A Simple Person’s Guide to the Value of Health IT.

A colleague who is preparing university grant applications to fund his planned research in Health IT has recently challenged me to identify what I see as the value that can be derived from the implementation of Health IT. He is concerned that until a case is made that can be understood by laymen (read smart people who just know very little about the specific area) his job in obtaining funds will be harder as will my more general task of trying to sell the proposition to Government and other potential funders of investment in the area.

It considering the answer to the challenge it seems to me the best place to start is to consider what it is we want from an ideal healthcare delivery system. I would suggest the following are at the top of the desiderata:

1. The system should be safe and should not cause any harm either through action or inaction.

2. The system should utilise evidence of treatment efficacy and quality to guide patient care.

3. The system should be as cost- effective and equitable as is reasonably possible.

4. The system should operate as a supportive and interesting environment and be as stress-free as possible workplace.

5. The system should provide a co-ordinated seamless experience in managing a particular episode or care or illness with all those involved having the information they need to do their part without continued reference to the subject of care.

6. The system should, as a result of care delivered and with minimal extra effort, generate the information required to support functions such as academic and clinical research, post marketing surveillance of drug side effects, treatment outcomes, systemic system errors, general health system management and delivery of public health and bio-terrorism services and warnings.

If it is agreed these attributes are about right where the question to be asked is there Health IT in its generic form can make a difference.

Health IT can provide clinical decision support to those making the ‘life and death’ decisions and improve both the consistency and quality of the decision making – reducing errors of all sorts - saving both lives and money.

Appropriate use of Health IT can improve the accuracy of a clinician’s recall of a patient’s important attributes (allergies, current and past illnesses, medicines being used etc), ongoing clinical record keeping and in the process assist in the sharing of information between carers while also making available vital information for use in areas mentioned in point six. This is part of the importance of the electronic health record (EHR).

Health IT can assist in the increasingly overwhelming task of managing clinical knowledge and providing this information to those who need it. Increasingly the stream of information being generated by research and clinical trials is exceeding the capacity of clinicians to absorb the available information and to navigate available knowledge without help.

Health IT, as it has been seen to do in so many other fields of endeavour, can also replace much of the repetitive and drudge activities of the operation and delivery of health services. Service departments (laboratories, pharmacies etc) can be automated to maximise efficiency and quality of service, routine accounting and supply chain management can be optimised, photographic film can be replace by digital imaging improving both ease of use and eliminating film costs etc.

Health IT when combined with appropriate communication technologies can provide the information needed for safe consistent and properly co-ordinated care no matter where the patient is – from the surgery to the hospital to the home.

The combination of EHR technology and its implementation and use by the majority or practitioners, will provide the data-bases required to address the needs of research, management and all the other interests mentioned in point six above.

Of course there are potential risks, barriers and problems that need to be addressed. These include management of the security and privacy of identified clinical information, obtaining the proper levels of investment in appropriate technologies, having adequate trained practitioners to ensure proper system use, the proper allocation of the benefits flowing from Health IT deployment and use between all the stakeholders and having pragmatic standardisation of key areas of the technology to ensure effective system interoperation.

Nevertheless the benefits are demonstrable in all the areas mentioned above, have been proven to be there for the taking at a reasonable level of expenditure. All the risks are manageable and it is essentially just time to get on with it!

David.

Guest Article on Secure Pathology Messaging III - The AAPP Perspective.

Ann Webb, Deputy CEO of the Australian Association of Pathology Practices has written another contribution to the secure messaging discussion, from their perspective. This adds another view to the discussion for all to consider.

Ann writes:

In a guest blog by Dr Ian Colclough (see the entry for February 1, 2007) there were a number of references recently to the position held by pathology laboratories with respect to electronic messaging. It was felt by the Australian Association of Pathology Practices (AAPP) that a clarification of that position may be helpful.

Members of the AAPP provide more than 90% of community pathology services in Australia. The AAPP has a working group in health informatics and has supported standards development in this area for more than 10 years.

Among these standards were the Australian standards that Dr Colclough references for the implementation of HL7 v2.3.1. Indeed the Standards Australia Working Group that has carriage of the AS4700.2 and HB262 was formed through the efforts of AAPP member’s and in 1996 after formally reviewing the available messaging standards (including EDIFACT and PIT) unanimously endorsed the adoption of HL7 and immediately set about producing an Australian implementation guide.

This was first published in 1998 five years after PIT, a simple document format, was developed and adopted by pathology practices to get electronic messaging started in Australia. The structure for reports provides for the transfer of both atomic results and a rendering to ensure that the intended pathology report is conveyed with high fidelity.

Pathology practices have since been involved in a number projects to test and improve these messaging standards including communications with cancer and communicable disease registries.

The AAPP endorses ‘AS4700.2-2004 Implementation of Health Level Seven (HL7) Version 2.3.1 Part 2: Pathology orders and results’. Furthermore member practices currently provide significant volumes of electronic messaging to GPs, Specialists, and hospitals using these standard messages. The Pathology Practices are not the source of delay in widespread standardisation and would be pleased to move their customers to this mode of messaging now. The report receiving systems however have to be capable of managing these messages properly. That PIT is provided at all by AAPP member practices is because that is what their customers have asked for.

The AAPP continues to work closely with Standards Australia and NEHTA to establish the infrastructure that Australia needs for connected health care.

-----------------------------------------

Thanks Ann!

