I need to show how many wounds are closed after each time, to different groups. Can I use a survival graphic to show this data, or it is not indicated? What could be a better choice?
Of course yes, but attention I draw your attention to define what do you mean by healing carefully as an outcome, this phenomena should be will define as dichotomous variable before to involve survival analysis on the subject.
Yes, the statistical basis for analysing an outcome like "healed" will be the same as for an outcome like "survived", ie a Kaplan-Meier curve. This will likely be a sigmoid shape. This can then be subjected to probit regression analysis to look for significant contributory factors.
The vital thing, as Dr Khamis says, would be to clearly and explicitly define what is meant by "healed". Many studies use slightly incorrect definitions here and assign the term "healed" to basically unhealed states, like 90% or 95% healed. Don't fall into that trap! :)
Yes, the Kaplan-Meier analysis can provide this.
See eg Moulik et al. Diabetes Care. 2003 Feb;26(2):491-4. PMID: 12547887
- Badges
- Science topic