Here is Dr Wil Ward on the downsides of anecdote as data, He uses the LifeVest wearable cardiac defibrillator and VEST trial as an example. This is really good. JMM
Thank you for the article, but I find the part on SSRIs a bit catchy as the statement clearly says *can*,
and there is no doubt those drugs can save some childrens, and induce suicidal ideations in others (which is, let me point, different from death by suicide)
At a prior hospital system I spent lots of my time explaining to inpatients with newly diagnosed heart failure why they would not benefit from a Lifevest., how uncomfortable it can be, and how cost to them was unpredictable. These kinds of conversations are far more dificult than ones where a gizmo or device is recommended. Unfortunately, various members of the health care team had been brainwashed by LifeVest reps and had told my patients they should get the vest in order to save their lives. Occasionally, one of my patients would have the LifeVest process initiated because I was off on the day they were discharged. Invariably, this resulted in 3 months of hassle, inappropriate shocks and discomfort.
At this point in my career, I’m no longer surprised when things that *should* work don’t actually work when tested.
Like Mary said earlier, I sympathize with the cardiologist. It’s still really hard for doctors to get past the mechanistic thinking that’s drilled into us in training. But the reality is that humans are, by nature, unpredictable systems.
Excellent article. Mechanism, theory, and biologic plausibility are great…but they will never be a replacement for outcome evidence. To rely on “mechanism” when the outcome trial failed is inexcusable.
It is true that trials provide information on average effects, and it is always with some uncertainty when we try to apply that evidence to individual patients. But that comes from positive trials and the limits of external validity. To assert that treatment X will work on a patient when the trial failed to show a benefit, is completely without scientific basis.
Excellent article. I think some valuable clues to proper analysis of other therapeutic interventions are brought up here. The figures for all cause mortality (3.1% v 4.9%) were said to be statistically significant with a p=0.04 but not significant after "multiple comparison adjustment". Arrhythmic mortality was 1.6% v 2.4% with p=0.18. Actual risk reduction was 1.8% and 0.8% respectively and relative risk reduction about 36 and 33%. Obviously there are a lot of problems trying to draw accurate conclusions from data with very low endpoint incidence. But much of the data on other therapeutic modalities show only slight differences in magnitude of effects and are often presented with only the RRR figures even in some articles on Sensible Medicine. Some have even called these small differences robust. I have in mind, particularly, the numbers on statin therapy which are used to justify prescription of statins to a large percentage of the population. In my opinion, the only firm conclusion we can draw from these kinds of figures is that the people that populate the "official" medical policy boards are bought and paid for by the medical device and pharmaceutical companies.
I suspect physician non-compliance with published “guidelines”, particularly in a patient population characterized by sudden death, raises medical-legal and lawsuit concerns within physicians. Guidelines don’t just “inform”, they can also be inappropriately coercive.
I can sympathize with the cardiologist for encouraging the vest's usage (not for exaggerating its benefits or for over-ruling the patient's no, however). It is hard when something that we feel in our guts *should* work doesn't work. Especially if the cardiologist feels like he can point to specific patients that he thinks were helped (they got shocked and lived).
Medicine is hard! I want to be a rocket scientist instead! I think it might be easier.
So, what we’ve learned is that the cardiologist was a domineering a*****e, who ignored patient wishes. Wondering if he’s coincidentally a paid ‘consultant’ for LifeVest Inc. Patients don’t have to make decisions with which we agree. They make decisions, and it is NOT our place to ‘sell’ them on The Correct Choice. It is our responsibility to inform, advise, and intervene to the extent they choose. Period, end.
“More more more, longer, harder, because I Am The Expert” is no way to practice medicine. And yet, it has become pervasive. Industry wins. Society loses.
I think a fundamental issue and something that many die-hard evidence-based medicine types seem to forget is that there is a difference between the population information afforded by the RCT versus the individual. The RCT shows that in the vast majority of cases the VEST does nothing and isn't worthwhile. However, it is very likely that there is a subset of patients where the VEST may safe their life. The key, and this is where medicine is as much an Art as a Science, is knowing which patients will benefit and which will not. Juts blindly following protocols is simply bad medicine.
