Evaluating the impact of population-based and cohort-based models in cost-effectiveness analysis: a case study of pneumococcal conjugate vaccines in infants in Germany.

Abstract

Objective: The objective of this analysis is to evaluate the impact of model choice (closed single-cohort versus population-based) in cost-effectiveness analysis (CEA) using pneumococcal conjugate vaccines (PCVs) in infants in Germany as a case study.

Methods: Two Markov models were developed: one with a closed single-cohort model and one with a population-based model. Except for the design of the modelled population/cohort, all other inputs and characteristics were kept identical between the models. Comparators included PCV20 under a 3 + 1 vaccination schedule versus PCV13 and PCV15 under a 2 + 1 vaccination schedule. Health and economic outcomes were compared between the two models.

Results: The population-based model demonstrated that PCV20 was cost-saving and provided better health outcomes compared to both PCV13 and PCV15, indicating PCV20 as the dominant strategy with negative ICERs per QALY. In contrast, the closed single-cohort model showed PCV20 was associated with higher total costs compared to PCV13 and PCV15.

Conclusion: This analysis highlights the importance of accurately identifying the relevant population when conducting CEAs of vaccines. This is particularly crucial when a vaccine produces indirect effects in individuals who are not directly vaccinated, as this otherwise leads to an underestimation of cost-effectiveness.