One explanation for the observed correlation is related to the development of the mammalian egg. Mammalian oocytes (cells on their way to becoming an egg, but not yet there) begin meiosis in females before their birth. However, meiosis arrests at two specific stages in these oocytes. Meiosis is re-initiated from the second arrest by fertilization. An oocyte of a 40yo pregnant woman was therefore frozen in meiosis for somewhat greater than 40 years before its fertilization. Perhaps the spindle fibers holding on to chromosomes for that long weaken and break, leading to nondisjunction of homologous chromosomes in Meiosis I or sister chromatids in Meiosis II -- and, therefore, eggs with too many or too few chromosomes.
The second arrest of meiosis in mammalian oocytes occurs at which stage of meiosis? (probably requires a text or internet search on your part)