David.

What Might A Privacy Friendly Access Card Look Like?

In this short article I try to suggest there is a way to achieve the benefits of Smartcard technology and at the same time improve the service delivery and efficiency of the Commonwealth Human Services Department while avoiding much, if not all, of the present concerns regarding privacy and function creep.

The approach would involve the following:

1. Having an Access Card that has nothing more visible on it than its name (A Commonwealth Access Card) and a strip where the citizen can write their chosen ID (that could be their name but does not need to have anything to do with their actual name, address etc). (The card face has no photo, no name, no date of birth, no number etc).

An option, if required, for those who need to deal with services by phone, and need to quote a number might be to allow the ID number to be printed on the card at the specific request of the citizen. Normally most would not need this option as they would be obtaining a benefit at a point of service in person with their card.

2. The Access Card holds only four pieces of information electronically.

a> The card’s ID Number and

b> A quality photo of the card owner and

c> The unique biometric identifier code created from the photo and

d> A card expiry date.

3. The Commonwealth Secure Customer Database only holding the ID number, the citizen’s name and the biometric identifier derived from the photo (not the photo itself).

4. The Access Card being secured electronically so it is only usable by Commonwealth Government services authorised by legislation to utilise the Access Card using Commonwealth Government Access Card Readers.

5. The Access Card held information not being accessible by standard PC equipment or card readers.

6. Having the Access Card do nothing but act as a ‘key holder’ for Government services and nothing else (not a “mini-iPod”, e-Health Card, credit card etc)

How should the system be used?

First, when enrolling for an Access Card, high quality (“100 point”) ID is provided and temporarily stored against the ID Number that is to be allocated. At the same time the photo is taken, converted into a biometric, coded biometrically and also stored.

Next, once appropriate verification of the documentation is undertaken, the card is issued with the data mentioned above being stored on it. All information other than the name, the other data used by the electoral commission, the biometric ID code and the ID number are then removed from the secure database and destroyed. (This is necessary to prevent multiple cards being issued for the same person)

Internal Human Services Department systems use the ID number as their key and each collects all the other information they require for their operations when the card is first presented at say a Medicare, Veteran’s Affairs or CentreLink office. Each benefit thus has its own data-base to manage each benefit and linkage of these data-bases would be only permitted for reasons the public are content with.

All Government card readers will be photo display-enabled to facilitate display of the stored image of the customer, which is only held on the card and no-where else, and the name associated with the card. The ID Number is made available only electronically to Government systems and to no other requesters – even if they gain access to a Government card reader. Thus people delivering services can verify an individual’s ID and determine their ID number, but the card is useless to anyone else who is not an identical twin with the same name who can access a Government card reader and persuade the supporting system to provide its ID number. Still better, even if an ID number is known, it cannot be used without a matching card or specific consent from citizens who desire to access services over the phone and who are happy with the risks this involves.

If desired a PIN could be also used to maximise the security of access, even to the photo.

The benefits of this approach are:

1. There is no database created of every Australian Citizen that holds any more information than the electoral role with an ID number and a biometric ID (Note: no photographic image is held by government).

2. No photographic database of citizens is developed.

3. The card, having only a hand written “name” on it, cannot be used by anyone if lost. If use is attempted the embedded picture and the correct name will be displayed by the reader and this will disclose any attempted fraud to the Government operator.

4. If a citizen chooses to apply a PIN the card will be virtually useless if lost.

5. No one is going to ask anyone to produce a blank card to confirm their ID – especially when no useful information is available without a special Government reader. It will not become an instrument of control and oppression as the present proposal risks.

6. Function creep cannot happen except if the network of Government readers is extended and the citizen chooses to use the service that the Access Card enables.

7. The risk of a numbered card causing identity fraud disappears - there is no number on the card.

With this approach the Commonwealth has a strong link between the key it uses to deliver services and the presenting citizen seeking to use those services, thus it can control fraud etc but the card is – still as it should be – just an Access enabler.

The only downside of this approach is that, if the card is lost, full re-identification is required. That dis-incentive of itself should make most people pretty careful with their card!

The point of this commentary is to show it would be possible to design an Access Card system that would be privacy friendly and meet the Government’s objectives.

Why this has not been done is a mystery to me.

David.

Late Note:

In the Financial Review of the 13 Feb 2007 we now learn function creep is running on apace with, among other things, disputes breaking out with the banking sector of the recording of the Access Card ID in banking records and the rules for how such ID can be asked for. The banks say that having to get written consent to record and use the ID is too onerous. My position would be that an Access Card ID is no business of the banks whatsoever, consent or no! Extra function creep number two is that it seems a ‘voluntary’ Aboriginality flag is to be added – as requested by Medicare Australia. Will it never stop!

D.

What Should be the Practical Outcomes of Health IT Privacy Protections?

In the last year, and most especially since awareness of the Human Services Access Card Project and the NEHTA IHI Project has increased, there has been a lot of legal and technical debate regarding these projects.

What seems to have been missing, for me, is a discussion of what is actually desired as the practical outcome(s) of whatever privacy regime we put in place and why each of those attributes is important. This is to forget for a moment how these outcomes are to be achieved technically but to recognise that in the design of any technical solution privacy has to be an embedded, fundamental design requirement.

