Tuesday, May 20, 2014

We Have To Be Very Clear The Report Of The PCEHR Team Is Not Government Policy. However, What Does It Suggest and Will It Work?

I had an e-mail from the DoH E-Health Branch today  which makes this very clear. We are at the beginning of a journey.
“Dear colleague
I am writing to let you know that the Government has released the review of the Personally Controlled Electronic Health Record (PCEHR). A copy of the report is available at:
The Government supports a national eHealth record system, but it needs to be effective, functional and easy for all Australians to use, while being clinically relevant to doctors, nurses and other frontline healthcare providers.  
The review makes 38 recommendations and outlines opportunities to improve the usefulness of the system.
The Government will continue considering the recommendations over the coming months to understand the issues, their implications and in consultation with key stakeholders, determine the best ways to deliver on the intended outcomes,
As part of this commitment, the Department will be undertaking a range of consultation activities to consider options for responding to the Review recommendations .
I look forward to working with you over the coming months to improve the current and future capacity of Australia's health care system.
Linda Powell
First Assistant Secretary
eHealth Division”
I hope we get both consultation and listening this time!
With that made clear what about the report itself?
First - We have a Vision (p2)
“The electronic health record for Australians, will be a reliable, secure and trustworthy source of key clinical information. It will facilitate efficient and effective treatment of patients by health practitioners and enable consumers to access and manage their own health records in cooperation with their health providers to improve care. It will respect individual privacy but be clinically valuable to all areas of the health care industry. Interaction with the electronic health record will be highly automated and form a natural part of clinical workflows. The value of sharing health information electronically between healthcare professionals, will be demonstrated by enhanced efficiency and effectiveness of the delivery of healthcare, reduced hospitalisations and ultimately lives saved.”
What is made most clear here is an obvious shift which  really suggests a change of intent. As I read this what is now being sought is a record for use by clinicians to co-ordinate their care while providing access to allow the patient to be involved and be aware and comment on what is happening to them. However the issue of having a local and a parallel national system is still not clearly addressed.
Reading the report I noted the following:
1. The level of actual use of the system was actually going down despite continuing patient registrations.
2. The 2013 National E-Health Strategy Refresh is yet to be finalised. - but a (secret) summary exists.
3. The Panel felt the e-Health infrastructure (SMD, IHI, NASH, NPC ) were all in good shape. Some may be not so sure.
4. The benefits case is a total joke - conflating local systems and the PCEHR and pretending to understand the level of benefits possible (p9-11).
5. The barriers to implementation of e-Health slide (p11) certainly covers a lot of the issues.
6. The recommendations are summarised on p15-18 and available in a previous from a blog of May 19, 2014.
7. The rename to MyHR seems at odds to seeking more clinician use.
8. The governance recommendations seem good:
“Given the scale and complexity of the national eHealth work program, Australian Governments should continue to play a lead role in directing and coordinating national implementation activities. However, the move from a focus on nationally shared eHealth infrastructure to meaningful use of eHealth solutions by care providers and consumers, argues the case for a more broadly based involvement in the governance process – particularly extending to clinical and patient communities and private sector health operators.
Key to strengthening the current governance arrangements will be the establishment of an eHealth entity (created through the transformation of NEHTA) that is focused on coordinating execution of the national strategy and the nationally funded eHealth work program. To perform this role it will be necessary for the eHealth entity to have a Board made up of key parties beyond Government representatives (including strong care provider and consumer representation) and to oversight the building of close working relationships across the public and private health sectors and with the health IT vendor community.
Deloitte : National e-Health Strategy for Australia November 2013”(p20).
9. The Panel notes that NEHTA and DoHA just ignored most stakeholders most of the time (How true) (p21)
10. There seems to be an Independent Advisory Council for the PCEHR - members not listed and clearly hardly working if the last 2 years are any guide. (p26)
11. The move to opt out is reasonable but won’t happen in 7 months - and rushing it could be a big problem. (p29)
12. When the Panel says “The PCEHR should be considered a source of supplementary information.” They have it right in my view.  (p31).
13. It seems the Panel noticed patients like to be able to message doctors….”In addition, Secure Messaging is a closed network that operates between clinicians and it has not been designed to include providing Secure Messaging to patients and consumers. Action must be taken to expand the scope of Secure Messaging to a next generation service that ensures interaction between the medical profession and consumers for information that must be passed in a reliably secure manner to facilitate improved workflow and secure communication of private information.”(p40).
14. The NEHTA not listening again theme.
“It is acknowledged that a substantial amount of input was provided by various industry user groups during the development phase, and not incorporated into the PCEHR. Greater implementation of the recommendations provided during the development phase may have resulted in better acceptance and increased uptake of the PCEHR by both clinicians and individuals. Clearly, if the system is not designed around the realities of the clinical environment and workflows, then the uptake of the system will be limited.”
And:
“all medical specialties, other than General Practice, have been largely ignored”(p59)
15. More failure to do what was required from DoHA / NEHTA.
“The PCEHR Concept of operations refers frequently to the importance of governance. It quotes seven governance principles from the National eHealth Strategy including transparency, accountability, appropriate stakeholder representation, sustainability and the balance of local innovation and national outcomes. These principles were not followed, and the result is that a good plan failed.” (p78)
16. Accenture off on a bit of a frolic it seems from this.
“To ensure that the systems operator (in this case Accenture) does what it’s told by the system owner, which is to deliver the functionality specified by the system owner, rather than the operator delivering what it thinks/believes is best for the owner.”(p82)
17. The present PCEHR does not serve the key benefit opportunity. Bad design.
“Medicines management is a significant area of value that is not sufficiently supported within current PCEHR design” (p86)
18. Clinical risk is not really addressed properly.
“The current PCEHR has multiple safety risks including:
o Using administrative data (e.g. PBS data and Prescribe /Dispense information) for clinical purposes (ascertaining current medications) – a use never intended;
o Using clinical documents (discharge summaries) instead of fine- grained patient data e.g. allergies. Ensuring data integrity is often not possible within documents (e.g. identifying contradicting, missing or out of date data);
o Together the secreatean electronic form of a hybrid record with no unitary view of the clinical ‘truth’. Hybrid records can lead to clinical error by impeding data search or by triggering incorrect decisions based on a partial view of the record
o Shifting the onus for data integrity to a custodian GP avoids the PCEHR operator taking responsibility for data quality (a barrier to GP engagement and a risk because integrity requires sophisticated, often automated checking).
o No national processor standards to ensure that clinical software and updates (and indeed the PCEHR) are clinically safe.” (p88)
Overall p53 to p89 are the most valuable as collects all sorts of useful input and Panel .

Commentary.

My Overall Perspective On the PCEHR Review.

1. The key architectural flaws of the PCEHR are not understood and solutions for remedy proposed, neither is there any clarity as to what the system is to do and how it is not actually do it and for whom. It is also clear the panel really did not have a handle on what benefits the PCEHR can actually deliver.

2. If implemented as suggested the system will still not be successful as a valuable clinical tool I believe.

3. To do what is recommended and nothing more will be expensive and has the feel of ‘applying lipstick on a pig’.

4. The lack of real e-Health expertise involved in undertaking this review really shows.

5. Unless opt-out is implemented with great care and much effort around data quality and reliability disaster awaits. This means the change must be slow, incremental and frequently reviewed.

6. Clinical Safety issues are skated over.

7. The fate of the important AMT is not properly identified.

8. DoH and NEHTA have been hopeless in the Governance of Australian e-Health.

9. This is a seriously missed opportunity to make a sensible difference.

Sadly I could not go much above 2/10 for a score.

Sad about that. I look forward to seeing what the Government decides to do. There is no way at all this document could be described as a comprehensive plan for e-Health as Mr Dutton claimed in December.

David.

11 comments:

Anonymous said...

Well David, if today's Medical Observer is anything to go by we won't have long to wait.

Under a blueprint released by Health Minister Peter Dutton "ELECTRONIC health records will be created for each Australian by January under an opt-out model to replace the unpopular personally controlled electronic health record (PCEHR)".

