Obviously, you cannot help what one woman doesn't tell about her past. However, I don't see why a single study couldn't find women with abortions in their past, and their incidence of breast cancer as a population (all these questions are covered in any medical history). This incidence could easily be compared against women without abortion in heir past and thier incidence of breast cancer. It either is or isn't statistically significant.
Just because abortion may not lead to an increased risk of breast cancer doesn't make it ok. If you can't draw a link - let it go - it's just scare tactics and propaganda after that. If you want people to take you serious, then give them the facts and just the facts.
It sounds pretty, but I have a hard time believing that women on either side of the aisle would be completely forthright, enough to create a statistically accurate study. How do you know if women who say they have NOT had an abortion - for the sake of the other study group - are actually telling the truth? What will they say when recruiting women for the study? Apply only if you can be honest? And that's not how statistics are gathered usually anyway, right - how often are formal study groups used as opposed to clinical information gathered from non-human sources like records? I'd like to think it could be done, but I'm not convinced.
I don't think at this point it is a matter of being able to draw a link or not. The suspicion is valid, based on what I've read, and proving the link is still in its beginning stages. Besides that, there's considerable opposition to these studies and their implications. It'll take time either way, and dismissing it out of hand is unwise, imo.