Bar graph of qSOFA score vs probability of in-hospital mortality

[Abdullah Algarni]

Dear Statalists,

I'm working on a prospective cohort study investigating the predictive accuracy of a sepsis score (qSOFA) to predict 30-days in-hospital mortality. The score has four possible values (0–3).
I want to plot qSOFA score results (0, 1, 2, 3) vs. the proportion of in-hospital morality in a bar chart exactly like the attached image (see below, please).

My data structures are summarized below.

Any help would be so appreciated.
Thanx
Abdullah


. dataex qsofa died_30in, count(10)

----------------------- copy starting from the next line -----------------------

* Example generated by -dataex-. For more info, type help dataex
clear
input byte(qsofa died_30in)
0 0
1 1
0 0
1 0
1 0
0 0
1 0
2 0
0 0
0 0
end
label values died_30in YesNo
label def YesNo 0 "No", modify
label def YesNo 1 "Yes", modify

------------------ copy up to and including the previous line ------------------

Listed 10 out of 2220 observations

.

[Justin Niakamal]

Here's a start

statsby, by (qsofa) clear: ci proportions died_30in
twoway bar mean qsofa, barw(0.6) bfcolor("192  181 123") yscale(range(0, .1 )) ylabel(#5) plotregion(margin(0)) ||  rcap lb ub qsofa, leg(off) lcolor(black) xtitle("qSOFA")

[Abdullah Algarni]

Thank you so much, Justin, it works!!
Would you allow me to acknowledge you upon publication of my research?

[Justin Niakamal]

Hi Abdullah,

Admittedly, I don't think I did enough for acknowledgment. You're more than free to acknowledge me, but not required to do so. Most of the active community here posts because we're happy to help where we can!