Help with audit analysis please, ? Do i use Chi squared

#1
Hi,

I have done an audit of 30 patient. I reviewed the notes to see how many had a diagnosis of depression written in the notes =5 patients. I then reviewed these same 30 notes and i was able to make a diagnosis of depression in 12 patients. Ie we are missing the diagnosis of depression. I also checked the discharge summaries and only 3 of the same 30 patients had the diagnosis of depression written on their discharge letter.

All 3 patients with it on the discharge letter, were part of the 5 with it written in the notes and all those 5 with it in the notes were included in the 12 diagnoses i made if that makes sense),

I want to do some analysis to see if statistically significant but struggling how to go about this. The data is not comparing 2 different groups, its the same group of patients

I tried chi squared to compare the 5/30 with the 12/30 but i unsure which number is the observed and which is the expected?

Please could you tell me how to work this out?

Thanks
 
#3
Thank you for your response,
I think I will just present my raw data as it is, rather than struggle on trying to analyse it.

Thank you
David