› 
eHealth Records – Where Are They?

Malcolm Crompton's picture

We have been talking eHealth records for years.  Almost worldwide, the topic is treated by many in the health informatics arena as a self‑evident necessity, by some health service providers as an investment and re‑training imposition with little direct return to their practices and by health consumers with suspicion. 

There appear to be countless reasons for this impasse, in Australia as elsewhere. 

With the forthcoming hospital revolution foreshadowed by the incoming Federal Government, is this an opportunity to break the impasse?  Oxygenating the health information bloodstream that will be essential to this revolution would appear to be essential. 

The Health Informatics Society of Australia (HISA) seems to think so.  Recently they published “A Vision for an Australian Healthcare System Transformed by Health Informatics” as their contribution to this debate.  The full report and Executive Summary are both available on HISA's website.

Two of the most interesting aspects of this debate are:

  1. Whether Government inspired and provided eHealth records (or at least, standardised summaries) are the way to go, as being espoused by NeHTA for example, or whether the focus ought to be on standards and inter‑operability requirements then letting the private sector provide
  2. How to earn genuine trust from health consumers

For this discussion forum, I will focus on the second of these, taking the HISA Vision as a starting point.   

Focus Area 5 in the HISA vision - "Managing Privacy Security and Confidentiality" - is the most important component for earning trust from the health consumer.  The emphasis on security in this Focus Area goes without saying – health consumers would not expect anything less.  The loss of 25 million records about individuals in the UK announced recently will reverberate around the world in its impact has proven (yet again) that neither governments nor the private sector hold a mortgage on being the most ‘trustworthy’ when it comes to security.  It has resulted in a strong attack on the Chancellor of the Exchequer that is also spreading to the previous Chancellor, now PM, Gordon Brown and leading to questions as to whether that government can ever be trusted with personal information, for security reasons alone (e.g. see “Brown apologises for records loss”, BBC News, 21 November 2007).

Thus in my view, the more interesting element of Focus Area 5, because of the nuance required, are the Transparency & Individual Control elements.  The paper does quite a good job of this as well.   

Nevertheless, it does suffer from one of the most common failings in privacy thinking, namely addressing enforcement and what happens at points of failure – what we term the Safety Net component.  If these elements are not addressed, we are simply engaging in an exercise of inappropriate risk shifting – asking the health consumer to bear too much of the enforcement burden and too much of the risk of failure.  If eHealth records really are the boon to medicine and health system efficiency that they appear to be, we can easily afford the world’s best compliance, accountability, enforcement and safety net arrangements. 

The remaining question is “who pays”.  Private medical practices, with some justification, have said “not us”. 

Is this a classic “economic externality” that needs to be addressed either by legislation or external funding, e.g. from government or the other major beneficiaries? 

Open Forum welcomes your thoughts.

Malcolm Crompton is Managing Director of Information Integrity Solutions (IIS), a globally connected company that works with public sector and private sector organisations to help them build customer trust through respect for the customer and their personal information.



Recommended Reading

On 19 September 2007, Global Access Partners (GAP) held a strategic workshop discussing the challenge of implementing a rational e-health system in Australia in Parliament House, Canberra. It featured a paper commissioned by the Australian Centre for Health Research and written by Professor Michael Georgeff, CEO of Precedence Health Care Pty Ltd and Director of e-Health Research at Monash University.

The paper "E-Health and the Transformation of Healthcare" discussed the cost to the nation and the individual of continuing with Australia's current disjointed chronic health care system. It is estimated that improved knowledge sharing and better care plan management for patients with chronic diseases could generate direct savings to the health care system of $1.5billion per annum. Savings to the community from associated non-healthcare costs are of the same order.

The full report of the proceedings is available here.

