Smiling All The Way

• Abstract
• Introduction
• The Hutt Valley District Health Board
• The Information Systems Function
• The SMILE Project Initiation
• First Implementations
• What Has Been Achieved
• Factors in Achieving Success
  • CEO Risk Aversion
  • CIO Leadership
  • Executive Support
  • Strong IS Governance
  • Clinical Involvement
  • Incremental Change
• Conclusions
• References

Abstract
This study provides an example of a successful information systems (IS) project in a clinical care setting in an exemplary New Zealand public hospital. Hutt Hospital, located in the lower North Island of New Zealand, has achieved good results with the implementation of an electronic patient record in terms of cost-efficiency, system quality and user acceptance. A number of factors have been identified as contributing to the success including strong IS governance, capable and experienced IS staff teamed with clinicians, a commitment to customer service expressed in clinical user involvement and a process of continuous incremental change. The subjective assessment of the success of the project is validated by research and this case shows that the lessons of New Zealand and overseas health organisations can be applied to clinical IS projects.

Introduction
In 2001, the Hutt Valley District Health Board (HVDHB), New Zealand, embarked on a programme of work to develop electronic patient record systems for its provider arm ^a. The progress made by HVDHB to date has been remarkable. Within two years, the organisation has moved from a situation where virtually no clinically relevant information was available to clinicians at the point of care to a high quality electronic patient record (EPR) which is well regarded and accepted by the majority of clinicians. This study documents the progress made by HVDHB and examines some of the reasons why the system implementation has been so successful.

The Hutt Valley District Health Board
The Hutt Valley District Health Board (HVDHB) serves a population of approximately 132,000 people within the region governed by the territorial local authorities immediately north east of Wellington. The major hospital centre it operates is Hutt Hospital, which opened in May 1944. Hutt Hospital has a capacity of 250 beds and provides general medical, surgical and emergency services to the population in the immediate catchment area, and specialist tertiary services for Rheumatology, Burns Plastic and Maxillo-Facial surgery for a wider regional area. The hospital handles around 13,500 inpatient and day patient admissions annually, representing a busy workload for its 900 clinical and administrative staff. The population demographics are similar to the New Zealand population as a whole, with a slightly higher than average Maori and Pacific Island population. The hospital staff is a close-knit community of health professionals who work in an atmosphere relatively free of the organisational politics common in larger organisations. In all, Hutt is an exemplar of a modern New Zealand public hospital.

The Information Systems Function
It is no coincidence that at Hutt Valley, IS stands for Information Services. The current CIO - IS Manager Tony Cooke - has a strong customer service ethic shaped in a commercial environment. Cooke, who has been in his current role for approximately four years, believes that his department should behave toward its internal customers exactly as if there were a commercial relationship, and works to ensure that everyone in his department adheres to that service ethic.

Cooke’s career path is a common enough one in the IS industry. He started out as a programmer in the 1980s, and moved up through systems analysis and project management to his current role. He has worked in health for over 10 years, and has been closely involved in some major health IS projects in that time. Cooke, thus, has a strong IS background combined with an excellent understanding of how a health care enterprise operates. He is, like many who have worked in health for a long time, fascinated by the challenges of the sector, and committed to the success of his organisation in terms of improved patient outcomes.

The Hutt IS department comprises 20 people, including the IS Manager. Six staff members are involved in the EPR programme. Support is provided to a network of approximately 750 client computers and 30 servers. The IS department has been successful in its efforts towards establishing and maintaining a stable, reliable technology infrastructure.

All IS departmental staff excepting two are employed by a local technology company, which provides their services on a full-time outsourced basis. The IS manager and an administrator are the only full time employees of the HVDHB working in the IS department.

The department is governed by the Information Systems Steering Committee (ISSC) made up of members of the hospital’s senior management team including the Chief Executive Officer (CEO). The IS Manager reports to the Chief Financial Officer (CFO) and describes the ISSC as the "executive of Information Systems". Cooke has a good working relationship with the CEO and with other members of the senior management team.

The SMILE Project Initiation
The SMILE project was initiated by the IS Manager in 2001. SMILE stands for System for More Information and Less Effort. The need for clinical information systems was outlined in the organisation’s Information Systems Strategic Plan (ISSP). It is uncertain whether there was much involvement by clinicians in the development of the ISSP. When he started work for Hutt, Cooke was given instructions from his executive management not to take unnecessary risks; the CEO warned him, "IT has been the graveyard of many a Chief Executive".

Cooke initiated the project in February 2001 by convening a group of interested parties and giving "a talk" on the subject of "Bridging the Gap Between People and Technology". The discussion, led by the CIO, focussed on comparing and contrasting the benefits of paper records and electronic records and was deliberately non-technical; there no were assumptions about any knowledge in the audience. The IS Manager was keen to show the group the limitations of the paper record and to describe how computer systems could overcome some of these. He was committed to closely involving a clinical user group in the developments. He was prepared to shelve the project if he was unable to get clinicians’ involvement in and commitment to the development of an integrated clinical system covering the whole organisation.

The approach worked. The IS Manager obtained a strong endorsement from the clinicians present at the discussion, and in particular obtained the support of the Director of Medicine, the Chief Medical Officer (CMO) of the hospital. The CMO "knew nothing" about computers, but could see that electronic systems would be necessary in the future to ensure accurate and effective communications between clinicians across the continuum of care. According to Cooke, the CMO "became a convert".

A Clinical User Group (CUG) was set up, chaired by the CMO. This group would meet once every two weeks for no more than one hour, outside clinical time. The CUG obtained one of the best turnouts of clinicians for any committee in the hospital, and the CMO worked hard to ensure that the meetings were useful and productive for clinicians. The CUG meetings occurred without fail every two weeks from the start of the project and were regarded to be crucial to the project’s success.

A key member of the CUG was Stephen Purchas, a general surgeon employed full-time by Hutt Hospital. Purchas is relatively knowledgeable about and interested in the use of computers in clinical care, and had been actively pushing for greater IS support for clinical care for some time.

The first order of business for the CUG was to assess what had been achieved in other New Zealand hospitals in the development of electronic medical records. The CUG sent a team consisting only of clinicians to three other hospitals, which were at that time considered to be the leading hospitals sites in New Zealand in their adoption of electronic systems. All three had taken different approaches and had developed different systems. This served to raise the understanding of the Hutt clinicians that there were many different ways in which these systems could be configured, and that there would be a considerable reliance on technical expertise to achieve a good result. Purchas indicates that this knowledge was foreign to clinicians "from a clinician’s point of view, [it was] quite hard to understand".

Following the exploratory phase, a detailed requirements specification was prepared, and a Request for Proposal (RFP) was issued to the vendors of the systems examined in the exploratory phase. These processes were provided by the CIO, but the actual execution of the processes was controlled by the CUG. (It had been determined that the CUG would have control over the recommendation that would go forward for executive management approval and funding.) Cooke describes this as a "bottom-up process". Executive management maintained awareness of progress via the ISSC, but deliberately chose not to influence the CUG in the decision-making process. This was to ensure that there would be strong ownership of the final decision within the hospital’s clinical community. There was also a commitment from management and from IS to support the decision that the CUG made. There was a degree of risk to this approach, says Cooke. "What if they had come up with the ’wrong’ recommendation?"

The risk paid off. The system that was chosen by the CUG was acceptable to both IS (from the perspective of technological "fit" with the existing infrastructure) and to the executive management of the organisation. The executive also found it easy to accept the recommendation as they had confidence in the CUG and in its processes of decision-making. The Board approved the recommendation in October 2001. The funding required was quite modest for projects of this nature, nevertheless, this was a key test for the leadership of the IS Manager - gaining Board acceptance of the proposal and thus obtaining the funding for the project was a validation of the credibility he had already established with the CUG. Clinicians perceived him as influential and capable of supporting their needs with funding and with actions.

By November 2001, less than a year after the project had been initiated, the system implementation programme commenced.

First Implementations
The CUG provided a forum wherein the work of prioritisation would be done. The first system to be implemented was a digital dictation system. This was seen as an important part of the infrastructure for the overall project, but also was a relatively simple system whose use would encourage clinicians to become comfortable with the use of computers, using their logins, etc. Then came the "portal framework", which initially provided laboratory and radiology results reports. This was quickly followed by a ward information system for use primarily by nursing staff.

Each module of the system was treated as a project in its own right within the overall programme, carefully executed according to the quality standards set down by the CUG, and reviewed post-implementation. The CUG was consulted regularly to determine priorities and to resolve matters relating to clinical usability in the design. The members of the CUG actively promulgated the project through their own intra-discipline meetings and supported the implementers at the grassroots level.

Programme management was handled by a project director, who also took responsibility for managing certain individual projects. Another project leader with a nursing background handled the remainder of the projects. The project director reported to the CIO, and decision-making within this team was consensual, so that there was strong buy-in to decisions at all levels. The IS Manager ensured that projects maintained their relevance and was not afraid to shelve or scrap something if it became apparent, once the detail had been worked out, that it didn’t contribute particularly well to realising the vision.

The relationship between the IS team, the CUG and the ISSC was healthy and co-operative. There was a high level of mutual respect and trust between these different groups.

What Has Been Achieved
The Hutt Hospital EPR consists of an integrated clinical workstation, known as the "clinical portal". This uses a web-browser user interface, with a single login identifier. Clinicians logging on to the system are presented with a customised display listing their current inpatient list, their outpatients and any patients they have selected to monitor regardless of their status within the hospital. The portal provides access to all systems of interest to clinicians and is designed to maximise the available information at a glance. The system integrates the Hospital Information System, laboratory and radiology results and electronic mail. Plans for clinical document management are well advanced and electronic prescribing is also planned for the near future. Clinicians can access details of all inpatient, outpatient and emergency department visits, community health contacts and mental health contacts. Discharge summaries can be created and electronically dispatched to the relevant general practice, either via HealthLink (an EDI network) or by fax. Cooke estimates that eventually, 90% of the information that clinicians need in a clinical care setting will be obtained from the system without looking at the paper record. There is no intention to eliminate the paper record at this stage but it is already becoming less of a clinical management tool and more of a pure record-keeping instrument.

Factors in Achieving Success
This case has illustrated several critical factors in IS adoption that have also been identified in the Information Systems literature generally. These are:

• CEO risk aversion
• CIO leadership
• Executive support
• Strong IS governance
• Clinical involvement
• Incremental change.

These points are discussed in detail below.

CEO Risk Aversion
In general, CEOs participate infrequently in IS-related activities, despite such participation being a known factor in IS success.¹ The Hutt Hospital CEO was acutely aware of the risks to the organisation that would result from a failed information system.

The fact that the CIO did not report directly to the CEO indicates that the IS area may not have been one that the CEO felt comfortable managing directly, or at the least it was not perceived as a priority area for CEO focus. Nevertheless, the CEO did maintain adequate involvement in the project. This was found by Javenpaa & Ives to be the determining factor. ¹  Participation of the CEO helps, but does not create significantly more of a positive effect than CEO involvement. Similarly, the relationship of the CEO to the CIO is important. In the health care sector, direct reporting of the CIO to the CEO is less important than a good working relationship between them. Over 70% of CIOs are one or more levels removed from the CEO in the hierarchy of their organisations, but CIOs who have a close relationship with their CEO reported a better ability to influence peers and be more effective. ² This effect is observable at HVDHB.

The involvement of the CEO was perhaps best expressed through the ISSC. It has been observed that "far too many companies still delegate essentially all IT governance responsibilities to the CIO". ³ This mistake was not made at Hutt, despite the fact that the CIO was trusted and was given considerable autonomy. The ISSC maintained strong oversight of all projects and was appropriately composed of a combination of IS management, executive management and clinicians. The CEO’s aversion to risk in the IS environment did not result in paralysis, but instead encouraged all participants to approach the challenge in a pragmatic and methodical fashion.

CIO Leadership
At Hutt, the knowledge and leadership of the CIO was a major factor in the success of the project. The IS Manager was not afraid to stand in front of clinicians in his organisation and tell them what he thought about the difficulties with paper records. All too often, IS personnel take a completely one-dimensional view of the world, perceiving everything in terms of technology. ^4,5 Cooke clearly has a more balanced worldview than this; he says:

humans have their limitations. We work and operate primarily in a social environment not a technical one. We need to motivate, lead and communicate.

Cooke also understands how people who do not have a great deal of skill or experience with the use of computers can feel when approaching technology:

If you are an IT person, the system will do everything, if you are a user then IT is an arcane world with magical and unpredictable properties.

Despite his willingness to lead, Cooke recognised the danger of taking on the total responsibility without first ensuring that the organisation was firmly supporting him. He ensured that the pre-existing IS governance represented by the ISSC was supplemented by the CUG and invested considerable effort in ensuring that the processes used by the CUG were appropriate to the task. There was some element of luck here as well - for example, the enthusiasm and leadership capability of the Director of Medicine was probably fortuitous - but Cooke appears to have a genuine commitment to empowering his user community. He is sceptical about what he calls the "token clinician" approach to user involvement, where:

. . . you may know a friendly doctor who is interested in IT, ask his opinion and if they support you, you can say that you’ve consulted with users - despite the fact that you have 1,700 employees in your organisation.

The Hutt Valley IS Manager is a good example of the well rounded, health care IS executive as examined by Smaltz. ² Without this kind of capability in the organisation, it is unlikely that Hutt would have made as much progress as it has.

Executive Support
It has long been observed that effective performance of the IS function is dependent on mechanisms outside the IS structure, such as steering committees. ⁶ Similarly, the notion that executive support is required to achieve successful IS outcomes in any organisation is well-established.¹ However, the extent to which these concepts have been advanced at Hutt is quite out of the ordinary. Executives empowered the CUG to make the key decisions and made compromises in productivity by allowing clinical time to be used for training and other purposes related to the project. Most notably, executives deliberately excluded themselves from the decision-making processes at a number of key stages in the process, apparently to avoid compromising the outcome. This is rarely seen. In fact it is more common for hard-headed business executives to be sceptical of the benefits of IS projects and want to control decisions, or even retrospectively justify to their organisations the decisions that they have already made.^7,8 At Hutt the executive involvement was formally established via the ISSC, and this forum did not interfere with the clinically driven processes of the CUG. The risks of this approach were acknowledged but have paid off handsomely. Not only have good results been achieved, but there is also a level of collegiality between the management groups and the clinical groups that is rare in health care. This probably originated in the strong sense of community prevalent in the hospital, but it is notable that the IS initiatives dovetailed into this culture, rather than disrupting it.

Strong IS Governance
Functional IS governance is regarded as essential in health care organisations. Organisations that integrate IS into the overall enterprise governance framework are believed to have the capability to guide, direct and support IS development more effectively.³ The IS Manager describes the HVDHB ISSC as having "helped enormously". The ISSC meets once a month and approves new projects, reviews project progress and examines post-implementation reviews. There is clearly a functional relationship between the IS group and the steering committee. The ISSC also supports the CUG in its processes of decision making, providing a channel by which the clinical groups can access the decision making power of the Board, while maintaining appropriate control over expenditure.

Clinical User Involvement
It has been observed that hospital management and clinical groups often have difficulty communicating effectively. Researchers have found that the dissimilar life experiences and beliefs of these different groups create fundamental barriers to communication and trust. This can result in flawed management assumptions on the one hand, and a belief by clinicians that there are hidden agendas that may jeopardise their professional modus operandi on the other.⁸ Executive management has a penchant for using its organisational power to force change, and often IS is the medium. Even when this is deemed successful by management, clinicians are not usually as sanguine about these initiatives. Steven Purchas, now a clinical advisor on IS matters at Hutt Hospital says:

. . . hospitals have always run pretty much on [management systems] because of regulations that [mandated] data which they’ve had to produce for the government. That’s really what the systems have been about, rather than what clinicians have needed.

There are risks associated with the use of formal project methodologies to develop clinical information systems. There is a natural tendency of project members who have been sequestered from their normal working environment to identify more with the project than with their erstwhile colleagues. This can cause people to alter their behaviour and orientation towards the clinical environment once they have been removed from it. The result can be that the project team, even one composed of clinically experienced people, is seen by practising clinicians as a tool of management.

At Hutt, the establishment and operation of the CUG avoided this problem. Clinicians apparently perceived the CUG as part of a series of normal hospital activities, albeit an interesting and different one, and not something imposed by management. IS staff were expected to "front up" to clinicians and explain themselves when things went wrong. Equally, the members of the CUG were expected to actively participate when it came to evaluating proposals, understanding new concepts and resolving issues. The leadership and professional seniority of the Director of Medicine, as chair of the CUG undoubtedly assisted in achieving this level of participation.

The combination of a CUG prepared to engage constructively with the IS group and an IS group prepared to listen to the CUG has probably been the most significant single factor in the success of the SMILE project.

Incremental Change
Studies have found that prior experience with computers using comparatively low-technology systems such as email and other standard systems is a predictor of acceptance of EPR by clinical staff.⁹ The early deployment of the digital dictation system at Hutt was intended to accustom clinical users to the use of computers in a general sense. Subsequent implementations evolved the overall system in a careful, methodical fashion. In Cooke’s words:

Incremental change is everything. Once you’ve put a system in, allow time for it to settle down and then move on.

The overall magnitude of the considerable change achieved in two years has not been immediately apparent to those participating in the change. Thus, the change has been widely accepted as normal, and rapidly absorbed.

Conclusions
This study provides an example of a successful IS project in a clinical care setting in an exemplary New Zealand Public Hospital. Good results have been achieved in terms of cost-efficiency, system quality and user acceptance.

The SMILE project was initiated by capable and experienced IS personnel who utilised their knowledge gained through previous experience to prevent some of the more common causes of failure. The success factors included:

• Strong IS governance underpinned by commitment of senior executives, including the CEO
• Capable and experienced IS staff teamed with clinicians willing to accept the need for professional skills that they did not themselves possess
• Commitment to customer service expressed in clinical user involvement through the CUG
• Incremental change starting with precursor systems to gain the confidence of clinicians.

The Hutt Hospital case deserves recognition as a good example of what can be achieved when the focus of IS is on people rather than technology. The subjective assessment of the success of this project is validated by research and this case shows that the lessons of New Zealand and overseas health organisations can be applied to clinical IS projects.

References

1. Javenpaa S L Ives B. Executive involvement and participation in the management of information technology. MIS Quarterly 1991; 15(1): 205-227.
2. Smaltz D H. The elevation of CIO roles: organizational barriers and organisational enablers. J Healthc Inf Manag Spring 2000; 14(1): 81-91.
3. Clark F C, Kimmerly W C. Strong IT governance: don’t even think about not doing it! Proceedings of Healthcare Information and Management Systems Society Conference, Session 28, Atlanta GA: January 2002: 27-31.
4. Covvey H D, MacNeill J E. Checklist for success: essential competencies for IS leadership. Proceedings of Healthcare Information and Management Systems Society Conference, Session 148, Atlanta: January 2002: 27-31.
5. Glen P. Leading geeks: how to manage and lead people who deliver technology. San Francisco: Jossey-Bass; 2003.
6. Brown C V, Magill S L. Alignment of the IS functions with the enterprise: toward a model of antecedents. MIS Quarterly 1994;18(4): 371-404.
7. Delany R P. Lest the geeks inherit the earth: managing organisational knowledge to achieve successful outcomes in information systems. Unpublished literature review. New Zealand: University of Auckland; 2003.  Available at http://www.webconsulting.co.nz/papers.htm 
8. Myers M D, Young L W. Hidden agendas, power and managerial assumptions in information systems development: an ethnographic study. Information Technology and People 1997; 10(3): 224-240.
9. Gamm L D, Barsukiewicz C K, Dansky K H, Vasey J J. Investigating changes in end-user satisfaction with installation of an electronic medical record in ambulatory care settings. J Healthc Inf Manag Winter 1998; 12(4): 53-65.

Footnotes:

(a). Under the New Zealand Health & Disability Act 2000, 21 District Health Boards (DHBs) were created throughout the country. Each DHB is responsible for both the funding and provision of services within a defined geographical area. Typically, public hospitals form a substantial portion of the provider operations of a DHB.