You would really have to do research and look to the primary study paper as Hart advises. Media, government and public health cannot be trusted, who really are these organisations accountable to? It is not the ordinary people, not one of these institutions should be getting payed beyond their salary and what is strictly allowed to them by a budget that has no outside money to influence them. We have no transparency in this country hence the corruption.
I don't see much of a problem. The BBC article makes it clear that this is research funded by the drug manufacturer. The drugs are already well established weight loss drugs. This study is simply exploring their effects on cardiovascular risk independent of losing weight - surely worth knowing.
As I am sure the Hart team appreciate:
* The problem with absolute risk reduction is that it varies according to the population. The riskier the population the higher the reduction. Relative risk reduction may also vary according to population but that is not a given. It is more likely to be a stable property of the intervention applicable to wider variety of situations.
* Comparing it to the chances of stopping the treatment because of adverse events is not very meaningful without knowing more about the adverse events. An adverse event may be a relatively minor side effect such as a patient who suffers nausea and decides to switch to another weight loss drug.
[Edit: I've left the bulk of this message to show that, even though I'm aware of relative vs. absolute risk reduction, I still "fell for" the deception of the "20% reduction!!!" lure. Thanks to other posters here, I "sobered up" and have updated my own essay to reflect what I think are the absolute risk reductions as well as the NNT and projected treatment costs.]
In fairness, the "real" study (NEJM) makes it look like a highly effective drug. Of course I'm saying that with the usual caveats that such reports are massaged to make the result look as good as possible. The handful of trials I've looked at were for statins or a few for aspirin as primary prevention; the benefits (in absolute risk reduction) are usually so tiny that in my opinion they fell into the "Why bother?" range. Throw in the downsides (cost, risk of adverse effects, etc.) and it was a no-brainer to go off these and politely decline each year when the nurse recommends these useless drugs. In contrast, IF the trial was honestly conducted (and that is asking a lot!!!), I’d say the claimed reductions in morbidity and mortality (in the 20% range), or the 19% drop in all-cause deaths – curiously not reported at all – would indicate semaglutide has potential as prevention. For amusement, compare and contrast some of the statins, which could legitimately claim to reduce (say) heart attacks or CVD deaths, but you have to dig into the numbers to find out that there were more fatal strokes and the overall death rate was unchanged compared to placebo.
It’s also worth mentioning that relative risk reduction is a big deal when the absolute risk is quite high. For example, the trial population seems to have a risk of cardiovascular event is much higher than the general population. Therefore a 20%-ish reduction in risk of that is of greater value. Conditional, of course, on the desired effects outweighing the negatives of adverse effects.
"It’s also worth mentioning that relative risk reduction is a big deal when the absolute risk is quite high. "
That is the point I am trying to make. The relative risk is a number you can apply to different situations to estimate the reduction in absolute risk. The absolute risk varies immensely depending on the population, the timescale (your absolute risk over 10 years is clearly much higher than over 1 year), and many other contextual factors. Absolute risk reduction figures need to be interpreted just as carefully as relative risk reduction.
“It is more likely to be a stable property of the intervention applicable to wider variety of situations.”
Really? You’re saying that something (for example) which reduces cardiovascular mortality by 10% in people who have had a heart attack wound be likely to do the same in “all comers”? I don’t think so.
The point is that promoting a RRR obtained from an at-risk population (also while ignoring all-cause data), and then gradually widening the population eligible to be treated to encompass people at less and less risk seems to be the MO of pharma these days and I’m rather inclined not to take that lying down.
As for 2, yes, would be great to know more. But no more was available as this is marketing by media soundbite and data isn’t being shared, which was the point being made. A net 8% stopping due to adverse events for a drug being pushed on people as something preventative should be a concern.