January 2015 - How easy was that? I wonder what all the fuss has been about.

Bernard Robertson-Dunn said...

David,

I totally agree with your perspectives, especially the first:

"1. The key architectural flaws of the PCEHR are not understood ...."

They need to get it into their heads that an eHealth record is all about information not technology or infrastructure or messaging or security or even interoperability standards. Unfortunately they don't even understand what that statement means.

The review team comprised a hospital specialist, a medical specialist and an IT specialist, the latter with expertise in Australia Post, Tabcorp and IBM.

Not one of the reviewers has expertise in information, let alone health information so it's not surprising that the real issues were not properly identified or addressed and ended up lumped into privacy and security.

My personal view is that Health Records and the Information contained in them is a bit different from Health Informatics but has a huge intersection. So HI is not just a good start, it is essential.

Anonymous said...

After a very careful read of the Review I have concluded:
(a) the review team were overwhelmed by the task and do not understand the underlying problems
(b) if the Gov't follows the review teams advice it will be business as usual albeit with the PCEHR being renamed MyHR
(c) if (a) and (b) hold true more money will be wasted and no progress will be made.

Bernard Robertson-Dunn said...

Further to the proposition that the review team have not understanding the issues, I just had a look at the vision statement in the review that David's referenced

“The electronic health record for Australians, will be a reliable, secure and trustworthy source of key clinical information."

Do they really mean that the health record is to be the source of key clinical information?

I'd agree with the use of the word "repository" rather than "source". This use of the wrong word is repeated throughout the review document (e.g. recommendation 17 and a quote from the AMA about the PCEHR)

And a bullet point included in Addendum 3 section 14. Governance / Standards / Terminology says:

"The PCEHR system is based on multiple data sources stored in multiple repositories ..."

Data sources are not stored in repositories, data is. A more correct phrasing would be "The PCEHR system uses data from multiple data repositories ..."

As with most complex issues, the devil is in the detail. Vagueness and imprecision suggest that bad decisions have been, or are about to be made.

Anonymous said...

Speaking at the HIMSS conference 19 May in Sydney Health Minister Peter Dutton said Medicare Locals will be consolidated down to 24 and renamed “Primary Health Networks (PHNs) to give them more flexibility to continue with existing programs, including those using eHealth technologies.”

What does this mean?

He said “… we don’t want to micromanage” PHNs ….. but if “they don't have success over the contract period then ….. we could allocate work to others that have been successful.”

He said “I think that will drive innovation through technology and it will remove government from the process and allow creative minds to come together to keep people out of hospitals, to stop re-admissions and to deal appropriately with our primary healthcare response.”


Does anyone believe government will be removed from the process?... that it will not micromanage? … that it will allow innovation to flourish?

Bernard Robertson-Dunn said...

Just to demonstrate that I'm not just whinging for whinging sake, here are my (hopefully constructive) thoughts on what should happen re health records.

Health records are an integral and essential part of any health care system.

If you are developing a national health record system, you should start with the information. One of the most critical questions is: who needs access to the data and where are they located?

For an eHealth record the answer is: The primary users are those who deliver health care services and they are usually in close proximity to the patient.

Here are some more relevant questions/answers

Are there health record systems already in existence? Yes. Existing Health record systems are generally local to GPs and other health care professionals. They are run by health professional organisations, they are already compulsory (i.e. neither opt in or opt out) and their access is restricted to a small set of users, usually excluding patients.

Are they integrated? It varies, but usually, no.

Are they accessible from the internet? Not usually. They are only accessible by health care professionals on closed networks.

Is some relevant health information held nationally? Yes but mainly, if not exclusively, it is health funding related, rather than health care related.

Could access to this data be made available over closed, secure channels to local systems? Yes.

Now here’s the big question: What value does a national database of summary information, run by the Federal Government and exposed to the internet add to the existing local health record systems? As far as I can tell – it adds very little of value to health care professionals but increases risks of unauthorised access.

Could a distributed system, which includes health funding data repositories and which is largely if not completely disconnected from the internet deliver better health care use and value at lower cost and risk? My feeling is - yes.

