5
$\begingroup$

I am having a bit of a tough time with some logistic regression terminology. I have performed a multivariable logistic regression analysis where I have regressed a binary variable (death, where 1 = dead and 0 = alive) on some covariates, one of which is age (measured in years). I am then looking at what happens to the outcome when I increase age by 10 years, holding all else constant.

The resulting model gives me some parameter estimates. The one for age is $\hat{\beta}_{age}=-0.015$. So I multiply this parameter by 10 (years) and exponentiate the result getting $exp(10*-0.015)=0.86$. So, I'm a bit confused as to how to report this. Which of the following is correct?

  1. A 10 year increase in age is associated with a multiplicative $(1-0.86)*100% = 13.92$% decrease in the odds of death, holding all over variables constant; or
  2. The odds ratio is decreased a multiplicative $(1-0.86)*100% = 13.92$% for each 10 year increase in age, holding all other variables constant?

Essentially, I'm confused over whether the result is a 13.92% decrease in the odds or odds ratio? Can you help me clarify.

Thanks for your help!

$\endgroup$

1 Answer 1

4
$\begingroup$

This can be reported as:

The odds of death are reduced by 14% with every 10 year increase in age, i.e. the odds ratio was 0.86, after controlling for other factors.

or:

After controlling for other factors, the odds ratio for 10 year increase in age was 0.86, i.e. the odds of death were reduced by 14% with every 10 year increase in age.

Hope that helps.

$\endgroup$
1
  • $\begingroup$ This is very helpful, @rnso. I really appreciate it. I knew I was missing something subtle here, but just needed someone to clarify for me. Many thanks! $\endgroup$ Jul 12, 2015 at 21:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.