Thank you Dr Mandrola for another great lesson. As a trainee GP/ family physician I've found your columns excellent education.
My colleagues and I wrote a similar study of endpoints in RCTs a few years ago that you may find interesting. The summary is that trials often don't report their primary endpoints and that secondary endpoints are not very reliable.
Commen sense but it was nice to prove it.
Underreporting of secondary endpoints in randomized trials
JH Matthews, S Bhanderi, SJ Chapman, D Nepogodiev… - Annals of surgery, 2016
Great format for the clinically naive reader! I already hold a pretty comfortable understanding of this particular topic, but looking forward to seeing more important information broken down in such a digestible manner.
Having lost 5 years of my life to generic risk analysis of AVD without appropriate consideration of the ejection fraction or operator error in ultrasound among numerous other statistical decisions unrelated to my specific condition, nothing surprises me about bogus cardiology. Back to cycling 80 miles a week.
Wow, am I glad to see your thinking in my in-box again. I'm surprised if I missed some articles, but at 85, black swans happen.
I recently had an atrial ablation, and 6 months out am very happy. But a 7 day monitor 2 weeks after the surgery found 200+ heartbeat pauses, and a pacemaker was recommended.
I sure will look in past articles for what you may have said about pacemakers. I save what you say.
As a former longterm academician (who in years past gave countless national medical talks to primary care audiences) — the “spin” placed on Patiromer in the DIAMOND Trial makes me angry. It is pure deceit with sole goal of the study clearly being to increase the profits of the makers of Patiromer. The BIG Question = Any of the authors of the study being sponsored by the makers of Patiromer? IF SO — then the drug should not have been approved for this indication (and any responsible peer review should not have approved their study. THANK YOU John for your continued EXCELLENT work!
I think 2 years from now there needs to be a study on research published that was carried out during the pandemic that is NOT Covid related. Wonder how much of this sort of short cutting has resulted in misconstrued results. I was involved as an assistant in research and quit the job when I noted sloppy changes to put together data skewed by Covid and a deadline. Terrible science. 😒
I think your idea of helping laypeople to think critically about medicine is wonderful! Keep up the good work! But I do still find your posts contain a lot of technical terms. (e.g. hyperkalemia) That's O.K. for me, because I'm willing to take the time to look up their definitions. But you will lose some of your intended audience for that reason. I appreciate that it's very difficult to reduce your post to 750 words without using technical terms, but ... perhaps you could find a way. Anyway, thank you so much for writing these fascinating posts!
"Endpoints. Always look at endpoints. Don’t rely on news coverage or the authors’ conclusions. "
Similar to rules for finding fraud in corporate reports or loopholes in legislation. The dirty details are buried in the footnotes explaining some voodoo equivalent of moving endpoints and substituting actual facts for a "substantial equivalent" modeled pie in the sky metric. Enron accounting in Beltway Bandit parlance & highly prized, richly rewarded skill for Federal contractors & lobbyists.
I think that the concerns raised by many here, certainly deserve to be placed in the perspective of the scientists. Medical research, in my opinion, is smack-dab in the middle of the slippery slope. I don't believe that there are many researchers who've sold their integrity out to the corporate world; but I don't believe they're beyond making handy excuses to mollify their consciences.
I was on a track to go about medical research, and I completely changed my direction due to several reasons. One was the change in the National Science Foundation and National Institutes of Health. When once the Federal agencies cut a broad swath to enlist younger researchers with independent ideas, the Federal Government mostly withdrew funding of basic science research for financial reasons, deferring the responsibility for research to corporations, and thus allowing them to direct the questions under study. Now, the publications are pretty much "How does drug X work?" rather than basic science in medicine.
Also, the pressure to publish has climbed to the point that those in the good graces of the mega-study leaders are benefited and rewarded with publishable information. Being a "good study buddy" is certainly rewarded with attention and approval; producing "good results" can cause favorable "hub" bias towards a "spoke" researcher engaging in a study, and one who produces null results, a negative bias.
Now, many of the research "hubs" are dukedoms at a university, headed by a Chief with dozens of prole-bots grunting away to produce publishable data. Not everything they do is funded by Pharma, but enough to stay in Pharma's good graces. Where once the 30-year-old researcher could step out independently, prole-bots churn through "postdocs" and "research fellowships" into their 50's, with hopes of climbing the mountain to a deputy chief position.
