Advancing Mental Health Pathways

• Abstract
• Introduction
• Failure in Care Delivery
• Mental Health Perspective
• Mental Health Services in the Auckland Region
• Knowledge Management is the Solution
• Project Initiation
• System Objectives
• Limitations
• Knowledge Processes
• Information System Methodology
• Conclusion
• References

Abstract
In the global health sector there are widespread, serious shortfalls in the procedures for accessing key clinical information which contribute to very high numbers of adverse health events.

In New Zealand, the Auckland region which is serviced by three District Health Boards is not immune to this phenomenon. Incomplete clinical information exchange processes and fragmented policies and information systems have been highlighted as risks to care continuity in a national Mental Health Commission Review.^{[ 1 ]} Partially in response to the review, and partially because it was technologically feasible, a system was implemented in March 2005 by a small project team made up of the three DHBs. This system allows appropriate providers of mental health services within the Auckland region to identify whether a patient is under the care of another mental health service in the region. It utilises an existing technology platform to share high-level, non-clinical regional patient details. The system facilitates timely and co-ordinated care with access to explicit and tacit clinical information through existing communication channels (primarily fax and phone) at the point of decision-making.

Introduction
There has been growing pressure from consumers, providers and funders of health care and governments throughout the world to minimise adverse health events.^{[ 2 ]}

Having key information readily accessible at the point of decision-making can maximise correct care decisions, decrease errors and omissions and improve outcomes.^{[ 3 ]} Recognising and adopting Orr’s^{[ 4 ]} F.I.R.S.T (fast, intuitive, robust, stable and trustworthy) mnemonic promotes the acceptance and use of such an innovation.

Failure in Care Delivery
There is increasing international evidence of adverse health events which are caused by a failure to deliver services when essential information is unavailable, as summarised below. The World Health Organisation’s World Report on Knowledge for Better Health^{[ 5 ]} acknowledges advances in health through science but says that “Disparities and inequities in health remain major development challenges in the new millennium and malfunctioning health systems are at the heart of the problem. Half of the world’s deaths could be prevented with simple and cost-effective interventions. But not enough is known about how to make these more widely available to the people who need them." It recommends putting more weight on" translating knowledge into action to improve public health by bridging the gap between what is known and what is actually being done."

Chu^{[ 2 ]} argues that "The health care industry has been under increasing pressures to reduce health care practice variability and improve quality of services" and the escalation of such pressure comes from two reports by the American Institute of Medicine:

1. "To Err is Human: Building a Safer Health System"^{[ 6 ]} recommends better access to information and illustrates the alarming statistics of preventable adverse health events in the US, where these are a "leading cause of death". Two studies imply "that at least 44,000 and perhaps as many as 98,000 Americans die in hospitals each year as a result of medical errors."
2. "Crossing the Quality Chasm"^{[ 3 ]} reveals care deficiencies arguing "the care delivered is not, essentially, the care we should receive" due in part to the best scientific knowledge not being accessed.

Davis et al^{[ 7 ]} verified that New Zealand is in a similar position.

These are examples of evidence that there are deficits in clinical processes that result in some level of failure to deliver care due to a lack of access to essential information.

Mental Health Perspective
Acute mental health patients are at risk of a breakdown in their care pathway with consequent adverse outcomes for their health when crucial information cannot be found at the time clinical decisions are made.

In Australia, de-institutionalisation, mainstreaming of mental health services and measures to reduce mental health stigma resulted in hospital emergency departments experiencing a dramatic increase in patients presenting with "primarily mental health problems".^{[ 8 ]} The emergency department environment in Australia is "directed and guided by specific formal practices and protocols … that do not lend themselves to the complexities of psychiatric illness".^{[ 9 ]} There is evidence that the majority of acute mental health presentations in England are at hospital emergency departments.^{[ 10 ]} Given the similarities in these health care systems to New Zealand’s , there is little reason to suspect that the situation in New Zealand would be different to those countries.

Croskerry^{[ 11 ]} considers that errors in emergency departments are often a result of "Cognitive Dispositions to Respond (CDRs)". Susceptibility of mental health patients to "psych-out error" (medical conditions are overlooked because the patient has a mental health label) is one such CDR.

In 2001, the New Zealand Mental Health Commission conducted a national review of privacy and information-exchange practices and policies and subsequently released its findings in a national statement.^{[ 1 ]} The Commission noted that the disparate systems of clinical care for the mentally ill underpinned the inability to access key information across services. These deficiencies led to the conclusion that policies should "ensure the relevant information is given to the right person at the right time." This was, and still is, pertinent to the Auckland region and its three District Health Boards (DHBs).^{[ a ]} Each has different management systems, in a geographically close-knit area which experiences patient inter-district flow. Medical errors can have an adverse impact on health outcomes^{[ 6]} and emergency departments are the point of key clinical decision-making. There is a risk of a variation to established pathways of care and a mental health patient, thus, becoming exposed to interventions that have adverse results (eg, drug reaction). King^[12,13] presents broad evidence for patients under mental health care having a higher suicide risk than patients without a recent mental health contact, and a recognition of the importance of identifying "medical and social risk factors in recent patients and effective monitoring of treatment."

