Frequent surveillance testing is recommended and routinely conducted in several disease settings. Although surveillance tests provide information about current disease status and present an opportunity to detect disease progression early, the complications and costs of frequent tests may not be justified for patients who are at low risk of progression. I will discuss our recently developed Personalized Risk-Adaptive Surveillance (PRAISE) framework, a method for embedding dynamic predictions into a sequential decision-making framework to determine the time point at which the next collection of patient data would be most valuable, as well as a preliminary application of the framework to develop more cost-effective surveillance strategies in cystic fibrosis. I will begin with a short talk aimed at statistical audiences, followed by a presentation of the same work targeted to non-statistical audiences, and end with general tips on presenting technical work to non-statistical audiences.