Dr Eubank (and most of the medical community) seems to confuse two very different concepts: the probability of a “good” outcome and the probable outcome. Both NINDS and ECAS-3 have serious problems. Even if you accept the results as presented there is no clinically useful evidence of benefit. TPA appears to put 7% more patients in the “good outcome” category but it also causes ICH in 10%. As a clinician I want to be able to tell my patient the net balance of benefit and harm; this information doesn’t allow me to do so. The difference in mean stroke scores between treatment and control groups or even better the mean change in stroke scores in the two groups would allow me to advise my patients. This information has been withheld. A good analogy is the Minnesota State Lottery. It collects about $400 million per year and pays out $239 million. It’s true that the probability of a good outcome (winning millions) is much higher (OR is infinite) in the group that buys a ticket but the probable outcome is that you will lose 40% of what you spend. The probability of a good outcome is useful for marketing purposes (“Buy a ticket and win millions”) but it’s still a sucker's bet. The probable outcome is that you lose money.
The fact that these manufacturer supported studies resorted to this clinically irrelevant outcome and didn’t reveal the mean stroke scores or the mean change in stroke scores leads me to assume that there was no evidence of benefit. Perhaps, as responsible clinicians, we should ask for an independent clinically meaningful analysis of the data before subjecting our patients to a potentially harmful and expensive therapy.
Dave Ellison, MD
To join the conversation, you need to subscribe.
Sign up today for full access to all episodes and to join the conversation.
A. Terry - October 14, 2011 10:06 AM
Great stuff, but Kevin Klauer is a DO, not an MD, as the Contributor list states.
Peter W., M.D. - October 16, 2011 9:40 AM
:?jdge
tom w. - October 22, 2011 12:48 PM
The over site in Dr. Klaur's title has been fixed. Thank you for letting us know.
Dave E. - November 15, 2011 4:40 PM
Dr Eubank (and most of the medical community) seems to confuse two very different concepts: the probability of a “good” outcome and the probable outcome. Both NINDS and ECAS-3 have serious problems. Even if you accept the results as presented there is no clinically useful evidence of benefit. TPA appears to put 7% more patients in the “good outcome” category but it also causes ICH in 10%. As a clinician I want to be able to tell my patient the net balance of benefit and harm; this information doesn’t allow me to do so. The difference in mean stroke scores between treatment and control groups or even better the mean change in stroke scores in the two groups would allow me to advise my patients. This information has been withheld. A good analogy is the Minnesota State Lottery. It collects about $400 million per year and pays out $239 million. It’s true that the probability of a good outcome (winning millions) is much higher (OR is infinite) in the group that buys a ticket but the probable outcome is that you will lose 40% of what you spend. The probability of a good outcome is useful for marketing purposes (“Buy a ticket and win millions”) but it’s still a sucker's bet. The probable outcome is that you lose money.
The fact that these manufacturer supported studies resorted to this clinically irrelevant outcome and didn’t reveal the mean stroke scores or the mean change in stroke scores leads me to assume that there was no evidence of benefit. Perhaps, as responsible clinicians, we should ask for an independent clinically meaningful analysis of the data before subjecting our patients to a potentially harmful and expensive therapy.
Dave Ellison, MD