Correlated Data with PHREG?

[aligator]

I am trying to analyze correlated data with PHREG. The clustering / grouping variable is drug class. What is the best way of getting accurate estimates? 

OPTION A

proc phreg data = data covsandwich(aggregate);
class route (ref="0") drug_class / ref=first;
model follow_up_time*discontinued(0) = drug_age route / rl=both;
id drug_class;
run;

OPTION B
proc phreg data = data covsandwich(aggregate);
class route (ref="0") drug_class / ref=first;
model follow_up_time*discontinued(0) = drug_age route /rl=both;
strata drug_class;
run;

In other words what is the difference between strata and ID when trying to account for correlated data?

Thank you very much.

[tamhane]

For correlated data, you would use ID statement.

See:  https://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/viewer.htm#statug_phreg_sect0...

When the proportional hazards assumption is not met, you would use STRATA statement.

See: https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_phreg_sect0...

and

https://www.cibmtr.org/meetings/materials/crpdmc/documents/2010/feb2010/loganb_coxproportion1.pdf

(page 7)