# Adjusting data for age and bmi

I am doing a study on the effects of bad blood sugar control on patients. BMI, age, type of diabetes (Type 1 and 2 or gestational diabetes) are all variables. Outcomes include C section, early labour etc. A reviewer told me

Comparisons should be adjusted to maternal age and baseline BMI.

How does one do this with an online calculator? I have been doing my calculations only with online calculators and have never ventured to use SPSS. Can anyone advise me?

-
Hi there, can you put in a link to the online calculator you have been using, and also let us know what your various analyses are? I'm assuming you have a few as you suggest more than 2 outcomes in your question. –  Michelle Feb 20 '12 at 9:13
–  FSNWRM Feb 20 '12 at 9:53
quantpsy.org/chisq/chisq.htm –  FSNWRM Feb 20 '12 at 9:55

With respect, if you need to ask this question you should not be attempting to "adjust" your analysis with an online calculator. Seek some statistical expertise.

To get you started, it sounds like you have dichotomous outcomes (had a c-section/didn't have one, had early labour/didn't have it) that would be analyzed using logistic regression. Logistic regression makes it easy to incorporate covariates that you wish to adjust for (provided you know what you are doing). This will never be as good as randomization for controlling covariates, but it will be the best you can do with your data (which I presume is retrospective collection?).

-
Thanks for your help . It is a retrospective study and BMI above 30 is actually suppose to be a risk for gestational diabetes. I have my data as an excel spreadsheet and have imported it to Spss. I am trying to control BMI and age . How would you do it in spss. Since i am new to this software. It is dichotomous variables C sec or natural labour. I have calculated odds ratio and just now exploring the adjusted odds ratio if someone could direct me on what tests to run in spss for multiple linear regression. Thanks to everyone –  FSNWRM Feb 20 '12 at 19:27