I said it last year, and I’ll say it again — I love the Canadian Medical Association Journal.
Last year, CMAJ published a very funny parody of review articles by using the format to argue that smoking is good for endurance athletes, which was on-par with an earlier article extolling the health benefits of the unmade bed, a parody of non-financial conflicts of interests (the authors didn’t like making their beds).
This year, CMAJ is back with an equally important article entitled, “The Mayan Doomsday’s effect on survival outcomes in clinical trials.” The authors divided their study population into two groups — a control group and an obliteration group — and noted the following:
. . . [Mayan Doomsday] leads to a statistically significant, and clinically relevant, difference in survival between the control and obliteration groups (we’re pretty sure that, were it calculated, p would definitely be something really significant, and certainly less than 0.05).
But the authors also note a disturbing and unexpected result from their studies:
Oddly, despite censoring for major known sources of bias (e.g., astronauts currently aboard the international space station, as well as zombies, the undead, the Grateful Dead, Dungeons and Dragons players, men who have read Fifty Shades of Grey and other similar beings likely to be unaffected by the apocalypse), the obliteration group does not fall to 0. We have dubbed this slow rise in the obliteration curve the “zombie repopulation.”
There are a few excellent shots taken at bureaucracies and ethics committees, and the outline of the paper is a thing of beauty, as well:
- Mostly true background
- Stuff we did
- Stuff we found out
- Chit chat
The recommendations related to the Mayan Doomsday are clear:
If we have been thinking clearly, then it is apparent that the end of the world will have catastrophic effects on statistical analyses of survival outcomes. We therefore recommend that all clinical trials should stop immediately, as [the Mayan Doomsday] will negate all potential trial results.
So, it’s time for clinical trialists to put their feet up until December 22nd, to see if their skills are still needed, or if they should be reoriented toward zombie research. Until then, enjoy reading CMAJ.
(Hat tip to SW for the pointer.)