In this study - where is sodium polystyrene resin? That's the gold standard which should be used for enteric potassium reduction agents. You do research against the gold standard. Here, it appears to be excluded. Why? Any polymer should do better than Kayexalate (sodium polystyrene), for starters.
The "endpoint" is merely what was a step in the protocol for results. If the medication did NOT lower potassium but DID have beneficial effect, therefore the beneficial effect would be of unknown cause. That's important to use in a study, but it's one of the first steps, not the top of a staircase.
Why they stopped here? I don't know.
Another cause of ivy-tower research isolation is the capitalization of research behind paywalls that are insurmountable to the informed reader - who has perhaps paid taxes for the research done. A large university library will subscribe to hundreds of paywalled journals. Joe Dr. Sixpack can't read the methods, just the press releases. Why?
I thought this comment early in the article was telling. "I know; this sounds convoluted: a drug to treat a drug side effect." In the context it seemed to imply that the readers would be surprised or confused at this. On the contrary, I think most readers realize just how widespread and insidious this is. Most doctors find it easier to prescribe a drug than encourage (demand?) that a patient change their lifestyle. If and only if, after changing and adhering to the required lifestyle change, they still need the drug will it be prescribed. Medical issues that might be controlled by lifestyle changes include joint pain, mild depression, heartburn, insomnia, overactive bladder, prediabetes, prehypertension, and obesity.
Do you have knee and joint pain? Here, try Celebrex. Works great. A few years later your blood pressure is up so you should start on a additional medication. A few more years go by. Low and behold you now are showing signs of reduced kidney function so lets add another medication. A few more years and you are having digestive problems. More medication. And on it goes. The average elderly person is taking 5-6 prescription medications a year.
Drug companies push the medications. Doctors prescribe them because it's easier to write a prescription than take the time to figure out the cause of the problem is and attempt to correct it.
And, Joe, you know what else? Now that Corporate Medicine has turned your doctor's visit into a Billable Unit, and are jamming people into 15 minute slots. Writing a single prescription causes the visit to bill at a Level 4 instead of a Level 3, which is what your provider bills at if they talk to you for 15 minutes and write NO prescriptions. Too many Level 3's, and the performance auditors have a sit-and-think with you to help you improve your performance. Because Four Pays More.
It doesn't have to be this way. Never let go of that understanding. It doesn't have to be this way.
Yeah, I thought the same thing. Maybe 20 years ago, polypharmacy would be a mysterious concept to someone's elderly parent. We've basically all gotten used to the idea now.
Last I saw, that average (5-6 prescriptions) was higher. My elderly father was on 25 separate prescriptions before I noticed. He's now on 3, and all it took was a little assertiveness and asking, "why?" and dropping the scrip if there's no good answer- turns out 65% of them were just follow-on prescriptions after discharges from the hospital that never ended.
Surely this kind of statistical gaming and propaganda didn't happen with any of the mRNA treatments for COVID-19, or in analyses of treatments using preexisting drugs.
It was just too important. A global pandemic. Everyone decided, "let's all come together and agree to be 100% honest and altruistic until this thing is over, and then we can go back to lying and stealing."
I'm almost always skeptical of a clinical trial where study design involves the pharmaceutical company and paid academic researchers rather than the company providing a sufficiently large grant for a more opaque (to the company) multicenter trial.
I'm personally confident in the mRNA trials after reviewing the protocols and data several times. DIAMOND would never have passed muster in the institutions I've worked with in the past.
And my favorite, Kayexalate, costs 25¢ a gram, retail; fifteen grams a dose, that's about four bucks. Pulls out 15 mEq of potassium. No way are we getting near $12,000 This is not medical advice, but I've seen hyperkalemia before, and I've personally never seen it to be lethal in a non-dialysis patient. Just my observation - I know it can be lethal. But in heart failure, I think of potassium as more salubrious than hazardous, and find panic about K+=5.5 a little dramatic.
Thank you Dr Mandrola for another great lesson. As a trainee GP/ family physician I've found your columns excellent education.
My colleagues and I wrote a similar study of endpoints in RCTs a few years ago that you may find interesting. The summary is that trials often don't report their primary endpoints and that secondary endpoints are not very reliable.
