I appreciate the author’s ideas but the 2 major problems I run into are:
1) Most patients still don’t get it when you go into the details of potential harm with unintuitive concepts like overdiagnosis or overtreatment. I’ve tried multiple approaches over 14 years. I take my time and am known for being a good communicator (patients and colleagues). Yet the most common patient response is either “So what should I do, doc?” or “Let’s do it, it couldn’t hurt.”
2) The elephant never acknowledged in the room is that while guidelines shouldn’t blindly rule out individualized clinical judgement, why does the later get a pass on proving net benefit? An imperfect guideline will have detailed references and explanations but still get torn apart while romanticized individualized clinical judgement gets the benefit of the doubt. Would love to hear someone with much more wisdom than me reconcile this.
Thank you for an excellent summary of the issues in screening. Personally, I paid little or no attention to guidelines over my decades of practice. The fact that they are now tied to compensation and often used for medicolegal prosecution makes them antithetical to good medical practice. Medical care is an intensely individualistic discipline and it is the duty of doctors to apply their acumen and judgement for the benefit of each individual patient. There is no such thing as "greatest good for the greatest number" in medicine.
"There is no such thing as "greatest good for the greatest number" in medicine."
But that's exactly what many doctors and Public Health professionals will argue. They for instance will tell a patient to take statins because their studies show population level benefits even if the patient is more interested in avoiding the potential side effects of statins than he is of getting their slight benefits (NNT=118 for middle-aged man to avoid hospitalization from heart attack).
That is why "public health" is an oxymoronic term. By the way, I don't care for NNT or other statistical metrics that tend to obscure the common sense results of clinical trials. I always advise those interested to just look at the real risk reduction percentages and use common sense to decide if there is any difference that is clinically significant. The statin trials are a beautiful example of this technique that is used to convince people that the drug is useful. The differences in mortality are only a couple of percentage points at most and that is in the major published studies that are commonly cited as proof of the benefits of statins---usually financed by the pharmaceutical companies and conducted by their satraps in the research community. The reports cited never mention another interesting finding in most of those studies. There was a total disconnection between "improved" outcomes and both the initial cholesterol level and the degree of cholesterol lowering attained. This means that the "benefits" of the statins were the same for those with both high and low baseline cholesterol levels as well as for those whose cholesterol declined a lot and those that declined much less. This disconnection is known as "lack of normal exposure-response" and can be interpreted as showing that the factor under investigation (cholesterol level) is not a true cause of the disorder.
I’m not a fan of guidelines, esp in recent years. Whereas they were once (perhaps in my naive opinion) truly a compendium of the state of the science, I find in the past 20 years (perhaps in my cynical opinion) that they’ve become unabashed cheerleading for industry. It doesn’t help that guideline writers are generally “subject matter experts” who also seemingly always come with conflicts of interest, the disclosure of which does nothing to mitigate them.
That said, and as you note, their purpose is directed at a population/ generic level, and by definition cannot be (anywhere close to) tailored towards individuals. That must be the job and purview of the individual clinician.
I think the atmosphere is also different, where in the US “guidelines” attain the level of commandments, with implications not only for remuneration but also for medicolegal risk management. In Canada, they remain only as suggestions, as they should be.
I think guidelines do skew the Sackett dynamic of EBM, where “scientific evidence” is but one part of the Venn diagram, and “patient values and preferences” should not be reduced to throwaway lip service.
In that way, I feel some Canadian guidelines strike a reasonable compromise. Rather than incorporating a patient’s values and preferences (which is of course impossible for it to do), instead it declares the writers’ values and preferences in making individual recommendations.
One of the harms I see in these recommended (aka "mandatory") screenings is that it prevents patients who want to opt-out of say, irradiating their breast every year (or now every other year, what an improvement!), from discussing the real issues that are plaguing them. In order to avoid having to be chastised for not getting your colonoscopy, breast irradiation squeeze, and who knows what else, we avoid seeing a doctor for things like hearing loss, brain fog, moodiness, and joint pain which may be destroying our relationships, making it hard to work, and keeping us from exercising....
While I cannot disagree with this fairly well documented argument, I am struck by the failure of this analysis, like so many others addressing this issue, to consider or even mention a major benefit of cancer screening: the not irrational reassurance that both the patient and his/her caring physician receives from a negative screening test, especially because the result of most such tests is negative. With a family history of prostate cancer, my repeatedly negative (not rising) PSAs, and admittedly uncomfortable digital reactal exams appropriately done after the blood is drawn, are far from a waste of effort and money, but a welcome annual reassurance. This substantial benefit of screening, especially of the overwhelming negative test results which contribute statistically to the anti-screening arguments, seem to me to be all too infrequently considered.