Yes, that's why ideally one would know what the are the real risks involved. Partially, that requires knowing the "claimed" relative and absolute risks (usually a reduction, but could be an increase, too.) E.g. a 50% reduction of a tiny risk is of comparatively little value. On the other hand, a 50% reduction of a disease that has an 80% fatality rate is a big deal. This question is not entirely academic, either. Allow me to give two examples from my personal life:
Although technically I was never offered it, I would never accept a mRNA Covid-19 “vaccine” because I know that even if the vaccine were effective (it’s not), it would provide me immunity against a disease that is rarely serious and even rarer fatal, to a healthy person. Add in the negatives of mRNA being novel technology with unknown and unknowable adverse events, and it’s a no-brainer to say “hell, no.” – as millions of wise people did.
On the other hand, I got bit by a feral cat. The cat escaped so no information was available whether it had rabies or some other disease. I was recommended, and did, choose to take a vaccine against tetanus and rabies. Now, I don’t have figures at hand, but I suspect that even in an area where rabid animals are common, that the chances of any one bite by any random animal actually causing either of those diseases might be one in a few thousand, but that’s not a risk I was willing to take against a disease that has fatality rates of 10% or 100%, respectively. Yet, if I’d had to pay all my costs out-of-pocket (a few thousand dollars for emergency treatment) I might have chosen to accept the risk.
"You’re saying that something (for example) which reduces cardiovascular mortality by 10% in people who have had a heart attack wound be likely to do the same in “all comers”? "
No. But knowing it reduces mortality by 10% in one population gives you some indication as to whether it is worth researching the effect in other populations - more likely higher risk populations. The Absolute Risk Reduction tells you zilch about the possible effect on other populations.
"The point is that promoting a RRR obtained from an at-risk population (also while ignoring all-cause data), and then gradually widening the population eligible to be treated to encompass people at less and less risk seems to be the MO of pharma these days"
Really? Do you have some examples?
"A net 8% stopping due to adverse events for a drug being pushed on people as something preventative should be a concern."
For a weight loss drug which has been in widespread use for over a decade? Some more information would certainly be a good thing but if there is a serious problem with these drugs then surely it would have emerged before now?
"Since then, additional studies have shown similar results. But they’ve also revealed that participants tend to regain the weight lost when they stop taking semaglutide." ( from UCLA Health 2024)
Do you think we'll find this little tidbit along with the cheerleading literature?
There should be no surprise about weigh regain; that's [probably] true of any diet plan (and semaglutide is, sort of...) Although I'd not offer it as dietary advice I do believe it's in Proverbs 26 where it is written "As a dog returneth to his vomit, so a fool returneth to his folly." Who would ever have guessed if you stop doing what works (the diet) and go back to whatever it was that got you in trouble in the first place, that the weight would come back?
Thank you for one of the more informative (skeptical) analyses I've read on this latest study. In fact, the commenters here inspired me to tidy up my own Substack essay and add these figures (I have a copy of the full study). With the assumption that I didn't make any errors, the NNT for semaglutide (to prevent death from any cause) is about 323. I used $100/mo. as assumed drug cost, that's nearly $388,000 to prevent (delay) one death. Broadly speaking, the prevention value is in the same league as a statin. I'm probably preaching to the choir here, but folks, if you really want to lose weight and keep it off and most likely enjoy lowered risk of CVD, a keto or Atkins type diet will almost certainly accomplish that and for a hell of a lot cheaper than the drug(s).
Nothing like a short term solution to a long term problem. I just wonder whether the prescribing physician proactively informs their patients that they will have to be taking injections for the rest of their lives in order to derive the same benefit as a lifestyle change or two. If there were full disclosure in the doctor's office there would be no need for "guessing".and the patient might realize the savings. I suppose in the current era it's asking too much of a physician to have a conscience. After all the doc.might be missing next years country club membership.
