OBJECTIVES: With the development of prophylactic interventions for respiratory syncytial virus (RSV), an increasing number of RSV cost-effectiveness models have been published in the literature, including both static and dynamic models. However, these models can produce varied outcomes partly due to different model structures, assumptions, input parameters, and country contexts. Our study aims to characterise and compare cost-effectiveness models of RSV interventions and to examine drivers for different model outcomes.
METHODS: We compared three static and two dynamic models using a common input parameter set for a hypothetical birth cohort of 100,000 and discounted the outcomes at 3% annually. Year-round and seasonal programmes were evaluated for both maternal vaccine (MV) and monoclonal antibody (mAb) interventions. Extensive sensitivity analyses were performed.
RESULTS: Three static models projected similar numbers of medically-attended (MA) cases averted compared to no intervention (MV: 1079-1073, mAb: 5,075-5,481), with the MV year-round programme directly saving ~€1 million medical and €0.3 million non-medical costs, while gaining 4-5 discounted QALYs annually in children <1 year, and mAb resulting in €4 million medical and €1.5 million non-medical direct cost savings, and 21-25 discounted QALYs gained. In contrast to these static models, (i) both dynamic models estimated fewer MA cases averted (MV: 402-752, mAb: 3,362-4,622), (ii) one dynamic model found residual infections/cases to undergo an age shift through herd immunity, (iii) the other dynamic model reported many more non-MA symptomatic cases averted, especially by MV (2014). Sensitivity analyses showed most differences can be explained by model type and model-specific assumptions on waning efficacy.
CONCLUSIONS: Our static and dynamic models produced overall similar results in terms of hospitalisations and deaths, but there were important differences, especially in non-MA cases. These explain most of the resulting differences in incremental cost effectiveness ratios indicating that the non-MA symptomatic RSV burden is important when evaluating RSV interventions.
