Bill, in my experience medical records retention is related to the type of
services/treatment received (outpatient, hospitalized, tuberculosis
treatment, etc...), and can vary from x years after last "service", to life
of the patient (often expressed in "actuarial" retentions of roughly 70
years). While a patients particular blood pressure reading may not
be separated from other parts of their history, the patients medical
record(s) may be defined by where, how, and what services were received.

Part of the purpose of going to electronic medical records is the idea that
silos as reflected above will begin to lower, allowing for a fuller, more
holistic view of the patient's medical needs - this is done by creating a
transportable record between different treatment enterprises. In addition,
the intent is to create a data store to allow for the analysis of
effectiveness of treatments, something which does not happen as often or as
effectively as one would expect. This is done through the provision of
de-personalized data to a data stored maintained at the national level (I
believe in an institution similarly structured to the CDC).

Because of these factors, I expect retention will be applied at an
enterprise or holistic patient level, resulting in retention defaulting to
its longest component within an enterprise. Privacy concerns may even drive
retention through multiple enterprises. This is  going to bring up
challenges of custodianship - in this regards, electronic health records
are similar to systems like SAP where a single data store can meet multiple
records needs. The "archival object" then becomes a flat file reflecting a
certain transaction set, or meeting a particular
legal/audit/operational/historic need.

-- 
Dwight Wallis, CRM
Multnomah County Records Management Program
1620 SE 190th Avenue
Portland, OR 97233
ph: (503)988-3741
fax: (503)988-3754
[log in to unmask]

List archives at http://lists.ufl.edu/archives/recmgmt-l.html
Contact [log in to unmask] for assistance
To unsubscribe from this list, click the below link. If not already present, place UNSUBSCRIBE RECMGMT-L or UNSUB RECMGMT-L in the body of the message.
mailto:[log in to unmask]