The same idea has been used by others in a less developed way, looking at the proportion of baseline tests which have P<0.05 and arguing that this should be 5%. Schulz et al. [1] reported that, in 125 published trials in which hypothesis tests had been used to compare baseline characteristics, only 2% of 1076 test results were statistically significant, lower than the expected rate of 5%, a significant discrepancy (P<0.001).