ABSTRACT: This 2006 survey of primary care physicians in Australia, Canada, Germany, New Zealand, the Netherlands, the United Kingdom, and the United States reveals striking differences in elements of practice systems that underpin quality and efficiency. Wide gaps exist between leading and lagging countries in clinical information systems and payment incentives. U.S. physicians are among the least likely to have extensive clinical information systems or incentives targeted on quality and the most likely to report that their patients have difficulty paying for care. Disease management capacity varies widely. Overall, findings highlight the importance of nationwide policies: Policy changes in the United States could lead to improved performance.

Comments

The Banking Example

Banks handle a huge amount of sensitive complex private data with a terrific degree of accuracy every day not just around the country but around the world. They co-operate with each other, provide services which people by and large trust and somehow managed to do all this without explicit Government planning or involvement. If health care was truly privatised, and e health records increased efficiency and therefore profitability or acted as a selling point to attract customers, then I think you'd find everyone had an e health card just as everyone has a bank card today. The problem with e health provision is not too little Government intervention on this issue but too much Government involvement in the economy overall.

Tony Abbott on Electronic Health Records

Here is what Tony Abbott has said in his recent exclusive interview with Open Forum on slow progress with the implementation of e-Health records in Australia:

Tony AbbottTony Abbott: "Progress has been embarrassingly slow because it's in no one's obvious interest to make it happen. Why should one doctor or one hospital make it easier for another to treat the patient next time? Why should health institutions empower patients to be more discriminating consumers? Privacy and professional ownership concerns are not insignificant but, in my view, are often excuses to avoid what obviously needs to be done. Then there is the tendency of policy-makers to want the "best" system rather than the system that is readily achievable.

In my view, the federal Government's best contribution to an ehealth system would be to make securely available through the internet patients' Medicare, PBS and immunization data. This has been regularly promised but never delivered, mostly because Medicare Australia is no longer part of the Department of Health and thus not amenable to health ministers' directions. The need for the data to be "translated" does not mean it should not be available.

I doubt that government could build a functioning ehealth system, any more than government could have built the internet. Still, government should do what it reasonably can to make information available in this way. Eventually, once people discover how useful it can be, an integrated ehealth record will start to evolve.

(Read the whole interview here)

A research paper by the Australian Centre for Health Research

I would refer you to a research paper The Australian Centre for Health Research commissioned which was completed in April 2007. There are two areas covered in the report which are relevant to the points raised by Mr Crompton.

The first is that a properly implemented information-sharing system would save healthcare providers more than $1.5 billion per annum, and add up to $4 billion dollars to the Australian economy due to better patient outcomes, and greater workforce participation.

It is, however, important to point out also that there is a mis-alignment between those who will shoulder the bulk of the cost of the service, and those who receive the benefit, a barrier which, the report suggests, can be breached through a series of incentive programs, as well as strong partnerships between government and industry.

The full report is available through the Australian Centre for Health Research website, and I invite anyone who is interested to look through the findings.

"e-Health and the transformation of healthcare" Author - Prof Michael Georgeff http://www.achr.com.au/

Evolution, not revolution

A network-enabled and regionally accessible digital health record system could be of benefit. The current public debate on EHR's and eMR's assumes this to be the case. There is, I believe, a need to critically assess this claim. For the rest of this discussion I will refer to both types under a single label (eHR) for simplicity.

Digitally stored records tracking patient care already exist in well over 80 percent of GP clinics around Australia. Patient information on a Word document or stored on a single instance PMS system is an electronic record. In some cases these are shared amongst health care networks, enabling test results, referrals and medication data to be accessed between healthcare providers on the one network. In a minority of cases there exists Internet (IP )networks transporting shared health records in a secure manner. As eHR's become more common issues of standards, common language and secure access will increasingly dominate the public debate.

To understand the future implications of eHR's it's important to focus on what we are trying to achieve by implementing such a system . Does it deliver improved healthcare for the individual and/or the population. Will spending tens of millions on an eHR system be the best investment at the current time?

Let's briefly explore the issue.

The average GP has neither the time (8 minutes per consultation), nor the need to access a patient's health care record in order to diagnose a current condition . In the main, they achieve this verbally. In the case of a hospital emergency admission, the clinician is unlikely to read through a patient 's eHR before providing treatment. I suspect that only two types of patients will benefit form an eHR; those with chronic conditions and mothers with young children. These two group require extensive consultation time and, in many cases, case management.

