Grant: Travel Awards - Institute Community Support

Understanding the progression of chronic diseases, such as COPD, is important to inform early diagnosis, personalize care and healthcare system management. Data from clinical and administrative systems have the potential to advance this understanding, but traditional methods for modeling disease progression are not well suited to analyzing data collected at irregular intervals, such as when a patient chooses to interact with the healthcare system. Many cases of severe illness can be traced back to a lack of appropriate and ongoing followup health care. To help understand the long-term causes of severe and preventable illness, we need to understand not only how many patients have severe illness, but also what their "care path" or "trajectory" was before they ended up with severe illness. Until recently, it was very difficult and costly to collect the patient data from paper-based records, making it impossible to efficiently track patient trajectories. With the advent of electronic health records, we can now collect patient data automatically. The patient data is still, however, very complex; to identify patient trajectories automatically, we need new statistical methods. From my previous work, I have developed a new statistical method via continuous-time processes to identify meaningful disease progression and transition patterns according to individual characteristics, which are interpretable by healthcare system managers and can be used to measure the performance of the healthcare system over time. Based on this work, I extended the model to cluster the patient trajectories to explain population-level variations in trajectories and to optimize clinical care strategies by using health care trajectories for dynamic treatment decisions. Unlike most clustering techniques, the approach I proposed allows the number of clusters to be learned as a function of the sample size.