Modelo multiestado paramétrico para el estudio del lupus: estimación de las probabilidades de transición a través de simulación.
Cargando...
Fecha
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
El análisis de supervivencia de pacientes puede enriquecerse mediante la consideración de eventos no fatales durante el seguimiento dando lugar a esquemas multiestado. En particular para el estudio del lupus, interesa estimar las probabilidades de transición entre los estados Diagnóstico de lupus, Complicación severa y Muerte mediante el ajuste de los denominados modelos multiestado. Mediante la reestructuración de los datos como objeto Lexis se ajustan modelos Poisson para obtener las tasas de transición. A partir de ellas se recurre a un método de simulación para la obtención de las probabilidades de transición. Entre los resultados hallados se destaca que durante los dos primeros años de seguimiento, la probabilidad de transitar a la muerte es prácticamente nula. A partir de ese momento, comienza lentamente a aumentar llegando, alrededor de los 8 años, a 0,04 para quienes pasaron previamente por una complicación severa y a 0,01 para quienes no transitaron por este estado intermedio.
The survival analysis of patients can be enhanced by considering non-fatal events during follow-up, leading to multi-state models. Specifically, for the study of lupus, it is important to estimate the transition probabilities between Lupus Diagnosis, Severe Complication, and Death through the adjustment of so-called multi-state models. By restructuring the data as a Lexis object, Poisson models are fitted to obtain the transition rates. From these rates, a simulation method is used to obtain the transition probabilities. Among the findings, it stands out that during the first two years of follow-up, the probability of death is virtually null. Then, it begins to slowly increase, reaching around 0.04 at approximately 8 years for those who previously experienced a severe complication and 0.01 for those who did not go through this intermediate state.
The survival analysis of patients can be enhanced by considering non-fatal events during follow-up, leading to multi-state models. Specifically, for the study of lupus, it is important to estimate the transition probabilities between Lupus Diagnosis, Severe Complication, and Death through the adjustment of so-called multi-state models. By restructuring the data as a Lexis object, Poisson models are fitted to obtain the transition rates. From these rates, a simulation method is used to obtain the transition probabilities. Among the findings, it stands out that during the first two years of follow-up, the probability of death is virtually null. Then, it begins to slowly increase, reaching around 0.04 at approximately 8 years for those who previously experienced a severe complication and 0.01 for those who did not go through this intermediate state.
Descripción
Citación
Aprobación
Revisión
Complementado por
Referenciado por
Licencia Creative Commons
Excepto donde se indique lo contrario, la licencia de este ítem se describe como Attribution-NoDerivs 2.5 Argentina

