Chapter 3

Triumph and disaster: The tumultuous history of health IT

Ray Delany

Following World War II, technological change drove an explosion of innovation in the delivery of health care. This worldwide trend placed considerable pressure on the funding of the New Zealand public health system, and undermined the assumption underlying the still-new nationalisation of that system that there would be a finite amount of healthcare to be delivered. It remains a fundamental economic problem to this day that the delivery of healthcare has the potential to consume all of the resources of an economy if left unchecked, and must therefore be rationed in some way. As a consequence, healthcare is possibly the most highly-politicised aspect of the New Zealand economy. Health is the second largest area of government expenditure after social development; there is more legislative activity each year within the Ministry of Health than any other government department, and health policy is almost always a key focus of any election campaign. All of this makes Health one of the most challenging and fascinating areas of endeavour.

Since 1984, when David Lange’s Labour government initiated a sweeping series of changes to New Zealand public services, health has been caught up in a cycle of regular re-structuring of the entire system. In the mid-1980s, over 30 local hospital boards were replaced by 14 area health boards. Successive reviews, notably a report produced under the leadership of businessman Alan Gibbs, led to the most radical of these re-structures in the early 1990s. The then National government announced a radically different market-driven system. This was underpinned by the belief, reinforced by the world-wide popularity at the time of monetarist economic thinking, that market forces would be more effective than the traditional public-sector style of management, by switching the focus from long term public benefit to near term cost efficiency, and ultimately encouraging greater involvement in the delivery of healthcare by the private sector. The Gibbs Report, entitled Unshackling the Hospitals, focused on the cost of hospital service delivery, arguing that the structures in place at the time were overly bureaucratic, and inhibited the development of more cost-efficient ways of providing services. The report was criticised for superficial analysis, the details and methodologies of which were never made public. Consequently, the report was seen by clinicians in particular as ideologically driven, rather than an objective analysis.

Although the Gibbs taskforce was commissioned under a Labour government, there was consensus across political parties that the hospital board system was not working. This belief was strongly and very publicly reinforced by a scandal surrounding the treatment of early-stage cervical cancer at the National Women’s Hospital, which resulted in the Royal Commission of Inquiry, led by Judge Sylvia Cartwright, and the establishment of the Health & Disability Commission.* A year after the Gibbs report and the Cartwright Inquiry, Helen Clark — then Minister of Health — sacked the elected members of the Auckland Area Health Board for alleged fiscal mismanagement. The Cartwright Inquiry and surrounding publicity had a negative impact on the public perception of the medical community within hospitals, and helped to pave the way politically for radical reform of the system.

Market-based reforms were initiated in 1993 by the National government. The principle underlying these reforms was what became known as the ‘funder-provider split’. Previously, area health boards were funded for both the planning and funding of health services in their regions, and the delivery of those services. In 1993, the reforms established four Regional Health Authorities (RHAs) to control the planning and funding of services, and 23 state-owned enterprises named Crown Health Enterprises (CHEs), which were to act as providers of services. In principle, the RHAs would contract health services from CHEs, and would monitor the outputs of the work undertaken by the CHEs. It was believed that this model would engender competition between publicly-owned CHEs and private sector companies, with the expectation that the latter would be more efficient, and thus encourage the CHEs to improve their own efficiency or risk losing their contracts. The reality proved far different: for the majority of acute health services, the CHEs were the only organisations geared to provide them, and the barrier to entry into the new ‘market’ for private sector companies was high. All semblance of national planning and consistency disappeared, and the new system rapidly became bogged down in administrative details.

Another restructure was implemented following the 1996 general election, which merged the four RHAs into a single Health Funding Authority (HFA), while directing a focus on national consistency, public service and equity of patient services. The CHEs were renamed Hospital and Health Services (HHSs), but little else changed on the provider side of the system. The HFA had barely started to function effectively before the 1999 election, when Labour regained power under the leadership of Helen Clark. Clark’s government rapidly implemented a new strategy, which established 21 district health boards (DHBs)* using the corporate structure of the HHSs as the nucleus of the new organisations. DHBs would be governed by publicly-elected boards, and would be both funders and providers of services within their districts. The DHBs operated under a new national health strategy which for the first time encouraged a greater focus on primary care services, rather than hospitals. The residual functions of the HFA were merged into the Ministry of Health. The directive from the Minister of Health was that the DHBs were to be formed without any increase in administrative overhead, a position that was viewed sceptically by the Treasury in particular. In the period between the formal establishment of the DHBs in 2001 and 2009, administrative overheads ballooned. This was due to the inherent inefficiencies of 21 similar organisations addressing various issues after their own fashion, parochial attitudes, and a lack of national leadership. Following a change of government in 2008, a Ministerial Review Committee was set up to review the system yet again. This committee reported its findings and recommendations in August 2009, and in October 2009, the Minister of Health announced that there would be another restructure of the system, creating a National Health Board to oversee sector back-office functions, including information systems.