Mental Health Services in the Auckland Region
Figure 1: Auckland Region

Auckland is New Zealand’s largest city with approximately 1.3 million people, nearly a third of the nation’s population, across 519,032 hectares (1,282,580 acres). The Auckland region is covered by three separate DHBs, (Auckland, Waitemata and Counties Manukau), each with a range of services, multiple sites and multiple teams.

Mental Health Services in the region covers a broad spectrum of service delivery. Services range from Maternal Mental Health, Child and Youth and their Families through to Adult Mental Health Services and Mental Health Services for Older Persons. The diverse cultural mix of the region is reflected in service delivery needs; there is particular emphasis on the needs of Maori, Pacific Island and Asian people and their specific epidemiology.

The services encompass primary and secondary, inpatient and outpatient services. These include Community Mental Health Care, Crisis Assessment and Treatment Teams, Emergency Departments, Rehabilitation and Housing, Forensic services, Community Alcohol and Other Drug services.

The literature regarding error and risk, the importance of the emergency department as the point of contact for mental health patients and the corrective capacity of the right information at the point of decision-making, constitutes an argument for the development of a regional information system to share key patient details.^{[ 1 ]}

Knowledge Management is the Solution
Orr^{[ 4 ]} illustrates that knowledge management is fundamental for the successful implementation of information technology innovations in health care. Harnessing and building from existing technological and communicative processes is essential in the provision of affordable, sustainable and effective care systems.^{[ 14 ]} Orr raises the risk of failure, financially and culturally, with implementing concepts that are beyond the capability of the current environment ("concept-reality gap").^{[ 4 ]}

There are several prerequisites for a knowledge management solution. The information and business service components include system architecture, common data sets, information management, privacy and security standards and alignment of differing business rules.^{[ 15 ]} Cultural and clinical components include the sharing of tacit (unrecorded, unwritten) and explicit (recorded, externally retrievable) knowledge, innovative development of processes from learning and integration of external knowledge. Together these form the common basic principles.

New Zealand is the only country in the world with a health index that operates on a national scale. The National Health Index (NHI) is a unique national identifier of people who use health services, enabling disparate patient management systems to share information.

In 1997, the New Zealand Ministry of Health launched a project to develop a national database which would collect patient activity information from the secondary mental health and alcohol and drug services. By July 2000, all DHBs were providing standardised data into the Mental Health Information National Collection (MHINC).^{[ 16 ]} MHINC is designed as a national epidemiological collection. In such a design, data is provided only in aggregate. In addition, the national privacy constraints resulted in the data being encrypted preventing patient identification by NHI. Thus, MHINC is not suitable for accessing individually identified records. However, the national data set does offer a common information exchange platform with the understanding of what each DHB is capable of sharing. In 2004, the three DHBs in the Auckland region developed a Regional Information Service Strategic Plan (RISSP) ^{[ 17 ]} , which proposes strategic alignment of information services and the adoption of regional standards and systems. This would provide a sound knowledge management foundation. The plan highlights mental health information exchange issues and prioritises the development of a single clinical information system, integrated across the region’s three DHBs. Because of the considerable resources required and the extended timelines associated with such a large undertaking, an interim system is being implemented, allowing authorised mental health workers to check whether a patient is being treated in the region.

The Auckland region has an existing health information infrastructure, hosting secure regional systems accessed through a medical applications web portal which provides a single access point to multiple, disparate information systems. This platform forms the backbone for developing regional information exchange processes.

Mental Health services in the Auckland region DHBs exchange essential information through multiple communication channels – predominantly facsimile and phone. This methodology is embedded into clinical culture and is efficient when information can be accessed at the time clinical decisions are made. Regional clinical data is saved in a variety of formats, some recorded, either on paper or electronically, and some tacit (eg, in the minds of clinicians). Recovering information from these diverse systems, especially urgently and from offsite, is problematic and risky. Knowing where to find this information is of great practical importance and has the potential to improve care.

Retrieval and use of such information also raises privacy issues and there is a dilemma between meeting the need to protect private information versus the need to share information, even when being used solely for an individual’s good. In New Zealand, personal information is managed within a national privacy framework that covers such issues as information sources, informed awareness, security, accuracy, use and disclosure of information, access and correction protocols and the data containment life.

