Respiratory
Real-world effectiveness of infant RSV preventative interventions: a target trial emulation study using a novel analytic approach Sara S Kim* Sara Kim Kim Kim Kim Kim Kim Kim Department of Epidemiology, Emory University Rollins School of Public Health
Respiratory syncytial virus (RSV) is the leading cause of infant hospitalization in the United States. In 2023, long-acting monoclonal antibodies (mAbs) and maternal bivalent prefusion F vaccination (RSVpreF) were approved to protect infants. To estimate real-world effectiveness of each intervention, we applied a novel target trial emulation (TTE) method that targets a similar estimand to matching-based studies designed to emulate phase 3 randomized controlled trials (RCTs) while leveraging data from all observed individuals without requiring matching to identify a randomization time.
Using claims data, we identified 165,516 infants eligible to receive mAbs or maternal RSVpreF and followed them from birth until an RSV-related event, insurance lapse, or study end. We estimated effectiveness of mAbs and RSVpreF against RSV hospitalization and medically attended illness (MAI) using a g-computation-based estimator. For each intervention, we estimated cumulative incidence with (ψ1) and without (ψ0) the intervention by fitting separate Cox proportional hazards models among infants who received the intervention and among those who had not yet received it and marginalizing over covariates and time of receipt. We defined effectiveness as (1- ψ1/ψ0). For RSVpreF, we assumed protection from birth and began follow-up at birth for all infants.
At 150 days post-intervention, mAbs reduced the risk of RSV hospitalization by 82% (95% CI: 74-88%) and RSV MAI by 75% (95% CI: 70-78%). RSVpreF reduced risk of RSV hospitalization by 96% (95% CI: 91-99%), but protection against RSV MAI was lower (48% (95% CI: 16-67%)).
Both interventions performed similarly as in their respective phase 3 RCTs, indicating sustained effectiveness in real-world settings and illustrating the utility of the target trial emulation approach.