Friday, June 03, 2011

Draft Blog for a Guest Posting on Australian E-Health And Its Impact.

I had in mind this rather high level post. Comments welcome!

Where Is The Consensus on Australian E-Health and How Can A Shared EHR Vision Be Progressed?

In recent times the National E-Health Transition Authority (NEHTA) has released a Concept of Operations for the proposed Personally Controlled Electronic Health Record (PCEHR). Submissions providing comment have been requested ending in early June, 2011.

Substantial funding ($450M+) for this initiative has been made available for an initial two years (ending June, 2012) and work is currently underway with a range of solution providers, NEHTA and the Commonwealth Department of Health and Ageing to make this rather grand vision (which originated in the Final Report to Government in 2009 of the National Health and Hospitals Reform Commission) a reality.

At the core of the proposal is that there would be an internet accessible national electronic health record system created in a series of national repositories and indexed in such a way that the patient had control of what information was to be made available to anyone (clinicians, friends, relatives etc.). The information to be held in the record for each individual was intended to be sourced from Medicare, diagnostic test providers, hospitals, specialists and primary care provider computers and assembled into a health summary and then a series of so called ‘event summaries’. There was also intended to be a facility for ‘consumer contributed information’ and commentary.

Review of the currently available submissions suggests that while the concept of an easily accessible and available summary health record to support continuity of care and care in emergency situations is universally supported (provided via either access to provider systems or a health information exchange) beyond this the ‘how’ this might be achieved is by no means clear and additionally it is recognised there are a very large number of barriers (including privacy and security, information quality, currency and reliability, overall system governance and management, consumer and clinician engagement, clinician workflow and incentive requirements and patient consent issues) to successful delivery of what seems like a simple and quite intuitive objective. It also needs to be pointed out that while intuitively it seems such a system would provide obvious benefits and usefulness the evidence supporting the value of such a care record is still by no means clear. This is to be distinguished from the very much stronger evidence for the clinical value of well-designed records intended for use by individual clinicians in their care delivery and decision making.

From this writer’s perspective Australian E-Health has rather got ahead of itself in attempting such an ambitious Shared EHR program without laying and proving the organisational and operational infrastructure as a number of initial and required steps.

These steps include:

1. Developing a Governance Framework and Leadership which properly engages clinicians, consumers and other stakeholders.

2. Recognition of the need to focus on achieving quality standardised implementations of effective clinician care support systems both in practices and hospitals .

3. Development and stabilisation of safe, secure, standardised clinical document sharing between each element of the health sector (as is now being undertaken rather slowly by NEHTA). This needs to cover clinicians, diagnostic service providers, hospitals, allied care providers etc.)

4. Evolution and training of an adequate workforce of clinically ‘savvy’ IT specialists to guide and manage the much increased level of activity.

5. Grasping some high impact ‘low hanging fruit’ in terms of projects that can make a real patient difference. (Consumer and clinical information portals and the like for example).

Only once these projects are well advanced, and the skills and governance in place, should consultation and development of a shared electronic health record be considered. At present the present proposals from Government very much put the ‘cart before the horse’ and in my view will have the effect, in failing to deliver a reasonable outcome, of severely impairing the chances of the Australian population receiving the benefits of technology implementation that has so improved other sectors of our economy and improved the overall quality of life for so many citizens. There is no doubt e-Health has a lot to offer but it needs to be delivered in a planned, considered, careful and consultative way. This is presently not what is happening in Australia.




Anonymous said...

I agree we have a governance problem. And since the pcEHR is the big play in the space (whether it succeeds or fails, it seems likely there won't be anything else national for a long time), it's something we need to take seriously.

It seems to me that the ConOps doesn't quite come to terms with the central problem we face trying to build a national EHR: we need to build distributed trust.

I say this because the national EHR is a grand play to reduce the amount of money - and more importantly, the time of doctors and nurses - to provide the level of care that we are used to. It will do this by reducing the amount of time that these people need to gather information - on the basis that collecting the information is what takes time. But we can only do that if the information that is in the system is trusted by the users (all of them). So the central problem with a really scaled EHR is building that trust.

