To control for history effects in an interrupted time-series study, a researcher includes a comparison group that does not receive the treatment. What is one drawback to using this technique