The same goes for Wil's comments regarding screening colonoscopy. Sure the NORDISK RCT showed no benefit for all cause mortality but it is obvious that in an individual with polyps, the colonoscopy coupled with appropriate follow-up colonoscopy will likely save their life or at least prevent them from dying with colon cancer. To put this in perspective. Let us say that patient X has colonoscopies every 5 years until after his 90th birthday and each time his/her colon was "clean" (i.e. no evidence of any disease). Then clearly that patient will not have benefited from all those colonoscopies and will have been subject to the risk of any invasive procedure and anesthesia. But let us say that patient Y comes in for a colonoscopy at age 45 and a number of polyps are found, then removal of those polyps will prevent his death/her death from colon cancer. What the procedure what do, of course, is prevent their death from heart disease or a car accident!
Really like Wil Ward’s guest columns! Would love to get one about doing consults (family med or internal med I guess?) on a psych ward and being unsettled by what he found (a comment he posted on Twitter). There’s such a lack of caution among psych “specialists” I feel like their prescribing is less trustworthy than PCPs.
Hear! Hear! As disinterested and scientifically honest practitioners, we should maintain skepticism on all interventions. --Not nihilism, but make the proponents prove their recommendations with convincing data.
The single underlying dynamic in this and the other examples in this article is the "More is better bias". This pervasive attitude is promoted by industry, their paid consultants/researchers, medical editorialists, the lay press, and general public. Two recent editorials in the Wall Street Journal are illustrative: They were critical of Marty Makary and Vinay Prasad because they didn't approve as many drug applications as the previous administration. And they were slower in making decisions. Did it occur to the editorial board that some drugs fail to make their case? And perhaps due consideration takes more time than a rubber stamp.
Wow--very scary indeed! Instead of using a LifeVest, I recommend taking N-acetylcysteine, D-ribose, omega-3 fatty acids, and other targeted supplements.
Thank you for the article, but I find the part on SSRIs a bit catchy as the statement clearly says *can*,
and there is no doubt those drugs can save some childrens, and induce suicidal ideations in others (which is, let me point, different from death by suicide)
Cardiology doing cardiology things; what else is new
This reminds me of when Vinay Prasad brought up the ludicrous study of offering lung cancer screening CTs to homeless people. Really?
A focus on the "big picture" and whole person is so, well, sensible!
At a prior hospital system I spent lots of my time explaining to inpatients with newly diagnosed heart failure why they would not benefit from a Lifevest., how uncomfortable it can be, and how cost to them was unpredictable. These kinds of conversations are far more dificult than ones where a gizmo or device is recommended. Unfortunately, various members of the health care team had been brainwashed by LifeVest reps and had told my patients they should get the vest in order to save their lives. Occasionally, one of my patients would have the LifeVest process initiated because I was off on the day they were discharged. Invariably, this resulted in 3 months of hassle, inappropriate shocks and discomfort.
What a great article, Wil!
At this point in my career, I’m no longer surprised when things that *should* work don’t actually work when tested.
Like Mary said earlier, I sympathize with the cardiologist. It’s still really hard for doctors to get past the mechanistic thinking that’s drilled into us in training. But the reality is that humans are, by nature, unpredictable systems.
Excellent article. Mechanism, theory, and biologic plausibility are great…but they will never be a replacement for outcome evidence. To rely on “mechanism” when the outcome trial failed is inexcusable.
It is true that trials provide information on average effects, and it is always with some uncertainty when we try to apply that evidence to individual patients. But that comes from positive trials and the limits of external validity. To assert that treatment X will work on a patient when the trial failed to show a benefit, is completely without scientific basis.
