Budget impact analysis of a pneumococcal vaccination programme in the 65-year-old Spanish cohort using a dynamic model.

BACKGROUND: This study aimed to assess the costs and clinical benefits of the 13-valent pneumococcal conjugate vaccine (PCV13) administered annually to the 65-year-old cohort in Spain versus the alternative of not vaccinating patients and treating them only when infected. METHODS: Cases of pneumococcal disease avoided were calculated through a dynamic model based on the work of Anderson and May (1999). Sixty-six percent of the 65-year-old cohort was assumed to have been vaccinated with one PCV13 dose (304,492 subjects). Base-case estimated vaccine effectiveness and serotype coverage were 58% and 60%, respectively. Disease-related costs were calculated based on published data. RESULTS: Over the 5-year period, a total of 125,906 cases of pneumococcal disease would be avoided. Net savings of €102 million would be obtained. The cost-saving distribution was not homogeneous, starting in the 2nd year and increasing through the 5th. To demonstrate model robustness, an additional scenario analysis was performed using extreme values of model parameters (vaccination programme coverage, vaccine effectiveness, discount rate and disease costs). Under those scenarios, net savings were always achieved. CONCLUSIONS: Based on the assumptions of the model, the 65-year-cohort pneumococcal vaccination campaign appears to be a cost-saving intervention in the Spanish population under different scenarios.

Los modelos de Markov probabilísticos en la evaluación económica de tecnologías sanitarias: una guía práctica / Probabilistic Markov models in economic evaluation of health technologies: a practical guide

ObjetivoLos modelos de Markov son el método estándar utilizado en los estudios de coste-efectividad para representar la historia natural de la enfermedad. El objetivo de este trabajo es mostrar los elementos clave en la construcción de modelos de Markov de tipo probabilístico.MétodosSe ha utilizado el ejemplo de un nuevo tratamiento para una enfermedad genérica. Para ello se ha construido un modelo de Markov con parámetros introducidos como distribuciones estadísticas para llevar a cabo el análisis de sensibilidad probabilístico mediante simulaciones de Monte Carlo. Los resultados se analizaron en forma de plano coste-efectividad y curva de aceptabilidad.ResultadosLa razón coste-efectividad incremental para el paciente medio es de 22.855€/año de vida ajustado por calidad (AVAC). En el análisis de sensibilidad probabilístico el resultado de todas las simulaciones se sitúa en el cuadrante nordeste, que corresponde a coste y efectividades positivas. El 67% de las simulaciones se sitúa por debajo del umbral de los 30.000€/AVAC.ConclusiónLa utilización de los modelos de Markov de tipo probabilístico requiere la integración de conceptos provenientes de la economía, la epidemiología, la estadística y la clínica. Algunas etapas del proceso, como la construcción y el procesamiento del modelo, la gestión de los riesgos absolutos y relativos, y el manejo de las distribuciones estadísticas, suelen plantear mayores dificultades, pero son necesarias para que el modelo reproduzca la enfermedad de forma válida(AU)

ObjectiveMarkov models are the standard method used in cost-effectiveness studies to represent the natural history of disease. The objective of this study was to show the key elements in building probabilistic Markov models.MethodsWe used the example of a new treatment for a generic disease. A probabilistic Markov model was constructed using statistical distributions. Monte Carlo simulations were carried out to obtain the probabilistic sensitivity analysis. The results were analyzed in terms of the cost-effectiveness plane and acceptability curve.ResultsThe incremental cost-effectiveness rate for the average patient was €22,855/quality adjusted life years (QALY). In the probabilistic sensitivity analysis, the results from all simulations were located in the northeast quadrant, corresponding to positive cost and effectiveness. However, 67% of the simulations were below the threshold of €30,000/QALY.ConclusionThe use of probabilistic Markov models requires the integration of concepts from economics, epidemiology, statistics, and the clinical setting. Some stages of the process, such as the construction and processing of these models, the management of absolute and relative risks and of statistical distributions, often pose major difficulties but are key steps required to reproduce the disease with validity(AU)

[Probabilistic Markov models in economic evaluation of health technologies: a practical guide]. / Los modelos de Markov probabilísticos en la evaluación económica de tecnologías sanitarias: una guía práctica.

OBJECTIVE: Markov models are the standard method used in cost-effectiveness studies to represent the natural history of disease. The objective of this study was to show the key elements in building probabilistic Markov models. METHODS: We used the example of a new treatment for a generic disease. A probabilistic Markov model was constructed using statistical distributions. Monte Carlo simulations were carried out to obtain the probabilistic sensitivity analysis. The results were analyzed in terms of the cost-effectiveness plane and acceptability curve. RESULTS: The incremental cost-effectiveness rate for the average patient was 22,855 euro/quality adjusted life years (QALY). In the probabilistic sensitivity analysis, the results from all simulations were located in the northeast quadrant, corresponding to positive cost and effectiveness. However, 67% of the simulations were below the threshold of 30,000 euro/QALY. CONCLUSION: The use of probabilistic Markov models requires the integration of concepts from economics, epidemiology, statistics, and the clinical setting. Some stages of the process, such as the construction and processing of these models, the management of absolute and relative risks and of statistical distributions, often pose major difficulties but are key steps required to reproduce the disease with validity.