Regional communication within and across health organisations
The R&D Project
1. Objectives
The health care sector
is intrinsically heterogeneous in its organisation. There are distinct,
formally independent, organisations or institutions that take part in the
delivery of health services: municipalities, counties, hospitals, private
health providers, (private and public) laboratories and private pharmacies.
Yet, the patients – the proper ‘value chain’ – cuts across these organisational
boundaries. There are huge problems with ensuring effectiveness through this process as a whole of receiving
different health services. These problems are essentially about finding ways to
communicating relevant clinical knowledge within and across the different
organisations which make up the health care value chain. This addresses
explicitly the fourth of VERDIKT’s calls on ‘communicating organisations’.
The problems with
inter and intra organisational communication, knowledge sharing and
collaboration give rise to notions like ‘continuity of care’, ‘shared care’ or
‘integrated care’. Reiterating prevailing thinking in organisation and
management science, health policy initiatives in the Western world emphasise
strongly the importance of dismantling ‘vertical’ boundaries in favour of
‘horizontal’ processes of work and sharing of knowledge.
Industrially, we are
in the midst of dynamic structural changes fuelled by ‘integration’ through
national and international mergers and acquisitions alongside structural
changes in the market through regionalisation of tenders.
The notion of
‘seamless integration’, then, in the context of health care has a deeply
ambiguous meaning. It marks the political and ideological commitment towards
integrated care, i.e. service integration as experienced by individual
patients. On the other hand, it is simultaneously used in the much narrower
sense of the technical integration of relevant clinical information systems.
Our overall aim is to
generate
G0 Operationally
relevant knowledge of product and process innovation to support health service
delivery across the value chain to increase efficiency and improve quality of
health delivery to patients;
More specifically, our
aims are:
G1 Identifying and
testing principal strategies and technologies for technical integration of the underlying information systems
G2 Exploring new business models supporting integration
G3 Developing methodological guidelines for management
of integration projects
G4 Establishing
experience forum for vendors, health
workers and researchers around challenges of integration
G5 Publishing
in Int. conferences/ journals: 2/0 (2008), 2/1 (2009), 4/2 (2010), 4/3 (2011)
2. State-of-the-art
There is literally a
jungle of information systems supporting healthcare providers today. This
abundance of different information systems is the mirror image of the enormous
variation in healthcare work along level, geography, professional groups,
agencies and specialization. Given this large number of partly overlapping,
complementary and interdependent information systems, it is hardly surprising
that considerable efforts have been poured into supporting a more
process-oriented service through tighter integration of underlying information
systems[1].
Yet the actual realization of the widely recognized potential of ICT has proven notoriously
difficult to achieve. The implementation – when “implementation” is recognized
to extend well beyond the mere technical realization – of information systems
in hospitals contain a disturbingly tall number of partial or complete
failures. The appreciation of how, to what extent and in what form, the
development and subsequent introduction of information systems raises
organizational, political, legal and technical challenges, is slow. No
systematic overview exists, but the emerging picture strongly suggests that
under-estimating or misconstruing the non-technical issue account for a very
significant portion of failures of information systems in healthcare[2].
This growing awareness
of the importance of non-technical issues focuses predominantly on development
and use of new applications, i.e. information systems aimed at replacing or
supporting manual or paper-based routines. Less attention is devoted to the integration of information systems. This
is unfortunate given the high expectations tied to integration. A key reason
for the lack of success for many health information applications is the way
these, at best, provide local or sub-optimal effects, but fail to achieve
benefits for the work processes as a distributed, collective whole[3]
Accordingly, the
integration of healthcare software systems has remained one of the most
prominent issues in healthcare software development[4].
Integration is expected to automate the medical processes, such as patient
admission, transfer and discharge, ordering of laboratory and radiological
examinations or medication.
An integrated solution
is supposed to give the physicians easy access to data from multiple
information sources, thus providing a complete
picture of the patient/client’s medical history[5].
The multiple information sources are accessed seamlessly from a single point of
end-user interaction[6].