[I found you from article at Conservative Woman] I first glommed onto the story thinking I might invest in Novo Nordisk. Being familiar with the writings of Dr. Malcolm Kendrick, I’m well aware of some of the games Pharma plays with clinical trials and how results are enthusiastically exported in the media in what amounts to little more than paid advertising. I was able to get a copy of the full study as it appeared in New England Journal of Medicine. As far as I can tell, the gushing praise in the mass media accurately reflects what the study found. The author reports there’s no mention of all-cause deaths; that may be true in what he looked at, but it’s certainly reported in the full study. And therein hangs a tale, although I’m not sure what to make of it. “Death from any cause” is given in a couple places. For instance, the graph Figure 1D on page 6. If you do the simple math, the annual death rate (placebo) works out to about 1.63%. Curiously, that's very near the death rate of the 62-year-old overall population (average age of test subjects). That seemed low, but now I’m not so sure, since the trial excluded diabetics and certain other conditions. But the real puzzler to me: “Hazard ratio 0.81” = a reduction in deaths from any cause of 19%. For comparison, “Death from cardiovascular causes” (Figure 1B) was “only” 15%. Now, it’s true both those are relative risk reductions; nevertheless those are grounds for bragging in either case. What’s tantalizing is that, so far as I can tell, there is ABSOLUTELY NO MENTION of the 19% drop in all-cause death rates in “Discussion” of the NEJM paper nor of any of the popular media. Normally such a result would have been heavily promoted.
Added: on second thought the NNT and the related relative risk reduction are not all that wondrous, that's pretty much in the neighborhood of statins and blood pressure meds for primary or maybe secondary prevention.
I calculate NNT = 323 (delay one death). At $100/mo. that's about $388,000. Pretty good deal for pharma if they can get it to be standard of care.
See my pathetic Substack for a bit more background if you’re curious.
I've started a deep dive into GLP-1 agonists. Two things stand out:
>Increased risk of pancreatitis, which in itself is linked to pancreatic cancer.
>Decreased beta cell mass. In plain English it means increased risk of adverse health effects if you stop suddenly, including type 2 diabetes mellitus.
Good work. Why can’t journalists do this? Isn’t it their job?
I believe lying is much more profitable for mainstream journalists nowadays. At least, it enables them to keep their jobs.
You would really have to do research and look to the primary study paper as Hart advises. Media, government and public health cannot be trusted, who really are these organisations accountable to? It is not the ordinary people, not one of these institutions should be getting payed beyond their salary and what is strictly allowed to them by a budget that has no outside money to influence them. We have no transparency in this country hence the corruption.
Semaglutide Deaths explained
https://geoffpain.substack.com/p/ozempic-semaglutide-deaths-endotoxin
But probably kill you by a stroke ! Thank you BBC but let’s flush you
I don't see much of a problem. The BBC article makes it clear that this is research funded by the drug manufacturer. The drugs are already well established weight loss drugs. This study is simply exploring their effects on cardiovascular risk independent of losing weight - surely worth knowing.
As I am sure the Hart team appreciate:
* The problem with absolute risk reduction is that it varies according to the population. The riskier the population the higher the reduction. Relative risk reduction may also vary according to population but that is not a given. It is more likely to be a stable property of the intervention applicable to wider variety of situations.
* Comparing it to the chances of stopping the treatment because of adverse events is not very meaningful without knowing more about the adverse events. An adverse event may be a relatively minor side effect such as a patient who suffers nausea and decides to switch to another weight loss drug.
[Edit: I've left the bulk of this message to show that, even though I'm aware of relative vs. absolute risk reduction, I still "fell for" the deception of the "20% reduction!!!" lure. Thanks to other posters here, I "sobered up" and have updated my own essay to reflect what I think are the absolute risk reductions as well as the NNT and projected treatment costs.]
In fairness, the "real" study (NEJM) makes it look like a highly effective drug. Of course I'm saying that with the usual caveats that such reports are massaged to make the result look as good as possible. The handful of trials I've looked at were for statins or a few for aspirin as primary prevention; the benefits (in absolute risk reduction) are usually so tiny that in my opinion they fell into the "Why bother?" range. Throw in the downsides (cost, risk of adverse effects, etc.) and it was a no-brainer to go off these and politely decline each year when the nurse recommends these useless drugs. In contrast, IF the trial was honestly conducted (and that is asking a lot!!!), I’d say the claimed reductions in morbidity and mortality (in the 20% range), or the 19% drop in all-cause deaths – curiously not reported at all – would indicate semaglutide has potential as prevention. For amusement, compare and contrast some of the statins, which could legitimately claim to reduce (say) heart attacks or CVD deaths, but you have to dig into the numbers to find out that there were more fatal strokes and the overall death rate was unchanged compared to placebo.
