How to combine or tabulate same values labels from different variables?

[Shaheen Ba-Ali]

I have searched wide and far, but can't seem to find a problem matching mine. How do I combine values from different variables that have the same value labels?

* Example generated by -dataex-. For more info, type help dataex
clear
input long Diagnosis1 byte(hp1_f_notype hp2_f_notype hp3_f_notype)
1  . . .
2  2 1 3
1  5 . .
2  5 . .
2  5 . .
1  5 . .
1  3 2 5
1 99 . .
1  2 . .
1  5 . .
end
label values Diagnosis1 Diagnosis1
label def Diagnosis1 1 "IIH", modify
label def Diagnosis1 2 "none-IIH", modify
label values hp1_f_notype Phenotype
label values hp2_f_notype Phenotype
label values hp3_f_notype Phenotype
label def Phenotype 2 "Migraine with aura", modify
label def Phenotype 3 "Tension headache", modify
label def Phenotype 5 "Other", modify
label def Phenotype 99 "Don't know", modify
label def Phenotype 1 "Migraine without aura", modify

I essentially just want to combine these value labels in one row with all the phenotypes.

[Nick Cox]

tabm from tab_chi on SSC may help. See e;g; https://www.statalist.org/forums/for...isions-to-tabm

[Shaheen Ba-Ali]

Thanks for the reply. I've tried tabm before, but I'm not sure how to combine it or code it, so that it shows the Headache phenotypes associated with the IIH and none-IIH like the picture shown above.

[Nick Cox]

You're correct. It won't do that.

[Shaheen Ba-Ali]

Do you know a command which would help me do that?

[Nick Cox]

See help table

(As you haven't declared otherwise, I am assuming that you are using Stata 17.)

[Shaheen Ba-Ali]

Yes, I'm using Stata 17, and I've tried reading through "help table", but I can't seem to figure out how to do it. I'm fairly new to Stata.

[Nick Cox]

Sorry, no free lunch here.

If you want a table just once, then it might be easiest to do the calculations in Stata and use your favourite word or text processing software to build the table using what you already know about that software.

If you want a customised table to repeat a hundred times with different data, the bad news may be that the code may take several hours or days to get right but that might be a reasonable expenditure of effort.

[Shaheen Ba-Ali]

Okay, thanks for the help nonetheless.