Non-inferiority (NI) trials compare new experimental therapies to stan- dard ones (active control). Since historical information on the control treatment is often available, a Bayesian approach to NI trials allows to exploit results from past studies and, eventually, to improve accuracy of inference. Here, we propose the use of a dynamic power prior: the active control treatment’s endpoint is modelled by a power prior distribution, whose informativeness is tuned by a measure of similarity between past and current information. The methodology is evaluated and compared to the frequentist method by simulation; an application to real drug data is available as well.
I test clinici di non-inferiorita` (NI) confrontano una terapia sperimen- tale e una standard (controllo attivo). Un approccio di tipo bayesiano ai test NI permette di sfruttare negli studi correnti i risultati degli studi passati e, eventual- mente, di migliorare l’accuratezza dell’inferenza. Qui viene proposta una metodolo- gia bayesiana in cui il parametro relativo all’effetto del controllo attivo e` modellato con una power priori informativa, la cui informativita` e` modulata secondo una misura di similarita` tra informazione storica e corrente. Mediante studi di simu- lazione, la metodologia e` valutata con criteri frequentisti; viene infine presentata un’applicazione a dati reali.
A dynamic power prior approach to non-inferiority trials for normal means with unknown variance / Mariani, Francesco; DE SANTIS, Fulvio; Gubbiotti, Stefania. - (2022), pp. 1191-1196. (Intervento presentato al convegno SIS 2022 - 51esima Riunione Scientifica della Società Italiana di Statistica tenutosi a Caserta).
A dynamic power prior approach to non-inferiority trials for normal means with unknown variance
Francesco Mariani
;Fulvio De Santis;Stefania Gubbiotti
2022
Abstract
Non-inferiority (NI) trials compare new experimental therapies to stan- dard ones (active control). Since historical information on the control treatment is often available, a Bayesian approach to NI trials allows to exploit results from past studies and, eventually, to improve accuracy of inference. Here, we propose the use of a dynamic power prior: the active control treatment’s endpoint is modelled by a power prior distribution, whose informativeness is tuned by a measure of similarity between past and current information. The methodology is evaluated and compared to the frequentist method by simulation; an application to real drug data is available as well.File | Dimensione | Formato | |
---|---|---|---|
Mariani_dynamic-power_2022.pdf
solo gestori archivio
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
1.06 MB
Formato
Adobe PDF
|
1.06 MB | Adobe PDF | Contatta l'autore |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.