lung cancer a national priority is a social justice issue. It's still the #1 cancer killer and the LEAST funded. The 5 year survival rate (15%)hasn't changed in more than 40 years. That is absolutely unacceptable. We literally have a public health epidemic on our hands and you believe your best counsel to prevent lung cancer in 2011 is to simply tell your patients to stop smoking? Smoking is only part of the equation. Please tell me you don't dismiss any of your patients who may be experiencing lung cancer symptoms because they "don't fit the profile?" Please tell me you do, in fact, further investigate and recommend a scan in necessary situations? At the very least, please tell me this. I am very disturbed by your approach and perception of lung cancer patients.
    Jennifer Windrum

  8. I am enormously sympathetic to people who have had non-smoking friends or relatives suffer and/or die from lung cancer. One of my friends died from lung cancer in his early 30s, having never touched a cigarette in his life. But these stories don't change the fact that the evidence that screening current or past smokers for lung cancer improves outcomes is weak, and that for screening non-smokers nonexistent. Consider what Dr. Len, deputy chief medical officer for the American Cancer Society, had to say about the National Lung Screening Trial:

    "This study does nothing to advance our treatment of lung cancer, which remains far from satisfactory. And it does nothing for people who are not smokers and develop lung cancer. And it doesn't say anything to us about people at lower risk of lung cancer, such as everyone who does not meet the criteria outlined for the study.

    So if you are a younger person, who has "only" smoked a pack a day for 20 years, this study tells you nothing. And if you are a non-smoker who works in a heavy smoking environment, such as a bar in some communities that don't have smoke-free laws, this study says nothing to you. ...

    The American Cancer Society does not have a formal recommendation at this time based on the results of this study. Remember, this is a very preliminary report and more information will be forthcoming. As will others, we will carefully examine the information once it becomes available and at that time will probably say something more definite. But for now we just don't have enough information.

    Having said that, if you are someone who meets the criteria similar to people who participated in this study, then you may want to consider getting a scan. But before you do that, it would be wise to have a conversation with your health professional and consider whether screening is right for you. You should especially consider your current health status and answer the question if something is found, would you be willing and able to get treated with surgery, radiation therapy, and chemotherapy, either alone or in combination?

    You should also be aware that there is a pretty good chance that somewhere along the line you may find a lesion in the lung that is not cancer. It has happened to me, and I am a non-smoker--so I know what the experience can be like, including all of the "follow-on" tests that would be recommended."

  9. In November 2009, the U.S. Preventive Services Task Force recommended that women age 40-49 NOT be routinely screened by mammography for breast cancer. Their concern was overdiagnosis, false positives leading to unnecessary biopsies, undue psychological stress for the patient, and the difficulty of seeing the "polar bear in a snowstorm" (cancer in dense breast tissue). There was quite an uproar about this recommendation, similar to some of the comments opposing your point of view. But mammography hasn't been withheld from women of this age group as a result. Apparently there is consensus that mammography's benefits outweigh the risks, and the general population of women, regardless of risk factors, is routinely screened.

    As I read your post, I see the same concerns that USPSTF expressed regarding mammography. So why, for the #1 cancer killer, shouldn't we screen people with high risk factors? The survival rates for people diagnosed in early stage lung cancers is nearly 50%, compared to the meager 3% that survive a late stage diagnosis. As a former smoker, shouldn’t I be given the opportunity to be screened for cancer, when the benefits outweigh the risks?

    I have been annually screened for breast cancer with mammography and ultrasound since I was 30 years old, with no risk factors, and I have NEVER been diagnosed with breast cancer. I have NEVER been screened for lung cancer, with risk factors, and I have been diagnosed, late stage.

    It's time to begin screening high risk individuals annually for lung cancer.

  10. I am a pulmonologist at a large academic center, and participated in the NLST. While many will chastise you, most of your points are valid, and highlight the nuances of "screening" that most just don't or won't try to understand.

    However, you referenced a study that suggests a 1 in 1000 lifetime risk of cancer from a CT scan, but this was a deeply flawed study, basing risk on data taken from large exposures (assuming incorrectly that the risk is linear AND can be extrapolated to much lower levels of ISN'T, and it CAN'T). The dose of radiation from a CT as done in the NLST is considerably lower (by a factor of 5-10) than 100 chest X-rays.

  11. We all have anecdotal evidence for both "sides" of the discussion. That is why it is so important to truly study low-dose screening CT for lung cancer.
    The NLST was carried out in a high risk population of 55-74 year old current or former smokers without a standard diagnostic or treatment protocol. A large scale demonstration screening program is required to develop and implement the most effective, safe and efficient risk assessment tools and diagnostic and treatment protocols.
    As for the ACS, what do they offer to lung cancer patients (only 20% of whom are smokers) other than smoking cessation?

  12. Risk of Cancer from the low dose CT? A minimal risk if any in my opinion. Biopsies and second opinions should help eliminate over diagnosis. I can see insurance companies balking at he idea. If anything, the misdiagnoses might scare the addiction out of the smoker.