A completely integrated IT solution is a clear goal, both within hospitals as
well as between the different organisational levels in the healthcare system:
“A scalable I-EHR
[integrated electronic health record] would provide
the means to access all available clinical
information, at a corporate, regional, national or even international level” [7]
However, despite the
high aspiration of integrated solutions, Berg (1998: 294) fairly accurately
characterises the situation when he maintains that “fully integrated [PICS] …is
hard to find” and reiterated in the action plan for Norwegian health care[8].
One cause is that many
software products have been built and acquired from heterogeneous sources
during a long period of time, and the systems have differences in
implementation technologies and architectures[9].
Other causes may be ascribed the relatively autonomous domains of the Norwegian
healthcare system (hospitals, primary care, nursing homes and home care
services) which by its autonomous character pose several challenges
(technically, legally and organisationally) to effective integration.
The IT Committee of the Secretary of State
(“Statssekretærutvalget for IT”)
published in
“The healthcare
services will be improved and the quality increased by the coordinated
introduction of information technology in all segments of the healthcare action
chain. Local, regional and national health networks will strengthen cooperation
and resource management in the healthcare sector”[10]
“[T]he present flow of information is
unsatisfactory. Nurses in home nursing services and nursing homes often do not
have access to essential information from other service providers about their
patients”[11]
In summary, there is
thus a broad consensus about the aim
of improving inter- and intra-organisational communication in the health sector
as pictured in figure 1:
General practitioner / Emergency ward Home care service Nursing homes
Figure 1. Flow of information between the different actors in the healthcare sector
The approach, however,
to how to achieve improved
communication flows is fiercely debated. There are two, principal strategies
employed as characterised in table 1 below:
|
Monolithic |
Heterogeneous |
|
Centralised |
Decentralised |
|
One
principal vendor |
Multiple
vendors |
|
Relative
stability in external environment |
Fluctuating
environment |
Table 1. Characterisation of the two, principal approaches to improve communication flow
Our proposed research
project explores concrete manifestations of these strategies as they, starting
from the left/right hand side of the table, are challenged to move in the
opposite direction.
3. The Research Challenge
Table 1 outlines
schematically the two – operational as well as analytic – challenges around
which our project is organised: strategies, technologies and methodological
guidelines for developing and deploying seamlessly integrated health
information systems. To better ground and methodologically restrict our aims,
we have consciously targeted two distinct empirical settings (hospitals;
regions) that traditionally have employed very different strategies to
integration (tight and loose, respectively).
A. WP1: From monolithic to heterogeneous regionalisation
Traditionally,
information systems within hospitals have been tightly coupled. Each of the 85
Norwegian hospitals made individual contracts with the Norwegian vendors (some
international as well) health information systems. From 2002, the central
Government undertook the ownership of the country’s 85 hospitals from the
counties. The former five health regions were replaced by five regional health
enterprises, aiming at running the hospitals as private enterprises. The former
autonomy in the hospitals is thus replaced by more centralised control,
subsequently resulted in a new kind of demands and challenges.