So there is a choice between a closed, distributed logical information architecture and a mixed distributed/centralised, internet accessible logical information architecture.

IMHO, a distributed, closed architecture would be far more useful, resilient, responsive, reliable, maintainable, secure and cost effective. Users could still be involved in their own health care and could see their health information but would have to do so through intermediaries (such as their GP). Much the same as with their credit histories. Unusual, non-local or infrequent access to a patient’s health information could be facilitated via interoperability mechanisms.

Much of the existing eHealth infrastructure could be resused, only the central databases and internet access would need to be discarded. In terms of central responsibility, there would need to be a Health Information Architecture body whose primary responsibility would be for information interoperability and standards. Other issues such as legal, privacy, safety etc could also fall under that body. They would have no responsibility whatsoever for technology.

DHS would only need to support access to its existing systems, which would be nodes in a distributed system. Health and NEHTA would be out of the loop altogether, although Health would continue to provide policy advice to the Minister when dealing with the states.

I do hope the government is asking itself these questions and comes up with answers that are acceptable and rational.

The answers should also be compatible with this government’s claim that it has no part to play in the delivery of health care – that health is a matter for the states.

Anonymous said...

>Now here’s the big question: >What value does a national >database of summary information, >run by the Federal Government >and exposed to the internet add >to the existing local health >record systems? As far as I can >tell – it adds very little of >value to health care >professionals but increases >risks of unauthorised access.

>Could a distributed system, >which includes health funding >data repositories and which is >largely if not completely >disconnected from the internet >deliver better health care use >and value at lower cost and >risk? My feeling is - yes.

I know people who work in Emergency Departments who normally get no information at the moment about patients. They are very interested in having real time information available about their patients. I'm not sure a disconnected system would work for them.

Thus, I think it depends on the role of the Health Practitioner as to what would work and not work.

Bernard Robertson-Dunn said...

"I know people who work in Emergency Departments who normally get no information at the moment about patients. They are very interested in having real time information available about their patients. I'm not sure a disconnected system would work for them."

Firstly, I'm not talking about disconnected systems, I'm talking about distributed systems. There is a major and significant difference.

Secondly, real-time (and reliably accurate) information can only come from tests and diagnosis undertaken at that time.

A distributed system that provides access to historical information in a health record can be just as responsive as in a centralised system - it's just delivered differently.

It seems silly, costly and riskier to design a centralised system that might be seen as optimal for an ER when most of the use and value comes from supporting the health care of patients with day-to-day, longer term and chronic health issues.

Anonymous said...

Bernard - I think the Northern Territory has done something close to what you describe. Perhaps we should have built on that?
Page 36 of the PCEHR Review report says:

"The panel was fortunate to be able to visit the Northern Territory and see MyEHR in operation supporting predominantly the Aboriginal and Torres Strait Islander communities.
It is clear that the above functionality is a very useful clinical tool to the point where the Panel was informed that it would be impossible to effectively function without the full functionality that they already have including diagnostic imaging and pathology and clinical measurements. The importance of clinical measurements was very clear during our visit to the renal unit where the simple recording of dry weight can inform the clinician about the level of acuity of a renal patient presenting in any one of the nodes served by MyEHR."

We could have cloned it and put one in every state.

Anonymous said...

What would Accenture have charged for that approach??

Bernard Robertson-Dunn said...

I was aware of the NT work and I believe there is a group of 80 hospitals in SA who are also developing a shared eHealth Record, although I don't know its status.

http://ehealthspace.org/casestudy/case-study-south-australia-goes-large-transform-care

There may well be other eHealth record projects around the country.

Given that many/most large scale IT projects go belly up, I would have thought that a distributed system whose parts could could grow and evolve independently is a better approach.

Getting back to the report on the SA project here's a quote:

"One of the strengths of the new solution will be the availability to clinicians of the statewide patient EHR at point of care. Mr Johnston distinguished the state EHR as being much more information rich than the national personally controlled electronic health record (PCEHR)."

So I ask my question again - what health care value does the PCEHR add? The data in the PCEHR is a subset of that which is held somewhere else.