I don't see why the negative test result would be so reassuring. Take a 55 year-old man. His life expectancy is 25 years. If he gets a negative result from his PSA test, well he's probably not going to die from prostate cancer this year. Big deal. Even if the test is positive, he's not going to die from prostate cancer this year. But if the test is positive then he might end up going down a road of unnecessary cascading care. Pro-PSA advocates argue Joe Biden should have been PSA-tested even after age 70. He had a negative PSA in 2014. If he was tested after that, he probably would have gone positive around 2016. It likely would have prevented him from becoming President (not that I voted for him!) But as a patient he would not have been served by a positive PSA test worrying him and likely sacrificing his career! Now he's made it to almost 83 even with advanced prostate cancer.
Though one could argue that even though the PSA didn’t make him healthier, he greatly benefitted from a PCP who was willing to listen and empower him to decide. That is something that I think both of you can feel great about.
I picked up many of my patients from a doc who seemed to do shared decision making for PSA screening by telling people it was a useless test, and told them not to do it. It's been interesting to do my tap dance of pros and cons and see what people choose. It turns out I am very bad at predicting who will opt for it and who will not.
A further challenge for me is the 80+ year old patient who feels he needs a screening PSA. I do the whole talk, including the part about how PSAs tend naturally to rise as men get older, and we don't really know how to interpret PSAs of 83 year old men. And they still want it. It comes back slightly elevated and I send them to see a urologist who says, "It's fine, but let's keep an eye on it." They feel vindicated because "See! he said it's important to keep an eye on this!" This is not better medical care. He's not healthier; he's slightly less healthy (although he can afford the copay) and feels the hour spent at the urologist office was a good use of the hour. Certainly, he *feels* more healthy and maybe that benefit is enough to justify the test and the hour of urologist's time. (Thankfully, none of these has caused a patient to get a biopsy.)
TL;DR: I find it tremendously challenging to explain the rationale for stopping PSA screening at 70 to my patients.
Very much appreciate your time and writing. I like the idea of giving my patients all the information I have and then letting them decide but telling them no matter what their decision I will support them.
In fact, this is one of the major tenets of EBM: "Evidence is not enough". Values & preferences matter more.
I appreciate the author’s ideas but the 2 major problems I run into are:
1) Most patients still don’t get it when you go into the details of potential harm with unintuitive concepts like overdiagnosis or overtreatment. I’ve tried multiple approaches over 14 years. I take my time and am known for being a good communicator (patients and colleagues). Yet the most common patient response is either “So what should I do, doc?” or “Let’s do it, it couldn’t hurt.”
2) The elephant never acknowledged in the room is that while guidelines shouldn’t blindly rule out individualized clinical judgement, why does the later get a pass on proving net benefit? An imperfect guideline will have detailed references and explanations but still get torn apart while romanticized individualized clinical judgement gets the benefit of the doubt. Would love to hear someone with much more wisdom than me reconcile this.
Thank you for an excellent summary of the issues in screening. Personally, I paid little or no attention to guidelines over my decades of practice. The fact that they are now tied to compensation and often used for medicolegal prosecution makes them antithetical to good medical practice. Medical care is an intensely individualistic discipline and it is the duty of doctors to apply their acumen and judgement for the benefit of each individual patient. There is no such thing as "greatest good for the greatest number" in medicine.
"There is no such thing as "greatest good for the greatest number" in medicine."
But that's exactly what many doctors and Public Health professionals will argue. They for instance will tell a patient to take statins because their studies show population level benefits even if the patient is more interested in avoiding the potential side effects of statins than he is of getting their slight benefits (NNT=118 for middle-aged man to avoid hospitalization from heart attack).
That is why "public health" is an oxymoronic term. By the way, I don't care for NNT or other statistical metrics that tend to obscure the common sense results of clinical trials. I always advise those interested to just look at the real risk reduction percentages and use common sense to decide if there is any difference that is clinically significant. The statin trials are a beautiful example of this technique that is used to convince people that the drug is useful. The differences in mortality are only a couple of percentage points at most and that is in the major published studies that are commonly cited as proof of the benefits of statins---usually financed by the pharmaceutical companies and conducted by their satraps in the research community. The reports cited never mention another interesting finding in most of those studies. There was a total disconnection between "improved" outcomes and both the initial cholesterol level and the degree of cholesterol lowering attained. This means that the "benefits" of the statins were the same for those with both high and low baseline cholesterol levels as well as for those whose cholesterol declined a lot and those that declined much less. This disconnection is known as "lack of normal exposure-response" and can be interpreted as showing that the factor under investigation (cholesterol level) is not a true cause of the disorder.