Health information

The complexity of health information systems reflects the environment in which they have been developed. In New Zealand, this inherent complexity has been exacerbated by multiple re-structures. It has frequently been the case that the projects generated by one re-structure have yet to be completed before the next re-structure is implemented.

Hospitals and health services are significant corporate entities. They employ large numbers of highly trained staff, and often administer large portfolios of land and other fixed assets. They must manage common logistical problems such as the procurement and supply of basic commodities (e.g. bed linen and food), as well as specialised areas such as pharmaceuticals, the sterilisation of surgical instruments and the disposal of hazardous waste. This means that they have at least the same information management complexity as any large commercial organisation, on top of which they must deal with the highly complex and sensitive personal information associated with patient care.

The 1988 Gibbs Report was scathing of what is referred to as the ‘troika’ of hospital management (Medical Superintendent, Director of Nursing and Hospital Administrator) and expressed astonishment over the lack of information, noting that expenditure on information systems was regarded as wasteful by the hospital boards.

In fact, in the 1980s, hospitals were reasonably well advanced in the adoption of computerised systems by the standards of the time. A significant project to bring all hospitals on to a single standard system had been implemented by the Muldoon government in the 1970s and early 1980s. As a result, all hospitals were connected to a centralised system for managing the basic administrative transactions associated with patients, known as Admissions, Discharges and Transfers. From as early as 1977, this system provided a National Master Patient Index — designed to ensure that all information regarding any individual was properly collated — and a Medical Warnings System incorporating adverse medical reactions and significant medical conditions, event summaries incorporating identification of the facility where the patient’s paper medical record was located, and donor information incorporating donor summaries and healthcare user contact details.

These systems were provided by Health Computing Services (HCS) a division of the Department of Health. In 1987 the government established a committee to review the financial viability of HCS and to recommend a future structure for the organisation. This was during a time when large segments of the New Zealand public sector were being privatised, so it was no surprise when the committee recommended that HCS be tendered for disposal to the private sector. In 1988, the organisation was sold to Paxus Information Services Limited,* a large provider of bureau services at the time.

By the early 1990s, hospitals were starting to implement more sophisticated computer systems for management of the high-cost areas of their enterprises. These started in the areas of greatest cost, where processes were already structured enough to benefit from computerisation. The three areas given the most focus were laboratory, pharmacy and radiology services. While these systems tended to operate well enough within their scope, and made an important contribution to departmental efficiency, problems started to emerge when these systems needed to share information across departments. Individual systems dealt with patient and clinician information in their individual ways, and there was a constant struggle to maintain consistency across the different systems. Some hospitals had also acquired products for integrating data from the different systems for clinical costing purposes, but these were cumbersome and expensive to operate, and provided outputs that were of little interest to clinicians.

The implementation of the corporate model in the early 1990s provided an incentive for Crown Health Enterprises to devise their own solutions to these common problems. It was understood that there was a risk of fragmented systems becoming the norm, so the Ministry of Health established the New Zealand Health Information Service in 1992 to act as the organisation responsible for national consistency of information and standards relating to information exchange and interoperability. NZHIS initiated a number of projects that were to have lasting impact on the health information landscape.

The national health index and national collections

One of the first projects undertaken by the newly-formed NZHIS was the establishment of the National Health Index (NHI) system. The NHI system was designed to replace the National Master Patient Index with a system that could interoperate according to a predefined set of standards with the new hospital information systems. The standards were represented in Application Programming Interfaces that vendors of hospital information systems could use to integrate into their systems. It was intended that the NHI would be the definitive source of patient demographic information throughout the country, and would be progressively linked to all types of patient related information so that these could be meaningfully integrated for informational purposes as required.

The NZHIS started out as a powerful entity with a strong mandate to enforce standards. There has always been a general consensus in the health informatics community that a universal healthcare identifier is an essential foundation for the effective integration of the various sources of patient information. By providing a common ‘key’ that can be used to link records in disparate systems, it was envisaged that the NHI would become the means by which all information relevant to patient care, and to the funding of that care, would be integrated.

From a technical perspective, the NHI was designed to the highest of standards. It remains a testament to the quality of this design that the system has continued to operate with very few modifications for over 17 years. The NHI was possibly the only system in New Zealand designed according to the Government Open Standards Interconnect Protocol (GOSIP) standard, and is certainly the only one still in use.

Underlying the NHI, a number of databases were constructed. With the implementation of the funder-provider split the RHAs needed a mechanism to make Crown Health Enterprises accountable for their contracts. Systems were set up to monitor hospital ‘outputs’ according to the World Health Organisation’s International Classification of Diseases coding system. Each CHE was expected to report their coded outputs at the end of each month, and in theory a straightforward commercial model would operate: the hospitals would be required to meet the outputs required of the contract or risk losing their funding.

The data derived from these systems were collected into databases operated by NZHIS, which became known as the national collections. These databases were intended to become a definitive source of nationally consistent and comparable data from all healthcare providers.

At the macro-economic level, the concepts and designs underlying these systems were forward-thinking and world-class, and firmly established New Zealand as a world leader in health informatics. However, while the national strategy driven by NZHIS was successful in many respects, it suffered from three systemic defects that eventually eroded its long term effectiveness.

The first of these was concern regarding patient privacy. New Zealanders had become distrustful of health authorities following the Cartwright Inquiry, and the idea of the government assembling large quantities of data and linking databases using the NHI to draw conclusions about health care use was not politically popular. This was at a time when there was a much higher level of general concern about privacy and the use of computers to store personal information than a generation accustomed to social networking systems would relate to. The 1993 Privacy Act imposed a health information privacy code that hampered the efforts of NZHIS et al to implement their vision, and there was much tension between NZHIS and the Privacy Commission (responsible for monitoring the Privacy Act and the associated codes) for many years. One consequence of this was that the public awareness of the NHI and associated systems that would have shone light on the privacy concerns never materialised, and the significance and potential benefit for patients and the general public of the NHI remains poorly understood outside of the health sector. As late as 2002, one expert observer from overseas commented that she

found the level of concern about protecting privacy extraordinary … If the popular view remains ‘Privacy at all costs’, then it must be recognised that one of those costs is ineffective and inefficient Public Health systems.*

A second factor was the disconnection between the hospital information managers and the NZHIS. From the outset CHEs were under-resourced, struggling to cope with massive budget deficits, and had neither the financial nor technical resources to implement their end of the systems to the extent the national strategies demanded. Moreover, the hospitals were preoccupied with understanding and managing what was going on within their own (very considerable) scope, and had little interest in the big picture, viewing the NZHIS and the remainder of the Ministry of Health as well as the funding authorities as remote and uncaring entities that didn’t appreciate the practical difficulties they faced. NZHIS had set up a system of annual updates to the national systems, governed by a series of committees populated by representatives from all HHSs and staff from the Ministry of Health and funding authorities. These committees would debate the details of the changes that were deemed necessary, and develop the specifications which would then be sent out to all hospitals to implement. This was viewed by the HHSs as unfunded mandatory compliance, and was thus implemented begrudgingly, if at all. In some cases, hospitals played the system by exploiting loopholes to maximise their funding, which they saw as legitimate, but was viewed with horror by NZHIS staff who perceived it as corrupting the long-term value of the national databases for short-term gain. These issues led to a lack of faith in the quality of data available from the national collections in general and the NHI register in particular.

The third factor was an underlying trend towards a more preventative style of public health management. It was becoming apparent that even with significant efficiency gains, hospitals were becoming increasingly incapable of dealing with the volume of work being generated by their district populations. While hospitals continued to be synonymous with healthcare in the public mind, the 1999 health strategy developed by the Labour government had a much greater emphasis on maintaining wellness rather than treating sickness, and primary care services started to receive a much greater focus. This shift was to have profound implications for the management of healthcare information.

Over time, as the patient systems in the hospitals and primary care organisations became more sophisticated, those responsible for the management of these systems at a local level became more influential. With the establishment of the district health boards, the DHB chief information officers (CIOs) began to insist on a greater say in health information systems policy, while the influence of NZHIS declined. Eventually, NZHIS was formally disbanded as a separate entity and its staff merged back into the Ministry of Health.

Hospital systems fragmentation

The national framework provided by NZHIS did not prove powerful enough to stop the hospital systems from fragmentation following the funder-provider split. The newly appointed management of the CHEs viewed information systems as a crucial part of their competitive advantage; collaboration between CHEs was actively discouraged in the period from 1993 to 1996, resulting in a rapid disintegration of any existing commonality of systems.

All of the CHEs embarked on ambitious programmes to gear up their infor­mation management capability. The approaches taken were almost as varied as the number of enterprises. There was some justification for this, in that CHEs varied greatly in size and complexity. Health districts tend to follow the boundaries of territorial local authorities. Despite the fact that all health enterprises operate according to the same principles in all areas, there was and remains a massive difference in complexity between major tertiary centres in Auckland or Christchurch and small rural entities such as those in Wanganui or Wairarapa. In addition, geographical distance is a much more significant factor in sparsely populated areas. The six largest CHEs adopted significantly varied approaches to the development of their information management capability. All were faced with massive challenges; according to one CIO, they had to start with the ‘cabling in the walls’.

Despite the individualised approach, the limited number of vendors willing and able to operate in New Zealand mitigated the fragmentation to some degree. Melbourne-based firm IBA Health succeeded in winning the largest number of CHE contracts, mainly for the smaller CHEs. Other contracts went to McDonnell Douglas Health, and the New Zealand-based Galen Group. Implementations frequently operated under the project management of one of the ‘big six’ accounting firms, which had large information technology implementation practices during that era. These organisations were incentivised to create long-term relationships with senior decision makers in the sector, and they hosted forums of various C-level executives, including CIOs, free of charge in exchange for the ability to listen in on the discussions. Beyond these forums, and the experience of the vendors, there was no mechanism to ensure any consistency between CHEs as to what was implemented, other than the mandatory connections to the NHI and national systems. CHEs were not required to develop Strategic Information Systems Plans, and few did so.

Hospital staff struggled to cope with the changes brought by the implementation of the systems. In many cases, significant process re-engineering was required to make effective use of the new systems, but efforts to engage clinicians in the decision making met with limited success. As a consequence, the systems were seen by clinicians as tools of the (unpopular) new management rather than supporting practical improvements to the delivery of patient care.

Against this background, public controversy erupted in the late 1990s around the implementation of systems provided by US-based Shared Medical Systems (SMS) at Wellington’s Capital Coast Health (CCH) and Health Waikato (HW) based in Hamilton, two of the large HHSs. The chief executive of CCH had previously operated as a consultant for SMS, and accusations of conflict of interest arose regarding the decision by CCH to purchase a suite of systems from SMS in 1997 (although an Audit Office investigation later showed no wrong-doing). Part of the support arrangement for these systems was conditional on SMS winning at least a second contract in New Zealand, and thus CCH was highly incentivised to ensure that this happened.

Around the same time, Health Waikato (HW), which had initially chosen to implement the Hospro-W suite of systems provided by Fujitsu, had become dissatisfied with the results, and went to the market for a replacement. Efforts were made to involve clinicians in the selection of the new system, and the system favoured by the selection process was provided by another US-based vendor, Cerner Corporation, which at that time did not have a presence in the New Zealand market. Despite this, in November 1998 the HW board announced its decision to purchase the SMS system in use at CCH. Clinicians at HW were furious at the decision, which they suspected was due to political interference, and the concerns escalated to the national media and to Parliament. Questions were asked in the House, and Opposition pressure put on the Minister of Health and various agencies to investigate. The HW project was subjected to an unprecedented level of scrutiny, and eventually, following the change of government in late 1999, the project was cancelled by the HW board in April 2000, provoking further controversy due to the costs of the project. Newly appointed Minister of Health, Annette King, who had been highly critical of the project in Opposition, requested that four members of the HW board resign due to their roles in the SMS purchase.

Meanwhile, things were not going well at CCH either. The SMS system proved to be less integrated than the purchasers had believed would be the case, and the effort required to implement in the New Zealand environment was proving too difficult. The support of clinical staff had not been achieved, there were many complaints about the new system, and SMS was not perceived as responsive enough to the various problems. Ironically, the SMS system had been promoted as a means to reduce costs and contain the organisation’s deficit.

Following the cancellation of the Waikato project, SMS announced it would no longer provide support for the New Zealand market and withdrew from the country, leaving Capital & Coast Health with an unsustainable support burden. In 2002, Capital & Coast DHB announced that it planned to abandon the SMS system.

The SMS fiasco, following as it did hard on the heels of another very public controversy over the INCIS Police System project, had the effect of convincing CEOs and politicians that information systems projects were high-cost, high-risk endeavours. Health Minister Annette King imposed a moratorium on health IT expenditure while a detailed strategy was prepared (eventually published as the Wave Report in 2001). This had the effect of slowing down further investment, as business cases of increasingly stringent requirements were demanded, and projects with ambitious scope were discouraged. A particular example of the latter was Auckland Healthcare Services (AHS), the largest HHS in the country. A project to select a new comprehensive integrated system from a single vendor, named RFP340 after the code in the organisations procurement system, was initiated in 2000. A new CIO had been appointed by AHS who had come from the primary-production sector where a world-wide integrated Enterprise Resource Planning (ERP) system had been implemented. It was not unusual for CIOs appointed from outside of the health sector to conclude that all of the problems of the organisation stemmed from the lack of integration, and that integrated ERP systems were the answer, as they had been proven successful in other sectors. In this respect, the AHS project was following a similar path to the failed projects at Capital Coast Health and Health Waikato. Consequently, the project was viewed with alarm by both peer organisations and central government. Through careful, low-key manipulation, the Minister of Health ensured that the project never got off the ground, to the disgust of several large vendors, who had spent six-figure sums in responding to the tender.

At the same time, a typical response from clinicians, frustrated by what they viewed as a lack of good quality information systems containing data relevant to their work, was to attempt to solve the problem for themselves. This led to the development of highly specific applications with very narrow scope. These were resourced through clinicians own efforts using end-user tools, by contracting resources using their departmental funds, external funding such as research grants or even personal funds. These systems sprang up all over the hospitals in particular, further increasing the complexity of the support problem for already over-burdened hospital IT departments.

Some of the HHSs, notably South Auckland Health (later Counties-Manukau DHB) had achieved considerable success through the using of integration engines to tie together information in disparate systems and present a browser-based ‘clinical workstation’ for use by clinicians. Efforts to interoperate with systems in primary care had also met with some success in pilot projects. These achievements were well regarded by both users and international reviewers, and served to maintain New Zealand’s reputation as a health IT innovator. However, adoption of these innovations was not widespread, and attempts to scale some aspects of the pilot projects to a nationally consistent level, notably the National Immunisation Register, were not considered particularly successful by clinicians. Individual DHBs continued to prefer their own choices over the adoption of others. One notable exception was Health Alliance, a shared services organisation established between Waitemata Health and South Auckland Health in 2001, formed, among other purposes, to leverage the benefits of the successful South Auckland Health IT projects into a peer organisation.

As part of the run-up to the establishment of the district health boards in January 2001, the Ministry of Health launched what became known as the Wave Project. Led by former Minister of Health, David Caygill, this project brought together representatives from government departments, hospitals and health services. For the first time, primary care organisations were invited to participate in the national debate on health information systems.

The 2001 Wave Report catalogued a litany of problems, including lack of integration, failed initiatives, decaying infrastructure and missed opportunities. The report was in many respects a regurgitation of problems identified in previous strategy documents. All of these reports found poor data quality, lack of standards, gaps in primary care data, interoperability problems, and weaknesses in the NHI and other national identification systems. The report also proved prophetic in its asser­tion that a failure to resolve these issues would result in the same strategy being re-written in five years’ time. Most of the Wave Report’s recommendations have yet to be fully implemented, some being severely curtailed due to lack of proper funding. It took until 2005 to craft and fund an integrated Health Infor­mation Strategy for New Zealand, and progress at the national level from 2005 to the general election in 2008 was limited.

In 2009, the Ministerial Review Group (MRG) noted that the ‘explosion’ of IT in the health sector had had both positive and negative impacts. The national systems development programme that was to have implemented the HIS-New Zealand was a year behind schedule, and strategic leadership of IT in the sector was low. The MRG recommended the transfer of responsibility for the management of health IT to the newly established National Heath Board.

Primary care

In New Zealand, primary health care is delivered almost entirely by what are effectively small businesses, collectively referred to as Primary Care Organisations or PCOs. The model is for each GP practice, retail pharmacy and non-governmental organisation (NGO) to be a lean organisation with low overheads, each organisation is able to operate with a no-nonsense pragmatic attitude while leaving considerable scope for entrepreneurial flair. Perhaps because of this, primary care has been in many ways more successful in the uptake of IT in the last two decades. In the 1980s, GPs rarely used computers in their clinics. Those that did tended to be technology enthusiasts rather than believers in any real clinical benefit of IT. With the establishment of the NHI and national systems in the early 1990s, the Ministry of Health had a desire to start capturing primary care information. This was sensible from a national planning perspective, since more people are seen in primary care every day than in hospital settings, but difficult to achieve due to the high volume of data in primary care that is distributed across many different organisations.

PCOs are contracted rather than employed directly; therefore, the approach used in the development of information systems has traditionally been indirect. The first of these was electronic subsidy claims in the 1990s. Over time, GPs were required to submit their claims for the government-subsidised component of their fees by electronic means, with electronic submission ultimately becoming the only path available. This required GP practices to acquire computer systems that they might not otherwise have needed, and initially met with fierce resistance, as the GPs felt that they were being expected to incur cost for the benefit of government efficiency. Over time, however, the presence of computers in general practices created a significant transformation. This was one area where market forces actually did create a desirable and arguably superior result. The market for practice management systems was initially cluttered, with over 20 different options. However, in the period from 2000 onwards, a clear leader emerged, MedTech Global, which came to dominate this market to the point where it was almost considered the de facto standard for a GP practice management system.

Primary Health Organisations (PHOs) were established under the Health and Disability Act 2000, with a mandate to administer a capitation-based funding system for primary care, which would replace the per-visit General Medical Subsidy. The PHOs had an incentive to further streamline the standardisation of systems within their catchments, and they accelerated the uptake of practice management systems and the trend towards a single dominant, if not mandatory, system. Gradually, the presence of computers in GP clinics became the norm, and the main repository for clinical notes, test results and prescription data. Prescriptions and lab test orders were being printed, itself a significant efficiency improvement for both doctor and patient, and lab test results were being sent via EDI directly to the ordering clinician. By the mid-2000s New Zealand had one of the largest uptakes anywhere in the world for use of computer systems for electronic medical records, as opposed to purely administrative tasks.

Attempts to make the NHI available for use in primary care were problematic. Policymakers in the Ministry of Health were enthusiastic about the potential of using the NHI as a key identifier in primary care in a similar fashion to the hospital systems, but they had little appreciation of the complexity of the task. Initially, the NHI was not made available to general practices, which were generally unable to afford the resources to connect in any case. The only way for most practices to obtain an NHI number was to contact a call centre at the claims agency, Health Benefits Ltd (later HealthPac), by phone or fax. There was no real incentive for general practices to get the correct NHI number attached to their record, since they had no use for it themselves. The rules for electronic claiming stipulated that a valid NHI number must be supplied, but there was no mechanism for ensuring that the NHI number supplied for any given claim was the correct one for the patient seen. Anecdotally, some practices simply supplied the same NHI number for every claim. As a consequence, the data collected via the subsidy claims system was effectively useless for any purpose other than the administration of claims payment.

Similar problems emerged with the systems for collecting lab tests and pharmaceutical prescriptions. Although the databases holding these data were huge in terms of transaction volumes, the utility of the data was low. Addressing these issues was a core focus of the recommendations of the 2009 MRG report.

All things considered, the change in primary care computerisation from 1999 to 2009 has probably proved more beneficial to patients and clinicians, and provided a greater contribution to the overall efficiency of the health system than any of the work undertaken in hospitals.

Interoperability challenges

No single aspect of the Health IT infrastructure highlights the legacy problems of interoperability more than the issue of secure network connectivity for health records.

In the 1990s, HealthLink, a company spun off from Telecom New Zealand’s EDI service, specialised in making sure that the electronic claims could be sent from any primary care organisation to the Ministry of Health’s payment systems. Although a messaging standard known as Health Level Seven (HL7) was mandated for these transactions, there was no consistent implementation of this standard, and so HealthLink became an information clearing house, able to receive messages from any organisation, in virtually any form, and forward them to their destination in the format that could be read by the receiving system. While this system functioned well within its limitations, new management at NZHIS in the late 1990s became uncomfortable with what was perceived as a key component of the infrastructure being held by a private business. There was also a view that the costs of the system were high, although this view was not shared by the general practices that used it, probably because of differences in pricing mechanisms for different organisations.

As a result, NZHIS launched a service which came to be known as the Health Intranet. This service purported to offer a level of security appropriate to the transmission of sensitive health data using managed network addresses and strong encryption. The uptake of the Health Intranet was not strong in its initial years. The system used a clumsy implementation of digital certificates, and was not suited for large HHS’s with their many thousands of individual computers, each of which had to have a certificate installed. The application and vetting processes were unnecessarily bureaucratic, the speed of the network was poor and the costs were high. While there was plenty of support in principle for a secure network, the reality was that few applications could make use of it. Some of the larger HHS’s (notably South Auckland Health) had initiated pilot schemes making hospital systems available to GPs, but for general practice the requirement was to have one integrated system, so using another organisation’s system, however, functional, was not seen as particularly useful, and again the experiences from pilot schemes did not scale up. There was much debate about the usefulness of the Health Intranet up until 2000.

Meanwhile, conflict erupted between HealthLink and NZHIS. HealthLink argued vociferously against the Health Intranet on the basis of cost, and enlisted the support of many of the GP users, who were also concerned that the Ministry of Health would impose another unfunded mandate upon them.

With the establishment of DHBs in 2001, it became possible for the Ministry of Health to enforce use of the Health Intranet on the DHBs. This was done by simply making the key contractual information DHBs required to operate available to them only by means of a Health Intranet application.

HealthLink continued to upgrade its services to enhance the levels of security being offered, and remains a key part of the health sector infrastructure.

Effect on IT industry

Over the past two decades, the health sector has provided significant opportunities for commercial IT companies to grow and develop by meeting the needs of this challenging sector. In the late 1980s, Paxus acquired substantial revenue through its acquisition of HCS. HealthLink has grown into an international company on the strength of its position in the New Zealand market, as has MedTech Global and a number of smaller firms. Other names, once prominent, have been absorbed into larger multinationals, including Paxus, Terranova, which supplied systems for Maternity care in hospitals, and Galen Group. Perhaps the most visible of the companies that are still New Zealand owned is Orion Health, which prospered initially through a close partnership with South Auckland Health, and a very down-to-earth, ‘get it done’ mentality. Orion has grown 20% annually for at least the last ten years, now employs several hundred staff and has a turnover in excess of $50m, most of which comes from software exports. There is scarcely a hospital in the country that does not use some part of the Orion stable of products.

These vendors and others participate in a small, but vibrant community of health IT professionals, characterised by a deep understanding of the unique nature of the environment, and a passionate desire to make a difference. A pivotal point of this community is the annual Health Informatics New Zealand (HINZ) conference. This event has gone from strength to strength since the establishment of HINZ in the late 1990s, and in 2009 the attendance and quality of material pre­sen­ted at the conference continued to grow strongly, despite the impact of a year of recession.

The future of health IT in New Zealand

Many of the core issues confronting the health sector in New Zealand are common to other developed economies. New Zealand is also not alone in having experienced health IT project failures, but fortunately has yet to experience problems on the scale of, for example, the UK NHS National Programme for IT. International benchmarks, such as the Commonwealth Fund, seem to indicate that New Zealand is relatively efficient on a cost of service provided basis.

It is a testament to the innovative capability of New Zealand society in general, and its ICT industry in particular, that despite decades of structural turmoil, the health sector has made significant progress in the development of its information systems. The use of these systems in primary care is among the most advanced in the world; most of the major hospital and health services have adequate electronic clinical systems, and the extent of interoperability between organisations is increasing; the National Health Index, for all its faults, has provided a robust foundation on which progress towards a truly national health information system can yet be built.

On the whole, clinicians, once regarded as IT sceptics, have now accepted that the future of the healthcare system will be heavily influenced, if not dominated by the extent to which information systems are successfully developed for use in clinical settings.

However, it is difficult to take a high-level view of the last 25 years of health IT in New Zealand without feeling a little pessimistic about the future. There have been many attempts by capable people to address the fundamental issues, and yet reports are regularly produced indicating how little has changed.

The latest structure being proposed does have considerable merits, but it remains to be seen whether those merits are enough to overcome the inertia of a sys­tem of massive complexity. The rate of progress against the much-canvassed issues has been glacial over the last five years, and yet there are significant opportunities to be grasped in the short term, if the vision and the will exists to do so.

Perhaps the most encouraging trend is social networking. There is quite a different public view now emerging about privacy in the context of the use of technology to maintain social links. The logical extension of this development is the use of these techniques for more practical purposes than simply remaining socially connected. There is considerable potential to engage youth in particular, and there have been some recent successes in this area, notably in the provision of mental health services.

Another growing trend likely to impact on health is software as a service (SaaS). Increasingly, all kinds of information systems that would once have been provided only via traditional desktop computing are being provided over the Internet. There is considerable potential for a ‘killer’ application in primary care in particular. Such an approach has the potential to simplify many of the interoperability problems that dog the sector. Again, some successful early work here shows promise.

Whatever is decided strategically, the details of the implementation of the strategy will be where the ‘devil’ will be encountered. This chapter provides a lesson for anyone contemplating the latest changes as to what has worked well for the nation in the past, and what has not. Projects run exclusively by government have not generally been successful. Nor is there a successful track record of adopting monolithic approaches imported from overseas. Only where there has been robust involvement of the innovation engine that is the New Zealand ICT sector, combined with an appropriate level of funding, and a governance mechanism designed to complete projects successfully has triumph been achieved. The good news is that our history also shows that when Kiwis get it right, the success can be phenomenal.

It is almost cliché to say that those that fail to understand the lessons of history are doomed to repeat it. The history of health IT is a lesson those focusing on the future of the health system would be unwise to ignore.

Glossary

AHS Auckland Healthcare Services

CCH Capital Coast Health

CHE Crown Health Enterprise

CIO Chief Information Officer

DHB District Health Board

ERP Enterprise Resource Planning

GOSIP Government Open Standards Interconnect Protocol

GP General Practitioner or General Practice

HCS Health Computing Services

HFA Health Funding Authority

HHS Hospital and Health Service

HINZ Health Informatics New Zealand

HW Health Waikato

MRG Ministerial Review Group

NGO Non-governmental Organisation

NHI National Health Index

NZHIS New Zealand Health Information Service

PCO Primary Care Organisation

PHO Primary Healthcare Organisation

RHA Regional Health Authority

SMS Shared Medical Systems Ltd

Ray Delany will celebrate 30 years of working in the computer industry in July 2010. Ray started his career in Ireland in 1980 as a programmer, working on Digital RSTS/E developments. He moved to New Zealand in 1984, and worked for several major New Zealand companies including Feltex, Moore Business Systems and Fisher & Paykel, in a variety of project and operational management roles. Ray’s passion for health systems started in 1998, when he joined Waitemata Health as Chief Information Officer, and was part of the senior management team during its transition to a district health board. He was subsequently appointed general manager of the New Zealand Health Information Service, where he was responsible for a number of nationally significant projects following the publication of the Wave report in 2001. As a consultant, he has assisted a number of organisations with information systems strategy, including district health boards and residential care organisations. He has developed innovative information strategies for organisations who wish to be thought leaders in this field, and has been actively involved in developing the implementation processes to shape strategies into reality. Ray has conducted research into models for ensuring successful IT implementations in healthcare, and has published on this and associated subjects. A former member of the editorial board of the Healthcare and Informatics Review Online Journal, he is the establishment Director of the National Institute for Health Innovation at the University of Auckland. Ray’s qualifications include Master of Management and Diploma in Business and he is a fellow of the New Zealand Computer Society. Ray resides in Auckland, where he is currently chief executive and part owner of Designer Technology Limited.

References

Auditor General (2006). Progress with priorities for health information management and information technology. Retrieved 5 December 2009, from www.oag.govt.nz/2006/wave/docs/wave.pdf

Bell, S (2002, 8 April). Health board dumps SMS for Citrix. Computerworld

Bland, V (2000, June). Is the health intranet just what the doctor ordered? MIS New Zealand

Cartwright, S (1998). The report of the committee of inquiry into allegations concerning the treatment of cervical cancer at National Women’s Hospital and into other related matters. Retrieved 1 December 2009, from www.nsu.govt.nz

Chamberlain, J (2002). Breastscreen Aotearoa — An independent review. Retrieved 3 September 2004, from www.moh.govt.nz/chamberlainreview

Delany, RP (2004). Evolution, not revolution: Measurement and management of health outcomes in New Zealand through efficient use of national information systems. Health Information Management, 32(3&4), 118–125

Delany, RP (2006). Fourteen years young: A review of the national health index in New Zealand. Retrieved 5 December 2009, from www.hinz.org.nz/journal/2006/06/Fourteen-Years-Young-A-Review-of-the-National-Health-Index-in-New-Zealand/936

Didham, R, Martin, I, Wood, R & Harrison, K (2004). Information technology systems in general practice medicine in New Zealand. NZ Med J

Easton, B (2002). The New Zealand health reforms of the 1990s in context. Retrieved 1 December 2009, from www.eastonbh.ac.nz

Gauld, R & Goldfinch, S (2006). Dangerous enthusiasms: E-government, computer failure and information system development. Wellington: Astra Print

Gibbs, A (1988). Unshackling the hospitals: Report of the Hospital and Related Services Taskforce. Retrieved 1 December 2009, from www.moh.govt.nz

Health Information Services (1991). National master patient index review: Archives New Zealand

Health Information Strategy Steering Committee (2005). Health information strategy for New Zealand, Wellington

Issac, P (2002, 1 July). Advise and consent; Surf’s up at the Ministry of Health. CIO New Zealand

Jackson, R (1997, 28 April). Capital coast CHE’s new chief targets $50m overspending. Computerworld

Law, G (1999a, 9 May). Bunkle demands probe into SMS buy. Computerworld

Law, G (1999b, 17 July). Hospital CEO ‘fixes’ SMS legal concerns. Computerworld

Ministry of Health (2001). The WAVE advisory board to the Director-General of Health. From strategy to reality: The WAVE project

Poskitt, N & Taft, S (2009). National immunisation register inaccuracies and duplications. NZMJ, 122(1288)

Shipley, J (1996). Health information strategy for the year 2000. Wellington

Treasury (1999). New Zealand government asset sales as at 30 September 1999 — Completed sales. Retrieved 5 December 2009, from www.treasury.govt.nz/government/assets/saleshistory/notes/#health

Williamson, M (1991). Health information strategy for New Zealand: A joint venture between the areas health boards and the Department of Health. Wellington