The Commonwealth HealthConnect program is becoming controversial in the health sector, but it is an issue that requires wider debate in the community. This article briefly outlines the program and considers what its future should be. To declare my position, I believe that the introduction (and use by all clinicians) of advanced electronic health records is an essential step in improving the safety, quality and efficiency of our health system. The lost savings of not using information technology to enable improvement in these areas are estimated to be up to $9 billion, if fully realised (this estimate is taken from an unpublished analysis undertaken by the author applying information developed by the Centre for Information Technology Leadership in the USA). This represents more than 10% of Australia’s $80 billion health budget. It is vital to move this initiative forward.
As most people are aware, delivery of health services is an information intensive activity with the ability to share and access information critical to the delivery of safe and effective patient care. While other information intensive industries (e.g. finance and banking) have willingly embraced information technology, such a transition has yet to occur in the health sector. Thus, we see that investment in Health Information Technology (Health IT) has typically been between 1-2 percent of operating budgets (about one half of what is considered necessary) compared with banking, where 10+ percent per year is typical.
Recently it has been recognised that Health IT can significantly improve the efficiency, safety and quality of care. The business case for dramatically increased investment in Health IT has been compelling and, in response, the national governments of the US, the UK and Canada are all ramping up their investments dramatically. To date, while there has been some increase in Health IT investment in Australia, it has not been of the same order of magnitude – especially at the Commonwealth level.
HealthConnect is Australia’s national Electronic Health Record program which broadly hopes to achieve these benefits. See here for details.
The overall Australian HealthConnect Program (previously Health On-Line and MediConnect) began in 1999 and has been undertaken in two broad stages – a research phase lasting about two years and a trial phase which has run for three years. Trials have been undertaken in Tasmania, Victoria, Katherine (in the Northern Territory) and North Queensland to date. These have involved a little over five thousand patients and approximately five hundred healthcare providers.
HealthConnect is envisaged as a largely passive repository for patients’ Initial Health Profiles (containing basic identifying and health information) and a series of ‘event summaries’, which are created each time a health service is provided. These two components go to form a lifetime electronic health record which is made available to healthcare providers under strictly controlled circumstances. The patient voluntarily decides to enrol and consents to provide access for all new providers or carers.
It is intended that access to this summary information will be made available through the local computer system of any treating clinician, subject to the patient’s consent. The desired outcome is that all clinicians will have access to up-to-date information on the individual patient, ensuring ongoing treatment will be well informed and appropriate.
This description may be subject to change, given the withdrawal from the HealthConnect website of both the Business and System Architecture documents which defined what HealthConnect was planned to be and how it should work. The latest information on the website supports the concerns that follow.
There are a number of issues with the current plans.
First, it is intended that the source for the ‘event summaries’ will be the computer systems used by healthcare providers as they deliver care to their patients. Given the present lack of use of clinical electronic health records, especially among specialists and in parts of the hospital sector, the consistent availability of these summaries is problematic. Secondly, a key business justification for automation of clinical care delivery is the prevention of errors at the point of care. However, if this is not addressed at a national level there is considerable risk to the anticipated benefits of the whole project. HealthConnect is not planned to provide interactive clinical decision-support, which would require advanced local IT systems for use by clinicians. The planned provision of simple web accesses to the HealthConnect repository or the use of less than advanced local systems will not deliver the desired benefits.
Thirdly, clinician adoption of such advanced systems has been problematic, both here and overseas, and it is clear that significant (usually financial) incentives are required, if substantial adoption is to be achieved. While the most recent HealthConnect documents recognise this point, there is no budget allocation identified, so the seriousness of Commonwealth must be questioned.
A key reason for slow adoption is that computer use typically requires extra time and work flow changes. This means clinician adoption of this technology in our ‘fee for service’ health system is likely to be slow if not glacial. Critically also, the benefits that do flow from e-health implementations are usually found to go to those who fund health system, rather than the providers who incur the costs and slowdown associated with the use of additional technology.
Fourthly, it is planned that patient participation be voluntary upon each encounter, rather than a total ‘opt-in’ or ‘opt-out’ of the program. This allows the patient to simply not mention they have enrolled elsewhere. Thus, clinicians will have no certainty about how complete is a patient’s record. They will, therefore, apply the cautionary approach of checking for all relevant information, thus mitigating any benefits of time saving and efficiency. (It is well known that patients with stigmatising diseases such as AIDS often use multiple clinicians and selectively reveal their medical history to different carers.)
The current strategy documents from HealthConnect (June 2005) recognise that the availability of suitable integrated clinician systems will not happen even in the medium term – up until then it is planned the clinician will use the Internet to access the central repository. When faced with the possibility of incomplete and maybe misleading records, how likely is it the clinician will spend the time required to switch to the Internet, log in, search for the consumer’s record and so on? Unless access to the patient record within HealthConnect is seamless, fully integrated, largely effortless, and provides real utility and value to the clinician, the repository will simply not be used. It is also a concern that the new implementation strategy specifically rules out provision of financial incentives to software developers to provide quality effective local systems.
Also of concern in the new plans are the suggestions that HealthConnect implementation will commence as a series of semi-autonomous regional projects with some national standards guiding development. Fully nationally coordinated information flows are to happen later. This will cause a long-term lack of integration. Lastly, the report on the legal aspects of HealthConnect identifies an array of jurisdictional and coordination issues which have yet to be addressed at a policy, let alone an operational or systems integration, level.
Overall, the present proposal does not seem to be aligned to or support the requirements for a more integrated national Health System. Consequently, there is a need to pause and carefully reconsider the program and lessons learnt to ensure that all of the policy, privacy, integration, standards, consent, workflow and clinician adoption concerns, of the proposed system are addressed.
What is to be done?
First, there needs to be a fundamental rethink of the approach being adopted by HealthConnect. This should begin with the full disclosure of all evaluation reports so far to allow detailed public scrutiny and comment on the value received for the $50 million invested to date. Based on the information gleaned from this disclosure, there should be a re-design of HealthConnect to ensure it is an integrated, standards-driven, holistic, National Health IT initiative that covers the clinician-used local systems, the communications and the repositories. This should involve a re-assessment of the budget and incentives required for such a complex and important initiative, and the development of high-level political sponsorship for the project.
Secondly, several related areas will also need funding increases if this project is to be successful. There should be a prompt increase for the National E-Health Transition Authority (NEHTA) to ensure all required standards are available as soon as possible. This should be followed by increased funding for developments in the area of clinician decision support and training to overcome the gap in expertise and skills needed to implement HealthConnect nationally. Also there will need to be careful design of an appropriate benefits realisation strategy that rewards those who do the work.
Finally, it will be imperative to consult widely in the community on the issues of privacy, identification and smartcards and to be frank in educating the public on the benefits that will flow to stakeholders with implementation of HealthConnect.
The introduction, and use by all clinicians, of advanced electronic health records is an essential step in improving the safety, quality and efficiency of our health system. None of the HealthConnect trials to date have yielded substantive or quantifiable benefits, despite the spin placed on the trial outcomes. HealthConnect should be thoroughly and consultatively re-designed to meet the suggestions above and then properly resourced to achieve the required high levels of clinician adoption and use, and the subsequent national consumer and clinician benefit.