Discharge Summary inertia
The inertia embodied in the Australian health system is immense. Attempts to change it have mainly foundered over the years. Portents of failure are writ large, almost everywhere one turns.
In this context, let’s look at a favourite topic of mine, the electronic discharge summary. I view electronic discharge summaries as a barometer of e-health capability. They transcend organisational and funding boundaries. They transcend the primary and acute care sectors. They transcend many different clinical information system applications. They signal a genuine capability to share clinical information between systems at a semantic level and therefore signal a capability that many would ascribe to useful shared electronic health records.
At least they would if they existed! Unfortunately, there are no discharge summaries of this ilk to be seen.
Over the past decade or so we have witnessed countless local, regional, and national discharge summary ( or separation summary or discharge referral ) projects undertaken in Australia. In late 2002, under Healthconnect funding, the Clinical Information Project was established. Led by Dr Frida Cheok, it undertook amongst a number of other tasks determined by the Australian Health Ministers’ Advisory Council, to produce a “standard” national hospital discharge summary. In April 2004, after considerable consultation and analysis, it published a specification for the content of hospital to GP discharge summaries. In 2006, NEHTA reworked and republished the specification. NEHTA did so again, in 2009.
Some 8 years ago, South Australia already had “electronic” discharge summaries being created in the major metropolitan hospitals. But they were electronic only in the sense that a computer application helped the relevant medical officers to compile them. They weren’t sent electronically. They were printed, faxed and posted to the GPs. Eight years on they still are! And South Australia is probably ahead of most other states. Yet we are so very, very far from being able to deliver these electronically and have them incorporated into each GP’s health record system, replete with the ability for those systems to reprocess the data reliably, consistently and safely. It will not happen in one year, nor two. Probably not even in ten years!
There is far too much inertia to be overcome. Unfortunately, simply labelling something “electronic” does not overcome the inertia.
If Medical Director can receive NEHTA standard discharge summaries then the advent of fully electronic distribution is closer than you think. There are hospital systems that are eminently capable of producing compliant summaries now and plans to implement even more capable systems in the future.
I wouldn’t be surprised to see something happen in this space in at least 1 state within 2 years.
What I said was: “we are so very, very far from being able to deliver these electronically and have them incorporated into each GP’s health record system, replete with the ability for those systems to reprocess the data reliably, consistently and safely.” I absolutely stand by that. I cannot believe that Medical Director, nor any other GP system is remotely close to being able to digest NEHTA discharge summaries conforming to NEHTA’s CDA implementation guide “100211 29 April 2010.pdf” . To do so they would have to incorporate the AMT, SNOMED, LOINC, ANZSCO, a raft of HL7 codesystems and some (as far as I believe) unpublished NEHTA codesystems, such as 1.2.36.1.2001.1001. They would also have to parse and process NEHTA’s quirky CDA.
I’m also sure no hospital in Australia could produce conformant summaries, let alone eminently. NEHTA’s own CDA sample fragments in the implementation guide are riddled with flaws and couldn’t conform. If you could name one hospital, or one GP system that might go close, I’d like to hear of it.
In NZ (I am not from there), I understand that there are discharge summaries being electronically compiled in hospitals, and then sent electronically to GP systems, but of course most of the data in these is blobs of text (a blob about pathology, a blob about medications, a blob about procedures and diagnoses etc). However, the overall blob (the discharge summary ‘document’) can be incorporated into the GP system and held against the patient’s medical record there, because they have a national patient identifier. This is not perfect, and is not CDA, but better than paper, and better than nothing. As LOINC and the AMT become more commonly used, then perhaps the blobs can start to be more atomic and use common coding/terminology. Baby steps, but steps in the right direction… Perhaps we will see some of this in Australia soon, as the patient identifiers start to be incorporated into medical applications?
@anonymous – electronic discharge summaries are available in Australia now. Like NZ, they are blobs. Unlike NZ, the GP must manually allocate them to the correct patient (though they’re used to doing this for path results, and they are helped by their systems reading the HL7 and bringing up suggested matches).
Eric, I fully support your call to look into the real-world viability of CDA. The challenges I’ve seen in getting a decent eDS going are immense. If CDA can meet its promise and build momentum for adoption, the potential for more effective handover of care (and resulting improvements in both hospital and GP software) would be significant.