Project Initiation
In 2003, recognising the potential of the right information at the right time, Waitemata DHB investigated the concept of an "electronic whiteboard" (manual whiteboards are used to record patients movements within some wards). Developers of the RISSP accepted the need to implement this as an interim system to the planned regional clinical information system. They initiated the Current Consumer Viewer (CCV) project which allows appropriate providers of mental health services within the Auckland region to identify whether a patient is currently accessing services from another mental health service in the region, for the purposes of facilitating timely and co-ordinated optimal care.

CCV is an integral part of the overall knowledge management system. It is not intended to replace clinical decision-making but to become part of the wider knowledge framework.

System Objectives
The objectives of the CCV project are to:

• Facilitate timely and co-ordinated optimal care.
• Be affordable and sustainable.
• Integrate essential high level health event data items (NHI number, patient name, gender, date of birth, service, care co-ordinator, last contact date and current legal status) of current regional mental health patients from the disparate patient information systems across the Auckland region.
• Provide fast, secure access to patient information through the web portal, returning patient health events summary details (evidence of contacts not the details) through the National Health Index (unique patient identifier).
• Capture the information daily in a standardised format from each DHB into a regional data warehouse with no additional data entry.
• Explore the possible benefits of information sharing to ensure best practice for the patients and families who move throughout the Auckland region.

The framework for the collection, exchange and management of health information about identifiable individuals by this system falls within the provisions of the New Zealand Health Act 1956, the New Zealand Privacy Act 1993 and the New Zealand Health Information Privacy Code 1994 (’HIPC’). The HIPC, in particular, provides a broad framework of controls for the management of information about identifiable individuals. The information is being disclosed in accordance with the purpose for which it was obtained and in accordance with the three DHB’s policies on the collection, recording, use and disclosure of information.

Limitations
Any information system is only as good as what is recorded electronically. The advancing of mental health pathways requires that the right information is exchanged through the existing communication channels.

Late or no data entry into the disparate information systems feeding CCV will create gaps in the information and clinician’s dependent on the CCV may mistakenly suppose the patient is not being seen within the region.

CCV needs to be accessed to benefit the patient. When this is not instilled into the culture of a clinician’s daily processes, exchange of key clinical information could be neglected.

The above problems are being mitigated by:

• dynamic analysis and feedback processes to align data collection processes and subsequent data quality
• training being embedded into existing processes
• iterative group discussion, reflection and review.

Knowledge Processes
The fundamental basis of knowledge management is the sharing of knowledge to enable staff to improve performance.

Indeed, the development of the CCV concept epitomises Davidson and Voss’s I-R-S-A model (Identify, Reflect, Share and Apply).^{[ 18 ]} The source and exchange processes of knowledge were identified, the gaps in the processes reflected on and subsequent improvement measures shared across the region generating the CCV application.

The CCV provides access to those people who have further sets of tacit and explicit information on a patient, improving the clinician’s capacity to act. Furthermore, the system minimises costs by making use of existing technology.

For example, a patient seeks treatment at a hospital in the Waitemata DHB when their care coordinator is with the neighbouring Auckland DHB (this occurs frequently in the geographically close-knit region). The health professionals at Waitemata use CCV to find out the name of the patient’s care co-ordinator and the service name at Auckland DHB providing care. The care co-ordinator is contacted and the alert of exploratory psychotherapy causing suicidal reaction is shared along with further key information about care and treatment. A potential adverse outcome though a breakdown in the care pathway is prevented with CCV facilitating timely and coordinated optimal care.

It is important to highlight the fact that the system itself does not contain key clinical information or change the current information exchange channels. It maximises the use of current information by providing a direct path to the information for patients currently accessing services from another mental health service in the Auckland region. This core piece of information allows timely access to a range of secondary information that has the potential to be converted into knowledge and facilitate timely and co-ordinated optimal care.

Action participation methodology will be used to implement CCV and evaluate parameters of CCV making a "healthy difference" (included within the Waitemata Health vision). This involves iterative group discussion, reflection and review by the Waitemata Mental Health knowledge management team. Once this process is established, a continual evaluation of performance against knowledge management principles can be practised.

Limiting changes to current processes and working within existing knowledge parameters kept the project simple and mitigated risk of failure.

Information System Methodology
Using existing technology and expertise was critical to the success and affordability of the project.

System management requirements were embedded into current processes for data warehousing, database administration and data management expertise. No development on the regional network infrastructure was required and an existing regional server and database accommodated CCV requirements. The only capital cost was the development of a set of screens within the web portal.

