By "different set of information" do you really mean information from
different fields, or information for the same fields from a different
record.

The latter does not require the mergefields to be changed.  If I was faced
with the former, I would probably use a query in Access to create a dataset
with field names that matched the mergefield names already in the document.
and use that query as the datasource.

First, I was going to say the same thing as Doug re. using Access queries
(or similar) to "hide" underlying column name changes. The way I see it is
that if you have a serious data management problem it is probably better to
try to solve that problem using data management tools than by creating what
I suspect could be large numbers of variant Word documents which might be
even harder to manage longer term.

As for the VBA side, I would suggest that you have a look at the bottom
section of

http://word.mvps.org/Tutorials/index.htm

for starters, then perhaps look at the Macros/VBA tab at

http://word.mvps.org/FAQs/index.htm

There is an example of how to process a complete folder of .doc files at

http://word.mvps.org/FAQs/MacrosVBA/BatchFR.htm

but I suspect I would probably /not/ use find/replace but use a loop to look
through all the merge type fields in each document and modify the fields
that way. There are sample code snippets in this group (which you can search
using the Groups tab at Google).