how can you extrapolate on a total population from such a small sample and not reviewing correlations with other preexisting conditions of SAE