To watch the webinar video, please click on the picture below: |
|
|
Webinar Talk 1: The Planning and Conducting of a Prognostic Research
Prognostic Studies
- Combination of predictors that give you quantitative knowledge on occurrence/complications in absolute risk (hot relative risk)
- the aim of true prognostic study is to provide absolute risk → not to confuse prognostic with aetiological study
Model development
- validation → prove the score able to predict outcome accurately in real life situation.
- Good discrimination model:
- as much the positive, less false positive model
- AUC (Area under curve): 0.70 ( The closer to 1.0; the better the prediction of the probability)
- To adopt any prognostic model, must always validate in targeted population.
- Clinical implication - need to think also the practicality in real life.
SUMMARY
- Prognostic research is multivariable in nature
- follows usual clinical care
- It can actually work in real life
- Absolute risks are required
- validation
Further recommended readings:
- Karel G M Moons, Douglas G Altman, Yvonne Vergouwe, Patrick Royston. Prognosis and prognostic research: application and impact of prognostic models in clinical practice. BMJ. 2009 Jun 4;338:b606. doi: 10.1136/bmj.b606. https://www.bmj.com/content/338/bmj.b606.long
- Riley RD, Ensor J, Snell KIE, Harrell FE Jr, Martin GP, Reitsma JB, Moons KGM, Collins G, van Smeden M. Calculating the sample size required for developing a clinical prediction model. BMJ. 2020 Mar 18;368:m441. doi: 10.1136/bmj.m441. PMID: 32188600. https://eprints.keele.ac.uk/7880/1/bmj.m441.full.pdf
|
Updated:: 21/07/2023 [intanbasirah]
MEDIA SHARING