That’s not how these kinds of trials work. The 13% figure comes from comparing the control and intervention groups, which were observed over the same period of time. The change in baseline mortality over time of the entire population isn’t relevant.