My views have been formed from a range of sources, including a range of contributions to the Access Card Debate from the Privacy Commissioner and Prof. Alan Fels and his Taskforce, but I have additionally had the experience of consulting a range of interest groups in this area directly and the strength of their feelings has had a significant effect on my position. In saying this I must point out that with the exception of one or two ‘extremist privacy zealots’ who would require that a clinician were not to take any clinical notes and be subjected to a ‘brain wipe’ the moment a consultation was concluded, most groups and their representatives were balanced and totally reasonable in their concerns.

If I was to try and summarise what I have heard and continue to hear in a sentence or two it would be that people desire to have control over disclosure (to anyone) of health information they see as sensitive for what-ever reason and that if they feel they lack that control they become concerned or worse. There is also a feeling (belief) that health information – especially historical and diagnostic information – is different to financial and other private information in the sense that once disclosed it may not age or become less important over time (knowing a person’s bank balance or credit card number is likely to be only of importance for a short period of time – not so that an individual who has a genetic illness, has had a mental illness or has HIV / AIDS).

While not yet an issue of significant public concern it seems to me that, in an age where discoveries are being made daily, the control and sharing of genetic information will also become a major health information privacy concern over time.

There is also an increasingly pervasive feeling that exercise of control over one’s private information is becoming more difficult in the electronic age and that there is a reducing tendency to trust assurances as to privacy as individuals see their ‘private information’ turn up, unexpectedly, in the hands of all sorts of marketing entities and the like. Trust is also being eroded with the frequency of stories about improper access to information given in confidence to agencies such as the Australian Tax Office and CentreLink.

As yet it seems that most people are happy to share virtually all private information with the key individuals involved in their care on the general, usually unstated, proviso that they will be asked or give consent in some way and that they understand why such sharing is required. (Chronic disease advocacy groups are among those most keen for controlled but necessary information sharing). There are however some groups – especially those who could be described as having either stigmatising or embarrassing illnesses – who desire very considerable control of the use of their information at almost a ‘line by line’ level. If such groups are not provided with such control they will typically withhold information from carers, with the risk of poor or dangerous care being provided.

There is little doubt that individuals with HIV/AIDS, a history of mental illness and others are a major group of owners of multiple Medicare Cards. This permits them to receive care they may otherwise miss out on without disclosing to routine carers information they wish to keep to themselves. Just how this group will be catered for, reasonably, in the present Access Card and NEHTA proposals, is not yet clear to me.

The use of such ‘multiple identities’ is more common than is widely understood and is based on very well grounded fears of discrimination in all sorts of areas of life if some secrets are not kept. Latrobe University has, in a number of reports, highlighted the range and depth of discrimination and disadvantage suffered by many HIV/ AIDS patients.

It should also be noted that for many chronic disease groups the prospect of the use of de-identified information for purposes such as research and public health is strongly supported – just as long as there is no risk of re-identification of the information being possible.

Another poorly recognised issue on the part of city based health policy developers is the impact of even relatively innocuous unauthorised disclosure of information in smaller communities. (The ‘only gay in the village’ effect.) The impact of such disclosure can result in people needing to re-locate or worse and must be carefully addressed in any system design.

Of note also is the need to recognise that health information typically is associated with detailed demographic information (age, sex, address, date of birth etc) and that this information is true ‘grist to the mill’ for those who wish to trace people for whatever reason or who wish to steal identities.

Another issue not often discussed is the need to ensure flows of information to employers, government agencies etc is not prejudicial to gaining or retaining employment or receiving appropriate benefits and payments. Employers are certainly not equipped, in general, to understand or make sense of private health information, and should not receive it without specific consent. The same also applies to insurance companies, recognising that the citizen must understand that a failure to disclose germane information will most probably mean the coverage is void.

All in all it seems to me there is a perception emerging that control of one’s private health information is being steadily eroded and that this is not beneficial to anyone. As I have said before trust is lost if the citizen does not have the final say as to how, within reason, their private health information will be shared and used, and if that is permitted to happen E-Health will struggle to succeed in the way it should.

It is up to the proponents of the Access Card and NEHTA to ensure these ‘unexpected consequences’ of their initiatives are properly handled, that reassurances provided are indeed true and that we don’t create a class of disadvantaged and discriminated against ‘health information privacy refugees’.

David.

Where Could NEHTA Actually Add Value?

In this article I attempt to identify the tasks and objectives that NEHTA should be addressing rather than what is presently happening. The strategic perspective I am adopting is one that says there are two key priorities for virtually all health service organisations and that NEHTA’s efforts should be predominantly focussed on supporting, enabling and facilitation these two (urgent) priorities.

The two key priorities are that first health services should be safe and effective and second that health services should be economically efficient while being equitably accessible and distributed where ever that is possible.

The implications of ‘safe and effective’ include that no unnecessary harm is done, that needed treatments are not omitted and that where possible the care provided is based on up-to-date evidence. There is an implicit implication here that recognises that modern clinical practice is so complicated that only with automated support and well designed systems (both manual and electronic) can ‘safe and effective’ care be demonstrably delivered. Another clear implication is that all deliverers of care must be provided with the best possible tools and circumstances to ensure their patients have the best outcomes.

Efficiency and equity of access I would suggest are matters of common sense and common humanity. We should not be wasteful in delivery of care and we should ensure that all who need care can receive it without unreasonable delay.

Secondary priorities include the collection and management of information that both permits and enables the health system to operate in a coordinated and coherent fashion, be well managed and have a sensible balance between preventative and curative care.

