You could try a couple things....
<META HTTP-EQUIV="expires" CONTENT="0"> or
<META NAME="expires" CONTENT="0">
is designed to force the browser to reload the page at every visit.

If that fails, you could try using an incremented GET value to change the url each visit... something like
http://your_Url.com/page.html?1137513119 (or whatever the current time is.)

Those META tags will often, but not always, be respected. No agent is forced to respect them to be compliant with HTTP standards. Better to include these in the HTTP header directly, right by your 'Content-type' line. If you are using CGI::Simple or similar (and you really should be):

print $cgi->header( -expires => 0, -nph => 1, -type => 'text/html' );
[download]

If you aren't using one of the CGI modules (again, you probably should be), you can also do:

print "Expires: 0\n";                # Expire immediately
print "Pragma: no-cache\n";          # Work as NPH
print "Content-type: text/html\n\n"; # <- mentioned as context
[download]

Any valid HTTP/1.1 (or higher) client will respect these headers as expiring in the past, and therefore require a reload the next time the page is viewed.

<-radiant.matrix->
A collection of thoughts and links from the minds of geeks
The Code that can be seen is not the true Code
I haven't found a problem yet that can't be solved by a well-placed trebuchet

For a somewhat different approach to the existing answers, you could use CGI::Ajax or an equivalent to dynamically update sections of your existing page without actually having to reload the page as a whole. You'd need to ensure that Javascript is enabled for your clients though, or have a fallback to a non-Javascript-dependent method.

Please, next time try with Super Search: http header / browser caching issues.

For the record, I did try Super Search (as I always do before posting a query). I did not try with those keywords, and of course that is the problem with keyword searches.