At a population health level eHR data could provide useful contributions to population health in the area of disease patterns and responses to medication side effects. Are these two benefits enough to justify the investment?

eHR systems and their associated standards will evolve as the need develops not as an abstract service with little demonstrated patient or clinician need. Delivering accurate information at the point of care is the goal but I remain unconvinced that eHR's are a priority. I suggest that the effort, time and money involved in building eHR's would be much better spent on areas like the management of chronic illness, telemedicine, mental health and paediatrics.

It is likely that by redirecting time and energy towards these critical care areas, digital eHR systems will evolve naturally on the basis of use and relevance.

eHealth will struggle to achieve wide uptake...

eHealth will struggle to achieve wide uptake and Return on Investment if we continue with conventional thinking around Electronic Health Records. If eHealth is to realise its potential to transform healthcare, there are three keys to success.

First, the incentives must be there. It is critical to ensure that there is a significant value proposition for the major stakeholders and to make this intrinsic to any eHealth initiative. It is not sufficient to have benefits accrue to the system as a whole. Because health is not a normal market, this means that there needs to be a comprehensive framework of government designed incentives to encourage best practice, effective care.

Second, the focus needs to be on connectivity, not data. It is critical to get information flowing among the highly heterogeneous and distributed silos of care - among healthcare providers and consumers and software systems and services. It is not essential that data all be standardised or centralised -- simply that whatever information is available be able to be shared across the continuum of care.

Third, the business models should reflect the knowledge economy, not the industrial enterprise. This means moving away from large, centralised systems towards smaller, more flexible and more dynamic services, typical of what one today finds on the Internet. With the right incentive framework, and the infrastructure to support connectivity, these kinds of business model will drive private investment and innovation into health care, reducing the cost to government and opening up opportunities for transformation.

There are plenty of things we can do to get moving in this direction. We can provide incentives that make purchase of interoperable, open clinical desktop software more attractive than today's closed systems. The Commonwealth could ensure that a consumer's Medicare records are available electronically rather than, as currently, only in paper form. (It is interesting to see that the former Minister for Health, Tony Abbott, proposed something similar but was unable to get it effected). Governments could provide incentives and infrastructure to allow referrals, discharge summaries, results and follow ups, in whatever form, to be sent electronically to care providers and patients. On top of this, private enterprise and other organisations could then build systems for managing chronic disease, with the potential for massive ROI and dramatically improved health outcomes.

eHealth is a much simpler and much less costly than we currently think. Many are still thinking conventional IT solutions, large centralised systems, and national electronic health records. Taking an Internet approach, focussing on connectivity, getting the right incentives in place, and letting the market drive the outcomes would be far more effective.

Self-evident or mythbusting?

Having spent a few decades working in different parts of the health sector, I thoroughly concur with the notion that e-Health records are treated by many in the health informatics arena as a self-evident necessity as Malcolm suggests.

Nevertheless, having once been told that scepticism in "health" is considered to be healthy, and often considered myself to be quite healthy in that regards, there appear to be one or two major challenges that have been, and may well continue to be problems. They may in fact be major factors that have manifested in the lethargic uptake of the technology, certainly in Australia, but perhaps even overseas. Moreover, it does not seem unreasonable to consider that if uptake over more than two decades has been exceedingly slow, then perhaps there are some elemental problems that have not yet been adequately dealt with.

The first of these might be alluded to in Tony Abbott's comments on the Open Forum site about the difficulty with the business case for one of the major stakeholders - government. In fact, there may well be grave difficulties in the business case for all of the end-users, and possibly because the outcomes seem very unclear.

A second challenge I would theorise, and this is certainly not accepted by everyone in the industry, is that it is not equally self-evident that the EHR has actually been developed to meet the work practices and service needs of those who actually deliver health care services. The hopes for EHRs are that they will provide the connectivity and availability of information between health service providers and that this, almost of itself, will improve decision making, quality of care, and access to important reference information like past history, previous treatments and diagnostic indicators.

