Timeline for Which models are available to estimate relative life expectancy (in years, not in hazard ratios)?
Current License: CC BY-SA 4.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 19 at 10:44 | comment | added | DianaS | Agree on custom-made models requiring extensive coding not being worth it in many cases :) | |
| Mar 19 at 10:29 | comment | added | DianaS | Thank you for so many good points on models' pros and cons. I haven't thought on hazard ratios (HR) vs additive hazard in terms of generalisability, though it's important. The problem with HR in diverse populations is that we multiply by the same number different bases, e.g. if smoking HR is 1.5x, then for a person with high baseline risk that would be a higher impact on mortality than for a person with low risk. Additive hazards would result in the same increase in hazard. Clinically, I may think of both situations being possible (and any model having difficulty for a diverse population) | |
| Mar 18 at 16:50 | history | answered | Björn | CC BY-SA 4.0 |