  13. Ok, so I'm to understand that this type of screening shouldn't be done because it is more dangerous than the cancer itself? That even if a person is screened, and they have lung cancer, they are going to die anyway, so where is the benefit? I suppose these studies are really leaving things up to God. No help, no intervention, no treatment. Just walk around until you drop dead from an undiagnosed cancer, or even another disease or condition, as these tests have shown to be very helpful in diagnosing and treatment follow-up. Ask any cancer survivor if they would like to go without these tests after a course of treatment, not knowing if the treatment worked. I'll bet they will choose the risk of developing cancer from the scan versus walking around not knowing.
    Just think of the savings, also! If one doesn't know one has lung cancer, then one would not require treatment, except for perhaps a week or so of pain meds when dying.
    Instead of funding studies like these, the money would be more humanely spent on research for earlier detection, better therapies and may I be so bold as to utter the word cure! I hope the humane medical world ignores this study and continues to do what they took an oath for.
    I believe that the decision should belong to the patient as to what process they choose to follow once diagnosed. Some may choose to forgo treatment and live out the remainder of their life as best they can. Some may choose to use every means available to prolong their life, and once again dare to hope for a miracle.
    I am beginning to suspect that the "stigma" of smoking and lung cancer is not what really lies at the bottom of this. Oh, it has been a "just" excuse for many years, for many. We've all seen the stats on how much it costs to diagnose and treat someone with lung cancer. And the stats on how lung cancer is the worst killer among cancers. So what if we say it's not safe to screen for early detection, and the insurance companies agree. No early detection, no long term treatments. Imagine the savings! Where do business' cut back when times are tough? They cut back on the higher cost items.


  15. The following is a report by one of the sites conducting the NLCST. This takes the discussion to the source.

    Cincinnati—Computed tomography (CT) scans are an effective lung cancer screening tool for a high-risk patient population, if done with a strict clinical protocol in place and with the input of a multidisciplinary care team, according to University of Cincinnati (UC) research.

    Led by Sandra Starnes, MD, the Cincinnati-based clinical trial was conducted to determine if lung cancer screening could be done effectively among a high-risk population living in a geographic area with rates of histoplasmosis three times higher than the national average.

    Histoplasmosis is a fungal infection that enters the body through the lungs and increases the likelihood of lung nodules. These irregular sections of lung tissue can look like lung cancer on CT and lead to unnecessary biopsies.

    "Despite having a 60 percent nodule rate, we were able to avoid doing benign biopsies and not miss any lung cancer diagnoses if the protocol was strictly followed. No one was diagnosed at a stage where the lung tumor could not be surgically removed,” says Starnes, director of thoracic surgery at the UC College of Medicine and a surgeon with UC Health.

    Starnes and her colleagues report their findings online ahead of print in the Journal of Thoracic and Cardiovascular Surgery.

    "The Ohio River Valley’s high prevalence of histoplasmosis makes lung cancer screening very challenging because so many nodules show up on the imaging studies. By using three-dimensional CT imaging, we are able to look more closely at the anatomical structure of these nodules and identify concerning lesions,” explains Starnes.

  16. dr. lin, i hope none of your patients take your advice. i am a never smoker who has stage iv lung cancer. i never had a respiratory symptom. if people know they are at risk due to a smoking history, or second hand smoke history they need to be screened. and, we need to find a way to screen those of us who have never smoked so that our lung cancer can be found early and be curable

  17. It is very regrettable that physicians (a group to which I belong) persist in the very damaging use of 'survival times' as a measure of success in screening for either lung or breast cancer.
    Why is it damaging?..because it gives people a completely false hope. The very phrase "survival time" indicates that at the end of that time death will occur and it is vitally important to realise that it will occur on the day or month it would have occurred without the 'early' diagnosis, the only difference being that the victim knows much earlier on that he/she has cancer and carries that burden of knowledge from then until death supervenes. No imaging test that only identifies the morphology (size and shape) of a tumor will ever be able to reduce the death rate
    from lung cancer do that requires knowledge (by imaging or othwrwise) of the molecular behaviour of tumors..fortunately a huge amount of research is being done in this area and there is good reason to hope that diagnoses can then be made at a stage so early in the tumors growth that it can indeed be "cured"...Scarce resources should not be wasted on CT screening (a huge expense) but should be concentrated on finding the earliest possible molecular/biochemical changes...There IS hope and evidence to support this hope, that death from lung cancer will be defeated by this approach..but certainly not by increasing "survival times".

  18. Of the many people that I have known that had cancer all but a couple died. In almost every case the doctors said the cancer was completely gone only to find out months/years latercancer spread throughout their bodies. I think we certainly have the ability to keep many cancer patients alive much longer. I say that because my father went to 2 hospitals that gave him less than 30 days to survive because he had late stage liver cancer. A week after being sent home to die he received a call from a doctor doing experimental cancer treatments in Iowa City, IA. They said his family doctor noticed he had insurance that covered experimental treatment. He was given a pill monthly until he finally died of liver cancer almost 10 years later. He felt bad that so many people died because they didn't have that insurance. Why aren't cancer patients instructed on eating healthy Foods and more emphasis pit on what caused the cancer? Commom sense tells me it's that same reasons beef producer wouldn't tell you to only eat chicken. It's a business right? I hate being negative, but for 30 years I held many family and friends hands while they died. I know Doctor's only know what they are taught and most truly wish they could help, and treatment they are allowed to perform on their patients. I certainly understand what Doctor Kenneth is saying and appreciate his honesty. I've been through the multiple tests because of a false read and the scare that goes with it. Never got a sorry, just a large bill and the words, congrats your results have all come back good. There are treatments for cancer. Never heard of a cure.