# Thread: Clinical research, Adequate statistical approach?

1. ## Clinical research, Adequate statistical approach?

Dear all!

Having completed basic science projects, entering clinical research demands a whole set of other statistical approaches. Please help :-)

The clinical questions are:

Which variables are chief responsible for the complication rate after a surgical treatment?
1) overall complication rate (any of the 11 types of complications): yes/no
2) the specific 11 types of complication (yes/no)

Clinical suspects:
- age [years]
- gender (2 subtypes: M,F)
- type of medical condition (9 subtypes: 1,2,3,4,5,6,7,8,9)
- delay of operation [days]
- surgeon's experience [4 subtypes: 1,2,3,4]

No missing data at all.

Available statistical software: STATA, Graph Pad Prism 6.

Looking forward to hearing from you! Thanx in advance!

2. ## Re: Clinical research, Adequate statistical approach?

So you want to find predictor of 1 & 2. For 2, do you want to examine each individually as well?

3. ## Re: Clinical research, Adequate statistical approach?

I am sorry for joking with you, DrDan (in a serious area) but it seems like you know much more about your colleagues experience than about your patients gender and age. Because in the file I saw there was a lot of missing values there. Then I saw:

No missing data at all.
So maybe there is something wrong with the file you sent, or I could not read it in properly. Since I know that many here are reluctant to open an excel-file (this was an exception for me) I just show a copy of the first 15 rows (out of DrDan's about 750 rows).

Spoiler:

I guess that when there is a "1" in say C3 then there has been a complication there. And all the zeros are not included. So it is a quite sparse matrix.

I could not see any data for
- type of medical condition (9 subtypes: 1,2,3,4,5,6,7,8,9)

This is an interesting question. How should this be modelled?

What is hlsmith talking about? Is my data crazy?

4. ## Re: Clinical research, Adequate statistical approach?

hlsmith did not open the mystery file. So your sanity may be unblemsihed.

5. ## Re: Clinical research, Adequate statistical approach?

Originally Posted by hlsmith
hlsmith did not open the mystery file. So your sanity may be unblemsihed.
I did not hypothesize that I was crazy, only that my data were.

But I didn't understand were you got the 1 & 2 from.

So you want to find predictor of 1 & 2.

6. ## Re: Clinical research, Adequate statistical approach?

Just used the assumed data behind the person's "1)" and "2)".

I would question the type I and type II error rates around lunacy hypothesis based on self-study.

7. ## The Following User Says Thank You to hlsmith For This Useful Post:

GretaGarbo (09-29-2015)

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts