Flag the first value (1) then change the following values by groups.

[sandyzman1]

Hi,

I have the following dataset. I flagged the observation that has HbA1c <6.5 and if months is less or equal to 6 months.

patid	hba1c	months	flag
1	6.6	0	0
1	6.1	3	1
1	7.9	5	0
1	6.3	3	1
2	6.6	0	0
2	6.4	4	1
2	6.4	3	1
3	6.4	0	1
3	8.8	6	0
3	5.3	3	1
3	5.5	4	1
3	3.2	6	1
3	7	8	0
3	7.5	3	0
3	7.6	5	0

However, when there is more than one consecutive flag (1), I want to keep the first Flag (1) as 1 (remission) and the following 1's as 0 (remission) until the next value is 0 (Highlighted red), by patid. 

Want:

patid	hba1c	months	flag	remission
1	6.6	0	0	0
1	6.1	3	1	1
1	7.9	5	0	0
1	6.3	3	1	1
2	6.6	0	0	0
2	6.4	4	1	1
2	6.4	3	1	0
3	6.4	0	1	1
3	8.8	6	0	0
3	5.3	3	1	1
3	5.5	4	1	0
3	3.2	6	1	0
3	7	8	0	0
3	7.5	3	0	0
3	7.6	5	0	0

Could anyone please help me with the code? I tried to search but I don't seem to get the right keyword for search. Thanks.

Sandyzman1

[maguiremq · Accepted solution]

data want;
	set have;
	by patid;
		if hba1c < 6.5 and months <= 6 then flag = 1;
		remission = flag;
		if not first.patid and lag(flag) = 1 then remission = 0;
run;

patid hba1c months flag remission 
1 6.6 0 0 0 
1 6.1 3 1 1 
1 7.9 5 0 0 
1 6.3 3 1 1 
2 6.6 0 0 0 
2 6.4 4 1 1 
2 6.4 3 1 0 
3 6.4 0 1 1 
3 8.8 6 0 0 
3 5.3 3 1 1 
3 5.5 4 1 0 
3 3.2 6 1 0 
3 7.0 8 0 0 
3 7.5 3 0 0 
3 7.6 5 0 0 

[sandyzman1]

Thanks, maguiremq!

Awesome, this works perfectly well. You are the best.

Sandyzman1