Smoothed Change-Point Renewal Process Framework for Modeling Hazards of Pulmonary Exacerbations in Cystic Fibrosis

• Activity Number: 522 - Contributed Poster Presentations: Biometrics Section
• Type: Contributed
• Date/Time: Wednesday, July 31, 2019 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Rachel Johnson* and Elizabeth Juarez-Colunga and John Rice and Brandie Wagner and Edith Zemanick and Margaret Rosenfeld
• Companies: Colorado School of Public Health and University of Colorado Denver and Colorado School of Public Health and Colorado School of Public Health and University of Colorado School of Medicine and Children’s Health Colorado and Seattle Children's Hospital
• Keywords: change-point; gap time; survival analysis; recurrent events; renewal process

Abstract

Cystic fibrosis research often utilizes time to first pulmonary exacerbation (PEx) or total number of PEx as clinical trial endpoints. These outcomes fail to capture patterns or timing of PEx and how covariates influence risk of further PEx, but these can be captured by analyzing gap times between PEx. Renewal processes—often used in reliability or engineering to model recurrent events or gap times—model gap times and can be adapted to account for progressive nature of chronic diseases. A smoothed change-point hazard is included in the renewal process to model elevated hazards after PEx which later return to some baseline hazard. We conducted a simulation study to determine model behavior in different scenarios, demonstrated this method on data from children enrolled in the Early Pseudomonas Infection Control (EPIC) study, and provide a summary of models with different change-point and trend structures. We found that a change-point provided the best fit to the data and that hazard of a future PEx dropped after 1.8 years, and that delF508 mutation copies, female sex, and higher numbers of previous PEx are significantly associated with high risks of future PEx.