Abstract
Variation in lung function is associated with elevated risk of worsening events in asthma. As more high-frequency data becomes available in respiratory clinical trials, new methods can enable faster and more accurate estimation of treatment effects. This presentation will feature one paper in which a mixed-effects hidden Markov model is developed for home-measured peak expiratory flow. A modified version of the stochastic-approximation expectation-maximization (SAEM) algorithm is implemented and evaluated. We demonstrate the model’s ability to detect and quantify treatment effects in a clinical trial in asthma. The presentation will further discuss ongoing and future projects.
For further information, please click here.