Bio Stat Question:

Question

1. Is there a difference in age between patients on the fast-track protocol, and those

receiving standard care?

2. Is the proportion of patients who are female in the fast-track group significantly different

from the proportion female in the standard care group?

3. Is there a difference in weight gain between patients on the fast-track protocol and those

receiving standard care?

4. Is the variability of length of stay similar among patients on the fast-track protocol, and

those receiving standard care?

5. Is there a difference in hospital length of stay among patients on the fast-track protocol,

and those receiving standard care?

I have data for 50 patients, however, I am missing previous weight for 2 of the 50 patients. When I have to do the above analysis, do I automatically eliminate the 2 patients for all the questions? Or do I use their data for Questions 1, 2, 4 and 5 but not 3 since I can't determine if they have any weight changes. If that's the case, then for some question n=50 and for question 3 n=48???????

Answer 1

I would go with N= 48 as well. Unless your instructor wants you to account for "drop out" or account for outliers - there is no need to include them. You simply state no data was available for those two. 

Answer 2

You could do it either way, but I'd learn towards n=48 for all questions since you are repeatedly referring to the 'fast track group'.  There should certainly be an expectation that not all candidates' data can be used; in the 'real world' you'll be tossing data if it is unusable.