The Fda, Statistics As Well As Fundamentalism
Yesterday the FDA panel heard as well as voted on diverse matters that may or may non atomic number 82 to approving of Lemtrada, a Multiple Sclerosis drug. [The press is instantly example) as well as it was reckon that the drug may non hold upward approved.
Safety was no surprise. The statistical anomalies were.
Today the FDA Panel heard the case. The best coverage came from Twitter - as well as I am grateful to Sarah Karlin for her wonderful Twitter account. The panel voted on many things, nevertheless I desire to highlight only a couple:
Question 1: are the trials adequate as well as well-controlled?
Vote: eleven no, vi yes. 1 abstained.
Question 2: Has the applicant provided substantial evidence of effectiveness of alemtuzumab [Lemtrada] for the handling of patients amongst relapsing forms of MS?
Vote: 12 yes, vi no.
On enquiry #2, later the panel voted, the FDA staff asserted that if yous voted no on enquiry #1, yous can't vote yep on #2.
The commission chair said they voted yep based on gut feeling which I am non sure as shooting is adequate - but the FDA staff assertion is incorrect as well as I believe incompetent. It is perfectly acceptable to order that the trials are non "well-controlled" but at that spot is substantial evidence of effectiveness. I acquaint to you, yet again, the famous "parachute paper".
In that newspaper Gordon Smith as well as Jill Pell uncovering (with appropriate irony) that at that spot has never been a good controlled double-blind report of parachutes every bit a device to "prevent decease as well as major trauma related to gravitational challenge". Indeed because an adequate as well as well-controlled essay out is a requirement for approving parachutes are non approvable.
This is bunk.
We know why a parachute works.
We empathise the physics well.
Influenza A virus subtype H5N1 unmarried somebody who jumps out of an flat as well as lands unharmed amongst a parachute is a convincing essay out that parachutes are effective.
Bluntly, yous alone postulate double-blind stats when yous don't empathise the physics.
The FDA staff direct keep (as parodied inwards the parachute paper) raised their statistical mystique to a cult - every bit if it were the alone valid degree of epistemology. But at that spot are other ways of knowing, they are to a greater extent than valid (as per the parachute example) as well as at that spot is a deep underlying incompetence inwards the FDA cult of double-blind statistics as well as inwards the assertion highlighted above.
We are living inwards an historic menses where it is possible to direct keep deep noesis every bit to how the trunk works. We know for instance at that spot is a factor which codes for a poly peptide which is necessary inwards the metabolism of sure as shooting selective serotonin reuptake inhibitor drugs and the presence or absence of that factor has effects. Very modest sets of statistics that confirm the effects would hold upward valid epistemology just similar a real modest fix of statistics nearly people who jump out of planes amongst parachutes would enable yous to conclude that parachutes are a proficient reckon if yous desire to jump out of planes.
I read the statistics newspaper that the FDA staff position out on Lemtrada. It was non fun reading. It likewise convinced me that nosotros should hold upward circumspect nearly both our noesis and our methods of determining what nosotros know.
Circumspection nearly epistemological procedure nevertheless is difficult for a fund manager. It is fifty-fifty harder for a authorities department.
John
PS. I wrote that an involvement inwards epistemology would hold upward used every bit a criteria inwards assessing chore applicants at . I was non kidding.
Disclosure: at that spot is a traded contingent value correct whose value depends on approving for as well as sales of Lemtrada [ticker: GCVRZ]. We ain a modest quantity. We likewise about doubled the seat later the FDA staff newspaper was published.
Safety was no surprise. The statistical anomalies were.
Today the FDA Panel heard the case. The best coverage came from Twitter - as well as I am grateful to Sarah Karlin for her wonderful Twitter account. The panel voted on many things, nevertheless I desire to highlight only a couple:
Question 1: are the trials adequate as well as well-controlled?
Vote: eleven no, vi yes. 1 abstained.
Question 2: Has the applicant provided substantial evidence of effectiveness of alemtuzumab [Lemtrada] for the handling of patients amongst relapsing forms of MS?
Vote: 12 yes, vi no.
On enquiry #2, later the panel voted, the FDA staff asserted that if yous voted no on enquiry #1, yous can't vote yep on #2.
The commission chair said they voted yep based on gut feeling which I am non sure as shooting is adequate - but the FDA staff assertion is incorrect as well as I believe incompetent. It is perfectly acceptable to order that the trials are non "well-controlled" but at that spot is substantial evidence of effectiveness. I acquaint to you, yet again, the famous "parachute paper".
In that newspaper Gordon Smith as well as Jill Pell uncovering (with appropriate irony) that at that spot has never been a good controlled double-blind report of parachutes every bit a device to "prevent decease as well as major trauma related to gravitational challenge". Indeed because an adequate as well as well-controlled essay out is a requirement for approving parachutes are non approvable.
This is bunk.
We know why a parachute works.
We empathise the physics well.
Influenza A virus subtype H5N1 unmarried somebody who jumps out of an flat as well as lands unharmed amongst a parachute is a convincing essay out that parachutes are effective.
Bluntly, yous alone postulate double-blind stats when yous don't empathise the physics.
The FDA staff direct keep (as parodied inwards the parachute paper) raised their statistical mystique to a cult - every bit if it were the alone valid degree of epistemology. But at that spot are other ways of knowing, they are to a greater extent than valid (as per the parachute example) as well as at that spot is a deep underlying incompetence inwards the FDA cult of double-blind statistics as well as inwards the assertion highlighted above.
We are living inwards an historic menses where it is possible to direct keep deep noesis every bit to how the trunk works. We know for instance at that spot is a factor which codes for a poly peptide which is necessary inwards the metabolism of sure as shooting selective serotonin reuptake inhibitor drugs and the presence or absence of that factor has effects. Very modest sets of statistics that confirm the effects would hold upward valid epistemology just similar a real modest fix of statistics nearly people who jump out of planes amongst parachutes would enable yous to conclude that parachutes are a proficient reckon if yous desire to jump out of planes.
I read the statistics newspaper that the FDA staff position out on Lemtrada. It was non fun reading. It likewise convinced me that nosotros should hold upward circumspect nearly both our noesis and our methods of determining what nosotros know.
Circumspection nearly epistemological procedure nevertheless is difficult for a fund manager. It is fifty-fifty harder for a authorities department.
John
PS. I wrote that an involvement inwards epistemology would hold upward used every bit a criteria inwards assessing chore applicants at . I was non kidding.
Disclosure: at that spot is a traded contingent value correct whose value depends on approving for as well as sales of Lemtrada [ticker: GCVRZ]. We ain a modest quantity. We likewise about doubled the seat later the FDA staff newspaper was published.
No comments