The CCV architecture is broken down into three key components (Figure 2):

1. Data Transfer – Daily data extracts are uploaded to the Regional Database Server from the three DHBs. Criteria for the daily extract requires patients to either have an open episode of care (referral) or be a current inpatient (including those on leave). A referral closure or a discharge from an inpatient service will trigger the removal of information from the regional database in the daily extract. There are three jobs scheduled to run every day to upload daily extracts from each DHB. These jobs are monitored by Auckland DHB database administrators using Microsoft Operations Manager monitoring tool (MOM).
2. Regional Database – Based at and managed by Auckland DHB, the Regional Database Server processes the extracts and provides secure access to approved users from across the region through the web portal. The CCV regional database is updated each morning from the three DHBs.
3. Data Access through the Web Portal – A set of search, display and result screens has been developed for separate secure DHB access to the regional dataset.

Figure 2: CCV system architecture

Conclusion
A sound technical foundation met all of the facets of Orr’s^{[ 4 ]} F.I.R.S.T. mnemonic. Existing technology, infrastructure and expertise provided the platform. Adding functionality to the existing web portal presented an intuitive system, easily embedded into training processes with minimal training for existing users.

A direct path was developed to the existing communication channels that exchange key clinical information. Access to regional care providers has enabled the rapid transfer of explicit and tacit knowledge at the point of care forming a critical part of risk and initial assessment and, more importantly, maintaining care continuity.

Feedback in the pilot stages of CCV use has been positive:

• A Waitemata DHB Psychiatry Liaison team clinician treated a patient in the Emergency Department who was under the care of an Auckland DHB team. Knowledge shared from contacting the care co-ordinator at Auckland formed a key part of the risk assessment.
• A general practioner (GP) rang when a patient in her rooms was hallucinating and threatening suicide. Access to CCV showed the client under an Auckland service. The GP contacted the patient’s psychiatrist, they were able to discuss the current mental health state and start psychotropic medication. 
• Participants at an Auckland CCV pilot site, who were at first sceptical (because of the lack of clinical information contained in the CCV), now find it effective for accessing the information they require at the point of decision-making.

Risks identified in the Mental Health Commission Review^{[ 1 ]} have been addressed with the tightening of key clinical information exchange processes and a ground-breaking regional privacy compliance policy. Together, with the learning from the continuing evaluation of CCV use and the analysis of the different types of information exchanged, they will provide a platform for the development of a single regional clinical information system.

References

1. Tretheway R, Hines P, Ellis S, Donovan M, Judson N, et al. Expectations for information-sharing and related practices in mental health services. 2005 Available at: DHB Privacy and Information Sharing Policy Review
2. Chu S. Guideline representation formalism and electronic decision support systems: Addressing the guideline – implementation gap. Health Care Informatics Review Online. 2005.
3. Institute of Medicine. Crossing the quality chasm: a new health system for the 21st century. Washington, DC: National Academy Press. 2001.
4. Orr M, Day K. Knowledge and learning in “successful” IT projects: A case study. Health Care and Informatics Review Online 2004; Jun 1.
5. World Health Organisation. World report on knowledge for better health – Strengthening Health Systems, 2004, xvi + 146 pages, ISBN 9241562811.
6. Kohn LT, Corrigan JM, Donaldson MS (eds). To err is human: building a safer health system. Washington, DC: National Academy Press; 2000.
7. Davis P, Lay-Yee R, Briant R, Ali W, Scott A, Schog S. Adverse events in New Zealand public hospitals: occurrence and impact. NZ Med J 2002; 115:1167.
8. Kalucy R, Thomas L, King D. Changing demand for mental health services in the emergency department of a public hospital. Aust NZ J Psychiatry 2005; 39:74-80.
9. Heslop L, Elsom S, Parker N. Improving continuity of care across psychiatric and emergency services: combining patient date within a participatory action research framework. J Adv Nurs 2000; 31 (1):135-143. 
10. Cassar S, Hodgkiss A, Ramirez A, Williams D. Mental health presentations to an inner-city accident and emergency department. Psychiatric Bulletin. 2002; 26: 134-136.
11. Croskerry P. The importance of cognitive error in diagnosis and strategies to minimize them. Academic Medicine 2003; 78 (8).
12. King EA. The Wessex suicide audit 1988–1993: a study of 1457 suicides with and without a recent psychiatric contact. Int J of Psych in Clin Pract 2001; 5(2):111–118.
13. King E. Suicide in the mentally ill – an epidemiological sample and implications for clinicians. Br J Psychiatry 1994; 165(5): 658–63.
14. Sambamurthy V, Subramani M. Special issue on information technologies and knowledge management. MIS Quarterly 2005; 29(1):1; 7p.
15. Rillstone M. Information management and information technology (IM/IT) for integrated care; Health Care Review Online 1998; 2(12).
16. Mental Health Information Collection (MHINC). 2001.
17. Counties Manukau and Waitemata District Health Boards. Auckland health information services strategy plan. Auckland, New Zealand, 2004.
18. Davidson C, Voss P. Knowledge Management in Action. Knowledge Management 2002; p118, Step One.