Extrapolating from one death + one failed suicide attempt + a large number of mood disorders. ezzaf is quite right that you shouldn't extrapolate from the one

Extrapolating from one death + one failed suicide attempt + a large number of mood disorders. ezzaf is quite right that you shouldn't extrapolate from the one death alone, since that might have simply been a statistical fluke (although even one suicide is outside of the 95% range for this population and duration of study), but it becomes far more statistically powerful when supported by other side effects which are known to strongly correlate with suicides.