I’m not a fan of guidelines, esp in recent years. Whereas they were once (perhaps in my naive opinion) truly a compendium of the state of the science, I find in the past 20 years (perhaps in my cynical opinion) that they’ve become unabashed cheerleading for industry. It doesn’t help that guideline writers are generally “subject matter experts” who also seemingly always come with conflicts of interest, the disclosure of which does nothing to mitigate them.
That said, and as you note, their purpose is directed at a population/ generic level, and by definition cannot be (anywhere close to) tailored towards individuals. That must be the job and purview of the individual clinician.
I think the atmosphere is also different, where in the US “guidelines” attain the level of commandments, with implications not only for remuneration but also for medicolegal risk management. In Canada, they remain only as suggestions, as they should be.
I think guidelines do skew the Sackett dynamic of EBM, where “scientific evidence” is but one part of the Venn diagram, and “patient values and preferences” should not be reduced to throwaway lip service.
In that way, I feel some Canadian guidelines strike a reasonable compromise. Rather than incorporating a patient’s values and preferences (which is of course impossible for it to do), instead it declares the writers’ values and preferences in making individual recommendations.
Thank you for this article.
One of the harms I see in these recommended (aka "mandatory") screenings is that it prevents patients who want to opt-out of say, irradiating their breast every year (or now every other year, what an improvement!), from discussing the real issues that are plaguing them. In order to avoid having to be chastised for not getting your colonoscopy, breast irradiation squeeze, and who knows what else, we avoid seeing a doctor for things like hearing loss, brain fog, moodiness, and joint pain which may be destroying our relationships, making it hard to work, and keeping us from exercising....
While I cannot disagree with this fairly well documented argument, I am struck by the failure of this analysis, like so many others addressing this issue, to consider or even mention a major benefit of cancer screening: the not irrational reassurance that both the patient and his/her caring physician receives from a negative screening test, especially because the result of most such tests is negative. With a family history of prostate cancer, my repeatedly negative (not rising) PSAs, and admittedly uncomfortable digital reactal exams appropriately done after the blood is drawn, are far from a waste of effort and money, but a welcome annual reassurance. This substantial benefit of screening, especially of the overwhelming negative test results which contribute statistically to the anti-screening arguments, seem to me to be all too infrequently considered.
Gregory Bulkley M.D.
I don't see why the negative test result would be so reassuring. Take a 55 year-old man. His life expectancy is 25 years. If he gets a negative result from his PSA test, well he's probably not going to die from prostate cancer this year. Big deal. Even if the test is positive, he's not going to die from prostate cancer this year. But if the test is positive then he might end up going down a road of unnecessary cascading care. Pro-PSA advocates argue Joe Biden should have been PSA-tested even after age 70. He had a negative PSA in 2014. If he was tested after that, he probably would have gone positive around 2016. It likely would have prevented him from becoming President (not that I voted for him!) But as a patient he would not have been served by a positive PSA test worrying him and likely sacrificing his career! Now he's made it to almost 83 even with advanced prostate cancer.
Thanks Mary - I, too, find that challenging!
Though one could argue that even though the PSA didn’t make him healthier, he greatly benefitted from a PCP who was willing to listen and empower him to decide. That is something that I think both of you can feel great about.
I picked up many of my patients from a doc who seemed to do shared decision making for PSA screening by telling people it was a useless test, and told them not to do it. It's been interesting to do my tap dance of pros and cons and see what people choose. It turns out I am very bad at predicting who will opt for it and who will not.
A further challenge for me is the 80+ year old patient who feels he needs a screening PSA. I do the whole talk, including the part about how PSAs tend naturally to rise as men get older, and we don't really know how to interpret PSAs of 83 year old men. And they still want it. It comes back slightly elevated and I send them to see a urologist who says, "It's fine, but let's keep an eye on it." They feel vindicated because "See! he said it's important to keep an eye on this!" This is not better medical care. He's not healthier; he's slightly less healthy (although he can afford the copay) and feels the hour spent at the urologist office was a good use of the hour. Certainly, he *feels* more healthy and maybe that benefit is enough to justify the test and the hour of urologist's time. (Thankfully, none of these has caused a patient to get a biopsy.)
TL;DR: I find it tremendously challenging to explain the rationale for stopping PSA screening at 70 to my patients.
Very much appreciate your time and writing. I like the idea of giving my patients all the information I have and then letting them decide but telling them no matter what their decision I will support them.