Same db restored to test. Resultset is the same apart from a few extra rows inserted into the production server.

What seems to be the issue is that SQL made a decision on the production server to spool out the record set to Worktable and this dramatically increased the IO and, consequently, cpu time. Some things I've read about this indicate bad joins being a factor. I can't find what triggers this behavior though. I've look at wait stats but not sure what is exceptional or not.

Clive