So it would seem that they deliberately excluded the category for which there was a statistically significant difference, and substituted it with a closely-related category (barely) lacking a statistically significant difference.However, this contradicts their 2011 Lancet paper which states that "There were more serious adverse events in the GET group than there were in the SMC group (p=0·0433)."
They do it with no explanation of that omission, no mention of such a switch, and then label it as if it were the same category in which there was a significant difference.
Last edited: