Child abduction deaths

In Cecil’s recent column How many kids are found as a result of missing-child posters? he offered some interesting statistics about child abductions in the U.S.

One question that I’ve always wondered about, though, is what percentage of “nonfamily abductions” result in the death of the abductee?

I guess I’ve always assumed that the vast majority of children abducted by strangers end up dead unless they escape or are rescued very quickly. On the other hand, a child “abducted by a family member” tends to be in considreably less serious danger of losing his or her life.

Is this really the case? Do the NISMART statistics (or any other studies) shed any light on this?