20885 Electronic Lab Reporting (ELR): Murphy's Law in Action

Wednesday, September 2, 2009: 1:50 PM
Hanover C/D
Hyung-Suk (Sue) Lee, BS, MT(ASCP) , Automated Disease Surveillance Section, Los Angeles County Department of Public Health, Los Angeles, CA
Raymond Aller, MD , Acute Communicable Disease Control, Los Angeles County Department of Public Health, Los Angeles, CA
Heather Capel, BS , Atlas Public Health/ELR, Atlas Development Corporation, Calabasas, CA
Susan May, BS, MT(ASCP) , Automated Disease Surveillance Section, Los Angeles County Department of Public Health, Los Angeles, CA
As of April, 2009, Los Angeles County (LAC) ELR systems have live feeds from 20 laboratories serving 21 hospitals and 4 referral laboratories, plus several labs in testing.  Implementation and maintenance of laboratories running diverse laboratory information systems (LIS) has been challenging, and we have encountered many unanticipated issues, that can be daunting and timeconsuming to resolve. We have implemented both LIS-integrated translation/filtering systems, as well as separate “edge servers” to perform filtering and translation. LIS vendors tend to create many different “flavors” of a single HL7 version, and we must transform such records to a more standard HL7 message that can be ingested by the electronic disease surveillance system.  Determining reportability (“is this borderline reportable – or not?”) is challenging, and results from an external reference labs, processed through the hospital LIS, and passed on to PH can be particularly problematic. We now have well established data flows to acute communicable disease, and to the sexually transmitted disease programs.  As we work with additional surveillance programs within our department (TB, HIV Epidemiology), we have found that their systems lack the capability of directly ingesting HL7 – so are forced to adapt to other non-HL7 formats. 

After live operation, problems don’t evaporate.  Labs continually make changes - move to different analytic equipment, change send-out labs.  Furthermore, our own public health staff modifies their expectations for results they need to see.  If the sending labs lack rigorous procedures as they update their test offerings (adding tests, modifying methodology, even adding patient locations), key reportable findings fall out of the reporting basket;  we have had three instances – involving dozens of cases – where a lab failed to report certain hepatitis results to us for several months.  We describe some of our methodology for anticipating, avoiding, and if need be recognizing such errors.

<< Previous Abstract | Next Abstract