Un-stated, but implicit in all of the above, is that everyone touching the health system is treated with respect, compassion and that their personal privacy and individual autonomy is fully respected and guarded.

NEHTA’s mission should be identify, specify, define and recommend the attributes of the Health IT systems required to optimally support the overall health system which has these objectives and goals. This it should be doing by working with the relevant stakeholders (including clinicians, relevant service providers (Pharmacists, Laboratories, Radiologists etc) ,Health IT providers, Health Departments and Institutions and consumers) to identify out what is needed and then develop innovative ways to have such systems delivered and a coherent fashion.

It seems to me, based on the Rapid Learning Approach identified in two recent blog entries, that where the main value lies is not so much in having systems, but having them actually used in the delivery of care. This means we need usable, quality systems, with rich functionality in decision support etc, in use in our General Practices, Specialist Offices and Hospitals. We also need to have them communicating successfully and safely the clinical information needed both to deliver care and understand what is happening out in ‘the field’. The systems also need to be able to ‘take care of business’ and thus appropriate links to Medicare Australia and other funders are important.

Additionally careful analysis of referral and prescription information and the associated systems is important. Also it is important not to ignore the needs of the public health sector in monitoring illness and warning of possible bio-terrorism. Analysis of where technology can assist in delivery of mental and aged care services is also needed urgently – although much is already known much of the relevant information seems a little fragmented at present in Australia.

On the basis that we know all provision of all the necessary systems is doable the first major task for NEHTA should be to develop a the National E-Health Strategy and Business Case and Broad Implementation Plan that, treating the health system holistically, maps a practical achievable and incrementally achievable roadmap of activities and investment.

Secondary activities should centre around utilising the work done both here and overseas to define and specify functionally the minimum standards for GP, Specialist, Hospital and Ancillary Systems. The CCHIT model in the US is one to consider with relevant changes to suit local conditions and business requirements for certifying systems once the system capabilities and connectivity is defined.

With hospital systems it may be valuable to develop common specifications against which State health systems can conduct procurements. Such systems are major investments and no doubt support from a skilled national entity would be welcome.

In essence what I am suggesting is that NEHTA should be working to ensure, as quickly as possible, quality systems are certified and health providers are able to procure / purchase them in the confidence they will be fit for purpose and deliver the benefits expected while having the basic levels of interoperability needed to enable practical information flows and clinical messaging around the health system. Note I am also keen that key enabling work, terminologies etc, also continue and that NEHTA be resourced to ensure deadlines promised to the Health IT community are actually delivered. That way the available products will be so much better!

It is up to the Health IT system providers to offer competent certifiable products and services and the health system to address how best to facilitate the required investment and ensure benefits are distributed to those who incur the costs and undertake the additional work.

I recognise that this outline is very high level and will require major change management. However, nothing here ‘rocket science’, the technologies are proven and implementable and the risks can be managed reasonably easily. Additionally the implementation can be phased and incremental so mistakes are learnt from and risk further minimised.

An approach of this type could, in a five year time frame, dramatically improve the safety, effectiveness and manageability of our health system. It would be good if NEHTA just forgot about identity management (others are doing it), academic interoperability frameworks, shared EHRs and the like for this period and facilitated getting the basic proven technologies implemented and used. Once the local operational systems of high quality are in place the challenge of enhanced information sharing can be addressed incrementally over time – just as is happening elsewhere in the world (witness the ground up Regional Information Network Approach in the US and the scaling back of the information content on “the Spine” in the UK NHS)

David.

Australian E-Health Policy – Is NEHTA’s Board Steering in the Right Direction?

On the 30th January 2007 The Australian published a long article entitled “ Doing the numbers on e-health” by Karen Dearne which was described as the “world according to Dr Ian Reinecke” CEO of the National E-Health Transition Authority (NEHTA).

It seems to me that a sensible way to approach the article is to endeavour to distil what was being said or implied and to offer commentary on that basis. As the comments were made by the NEHTA CEO it seems fair to assume he is accurately reflecting the views and understanding of the NEHTA Board and it is my plan to address the apparent policy directions and choices being made by that Board, based on the position reflected in the comments documented in the Australian.

First it was obvious throughout the interview that NEHTA has no sense of urgency about advancing the e-health agenda. We are told that the business case for the National Shared Electronic Health Record (SEHR) may be ready for Council of Australian Governments (COAG) in 2007 (or if not 2007, hopefully 2008) although it was not clear, even if ready, whether it could actually make it onto COAG's agenda. Elsewhere we are told that (presumably) after approval has been given it will take three years (at least) to develop before it can be progressively implemented (over how long is not specified). What this says is that implementation may start in 2011 at the earliest and then proceed at presumably an equally relaxed pace.

Given Australia has been working on a variety of Shared EHR proposals since a House of Representatives Report in 1997 it is just flabbergasting that it can take a decade to get to the stage of developing a business case for e-Health which is yet to be completed or released. (The UK, the US and Canada did this years ago.) Worse it seems they are then prepared to wait another un-specified number of years to actually begin to deliver benefits in terms of patient safety and the quality of care – to say nothing of health sector efficiency. We know from studies overseas the opportunity costs of not implementing Health IT are billions of dollars in benefits forgone per annum (Canada thinks about six billion per annum in their country, which is similar in size to Australia) but it seems clear the NEHTA Board is content to just ignore such costs and proceed at the current leisurely pace. NEHTA itself, in documents published almost a year ago, has shown it understands this opportunity cost – so where is the action?