Commen sense but it was nice to prove it.
Underreporting of secondary endpoints in randomized trials
JH Matthews, S Bhanderi, SJ Chapman, D Nepogodiev… - Annals of surgery, 2016
https://www.ingentaconnect.com/content/wk/sla/2016/00000264/00000006/art00024
Great format for the clinically naive reader! I already hold a pretty comfortable understanding of this particular topic, but looking forward to seeing more important information broken down in such a digestible manner.
And what about the use of “relative” improvement from a drug vs “actual” improvement (eg statins)?
Having lost 5 years of my life to generic risk analysis of AVD without appropriate consideration of the ejection fraction or operator error in ultrasound among numerous other statistical decisions unrelated to my specific condition, nothing surprises me about bogus cardiology. Back to cycling 80 miles a week.
Dr Mandrola:
Wow, am I glad to see your thinking in my in-box again. I'm surprised if I missed some articles, but at 85, black swans happen.
I recently had an atrial ablation, and 6 months out am very happy. But a 7 day monitor 2 weeks after the surgery found 200+ heartbeat pauses, and a pacemaker was recommended.
I sure will look in past articles for what you may have said about pacemakers. I save what you say.
As a former longterm academician (who in years past gave countless national medical talks to primary care audiences) — the “spin” placed on Patiromer in the DIAMOND Trial makes me angry. It is pure deceit with sole goal of the study clearly being to increase the profits of the makers of Patiromer. The BIG Question = Any of the authors of the study being sponsored by the makers of Patiromer? IF SO — then the drug should not have been approved for this indication (and any responsible peer review should not have approved their study. THANK YOU John for your continued EXCELLENT work!
I think 2 years from now there needs to be a study on research published that was carried out during the pandemic that is NOT Covid related. Wonder how much of this sort of short cutting has resulted in misconstrued results. I was involved as an assistant in research and quit the job when I noted sloppy changes to put together data skewed by Covid and a deadline. Terrible science. 😒
I think your idea of helping laypeople to think critically about medicine is wonderful! Keep up the good work! But I do still find your posts contain a lot of technical terms. (e.g. hyperkalemia) That's O.K. for me, because I'm willing to take the time to look up their definitions. But you will lose some of your intended audience for that reason. I appreciate that it's very difficult to reduce your post to 750 words without using technical terms, but ... perhaps you could find a way. Anyway, thank you so much for writing these fascinating posts!
Wonderful, and fluently explained, which is a high art on its own. Thank you!
"Endpoints. Always look at endpoints. Don’t rely on news coverage or the authors’ conclusions. "
Similar to rules for finding fraud in corporate reports or loopholes in legislation. The dirty details are buried in the footnotes explaining some voodoo equivalent of moving endpoints and substituting actual facts for a "substantial equivalent" modeled pie in the sky metric. Enron accounting in Beltway Bandit parlance & highly prized, richly rewarded skill for Federal contractors & lobbyists.
I think that the concerns raised by many here, certainly deserve to be placed in the perspective of the scientists. Medical research, in my opinion, is smack-dab in the middle of the slippery slope. I don't believe that there are many researchers who've sold their integrity out to the corporate world; but I don't believe they're beyond making handy excuses to mollify their consciences.
I was on a track to go about medical research, and I completely changed my direction due to several reasons. One was the change in the National Science Foundation and National Institutes of Health. When once the Federal agencies cut a broad swath to enlist younger researchers with independent ideas, the Federal Government mostly withdrew funding of basic science research for financial reasons, deferring the responsibility for research to corporations, and thus allowing them to direct the questions under study. Now, the publications are pretty much "How does drug X work?" rather than basic science in medicine.
Also, the pressure to publish has climbed to the point that those in the good graces of the mega-study leaders are benefited and rewarded with publishable information. Being a "good study buddy" is certainly rewarded with attention and approval; producing "good results" can cause favorable "hub" bias towards a "spoke" researcher engaging in a study, and one who produces null results, a negative bias.
Now, many of the research "hubs" are dukedoms at a university, headed by a Chief with dozens of prole-bots grunting away to produce publishable data. Not everything they do is funded by Pharma, but enough to stay in Pharma's good graces. Where once the 30-year-old researcher could step out independently, prole-bots churn through "postdocs" and "research fellowships" into their 50's, with hopes of climbing the mountain to a deputy chief position.
