Donald F. Klein’s reply to Leslie Morey’s comment
Dr. Morey's comment is somewhat difficult to follow. (Perhaps my trouble.) At any rate we agree that the tables presented by Dr. Katz present labeling problems. I disagree about the singular importance of NPV (NPP).
It is unfortunate that Dr. Morey did not relabel the 2 x 2 tables from which he derives his indices. With regard to the Hamilton scale my reconstruction is:
<50% > 50%
< 20% 15 2 PPV = 25/(25+8) = 0.76
> 20% 8 25 NPV = 15/(15+2) = 0.88
Recovery rate 27/50 = 0.54
Dr. Morey presents an NPP of 90.9% which seems close to my estimate of 0.88, but for such tables there should be no difference at all. Further, I can’t follow the statement“If those non-early responders (on the HAM-D) had all been discontinued early in the trial, only 9.1% of these patients would have ultimately responded by the end of the trial”.
Dr. Morey seems to feel that a NPP approximating 90% justifies discontinuation early. Certainly not in a clinical trial measured by ITT standards. Nor in the clinical case that often bears little resemblance to the study cases.
A close reading of Katz indicates that he was referring to a negative outcome across the range of measurements, rather than on a single measurement, to predict a bad outcome.
Finally, it should be noted that this is a single index and that no confidence limits are shown or suggested. Nor is it clear whether 95% limits or 90% limits, or broader, are in order.
Dr. Morey concludes, “I believe that Dr. Katz is still correct—namely, that a decision to discontinue treatment following the lack of an early response is likely to be correct (based upon these limited data, and on the HAM-D) about 90% of the time.”
In my last discussion of Marty’s statements, it is pointed out that the outcomes, based on his reported Hamilton scale measures can find no significant difference between “"Active drug” and placebo. Therefore, this is a failed trial by FDA standards. Drawing any conclusion from it is most dubious.
Donald F. Klein
March 31, 2016