The lack of urgency with which NEHTA is proceeding becomes even clearer when it is recognised that responsibility for development of the SEHR was passed to NEHTA from the Commonwealth Health Department in April 2005. What progress has been made since then has not been obvious to this observer. Very little - is likely to be the answer given the interview comment like “A lot of that detail will start to move into the public domain, and (the SEHR) will be the subject of intensive work over the next 12 months”.

Second we are told by Dr Reinecke that “My understanding is that the access card will operate quite separately from the healthcare identifier. At this point there's really no place where the projects intersect.”

Just how can this be right?. The Access Card is going to allocate an identifier for use in the access of Health Services - how then can it be that this doesn't intersect with the identifier NEHTA plans to allocate for use in the Health Sector? If this proceeds all it means is that essentially every user of the Health Sector will have two identifiers on their health records – one from Minister Ian Campbell and one from NEHTA. I wonder does the Board understand just how bizarre and potentially wasteful this is?

Worse than that, the NEHTA identifier is likely to be the less robust of the two and yet it is the one that is intended to assure reliable linkage of health records rather than the much more carefully verified Access Card identifier!

Even more difficult to follow is that NEHTA thinks it can provide a health identifier for $15 million a year over three years ($45 million) and the Access Card project is costed at over $1.1 billion over about four years. Someone surely has their costings badly wrong?

Third we are told that it is NEHTA's preference to change the law so they can mine the most reliable of the Medicare client databases for names and addresses to populate their identity database.

I wonder what Professor Alan Fels of the Access Card Privacy Task Force and Ms Karen Curtis of the Commonwealth Privacy Commission would think of this proposal to covertly, and without consent, mine a Medicare database for details provided by citizens in good faith. My suspicion is that they would be gravely concerned. The Board should certainly make sure they are consulted (if it hasn’t) before any more money is spent.

Last we learn that NEHTA is confident the Standards it needs will be ready and that great progress is being made. This is true to the extent that we see the emergence of pragmatism and a sense of urgency in the US and also work in the UK to simplify implementation of HL7 V3.0. While progress is being made at a good pace on the Healthcare Services Specification Project Overview (a collaboration between HL7 and OMG) this work is unlikely to be ready for “prime-time” anytime soon.

It should be remembered that when NEHTA was conceived its job was to *quickly* choose amongst the available (and evaluated) standards - now it is trying to write the standards, which will probably take a decade to agree upon and even longer to generate the evidence that says they actually work. The much lower risk profile for using proven currently working standards versus scrapping them for their obvious defects (they will have these, by definition, of course) and going for long term high risk standards development has never been discussed, nor acknowledged. It is a fundamental shift in NEHTA's program, and seems to almost have happened by accident. Yet it is a profound shift. For this to happen without anyone worrying about it, seems to show an apparent loss of control on the part of the Board in its oversight of NEHTA’s direction.

Without rehearsing again all the other matters I have raised over the last month or two it seems to me that the NEHTA Board has not demonstrated (after two years) it is up to the task of providing National E-Health Leadership and delivering the required oversight to NEHTA. If NEHTA’s CEO reflects the Board’s priorities and views accurately, and I am sure he does, it is my opinion that, on the public evidence to hand, the Board is strategically inept, ill-informed, technically out of its depth, out of touch with the e-health community and stakeholders and their requirements and simply inadequate to the task.

It would be of value to the whole E-Health domain in Australia if the NEHTA Board was to produce a paper documenting their understanding of the issues raised herein and their proposed strategic action plan to address them.

I would also suggest a lot of the problems we now see are related to the loss of the Australian Health Information Council (AHIC) and its role in providing technical and strategic governance of NEHTA. AHIC is apparently to be re-constituted in the next few months (after a two year hiatus) but unless it is properly constituted, resourced and empowered this may be more of a hindrance than a help. We will see.

What I was hoping to hear, and didn’t, was that work on a National E-Health Strategy was well underway and that calls from a large number of stakeholders for improved pragmatism, responsiveness, transparency and consultation from NEHTA’s stakeholders and customers were being taken seriously and actioned. I would also have liked to hear that NEHTA was going to push for rapid progress in implementation of the Health IT applications and approaches that are known to work in saving lives and improving efficiency while pursing its long term R&D agenda. No such luck it seems!

The E-Health Domain is one area where a Commonwealth takeover and Commonwealth leadership is to be welcomed!

David.

A Weekend Treat!

Just a short note to let everyone who has a moment on the weekend that the Robert Wood Johnson Foundation has posted a fabulous presentation on their web site entitled “Can Health IT Enhance the Pace and Power of Research? The Case for Rapid Learning Systems” dated Jan 23, 2007. The URL is:

http://www.rwjf.org/newsroom/activitydetail.jsp?id=10195&type=3

Those involved in the roughly one hour presentation are described as follows:

“Carolyn Clancy, director of the Agency for Healthcare Research and Quality joins national technology experts, including David Eddy and Lynn Etheredge, to showcase ways in which EHRs are making rapid advances in diabetes and cancer care, how rapid-learning capabilities will help accelerate personalized health care, and how, through rapid learning, doctors will do a much better job of advising patients. Also featured are John R. Lumpkin, M.D., M.P.H., RWJF senior vice president and director of the Health Care group and Joel Kupersmith, M.D., Chief Research & Development Officer, Veterans Health Administration. (HHS Secretary Michael O. Leavitt, originally scheduled, is unable to attend.)”

