> It's clear to me that scanning for DNA fragments cannot tell you anything about chromosome abnormalities, so I doubt the researchers said any such thing.
It's not at all clear to me; why would you think this?
To take trisomy 21 as an example (Down's syndrome cause), this means the fetus has three copies of chromosome 21. After a CVS or amnio, the lab will look at the cells under the microscope and look for two copies of each chromosome. If they find three instead, that's a chromosome abnormality. If you break open the cells and end up with DNA fragments floating around, it's possible to sequence the DNA but there's not any information that can tell you whether a cell had an extra copy of a chromosome. It's like asking whether the apples to make a cup of juice were all packed in the same box or multiple boxes.
Apparently this same researcher does have a down syndrome test, but it is based on measuring RNA ratios and does not seem to rely on sequencing the fetus's DNA at all.
Glad I asked! I was thinking about chromosomal rearrangements, which would be detectable using this method, but I agree with your point about nondisjunction.
It's not at all clear to me; why would you think this?