VERY different Odds Ratios for matched vs. stratified case control

[iressa1313]

Hello this isnt so much of a SAS question as it is a design question, very sorry but I NEED help. Using SAS I calculated my Odds ratios for a case control study using two different methods for sampling the controls; random and matched pairs. Pairs were matched with age as the linking variable since it is suspected to be a confounder. For the randomly selected controls I stratified by age afterwards in an attempt to control for age. My odds ratios for the randomly selected controls were significantly greater than the matched pairs, like 40 vs 2 (rounding). Why is this? What about the two different designs results in such extreme differences.

Is it because stratification can allow for residual confounding within strata that is not controlled for?

MANY THANKS FOR ANY HELP!!

[Reeza]

When you stratified by age, were the two group sizes very different?

Are the events/effects in each group significantly different?

[iressa1313]

yes, the group sizes between case and controls were very different. For example within one strata the case group had 25 and the control had 206. But this was also the case for the matched pair design with that a case group having 50 and the control 203.