An important strategy,
employed by several of the regional health enterprises, has been to achieve
improved regional communication by standardising – thus privileging one – on
one of the vendors (see
|
|
|
Regionalisation as privileging one vendor |
|
|
||
|
Prior to the construction
of the regional health enterprises, there was a significant variation in the
choice of vendors for EPRs, laboratory systems, PAS, radiology etc. in each
of the regions. This was a natural situation given that the different
hospitals in a region enjoyed significant autonomy in defining its technology
strategy. With the establishment of
regional health enterprises, this changed radically. The responsibly of the
technology strategy was centralised to the health region. This has rapidly
resulted in a situation where there has been enforced uniformity in the choice
of e.g. EPR in the whole region. As one IT managers in one of the regional
health enterprises states, “when all 10 other hospitals in the region has EPR
from vendor [X], it is obvious that also the regional university hospital
[bigger than all of the others together] should have EPR from vendor [X]”. Given the significant variations in needs and
preferences, the dilemma currently facing most of the regional health
enterprises is whether the preferred vendor – chosen for arguments of economy
of scale and standardisation – is able to satisfy the full range of
requirements. |
||
Health regional
enterprises today, then, need to respond to technological and organisational
changes to maintain its ambition of seamless integration. We emphasise the
increased unpredictability in
the Norwegian healthcare market, the increased
scale of vendors and customers, and finally standardisation of information systems and practices within
one region (see
|
|
|
Dynamically changing business environment |
|
|
||
|
The healthcare industrial
market in Moreover, the regionalisation of framework
agreements and tenders imply that a change of systems
in one health region does not just include one hospital and one system, but
will include all of the region’s hospitals and often many different systems
in order to be part of an overall integration strategy. As one of the
managers at Helse Nord RHF explains, “We see a shift from many small to a few
large contracts. These changes are extremely dramatic for vendors – it is an
all-or-nothing competition”. The Regional Health Enterprises demand standardised and common information
systems in the health region, making new demands on availability, robustness
and across-hospital integration so, the manager continues, “For each
specialized field [Bloodbank, Microbiology, etc], we want the same system in
the whole region [North Norway], and preferably in one database”. |
||
The activities in WP1
comprise of:
A1 Participatory
action research in either Health Region North or Middle to explore the
strategies of either Dips or Siemens, respectively, of how to accommodate more
heterogeneity in the region
A2 Participatory
action research with the international standardisation of laboratory modules
from either Dips or TietoEnator
A3 Test and evaluate strategies and
technologies
A4 Formulate methodological guidelines for
project management
B. WP2: From heterogeneous to monolithic/interlocked regionalisation
Traditionally the
communication between hospitals and primary care has run by ordinary mail. The
general practitioners have referred patients by sending referral letters and
upon discharge of the patients, they have received discharge letters. They have
also requested laboratory tests by filling out a requisition form which they
have sent to the hospitals laboratories, and got the response on email.
This traditional,
asynchronous communication form has gradually been supported by message-based
ICT. Electronic laboratory results and discharge letters have proven to be
relatively successful compared to the traditional way of interaction.
A condition for
success for the new electronic services has been an augmentation of the general
practitioners’ ordinary routines and integration with their GP-based EPRs. Some
GPs have warned that electronic requisitions of laboratory results imply an
increase in their workload compared to the traditional paper-based requisition
form.
Overall, the
implemented electronic collaboration tools (asynchronous requests/response)
represent no far-reaching change in work-routines. An incoming request, both
within hospitals and with the GPs) don’t impose an immediate action. Work
practices remain largely constant and the interacting systems are relatively
independent of each other: the systems are loosely coupled.
The current ambition
of improving the information flow in Norwegian health care makes a shift in
integration strategy from loosely to tightly coupling regionally (see Box 3 and
activity B1).
|
|
|
Stronger dependencies between hospitals and GPs |
|
|
||
|
New interactive services
are demanded between hospitals and general practitioners. For instance, Well
diagnostics are current involved in a large integration project with UNN
where the goal is to replace all paper-based routines related to requisition,
reception and management of laboratory referrals. The key point here is tracing. The general practitioner
should receive updated information about the analysis. Moreover, the system
is intended to support a two-way
interactive communication channel between the GPs in general practice and
the specialists at the hospital where they given an informal tool to
communicate. For their part, UNN sees this project as the opportunity to
streamline the information flow, reorganise the work routines and thus
improve efficiency in their laboratory services. In this way, they need a
tighter integration between UNN and general practice. Secondly, Norwegian health care authorities have
stimulated and initiated several electronic booking projects between
hospitals and primary care. Patients will be allowed to choose which hospital
they would like to go to and the date and time which suit them. This requires
a tight synchronous interaction between the hospital and the general practitioners.
