Calculating a two week difference in new cases based on a yes/no condition.

What have you tried? Can you provide a reproducible example?