The NNT is the estimated average number of patients who need to be treated to prevent one additional bad outcome (e.g. death, hospitalization). An NNT which is calculated by the manufacturer may be included in the APS provided the published paper conveys a statistically significant improvement in the corresponding absolute risk reduction (with multiplicity adjustments performed when required to manage risk of type I error).
The number needed to treat is the mathematical inverse of the absolute risk reduction (i.e. 1 divided by the absolute risk reduction). Note that the absolute risk reduction is the control event rate minus the experimental event rate.