Research in medicine and statistics: BIO ELECTRICAL IMPEDANCE ANALYZER

#1
Hello there.

First, I want you to know that English is not my first language - I am sorry for the bad English!

In the past 7 months I have been collecting data on a Hospital. I have been testing something called a Bioelectrical impedance analyzer (BIA).
This machine can estimate body fat, fat free mass, fat precentage, body water, body water distribution (extracellular = water outside cells) (Intracellular = water inside cells) i.e.

I have been testing this on patients above 65 y with pneumonia and I want to find out if it can be validated. The problem is that i am not very much experience in statistics so I am feeling pretty "insecure" on working with the data.


The validation is done with different test on the machine:

Test 1:
I have measured 36 patients on both the BIA and something called a DXA-scanner (which is a gold standard for measuring body composition; fat free mass, body fat and fat% - just like the BIA).
I have used spearman's rank correlation, which showed a nice correlation and low p-value on both fat free mass, fat mass and f%.
Also I have used the bland-altman plot which on the graph showed that the measurements was not systematic at all.
Question:
Do you guys have any idea what else i could do with these data and how to do this?
Also i could use a little help to interpret the data to make sure im on the right track.

Test 2:
I have measured 32 patients and done blood test when they were hospitalized and when they were released from the hospital again.
Here i want to know if the results from measuring body water can be validated. I have these variables:

Hospitalized:
total body water, extracellular water (outside cells), intacellular water (inside cells), fat mass, fat free mas.
Blood test: Creatinine and haemaglobin (which can be used to measure changes in extracellular water)

Released:
total body water, extracellular water (outside cells), intacellular water (inside cells), fat mass, fat free mas,
Blood test: Creatinine and haemaglobin (which can be used to measure changes in extracellular water)

Question:
I want to find out if there is any correlation or agreement with the changes between extracellular water , and changes in haemaglobin or creatinine from the 1st and 2nd measurement.



I have made some more testings - but i think i will keep it there for now and focus on test 1 and 2.

Any of you have comments, ideas or thoughts on these methods and statistics?


thanks in advance
Any help is appreciated!!!


/Daniel