building a categorial variable based on confidence intervalls

[Melanie Boekholt]

Dear all,

I try to part a continous variable in categories using the 50% CI for age groups.

Literture ref: "The alternative cut-off method defines as ‘fit for their age’ those with a FI below the lower 50% CI for the regression. ‘Frail for their age’ are those with FI above the upper 50% CI. The rest are ‘average for their age’." * (FI = value on the Frailty-Index)


For that i built age-categories (10yrs.) to define the "for their age"-part.
But i have no idea how to proceed from here.

I would be very thankful for any hint to what to look for or where!



*(Romero-Ortuno, R. (2013): An alternative method for Frailty Index cut-off points to define frailty categories. In: European geriatric medicine (5). https://www.ncbi.nlm.nih.gov/pmc/art...ihms519821.pdf)

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. Being able to replicate your problem makes it easier for us to help you.

I don't even know exactly what you're estimating with what routine. Is frailty you dependent variable or what? You'll also need to explain what you want in terms of standard statistical terms - I don't know about Frailty index but might be able to help you program if I knew what you were calculating. Just providing a citation is better than nothing, but your problem is not important enough to me for me to puzzle through the methods section of a paper in another discipline just to help you.

If you run a regression with categories, you can access the parameters in e(b) and use those to calculate things.