CD - The deviations you're referring to are the standard deviations of the two groups - Leron and Placebo. You need access to the raw data to calculate those. If the standard deviations (or, more precisely, the standard errors) are large relative to the means of the two groups, then the differences between the averages need to be larger for a statistically significant result; if they are smaller, then the differences in means can be smaller for significant result. Picture a scatter plot of the individual patient scores - if they are pretty much scattered over a wide range then standare deviations are larger; if they are closely clustered, then the SD will be smaller. Hope this helps.