Have you considered assessing the ratio of 'HF hospitalizations' to 'all-cause hospitalizations' according to age? Or perhaps analyzing it across multiple factors such as NYHA, EF and age simultaneously? I anticipate this could reveal an even lower weighted mean ratio, potentially in the range of 20–30%.
Additionally, the fact that this analysis was based on only 50% of the trials (as you mentioned in the limitations) suggests that the reported weighted mean ratio and its association with outcomes might be significantly overestimated.
Nvm “heart failure hospitalizations” ….which is soft and squishy enough. But very recent HF trials have even gotten into habit of including some version of “intensification of diuretics” into a “worsening HF” composite….which can mean as little as an increase in PO lasix. Such studies are transacting in surrogates of uselessness. I hope this project will start the drive to end such idiocy.
I will use an analogy at the end of my comment and it very well may not "click" with many other readers of Sensible Medicine, especially those with Ivy League credentials.
I can hardly wait to see your next installment regarding outcome measures when HF is the disease of interest. I am hoping to see you Lower The Boom on the BS tactic called "Using composite outcomes in clinical trials". Here comes the analogical punch line I will need: Using composite outcomes, followed by a dose of statistical Kung Fu, to interpret data from clinically compared therapies is almost exactly like what happens in a professional wrestling match: e.g. everybody knows what's highly likely to happen in the closing moments of the match -- the Bad Guy will be thrown over the ropes into the audience, but will then remove the prosthetic limb of some elderly patron (s/p BK amputation) sitting at ringside and then crawl back into the ring to use that device for clubbing the Good Guy's head so as to render him senseless just before the bell.
Although I think an AKA prosthetic would be much more dramatic, I think you are right about using a BKA. An AKA prosthetic would be unwieldy in a wrestling match due to its size and if you can’t effectively lock the knee component.
Excellent contribution. Agree that there has been a proliferation of the use of soft and questionable endpoints in cardiovascular trials that risks damaging the integrity of our field and jeopardizes our reputation for leading EBM. All-cause hospitalization and all-cause mortality are the right endpoints. Everyone should re-read Stephen Gottlieb's famous editorial on the EMIAT/CAMIAT trials, "Dead is dead - Artificial definitions are no substitute." True in 1997 and true today. https://pubmed.ncbi.nlm.nih.gov/9078192/
Very good and needed research.
Have you considered assessing the ratio of 'HF hospitalizations' to 'all-cause hospitalizations' according to age? Or perhaps analyzing it across multiple factors such as NYHA, EF and age simultaneously? I anticipate this could reveal an even lower weighted mean ratio, potentially in the range of 20–30%.
Additionally, the fact that this analysis was based on only 50% of the trials (as you mentioned in the limitations) suggests that the reported weighted mean ratio and its association with outcomes might be significantly overestimated.
Nvm “heart failure hospitalizations” ….which is soft and squishy enough. But very recent HF trials have even gotten into habit of including some version of “intensification of diuretics” into a “worsening HF” composite….which can mean as little as an increase in PO lasix. Such studies are transacting in surrogates of uselessness. I hope this project will start the drive to end such idiocy.
I will use an analogy at the end of my comment and it very well may not "click" with many other readers of Sensible Medicine, especially those with Ivy League credentials.
I can hardly wait to see your next installment regarding outcome measures when HF is the disease of interest. I am hoping to see you Lower The Boom on the BS tactic called "Using composite outcomes in clinical trials". Here comes the analogical punch line I will need: Using composite outcomes, followed by a dose of statistical Kung Fu, to interpret data from clinically compared therapies is almost exactly like what happens in a professional wrestling match: e.g. everybody knows what's highly likely to happen in the closing moments of the match -- the Bad Guy will be thrown over the ropes into the audience, but will then remove the prosthetic limb of some elderly patron (s/p BK amputation) sitting at ringside and then crawl back into the ring to use that device for clubbing the Good Guy's head so as to render him senseless just before the bell.
Although I think an AKA prosthetic would be much more dramatic, I think you are right about using a BKA. An AKA prosthetic would be unwieldy in a wrestling match due to its size and if you can’t effectively lock the knee component.
The only out come that actually matters is quality of life period unless of course it’s all about the money
Excellent contribution. Agree that there has been a proliferation of the use of soft and questionable endpoints in cardiovascular trials that risks damaging the integrity of our field and jeopardizes our reputation for leading EBM. All-cause hospitalization and all-cause mortality are the right endpoints. Everyone should re-read Stephen Gottlieb's famous editorial on the EMIAT/CAMIAT trials, "Dead is dead - Artificial definitions are no substitute." True in 1997 and true today. https://pubmed.ncbi.nlm.nih.gov/9078192/