It’s also worth mentioning that relative risk reduction is a big deal when the absolute risk is quite high. For example, the trial population seems to have a risk of cardiovascular event is much higher than the general population. Therefore a 20%-ish reduction in risk of that is of greater value. Conditional, of course, on the desired effects outweighing the negatives of adverse effects.
"It’s also worth mentioning that relative risk reduction is a big deal when the absolute risk is quite high. "
That is the point I am trying to make. The relative risk is a number you can apply to different situations to estimate the reduction in absolute risk. The absolute risk varies immensely depending on the population, the timescale (your absolute risk over 10 years is clearly much higher than over 1 year), and many other contextual factors. Absolute risk reduction figures need to be interpreted just as carefully as relative risk reduction.
“It is more likely to be a stable property of the intervention applicable to wider variety of situations.”
Really? You’re saying that something (for example) which reduces cardiovascular mortality by 10% in people who have had a heart attack wound be likely to do the same in “all comers”? I don’t think so.
The point is that promoting a RRR obtained from an at-risk population (also while ignoring all-cause data), and then gradually widening the population eligible to be treated to encompass people at less and less risk seems to be the MO of pharma these days and I’m rather inclined not to take that lying down.
As for 2, yes, would be great to know more. But no more was available as this is marketing by media soundbite and data isn’t being shared, which was the point being made. A net 8% stopping due to adverse events for a drug being pushed on people as something preventative should be a concern.
Yes, that's why ideally one would know what the are the real risks involved. Partially, that requires knowing the "claimed" relative and absolute risks (usually a reduction, but could be an increase, too.) E.g. a 50% reduction of a tiny risk is of comparatively little value. On the other hand, a 50% reduction of a disease that has an 80% fatality rate is a big deal. This question is not entirely academic, either. Allow me to give two examples from my personal life:
Although technically I was never offered it, I would never accept a mRNA Covid-19 “vaccine” because I know that even if the vaccine were effective (it’s not), it would provide me immunity against a disease that is rarely serious and even rarer fatal, to a healthy person. Add in the negatives of mRNA being novel technology with unknown and unknowable adverse events, and it’s a no-brainer to say “hell, no.” – as millions of wise people did.
On the other hand, I got bit by a feral cat. The cat escaped so no information was available whether it had rabies or some other disease. I was recommended, and did, choose to take a vaccine against tetanus and rabies. Now, I don’t have figures at hand, but I suspect that even in an area where rabid animals are common, that the chances of any one bite by any random animal actually causing either of those diseases might be one in a few thousand, but that’s not a risk I was willing to take against a disease that has fatality rates of 10% or 100%, respectively. Yet, if I’d had to pay all my costs out-of-pocket (a few thousand dollars for emergency treatment) I might have chosen to accept the risk.
"You’re saying that something (for example) which reduces cardiovascular mortality by 10% in people who have had a heart attack wound be likely to do the same in “all comers”? "
No. But knowing it reduces mortality by 10% in one population gives you some indication as to whether it is worth researching the effect in other populations - more likely higher risk populations. The Absolute Risk Reduction tells you zilch about the possible effect on other populations.
"The point is that promoting a RRR obtained from an at-risk population (also while ignoring all-cause data), and then gradually widening the population eligible to be treated to encompass people at less and less risk seems to be the MO of pharma these days"
Really? Do you have some examples?
"A net 8% stopping due to adverse events for a drug being pushed on people as something preventative should be a concern."
For a weight loss drug which has been in widespread use for over a decade? Some more information would certainly be a good thing but if there is a serious problem with these drugs then surely it would have emerged before now?
