PakMediNet Discussion Forum : Biostatistics : Functional outcome in patients with close fracture shaft of humerus treated non-operatively by coapt
Functional outcome in patients with close fracture shaft of humerus treated non-operatively by coapt
1.what study design i will have to choose
2.please send me the local reference of this study
3.how i will analyse my data statistically
Posted by: dr kakaPosts: 5 :: 16-05-2009 :: | Reply to this Message
Re: Functional outcome in patients with close fracture shaft of humerus treated non-operatively by coaptPlease u use prospective study design. T-test will be more suitable test in this case
Posted by: ibrahim_apPosts: 138 :: 19-05-2009 :: | Reply to this Message
Re: Re: Functional outcome in patients with close fracture shaft of humerus treated non-operatively by coaptsir thanks for co operation.now i m left with only sample size and reference of any study held in pakistan regading conservative treatment of shaft of humerus.plz help me in these issues as well.thanx
Posted by: dr kakaPosts: 5 :: 24-05-2009 :: | Reply to this Message
Re: Functional outcome in patients with close fracture shaft of humerus treated non-operatively by coaptsir plz help me in calculating the sample size.thanks
Posted by: dr kakaPosts: 5 :: 02-06-2009 :: | Reply to this Message
Re: Functional outcome in patients with close fracture shaft of humerus treated non-operatively by coapt
What is functional outcome? Will it be a a binary variable such as good or bad? Will it be a scale? If it is a scale, what will be the scale like? You may choose to consider scale (which is an ordinal measure) a continuous one and use mean(SD). You can also give frequencies.
I am not sure what you want to compare. There seem to be only one group; the one with close fracture treated with coapt. For a t-test, you will need to have two groups with outcome a continuous variable. Without knowing the second group, you can't do t-test. If there are more than 2 groups with a continuous variable, you can do ANOVA.
Unless you are sure what you want to study, you will not be able to calculate sample size or for that matter study design. You can calculate frequencies of the outcome in a cross-sectional study or with retrospective chart review of the data.
Bottom line, you HAVE to be clear what is your study question, it is not clear now.
Posted by: rqayyumPosts: 199 :: 24-12-2009 :: | Reply to this Message