Because infancy-related deaths have clearly declined most in recent decades and excluding them by some arbitrary cut-off still is reasonable methodologically? A

Because infancy-related deaths have clearly declined most in recent decades and excluding them by some arbitrary cut-off still is reasonable methodologically? A graph would have been better.