The ConOps doesn't grasp this issue. This is what is says specifically about trust:

"trust is critical for the success of the rollout and uptake of a PCEHR system"

Great. That's exactly what we need out of the EHR. But what kind of trust does the ConOps talk about?

"Trust is one of the many critical success factors for the PCEHR system, therefore it is essential to ensure that:
• people seeking access to information are whom they claim to be;
• information received from a claimed person is from that person;
• information transmitted across networks is appropriately encrypted and
has arrived at its destination point without being tampered with; and
• access to information is appropriately authorised.

A high level of data quality is another essential ingredient in trust around the PCEHR system. High levels of data quality are required to assist providers and individuals in making safe healthcare decisions."

Anonymous said...

(and part #2, since I found out about the limit *after* writing all this)

I agree with all that. And there's real challenges there.

But these technical issues are not the core issue - which is how we get people to trust each other. Because right now, trust is not a scalable asset we can rely on. Patients, Doctors, and Nurses trust each other as individuals - to a degree - but that that trust doesn't scale. These groups don't trust themselves as a group, let alone the other parties.

As long as all the parties insist on being gatekeepers for the pcEHR - so that information can only become part of the EHR if both the source agrees, and the patient agrees, and either can subsequently revoke their permission - this is not a trust based system.

When the ConOps talks about "high quality" - what does it mean? The discussion that follows it is very focused on the quality of the data, not the quality of the participations. There is a shout out at the end of the section to clinical governance, but still, this is about data quality, not about really establishing distributed trust.

The ACHI comments on the pcEHR say that eHealth change is really changing healthcare with a "e", and that's the real thing the ConOps doesn't grasp: that in order to build a system that can leverage trust, we need to build the trust, not only in the system, but also between the participants.

But perhaps this is what the pcEHR is about - it's called "patient controlled", because there's no other way for it to even exist as things are; perhaps it's all we can hope for - better than nothing. And there's certainly an argument that if this is all we can have, then let's have it, and hope that it grows into something useful in the longer term.

It certainly seems that we aren't yet ready for the hard discussions that need to occur in our society in order for us to build distributed trust - and these aren't technical, they're social. So I think that we'll have to watch other countries forge ahead and build working systems that they can leverage, while we squabble about the peripheral issues and whine that we can't have both our trust and our freedom at once.

Anonymous said...

@twopart_anonymous - nice thinking.

Perhaps the most difficult thing about a network which has trust across its membership is that the trust cannot simply be injected from the top from a central authority. It is up to the members to learn how to work together, and when and where to trust information.

eBay is perhaps the best example of being able to trust complete strangers based on what other complete strangers have said about them. The notion of reputation is key here, not just for the person but also the community.

It seems likely that the PCEHR will struggle to infuse its users with trust. IT takes time, it needs an established pattern of good behaviour, and (perhaps most importantly) it requires a critical mass of users all doing it so that we can say to new people (and patients) "that's how we do things in the PCEHR".

A community of people all acting similar goodwill and expectations seems to be exactly where eHealth needs to go. Will the current work get there in one year? Of course not. Is it establishing the groundwork by which it *could* get there? Perhaps, but it needs the like-minded community to actively support it, nurture it and build the core set of behaviours which can turn into a culture of trust.

Anonymous said...

We already have a community of trust within the medical profession where notes are forwarded between providers and have been for years. The rules wrt consent are well established. We even have a PKI infrastructure (HESA/Medicare) to support this and Australian Standards (HL7V2) to specify content.

Why that is not leveraged escapes me. I think there is way to much money available for grand plans that replace expensive existing infrastructure for something new and unproven and trendy. It will end in tears. To move to a Patient controlled EHR without the ability for providers to transfer records using existing privacy conventions and standards is pie in the sky drivel dreamed up by Hollow men and done for political reasons.