Hi Rachel,
I agree sample size calculations are more complex and that rigorous methodologies to evaluate ASP programs have not been the norm. But that can be one of your contributions.
As a start to being more rigorous, i would use the variance calculators (see my general comment posted to the whole group) to estimate the impact of auto-correlation on your simple sample size/power estimate. This is something you can include in your final presentation.
When actually designing the study, I would recommend involving a statistician to help you think through how best to use the 3 years of pre-data you have to estimate auto-correlation and trends, and to do a more sophisticated calculation that takes into account the number of pre- and post- observations you have, standard deviation, auto-correlation and trends.