In a nutshell, data that is collected in EHRs or transferred around HealthCare networks, does a lousy job of identifying where the data was collected. Brian and his team examined one health information exchange (HIE) in which more than 6 million records were examined for facility identification and found highly inconsistent results.
Why does this matter?
Brian was quoted as saying that;
“it’s really important if you’re an accountable care organization trying to analyze a large data set in a data warehouse, or if you’re doing health information exchange in your community and want to look for patterns of care delivery, where people are getting their care and what kind of outcomes they’re getting in different places,”
Ok. Population health and payment. Big issues for sure.
However, there’s a more fundamental issue at play and that’s provenance. If you can’t identify where the data was collected then how do you know it’s legitimate?
Who would fake their own medical data? Well, it does happen I’m afraid. Patient’s have modified documents to gain access to pain medications, or for psychiatric reasons, albeit nobody really knows how often.
My guess is that the author is talking about HL7 v2 messages where people can allow themselves to be lazy. Here’s a sample;
MSH|^~\&|GHH LAB|ELAB-3|GHH OE|BLDG4|200202150930||ORU^R01|CNTRL-3456|P|2.4<cr> PID|||555-44-4444||EVERYWOMAN^EVE^E^^^^L|JONES|19620320|F|||153 FERNWOOD DR.^ ^STATESVILLE^OH^35292||(206)3345232|(206)752-121||||AC555444444||67-A4335^OH^20030520<cr> OBR|1|845439^GHH OE|1045813^GHH LAB|15545^GLUCOSE|||200202150730||||||||| 555-55-5555^PRIMARY^PATRICIA P^^^^MD^^|||||||||F||||||444-44-4444^HIPPOCRATES^HOWARD H^^^^MD<cr> OBX|1|SN|1554-5^GLUCOSE^POST 12H CFST:MCNC:PT:SER/PLAS:QN||^182|mg/dl|70_105|H|||F<cr>
Ok, the PID segment is there and we can see the patient’s address. Hmmm … where is the facility? Well, it’s in the MSH, but it’s just a transmitting facility identifier and no use for analytics. The physician’s name and number are in the OBR, but no address or NPI.
The good news is that things are getting progressively better with respect to continuity of care and data provenance, thanks to CDA, which makes explicit provision for address and telecom fields for author and legal authenticators. The bad news is that these fields have cardinality [0..*] which effectively renders them optional (in both the standard and the implementation guide). Good vendors put them in, but lots of people don’t.
Amazingly, the records we receive via faxes in the clearinghouse, almost always have a header and a footer with the facility name and address.
Interoperability in action. Sigh.