Indeed, efficacy and safety data are normally reported separately, as if they had been obtained from different populations; moreover, the units of measurement frequently correspond to non-comparable scales of benefit and risk.
The surveys, pioneered in the United States by the Institute for Supply Management and conducted by Markit in many other countries, are normally reported with 50 as the break-even level, a number that indicates that half the companies that see a change in their orders are showing gains.
Class-size numbers are normally reported twice a year as citywide averages, showing small increases spread out over many classrooms and over many years.
Continuous variables that were not distributed normally were reported as median with interquartile range (IQR) and significance of differences between groups tested using the Mann-Whitney U test.
Variables (distributed normally) were reported as mean and standard deviation (mean ± SD) and otherwise as median and the 25th and 75th percentile (M (P25– P25)).
Variables (distributed normally) were reported as the mean and standard deviation (mean ± SD) or as the median and the 25th and 75th percentiles.
Group comparisons were made using chi-square tests for equal proportion, Student t-tests for normally distributed data and Wilcoxon rank sum tests otherwise, with normally distributed data reported as means ± standard errors and non-normal data reported as medians (interquartile range).