Excellent article. I think some valuable clues to proper analysis of other therapeutic interventions are brought up here. The figures for all cause mortality (3.1% v 4.9%) were said to be statistically significant with a p=0.04 but not significant after "multiple comparison adjustment". Arrhythmic mortality was 1.6% v 2.4% with p=0.18. Actual risk reduction was 1.8% and 0.8% respectively and relative risk reduction about 36 and 33%. Obviously there are a lot of problems trying to draw accurate conclusions from data with very low endpoint incidence. But much of the data on other therapeutic modalities show only slight differences in magnitude of effects and are often presented with only the RRR figures even in some articles on Sensible Medicine. Some have even called these small differences robust. I have in mind, particularly, the numbers on statin therapy which are used to justify prescription of statins to a large percentage of the population. In my opinion, the only firm conclusion we can draw from these kinds of figures is that the people that populate the "official" medical policy boards are bought and paid for by the medical device and pharmaceutical companies.
This was such a good article. Thanks so much Wil!
Adam
I suspect physician non-compliance with published “guidelines”, particularly in a patient population characterized by sudden death, raises medical-legal and lawsuit concerns within physicians. Guidelines don’t just “inform”, they can also be inappropriately coercive.
I can sympathize with the cardiologist for encouraging the vest's usage (not for exaggerating its benefits or for over-ruling the patient's no, however). It is hard when something that we feel in our guts *should* work doesn't work. Especially if the cardiologist feels like he can point to specific patients that he thinks were helped (they got shocked and lived).
Medicine is hard! I want to be a rocket scientist instead! I think it might be easier.
So, what we’ve learned is that the cardiologist was a domineering a*****e, who ignored patient wishes. Wondering if he’s coincidentally a paid ‘consultant’ for LifeVest Inc. Patients don’t have to make decisions with which we agree. They make decisions, and it is NOT our place to ‘sell’ them on The Correct Choice. It is our responsibility to inform, advise, and intervene to the extent they choose. Period, end.
“More more more, longer, harder, because I Am The Expert” is no way to practice medicine. And yet, it has become pervasive. Industry wins. Society loses.
I think a fundamental issue and something that many die-hard evidence-based medicine types seem to forget is that there is a difference between the population information afforded by the RCT versus the individual. The RCT shows that in the vast majority of cases the VEST does nothing and isn't worthwhile. However, it is very likely that there is a subset of patients where the VEST may safe their life. The key, and this is where medicine is as much an Art as a Science, is knowing which patients will benefit and which will not. Juts blindly following protocols is simply bad medicine.
The same goes for Wil's comments regarding screening colonoscopy. Sure the NORDISK RCT showed no benefit for all cause mortality but it is obvious that in an individual with polyps, the colonoscopy coupled with appropriate follow-up colonoscopy will likely save their life or at least prevent them from dying with colon cancer. To put this in perspective. Let us say that patient X has colonoscopies every 5 years until after his 90th birthday and each time his/her colon was "clean" (i.e. no evidence of any disease). Then clearly that patient will not have benefited from all those colonoscopies and will have been subject to the risk of any invasive procedure and anesthesia. But let us say that patient Y comes in for a colonoscopy at age 45 and a number of polyps are found, then removal of those polyps will prevent his death/her death from colon cancer. What the procedure what do, of course, is prevent their death from heart disease or a car accident!
Really like Wil Ward’s guest columns! Would love to get one about doing consults (family med or internal med I guess?) on a psych ward and being unsettled by what he found (a comment he posted on Twitter). There’s such a lack of caution among psych “specialists” I feel like their prescribing is less trustworthy than PCPs.
Hear! Hear! As disinterested and scientifically honest practitioners, we should maintain skepticism on all interventions. --Not nihilism, but make the proponents prove their recommendations with convincing data.
The single underlying dynamic in this and the other examples in this article is the "More is better bias". This pervasive attitude is promoted by industry, their paid consultants/researchers, medical editorialists, the lay press, and general public. Two recent editorials in the Wall Street Journal are illustrative: They were critical of Marty Makary and Vinay Prasad because they didn't approve as many drug applications as the previous administration. And they were slower in making decisions. Did it occur to the editorial board that some drugs fail to make their case? And perhaps due consideration takes more time than a rubber stamp.
Wow--very scary indeed! Instead of using a LifeVest, I recommend taking N-acetylcysteine, D-ribose, omega-3 fatty acids, and other targeted supplements.
This is really good and will be added to the collection of articles for my nursing students. Thank you