This second "challenge" then has to overcome two important obstacles. Linking data is a technical challenge which seems to be within the capacity of technical providers to achieve. There is a need though, if we are to have all the information we need available, to ensure that the data input too is adequate. For this to happen, one needs to be able to show that for the change in effort and work practices involved for clinicians to adoption EHR, they will see returns for their practices. It must be able to demonstrate they either make clinicians' work more efficient, or effective, or that the investment in time and energy required will actually deliver an improved health outcome in proportion to the size of the investment. It's not that health care workers are Luddites, or innately opposed to using eHealth systems to streamline their work patterns, or that they object to improvements in the way data is accumulated within the health care sector. The benefits, though, still need to be proved.

Also, or perhaps "more so", the output of the availability of linked data has to be in a format which allows clinicians to continue to manage their processes effectively - it has to support the way they arrive at clinical decisions and oversee clinical management, or they have to be prepared to alter their clinical decision making processes in order to better use the new levels of information available. It should be noted that changing behaviour is often slow to be achieved and not always successful.

And perhaps that is where I see the problems really start to build up. To say that health is a complex industry is an understatement and oversimplification. For instance, despite some of the best minds in the world being committed to the task, we don't even measure quality adequately: we haven't got measures and haven't been able to achieve systems which ensure quality of outcomes, or the sorts of process improvements in health that assure higher quality, as have been achieved in less complex settings, and haven't even achieved measurements of quality which would let us know where we should be concentrating our efforts at improvement. We know the quality is improving. In the developed world at least, people are living longer, and in better health. Yet as a clinician, for most problems, I have nowhere to go if I want to know what is the best way to treat a particular problem for my particular patient, or even to find out whether a specialist to whom I wish to refer has outcomes equal to or better than others. I can't find out complication rates or outcomes, let alone outcomes standardised for complexity. We can't even agree on what measures we might be able to apply to determine which treatment options should be chosen for which problems in which patients.

All that was trying to say that it is not absolutely clear that all clinicians apply the same thinking processes to make decisions, and not clear that there is an agreed correct way, or best way. And if there isn't, how will we bring together information to make it used and useful.

Of even more concern, the problem may well be more that the systems on offer do not really reflect the way in which even a reasonable proportion of health professionals use data and use records to solve clinical problems and modify management. EHRs may well assist with data storage, data retrieval, and potentially even with communication between the various healthcare silos we created and reinforce (public hospitals, private hospitals, GPs, Specialists, Allied Health professionals, community health, home nursing, HACC, and more). Even so, the business case for EHRs does not seem to have been adequately convincing based on those facets of information transfer alone, nor on the potential to be able to apply the more readily accessible information on offer.

By this time, I should probably apologise to those whose blood pressure has been pushed higher by such heretic questioning of the accepted "self-evident" truths about Electronic Health Records. I apologise too to those who know this to be of great and unquestioned value to their patients.

Yet in an industry which seeks to base practice on evidence, I look back over two decades - nearly three perhaps - and despite 17 years of the internet, I can only postulate that if the obstacles to the new world of accessible data have not yet been either argued out of existence or solved, there is something more to the problem than simply the question of privacy of data or the lack of willingness to invest.

Technology to create eHealth records, and digitally store this data has been around since the mid-eighties, although it has improved exponentially with the increased capacity of computing technology and telecommunications. Even so, it seems to me that if the systems were able to make the provision of healthcare more efficient and improve the quality of outcomes, and fitted in with the work practices of health care professionals, we'd already be using them after more than twenty years, and would not need to be involved in a debate.

The question arises then, as to whether I have more to contribute than "nay"-saying. Alright, let me put it that the theory is that if you connect up all the pieces of data collected by health care professionals at every point at which they interact with a patient, then make these available to all other health care professionals, you will save on duplication of tests, and processing, and therefore time, and reduce the potential for medical error. There may even be a contention that putting all the information available in one place will result in "better" decisions about patient management.

