2 个不同 dataframe 列中的比例

Question

After being treated with a first line of treatment that might be A,B,C(Drug code) what proportion of all cancer patients go on to be treated with a second line of treatment?

I am confused how to find that.

How should I find that through for loop? Would anyone show me the code for this? I am stuck there too.

I have tried implementing the code which i have pasted below

for i in q1b['PatientID']:
#here every patient ID of dataset(q1b) should go through dataset2 patient drugcode
 dataset2['DrugCode']
 #Now i need to find proportion like if some patient was treating for A then if drug a didnt work how many patients went for B as asked in question.i tried running loop i think we need nested loop

EXPECTED OUTPUT:

AS I HAVE 29 PATIENTS IN DATASET(q1b) I want to finD Proportion like how many patients go for 2nd line of treatment from drug code A to B, Suppose there are 3 patients who went for drug code from A to B so proportion would be 3/29 * 100 like this way I expect the expected output. dataset picture

Answer 1

So if I understand you correctly, you want to find the percentage of patients who have received drug A who have also received drug B. If so, you could do something like

codes = df.groupby('PatientID').apply(lambda x: set(x['DrugCode']))
codes[codes.apply(lambda x: 'A' in x)].apply(lambda x: 'B' in x).mean() * 100

If, moreover, you can assume that every patient starts out in 'A' , this boils down to

df.groupby('PatientID').apply(lambda x: 'B' in x['DrugCode'].values).mean() * 100