This is a very serious cast and this presentation is by far the best way to gain an initial appreciation of the key contents of the Health Affairs special issue on Rapid Learning Health IT and its implications.

I would highly commend this to all readers of my blog.

David.

Guest Article II - Supporting Diversity in Views.

Today's guest article is from a frequent commenter who asked for a little more space to present his views. Others are welcome to also ask for space!

Short Biography of Our Guest.

Dr.Ian Colclough has over 30 years experience in medical informatics in a career spanning medical practice, hospital administration and strategic and tactical sales and marketing in the corporate sector in Australia, South Africa and the United Kingdom. He has been involved in the research, development and design of hospital administration and departmental clinical systems, medical practice systems, the computerised medical record, patient profile systems and occupational health and safety systems.

Secure Health Messaging - Where to tomorrow?
An independent commentary on “A Connected Health Sector”.

That 'sensible' Roadmap

Tom Bowden touches on some very important points when presenting what he describes as “a sensible ‘roadmap’ for secure health messaging in Australia”.[1]

His arguments have merit and bear careful consideration. Is it a sensible roadmap? If yes – why has no-one come up with it earlier? If no – we need to know why it is not sensible. This is the sort of question NEHTA has been established to answer.

In posing his question - “However Can We Get to A Connected Health Sector from where we are today? ” - Tom echoes a presentation given by Ian Reinecke at the Health-e-Nation Conference in 2005: “e-Health – Getting There from Here”. Ian said that the new company (NEHTA) “will provide the critical standards and infrastructure required to support connectivity and interoperability of electronic health information systems across Australia”. How far have we travelled along that road?

The Pathology Sector

Many will argue strongly with Tom's view that “Australia's track-record in health sector automation (e-health) is among the worst in the developed world” [1]! Few, however will differ with his claim that the way to speed up implementation of much-needed health sector reforms is to put in place a universal communications framework. This is predicated on the proviso that the framework is independent of any one messaging vendor and accessible by all.

He asks “why Australia's health system (is) so substantially lacking in connecting its primary care sectors by comparison with Holland, New Zealand, and some of the Scandinavian countries” [1].In seeking the answer he points to the Pathology sector as offering the most expedient way forward.

Pathology laboratories are high-volume transaction-based entities totally dependent on sophisticated computer software systems to manage their large complex enterprises. In the 1970’s they were at the forefront of health care computing and over the last three decades they have continued to invest heavily in deploying ICT to deliver efficiencies and drive down costs.

Significant rationalisation of this sector occurred throughout the 1990’s. This led to the emergence of a small number of very large pathology providers such as Sonic, Healthscope and Symbion. Also, around 1992-1993 a small group of Australian laboratories collaborated to develop a way to deliver pathology reports by e-mail using standard print-file formats. This led to development of the Pathology Information Transfer (PIT) messaging format as an expedient way to solve the immediate problem at hand.

Pathology Information Transfer

PIT has served the pathology sector well to-date. It has however not been adopted elsewhere outside of Australia. Other countries have adopted either HL7 or some other alternative, such as EDIFACT which is used widely in Scandinavian countries. This has left Australia hostage to its own solution; one which developed locally through necessity and which, through its widespread adoption, ultimately became a de-facto standard throughout Australia.

Within this context Tom makes the point that the PIT message is “unable to be usefully incorporated within an electronic medical record” [1] (EMR) as it can only be displayed in its entirety and stored as an object. This means that, unlike an HL7 message, the PIT message cannot be broken down into separate data elements which can each be incorporated into the EMR. Hence PIT messages do not easily allow for follow-up recalls, the monitoring and display of trends using graphs and bar chats, and large scale epidemiological and other population-based research studies.

Clearly the widespread use of PIT as a de-facto Australian standard has ‘severely hampered adoption of internationally supported messaging standards within the Australian health sector”
[1].

The vendor community would agree that the best opportunity to initiate the adoption of widespread electronic communications throughout the Primary and Secondary care sectors is to drive into place as quickly as possible the deployment of automated secure messaging (based on internationally accepted standards) between pathology centres (and radiology) and their medical practice clients.

The argument for the immediate adoption and standardisation of HL7-based messaging is well founded and should receive widespread support.

HL7 Messaging Standards

In passing, it is important to note that the National E-Health Transition Authority (NEHTA) has stated that pathology providers “have indicated support for the open, web service standards (which) NEHTA has proposed and that they (the pathology providers) understand the benefits they would bring in a contestable market.” [2] That having been said however NEHTA has also noted that pathology providers “perceive a business imperative to address the issues arising from the current market situation” … and more importantly … “they are considering creating a message environment that will still form barriers to cross-sector connectivity and interoperability[2].”!

Regardless of whether they have a vested interest in maintaining the status quo or not the pathology service providers need to accept that PIT's days are over. They should move on without further delay. How might this be achieved? Should it be left to market forces alone to address, or to the State and Federal Governments, or to NEHTA? Or should the vendor community together with their clients, the pathology providers, drive the changeover? Ideally, Government should provide appropriate support to the vendor community to enable them to 'assist' their clients to make the change.

