Trigger Program Question

We have the same Trigger Program (in RPGLE) defined on multiple files,
which get updated from multiple different jobs. The program writes an entry
to a data queue. The total updates to all the files easily exceed a billion
a day.

I have a few questions. How is the program invoked on file update? Does it
always get a fresh copy? If not, is it possible to code something which
will execute only once?

What I am trying to do is this: The program is reading 4 data areas using
'IN' statement. I have found that the IN statement is about 7-8 times more
'expensive' than CHAIN. I am trying to figure out if there is a way to
increase the efficiency of the Trigger program. All the 4 data areas are
really static, but CAN be changed, hence the 'IN' statement for every cycle.

Any ideas will be greatly appreciated.

Vinay