In this study - where is sodium polystyrene resin? That's the gold standard which should be used for enteric potassium reduction agents. You do research against the gold standard. Here, it appears to be excluded. Why? Any polymer should do better than Kayexalate (sodium polystyrene), for starters.
The "endpoint" is merely what was a step in the protocol for results. If the medication did NOT lower potassium but DID have beneficial effect, therefore the beneficial effect would be of unknown cause. That's important to use in a study, but it's one of the first steps, not the top of a staircase.
Why they stopped here? I don't know.
Another cause of ivy-tower research isolation is the capitalization of research behind paywalls that are insurmountable to the informed reader - who has perhaps paid taxes for the research done. A large university library will subscribe to hundreds of paywalled journals. Joe Dr. Sixpack can't read the methods, just the press releases. Why?
I thought this comment early in the article was telling. "I know; this sounds convoluted: a drug to treat a drug side effect." In the context it seemed to imply that the readers would be surprised or confused at this. On the contrary, I think most readers realize just how widespread and insidious this is. Most doctors find it easier to prescribe a drug than encourage (demand?) that a patient change their lifestyle. If and only if, after changing and adhering to the required lifestyle change, they still need the drug will it be prescribed. Medical issues that might be controlled by lifestyle changes include joint pain, mild depression, heartburn, insomnia, overactive bladder, prediabetes, prehypertension, and obesity.
Do you have knee and joint pain? Here, try Celebrex. Works great. A few years later your blood pressure is up so you should start on a additional medication. A few more years go by. Low and behold you now are showing signs of reduced kidney function so lets add another medication. A few more years and you are having digestive problems. More medication. And on it goes. The average elderly person is taking 5-6 prescription medications a year.
Drug companies push the medications. Doctors prescribe them because it's easier to write a prescription than take the time to figure out the cause of the problem is and attempt to correct it.
The word “demand” creeps me out. How should it be enforced?
And, Joe, you know what else? Now that Corporate Medicine has turned your doctor's visit into a Billable Unit, and are jamming people into 15 minute slots. Writing a single prescription causes the visit to bill at a Level 4 instead of a Level 3, which is what your provider bills at if they talk to you for 15 minutes and write NO prescriptions. Too many Level 3's, and the performance auditors have a sit-and-think with you to help you improve your performance. Because Four Pays More.
It doesn't have to be this way. Never let go of that understanding. It doesn't have to be this way.
Yeah, I thought the same thing. Maybe 20 years ago, polypharmacy would be a mysterious concept to someone's elderly parent. We've basically all gotten used to the idea now.
Last I saw, that average (5-6 prescriptions) was higher. My elderly father was on 25 separate prescriptions before I noticed. He's now on 3, and all it took was a little assertiveness and asking, "why?" and dropping the scrip if there's no good answer- turns out 65% of them were just follow-on prescriptions after discharges from the hospital that never ended.
Surely this kind of statistical gaming and propaganda didn't happen with any of the mRNA treatments for COVID-19, or in analyses of treatments using preexisting drugs.
It was just too important. A global pandemic. Everyone decided, "let's all come together and agree to be 100% honest and altruistic until this thing is over, and then we can go back to lying and stealing."
Thank goodness.
and the FDA goes along with this charade.
I'm almost always skeptical of a clinical trial where study design involves the pharmaceutical company and paid academic researchers rather than the company providing a sufficiently large grant for a more opaque (to the company) multicenter trial.
I'm personally confident in the mRNA trials after reviewing the protocols and data several times. DIAMOND would never have passed muster in the institutions I've worked with in the past.
Great article . $12,000 perhaps the key.
Then a drug like this somehow sneaks into an algorithm for treating chronic hyperkalemia.
And my favorite, Kayexalate, costs 25¢ a gram, retail; fifteen grams a dose, that's about four bucks. Pulls out 15 mEq of potassium. No way are we getting near $12,000 This is not medical advice, but I've seen hyperkalemia before, and I've personally never seen it to be lethal in a non-dialysis patient. Just my observation - I know it can be lethal. But in heart failure, I think of potassium as more salubrious than hazardous, and find panic about K+=5.5 a little dramatic.