There is a further point about PIT which needs to be borne in mind. Part of its attraction could be that the reporting pathologist is legally liable for the content and accuracy of the pathology report. By sending information in the PIT format the pathologist has taken all legally necessary and reasonable steps to ensure the results in the report cannot be altered. (Just like an Acrobat .PDF file.) At the medical practice end the report is simply attached to the patient record as a document.

When dispatching HL7 format results, the reporting pathologist needs to be satisfied that the atomic data that will populate the patient clinical record, at the medical practice, is similarly protected. This requires some additional software complexity at the medical practice end to ensure that these results cannot be altered. The associated complexities of cross-mapping the incoming results definitions to those of the clinical desktop system also need to be addressed.

Nevertheless, HL7 V2.3.1 is the method of choice for transmitting pathology messages. There are sufficient numbers of organisations and people well versed in HL7 in Australia today to drive HL7 standards based solutions into place. The changeover to HL7 will be a formidable exercise given that the majority of pathology results being delivered today use the proprietary, non-standards-based, PIT message format. To bring this change about will require astute enlightened leadership underpinned by some very sophisticated strategies and a finely-honed understanding of all parties that will be impacted by the change and the obstacles and issues that will be encountered. The key to the solution lies in the strategies required to effect the change and the political will to bring it about. Careful analysis of the problem reveals that it can be resolved reasonably quickly.

Tom is not alone in expressing the thoughts of many in the vendor community when he says, “the really sad thing is that Australia has already spent huge amounts of effort and money developing many of the standards, support services, systems and capabilities needed to do the job” …. and … “that the basic tools to enable there to be a connected health sector do exist; they are just not being used properly. [1]”

He observes that “the National e-Health Transition Agency (NEHTA) is using very little of the immense amount of work that was already underway prior to its formation” …. and that …. “based upon the present rate of progress it is likely to be some years before we see any implementable results. [1]”

Standards can be so contentious

The standards argument is divisive and very complex. Yet, all would agree that work on developing standards and resolving the many conflicts and issues involved is very important. The unresolved issues however should not be permitted to impede the development of interoperability between software vendors. Nor should they be permitted to impede the step-by-step resolution of problems encountered in building an integrated, shared, longitudinal Electronic Health Record (EHR) for widespread use across all health sectors. [3]

On this subject David More presents a well reasoned view in his recent blog on Archetypes and Standards 4. He points to the "considerable successes achieved in the Health Sector with messaging technologies including EDIFACT and HL7 V2.x.". He also notes that "to-date - in the messaging arena - HL7 V3.0 and EN13606 are still in the process of development and tools and implementations are by no means common.[4]"

In terms of the ‘clinical data’ to be transmitted, it is clear from his discussion that much of the work based around standardising ways of storing and retrieving the atomic elements of information required to populate the EHR over a patient's lifetime is still in its infancy and evolving slowly.

It should be apparent that NEHTA has a major R&D role to play in the years ahead. That being the case the marketplace cannot and should not be made to wait until NEHTA has completed its deliberations on the many complex and elusive issues which it is trying to solve. Rather, the Government's and Health Departments should embrace David's call for a dual approach. One which supports "appropriate and promising R&D" ……. "blended with dramatically more energetic investment and deployment of systems which are already known to work and for which evidence as to their value is quite unarguable" [4].

Engaging 'the' stakeholders

NEHTA was established to drive forward critical e-health initiatives. Its governance is under a Board of Directors made up of CEOs from the Health Departments of the Federal, State and Territory jurisdictions across Australia.

The jurisdictions provide NEHTA's funding. As a consequence they are stakeholders and also NEHTA's only shareholders. However, they are not the only stakeholders to be impacted by NEHTA's work; far from it. The IT vendor community, medical, pharmacy and allied health care practices, hospitals, nursing homes, aged care and community health centres, and others, have a major vested interest as stakeholders in NEHTA's outcomes.

NEHTA's governing stakeholders are predominantly driven by imperatives arising from the highly politically sensitive public hospital (tertiary) sector. Yet, in many ways NEHTA has more to offer the Primary and Secondary care sectors than the Tertiary hospital sector. Hence, the absence of Primary and Secondary care sector stakeholder representation and influence at the highest level of NEHTA's governance is disconcerting; more so, should the immediate e-health needs of the Primary and Secondary care sectors become bogged down by the complex web of e-health and ICT related issues and politics so prevalent in the Tertiary care sector.

Fundamentally, the Primary and Secondary care sectors (organisationally, functionally and politically) are quite distinct from the Tertiary (hospital) sector; even more so when it comes to categorising health-ICT vendors into the particular health sector niche which each predominantly services.

A strong case can be made for allocating significant resources to focussing on addressing some of the more immediate e-health needs of the Primary and Secondary sectors separately from the needs of the Tertiary sector. The urgency of this is further reinforced by Ian Reinecke's comment that "we need to be carefully aligned with the powerful players internationally"[5]. Reflecting on this leads one to enquire as to which powerful international players NEHTA may have in mind. Could they be Cerner, Philips, GE Healthcare, and Siemens?

Whoever it may be it seems to indicate that NEHTA has blatantly failed to appreciate that the Primary and Secondary care sectors are not the domain of the powerful international players. Rather, they are serviced, professionally and competently, by a cohort of small, highly skilled, agile, homebred software vendors. Some have made very impressive inroads internationally. Some are at the cutting edge of standards-based implementation projects that test the operability of HL7 V3 and OpenEHR technology as an investment in advancing towards the EHR. They work in a market sector in which the powerful international players have shown little or no interest!

