Interrupted time series for opioid overdose

[Janinka]

I have data for opioid overdose before the pandemic and after pandemic. I have ran my data through regression with dummy variable being month without covid 0 and months with covid 1. 

But I am very new to sas and this week I been reading about interrupted time series using proc autoreg or poison regression. I want to try to run all my data again using the interrupted time series. For my data I am using quarters for the years 2017-2021 labeling them 1-20. Then for my dependent variable I am using number of overdose deaths. For months representing covid 1 and months without covid 0. And lastly pre-and post which pre is 0 and post months are 1,2,3….7.My data looks like this

6	1	0	0
5	2	0	0
8	3	0	0
9	4	0	0
5	5	1	1
7	6	1	2

that is just sample of my data.

first column my = y

second column = months

third column = pre covid and onset covid for the months 

fourth column = onset of covid

y=Bo+B1T+B2X+B3TX

i tried this code

Data y;
Input y time covid intervention;
Cards;

Proc autoreg data=y
Model y=time covid intervention/ nag=2 method=ml;
Run;

But I think I am doing something wrong because the results are off. Thanks for any help!

[Ksharp]

https://support.sas.com/kb/70498