PCEHR glitch: no lynch mobs please

by Charles Wright on January 24, 2012

It took a couple of days for The Australian and the usual suspects to catch up with the news of the problems with the specifications for the PCEHR which I reported on Sunday and again yesterday, which may explain the enthusiasm with which they have greeted confirmation of the story in a statement from NEHTA.

The Australian described it as “the latest blow for the Personally Controlled Electronic Health Record (PCEHR) project”. One can only presume they are scoring their own punches at the system – pretty well all of which have been low and wide of the mark – because I can’t recall any earlier blows. Concerns and criticisms surely don’t count, do they?

The PCEHR project hasn’t been perfect, largely because of the completely unrealistic timelines imposed by former Health Minister, Nicola Roxon, but compared to the record of IT development by federal and state bodies, it’s been remarkably smooth sailing.

That doesn’t mean that the problem isn’t serious. The fact that conflicting specifications seem to have been delivered last November is not something that instills confidence in the development and clinical community, but on the other hand, the fact that it was identified relatively quickly is surely a plus.

To say, as that well-known anti-NEHTA blogger David More does, that “the wheels have fallen off” and “NEHTA is utterly incompetent and is becoming the instigator of yet another failing over-reaching national ehealth program” is surely premature. Typically, of course, there’s a detectable edge of glee and self-congratulation, with More clearly including himself in the ranks of “the wise men who said you should learn to walk before running”.

Unlike More and The Australian, I’ve been speaking to people in the vendor and implementation world I regard as infinitely wiser. One of them described the situation as “not a show-stopper by any means”.

The developers were scheduled shortly to begin CAP testing – I presume that relates to clinical terminology testing under the SNOMED CT arrangements, but I invite correction – and that’s going to be put back by three to six weeks, assuming that there isn’t some serious issue that hasn’t yet been hinted at. The July 1 deadline is being seriously tested as a consequence, but that seems to me a political imperative, rather than one based on prudent judgment.

According to one of my vendor sources, “Some people have an agenda that they want to see the PCEHR fail. We want to see it work. The delay is understandable, and we don’t have a problem with it.”

Over at the Healthbase blog, Dr Eric Browne claims that there are critical failings with the HL7 Clinical Document Architecture (CDA) underlying the PCEHR’s messaging and calls for an end to all further work in this area, pending a thorough review.

“Having studied both the HL7 specifications in detail as well as dozens, if not hundreds of examples of CDA documents from around the world over the past 5 years,” he says, “I have come to the conclusion that there are significant safety and quality risks associated with relying on the structured clinical data in many of these electronic documents.”

His concerns arise from his study of six sample discharge summary CDA documents provided by NEHTA, and the fact that “data can be supplied simultaneously in two distinct, yet disconnected forms – one which is ‘human-readable’, narrative text displayable to a patient or clinician in a browser  panel;  the other comprising highly structured  and coded clinical ‘entries’ destined for later computer processing. The latter is supposed to underpin clinical decision support, data aggregation, etc. which form much of the justification for the introduction of the PCEHR system in the first place.”

According to Eric, the human readable data in the NEHTA examples bore no relationship to the coded data, to the extent that the entries looked fine in the discharge summaries, but according to the computer code the patients were all dead, one of them having been killed off at birth, according to the record.

It’s a rivetting piece of information, but one of my sources says that the data displayed in the live system is generated directly from the raw data. He suggests that the NEHTA samples were not representative of the way the system works. “They were samples that were not generated on a live system by a clinician.”

I would have thought that genuinely wise men would want to clarify issues like that before breaking the glass and pressing the big red button.

I understand that NEHTA’s clinical lead, Dr Mukesh Haikerwal, will be outlining the issues to medical colleges at meetings in Melbourne tomorrow. We’ll probably learn a lot more at that point. In the meantime, should we not perhaps be discovering whether the victim is deceased or even critically wounded, rather than trying to hang the accused?


Comments on this entry are closed.

Previous post:

Next post: