I am trying to use measurements from multiple studies. However, some report survival and some report mortality.
I know I can get survival rates just by subtracting mortality rates from 1. However, if I want to have an estimate of the SE in terms of survival as well, how would I achieve that?
Here You deal with a meta-analysis ? or what exactly? because the SE estimation would be directe if you want to estimate it for all studies.
Yes, sir. I am doing a meta-analysis. I am trying to get all of the measurements in the same terms (survival) because I want to compare them.
Yes dear، you can make all of them in terms of survival rate and estimate directly the standard error
I suppose I should clarify. How would I estimate the survival standard error from the mortality standard error? I don't have all the data to estimate the survival SE directly.
Hello,
Ahh ok dear; I misunderstood your first question,
In this case the SE for survival and mortality functions are the same; Why ?
I explain you; As you know:
survival function S(t) or survival rate (p) = [ 1- mortality function (F(t) or mortality rate (q) ].
As a property of variance, we have :
- Y= X+c, so V(Y)= V(X+c)= V(X);
- V(X) =V(-X)
Thus, SE(Y)=SE(X); We repalce Y= S(t) or (p) and X by F(t) or q;
so, V(p)= V(1-q), here , c= 1; give us, V(p)=V(-q) = V(q);
we make the square of the two side of equation, we have: SE(p) = SE(q).
As a final result, you can report SE for mortality rate as SE of surival rate.
Best Luck
- Badges
- Science topic
- Similar topics
- Biological Science
- Zoology
- Ornithology