I don’t know why they write that, but one of the papers they quote for that claim reported success with samples up to 19 years old and in turn cites a paper that reported “genotyping whole-genome-amplified DNA” from up to 25 year old blood spots. I don’t know what this means exactly, it definitely isn’t the methylenation thing, but I think it’s a good indication that blood much older than a decade is viable. Both of these papers also report no differences in results between older and younger samples, only between different extraction methods and amounts.