In February 2006, NEHTA discussed the consequences of a “proliferation of incompatible systems and technologies” and it noted that “harnessing the capabilities of software vendors and service providers in the health sector to provide this connectivity will be a key to success” [2]. In light of this the time has come to clearly distinguish between what activities should be classified as R&D and what should be classified as ‘implementable today’.

It is time to reconsider what benefits could flow from drawing upon and utilizing a lot more of that 'pre-existing effort and the skills and expertise' so readily available in the vendor community.

HL7Messaging – What is NEHTA’s view?

In order to achieve widespread standardisation of messages there needs to be a process in place to achieve consistency when implementing standards.

Few would disagree that Australia should “define one standard for each purpose and rigorously enforce it ……… to prevent proliferation of non-standard formats.” [1] The general consensus among industry experts seems to be that Australia should implement HL7 V2.3.1 (AS 4700.2) as the minimum health industry standard.

A strong case can be made for populating medical practice records with discrete results without further delay as opposed to continuing with the PIT format. As this can be achieved using HL7 V2.3.1 and transmitted under control of the HL7 messaging protocol it is clear that HL7 V2.3.1 provides a nationally consistent base from which to move forward building upon what is readily available today.

NEHTA’s view on messaging standards is unclear at this time although it would be enormously beneficial to everyone if it could be clarified quickly.

Australian Health Messaging Laboratory

Tom comments that the Australian Healthcare Messaging Laboratory (AHML) has an important role to play in helping to secure the consistent deployment of HL7 messaging standards by software vendors.

AHML, as part of the Collaborative Centre for eHealth (CCeH) based at the University of Ballarat, is unique in Australia. Having been accredited by the National Association of Testing Authorities (NATA) it has positioned itself to act as an on-line test-bed laboratory for the technical evaluation and certification of electronic messages in healthcare applications being deployed by vendors in more than 30 countries. It is therefore well placed to test for conformance and monitor compliance against Australian and International Standards.

State jurisdictions and the Federal Government have a vested interest in securing AHML's future as the central reference point for HL7 messaging standards[6]. It should not be too problematic to find the funding required for this to happen without undue delay.

This will likely necessitate AHML having to be constituted formally as a Reference Laboratory. Under this arrangement AHML's funding should be derived from a mix of Government support and self-generated revenue. Its staff and management should be focused entirely on the testing, certification and compliance of messages against standards. The Reference Laboratory will need to be seen to operate transparently and impartially, totally devoid of any potential for the development of conflicts which could arise should its management and staff elect to undertake consulting assignments where they may be compromised by competing against vendors whose messaging software is subject to AHML testing.

HL7 V2.3.1 is widely implemented, tried and proven. It offers substantial benefits over PIT and even more importantly, it provides a way for the entire e-health community in Australia to move forward. It is therefore time for a seriously concerted effort to be made to move the pathology sector away from PIT towards the adoption of HL7 V2.3.1 or above. This seems to be a very good way to go for the next few years and it can be done today; delivering enormous benefits with minimal pain by using standards which have a good track record of runs on the board worldwide. An important element in driving the uptake of standards is to implement strategies which will create a dynamic contestable market environment.

Is there a better way?

There has to be a better way to move the e-health agenda forward. Tom lists three strategic options for the way ahead, opting for some form of collaborative partnership in which “small, specialised service organisations are given government backing and modest financial incentives[1]” to achieve sector integration.

In broad principle the concept of a ‘collaborative environment’ has the most likelihood of succeeding. It should be appreciated however that the optimal selection and order of enrollment of ‘participants’ is not a straightforward exercise. It is absolutely imperative that the setting up of a ‘collaborative structure’ (consortium) is done thoroughly and properly at the outset in order to contain risk and ensure the consortium’s longevity and success.

Collaborative consortia are quite complex to construct, more so in the health sector than in most other industries. Building a successful Health-E-ConsortiumTM requires a unique business model and organisational structure. The partners must work in harmony together, where often they are in competition with each other. They must be committed, highly motivated, well led and well rewarded, and they must be prepared to work in an environment which promotes a culture that is based on trust, a concern for each other's profitability, and a mutual commitment to customer satisfaction.

References.

[1.] Bowden T, A Connected Health Sector – However Can We Get There From Here? (16 January 2007) Guest Blogger Article www.aushealthit.blogspot.com

[2.] NEHTA, Towards a Secure Messaging Environment – An E-Health Transition Strategy Version 1.0 – 2/02/2006 For Comment, page 6.

[3.] Colclough I, Comment attached to Implementation Really Matters (8 January 2007) Blogger Article www.aushealthit.blogspot.com

[4.] More D, Archetypes, Standards and All That Jazz - Part 2 (28 January 2007) Blogger Article www.aushealthit.blogspot.com

[5.] Dearne K, Doing the numbers on e-health - the world according to Dr Ian Reinecke, The Australian (30 January 2007) IT Business, page 3.

[6.] Australian Health Messaging Laboratory, Available Test Profiles, accessed 30 January 2007, http://www.ahml.com.au

© Copyright 1 February 2007
Dr Ian Colclough - Integrated Marketing and e-Health Strategies
ihsipl@smartchat.net.au, 0412 059 392

Note: These views are those of our guest commentary provider and should be read as such. This forum is to provide a space for comment and debate and I hope this contribution helps. Thanks Ian.

David.