"Since then, additional studies have shown similar results. But they’ve also revealed that participants tend to regain the weight lost when they stop taking semaglutide." ( from UCLA Health 2024)
Do you think we'll find this little tidbit along with the cheerleading literature?
There should be no surprise about weigh regain; that's [probably] true of any diet plan (and semaglutide is, sort of...) Although I'd not offer it as dietary advice I do believe it's in Proverbs 26 where it is written "As a dog returneth to his vomit, so a fool returneth to his folly." Who would ever have guessed if you stop doing what works (the diet) and go back to whatever it was that got you in trouble in the first place, that the weight would come back?
Thank you for one of the more informative (skeptical) analyses I've read on this latest study. In fact, the commenters here inspired me to tidy up my own Substack essay and add these figures (I have a copy of the full study). With the assumption that I didn't make any errors, the NNT for semaglutide (to prevent death from any cause) is about 323. I used $100/mo. as assumed drug cost, that's nearly $388,000 to prevent (delay) one death. Broadly speaking, the prevention value is in the same league as a statin. I'm probably preaching to the choir here, but folks, if you really want to lose weight and keep it off and most likely enjoy lowered risk of CVD, a keto or Atkins type diet will almost certainly accomplish that and for a hell of a lot cheaper than the drug(s).
Nothing like a short term solution to a long term problem. I just wonder whether the prescribing physician proactively informs their patients that they will have to be taking injections for the rest of their lives in order to derive the same benefit as a lifestyle change or two. If there were full disclosure in the doctor's office there would be no need for "guessing".and the patient might realize the savings. I suppose in the current era it's asking too much of a physician to have a conscience. After all the doc.might be missing next years country club membership.
[I found you from article at Conservative Woman] I first glommed onto the story thinking I might invest in Novo Nordisk. Being familiar with the writings of Dr. Malcolm Kendrick, I’m well aware of some of the games Pharma plays with clinical trials and how results are enthusiastically exported in the media in what amounts to little more than paid advertising. I was able to get a copy of the full study as it appeared in New England Journal of Medicine. As far as I can tell, the gushing praise in the mass media accurately reflects what the study found. The author reports there’s no mention of all-cause deaths; that may be true in what he looked at, but it’s certainly reported in the full study. And therein hangs a tale, although I’m not sure what to make of it. “Death from any cause” is given in a couple places. For instance, the graph Figure 1D on page 6. If you do the simple math, the annual death rate (placebo) works out to about 1.63%. Curiously, that's very near the death rate of the 62-year-old overall population (average age of test subjects). That seemed low, but now I’m not so sure, since the trial excluded diabetics and certain other conditions. But the real puzzler to me: “Hazard ratio 0.81” = a reduction in deaths from any cause of 19%. For comparison, “Death from cardiovascular causes” (Figure 1B) was “only” 15%. Now, it’s true both those are relative risk reductions; nevertheless those are grounds for bragging in either case. What’s tantalizing is that, so far as I can tell, there is ABSOLUTELY NO MENTION of the 19% drop in all-cause death rates in “Discussion” of the NEJM paper nor of any of the popular media. Normally such a result would have been heavily promoted.
Added: on second thought the NNT and the related relative risk reduction are not all that wondrous, that's pretty much in the neighborhood of statins and blood pressure meds for primary or maybe secondary prevention.
I calculate NNT = 323 (delay one death). At $100/mo. that's about $388,000. Pretty good deal for pharma if they can get it to be standard of care.
See my pathetic Substack for a bit more background if you’re curious.
I've started a deep dive into GLP-1 agonists. Two things stand out:
>Increased risk of pancreatitis, which in itself is linked to pancreatic cancer.
>Decreased beta cell mass. In plain English it means increased risk of adverse health effects if you stop suddenly, including type 2 diabetes mellitus.
Yes however... look in the mirror if you want to blame someone... heart disease is generally a CHOICE.
Let me explain: https://fasteddynz.substack.com/p/we-need-to-stop-hating-big-pharma
Damn!