> All our internet (multi-tier) applications use the strategy of
> personalization, that is once a clinician has authenticated to the
> system, we use pick lists that are created from the clinician's
> preferences. For example, with patients, a clinician can choose to
> maintain a list of patients themselves, or we can automatically generate
> a daily list of patients scheduled to appear in clinic or we can do
> both.
>
> The strategy is to reduce the pick list options to a manageable set
> through maintiaining information for each individual clinician. For
> orders in general we use the concept of order set's, which once again
> can either be instituitionaly defined off our best practices guidelines
> and/or customized by the clinician.
I think this is a quite workable solution for some systems. I haven't tried this yet
(but I will sap) - but the first thought that comes to my head: one more table for
each list for each doctor - a hell of a lot of tables. Probably each of them indexed
as well. Here goes my cache, my RAM..... =8-0 Wouldn't I be worse off at the end of
the day? Will tell you when I have tried.
Horst
