Salīdzināt metodes
Apskatiet izvēlētās metodes blakus; rindas, kas atšķiras, ir izceltas.
| Adaptīvā devas un atbildes reakcijas analīze× | Beijesas devas-atbildes analīze× | |
|---|---|---|
| Nozare | Epidemioloģija | Epidemioloģija |
| Saime | Process / pipeline | Process / pipeline |
| Izcelsmes gads≠ | 2000s (key papers 2005–2007; ICH E4 guidance 1994 for classical dose-response) | 1990s–2000s (Bayesian formalization) |
| Autors≠ | Frank Bretz, José Pinheiro and colleagues; foundational MCP-Mod framework | Developed from classical frequentist dose-response traditions; Bayesian formulations advanced by Dempster, Gelman, and colleagues |
| Tips≠ | Adaptive statistical design and analysis | Statistical modeling approach |
| Pirmavots≠ | Bretz, F., Pinheiro, J. C., & Branson, M. (2005). Combining multiple comparisons and modeling techniques in dose-response studies. Biometrics, 61(3), 738-748. DOI ↗ | Gelman, A., Carlin, J. B., Stern, H. S., Dunson, D. B., Vehtari, A., & Rubin, D. B. (2013). Bayesian Data Analysis (3rd ed.). CRC Press. ISBN: 978-1439840955 |
| Citi nosaukumi | adaptive DRA, adaptive dose-finding analysis, adaptive exposure-response analysis, adaptive D-R modeling | Bayesian DRA, Bayesian dose-response modeling, Bayesian benchmark dose analysis, BDR |
| Saistītās≠ | 6 | 3 |
| Kopsavilkums≠ | Adaptive dose-response analysis combines pre-specified dose-response modeling with planned interim looks that allow modifications — such as dropping ineffective doses or reallocating sample size — while maintaining statistical integrity. The most widely cited framework is MCP-Mod (Multiple Comparisons and Modeling), endorsed by the EMA and FDA as a fit-for-purpose methodology for dose-finding studies in drug development. | Bayesian dose-response analysis models the relationship between the level of exposure (dose) to a substance and the magnitude or probability of a biological response, embedding that model in a Bayesian probabilistic framework. Unlike frequentist approaches that yield a single point estimate with confidence intervals, the Bayesian framework produces a full posterior distribution over model parameters, allowing explicit quantification of uncertainty, incorporation of prior scientific knowledge, and principled model averaging. It is widely applied in toxicology, pharmacology, environmental risk assessment, and clinical dose-finding studies. |
| ScholarGateDatu kopa ↗ |
|
|