In theory it sounds great, but perhaps the evidence simply has not yet borne out the theory. And I was wondering if perhaps that is because the systems don't configure the information in a way that reflect optimal data entry processes at the coalface, or the decision making thought processes of clinicians at the output phase.

The GP who refers his patient to a dermatologist may well be able to pass on data on all the tests carried out over the last decade, but these will probably be of minimal relevance for the vast majority of the skin patients he refers. In most cases the dermatologist will be looking for a specific pattern of symptoms and more often signs, many of which may have developed more recently. Having looked at the way in which many GPs maintain records, and this has not been improved with the increase in computerised records, specifics of skin integrity would not usually be part of the GP's medical notes, and certainly not of computer records, at earlier consultations. How many patients would need to be treated to get a benefit from having an eHealth record in this scenario, and at what cost?

Moreover, how does accessibility to the new electronic records imitate the way in which clinicians seek to use records? Or does it? Will an infinite access to past records and past history be more than most clinicians want to inspect in order to inform their decisions? How easily will key themes or key thoughts be "bookmarked" or highlighted or brought together to allow clinicians to "flick" through the information to find the very specific items they might wish to look for in order to help make a decision? Does the human brain have a better "search" function, even through presumably disorganised or incomplete paper records, than have been developed to support EHRs.

That's not to say there are not pockets of people developing and using very interesting and creative programs for local use, but these are for the most part on a relatively small or simpler scale (even considering such notables as Kaiser Permanente that have a single organisation to contend with), and specific to pre-existing medical work practices or the needs of the relevant population. Many are organisation specific, or reflect work practices for particular specialty groups.

OK. There are some large HMO organisations in the USA that have records that cover their patients across hospital, primary and community care. We are told these have achieved various improvements in health care. Nevertheless, the systems developed have not been universally replicated, and in my mind, this at least begs the question "why?" Based only on my own experience, there are quite major limitations of any of the systems I've seen in being able to be easily used to support the way in which I track the important issues specific to particular patients, or have the same ease of search which I can use for written records.

Despite the scepticism, so far as I see it there are already two areas in which eHealth records could make a difference in Australia right away. In the areas of pharmacology and pathology, where digital processing is already part of the way the services are ordered and delivered, the leap to the next step of consolidating this data and making it available digitally is not all that far, so it would not be an unreasonable first step.

However, there is a substantial leap, in my mind at least, from the paper-based records, still used by doctors, hospitals and surgeries, or even of local computerised records, to the wide-spread adoption of electronic data entry devices, and shared eHealth records, because the actual physical processes of recording patient data digitally, and then retrieving it for patient care, fall outside the work patterns of the medical and paramedical professions.

Until we see an eHealth patient record system which actually combines streamlined work practices, with improved diagnosis and patient outcomes, uptake of a technology for which there is otherwise a "self-evident need" may remain slow.

I anticipate that there will be many who will now be eager to insistently and perhaps heatedly refute the doubt that I have cast, or at least the possibilities that questions remain unanswered. I have an interest in the area, but am not an expert, so I look forward to seeing the clinical outcomes and business cases being presented to show me how behind-the-times I am. Nevertheless, perhaps their energies may be better applied (and my own ego better protected) if they look at the failure to achieve the universal e-health record over three decades, and consider whether there might indeed be problems beyond the inertia of a large system. Some of you may recall the excitement of the revolutionary Sarich orbital engine that was going to achieve a major change in motoring technology. Did it? I have no doubt that EHRs will be the way of the future. I simply have the doubt that if the future hasn't arrived after 30 years, perhaps the product on offer needs some new thought, or perhaps this technology might spend a long time waiting for the Luddites to accept its inevitability.

And as for privacy and ethical issues, if the health outcomes case and business cases can be demonstrated without the use of the words "self-evident", "clear" or "obvious" then I have no doubt that those such as Malcolm could propose appropriate protections of both our privacy and ethical access to information.