PROJECT SUMMARY/ABSTRACT Despite rising incidence, prevalence and healthcare costs of inflammatory bowel diseases (IBD) in older adults, there continues to be paucity of evidence-based treatment guidance for management of these understudied and susceptible patients. The long-term goal is to inform evidence-based, risk-congruent treatment approach for older patients with IBD, balancing patients’ dynamic risk of disease- vs. treatment- related complications with specific treatment’s efficacy vs. safety. The overall objectives in this application are to accurately predict the short- and long-term risk of serious infections with immunosuppressive therapy in older patients with IBD, based on pre-treatment baseline clinical characteristics and time-varying treatment effect. The central hypothesis is that pre-treatment clinical characteristics can accurately predict short-term risk of serious infections in older patients with starting immunosuppressive therapy; this risk evolves over time, on therapy, and can be accurately predicted by accounting for time-varying impact of treatment effectiveness. By controlling disease effectively and decreasing the need for corticosteroids and/or opiates, an effective, yet potent immunosuppressive therapy will be associated with lower risk of serious infections over time. The rationale for this project is that accurate individualized risk prediction for treatment-related complications in older adults with IBD on immunosuppressive therapy will provide patients and providers knowledge to tailor therapy based on patients’ predicted risk of disease- and treatment complications. The central hypothesis will be tested by pursuing two specific aims: 1) develop and validate a model that accurately predicts the short- term risk of serious infections in older patients with IBD starting immunosuppressive therapy; and 2) develop and validate a model that accurately predicts the long-term risk of serious infections in older patients with IBD while on immunosuppressive therapy. Under the first aim, the applicant will develop and validate a risk prediction model to accurately identify patients at high- vs. low-risk of serious infections in the short-term (within 6 months of treatment initiation) based on pre-treatment baseline patient-, disease- and treatment characteristics. For the second aim, a separate risk prediction model will be developed and validated to accurately predict long-term (6 to 24 months after treatment initiation) risk of serious infections, accounting for time-varying treatment effect, in older patients from the same cohort who have been on immunosuppressive therapy for 6 months. The context for this proposal will be the comprehensive, longitudinal Danish nationwide register of all older patients with IBD in Denmark. The research proposed in this application is innovative, in the applicant’s opinion, because it incorporates time-varying treatment effect, in addition to patients’ intrinsic susceptibility to infect...