The hospitals must have free “hours” available (tightly coupled to both the
hospital’s PAS system and its EPR). Hospital physicians must have organised
their own time to fit with this hours and the GPs must book appointment
according to specific rules. In sum, the interaction is strongly regulated,
thus tightly integrated in both a technical and organisational way. Both UNN
and Well diagnostics are closely involved in this project: UNN as a user and
Well as the vendor as they recently has acquired |
||
A characteristic an
important aspect of the regional integration efforts is that the number and
type of stakeholders are numerous. It is no longer only a hospital – GP
interaction, but integration also includes nursing homes, homecare services and
pharmacies (see
|
|
|
Multiple actors involved in regional integration |
|
|
||
|
New demands for efficiency
and quality make it pressing to implement totally integrated systems in all
the different sectors in health care. A relevant example is the national
virtual medical card (also denoted “core journal” by the vendor Profdoc)
aiming at containing all information about any patient’s use of medications.
This presupposes integration of EPRs in hospitals casualty clinics, general
practice, nursing homes and the homecare services. The virtual medical card
is supposed to ensure that the different healthcare providers provide the
patients with consistent medication. For instance, nursing homes and homecare
services often lack information about what medication the patients are using.
Currently, most of the vendors in this research project participate in a
regional development project in Helse Nord RHF around this issue. This
presupposes a huge integration challenge as medication information in the
virtual medial card is supposed to consistent with all the EPRs located in
the different healthcare sectors: homecare, nursing homes, general
practitioners and hospitals. |
||
The activities in WP2
comprise of:
B1 Participatory action research at Well
Diagnostics and among its collaborating partners (Dips and Profdoc) and users
(hospitals) in order to identify strategies for establishing interactive services as well as suggest
how organisational routines may be changed in order to support the new
services.
B2
Participatory
action research (among the vendors, in hospitals, general practitioners,
nursing homes and homecare services) targeting the use of core journal to capture, share, and use clinical information (medication)
across institutional boundaries.
B3 Test
and evaluate strategies and technologies for regional integration
B4 Formulate
methodological guidelines for project management of regionally integrated information systems
4. Research Method and Results
The project is planned
to start August 2007 and run for four years, August 2007 – August 2011.
The main points of the
project design and general approach to the project are the following:
- Sub-projects are carried out in user and
vendor organisations, focusing on improving integration strategies among
healthcare systems. This involves cooperation between the users, vendors
and research institutions.
- Division into two work packages, one for
each of the two main themes outlined above. Each work package has
researchers who are responsible for coordination and progress.
- A common vendor/user arena that functions as a meeting place for
vendors, users and researchers. The meetings at arenas will focus on
various topics and themes in rotation. The vendors and the user
organisations will themselves choose if they want to be active in all
themes, or if they want to limit themselves to a few.
- A forum for the participating vendors where themes may focus on strategies for
integration between systems from different vendors and models for
increased collaboration.
- A common researcher arena that supports cooperation between the
participating research groups, and across the chosen themes, to insure the
overall integration between the themes (work package) and the vendor/user
arenas. This arena will also deal with contributions to educational
programs, and the development and implementation of a publication strategy
for the project.
Based in participatory
action research[12] we want
to approach the project also through qualitative, in-depth and longitudinal
studies, to better understand the many facets of systems development in general
and integration in particular
The research
participants have many years of project-experience with ICT in health research, focusing on co-generative learning
and close collaboration with demanding participants through for example the
KVALIS-project[13] and
NSEP[14].
The main research
activities to be utilized are:
- Specialist preparatory work. Literature
review. Objective: Clarification of the central theoretical positions and
research gaps on each theme.
- An introductory meeting for each work package at the vendor/user arena where experience is exchanged, the actual challenges facing both user organisations and vendors are described, research questions are identified and outlined, and the design of case studies is determined. Objective: Development and determination of research gaps and questions.
- Four research studies from the
participating organisations. Objective: At least 3 well documented studies
available for comparative analysis.
- Regular meetings at the vendor/user arena
to present preliminary results, to adjust focus and data collection in the
case studies, and to promote discussion to transfer results relating to
the challenges faced by individual companies. Objective: Quality assurance
in relation to relevance for the vendor and user organisations. Further
empirical evidence. Exchange of experience and cross-fertilization across
organizational boundaries.
- Regular meetings at the research arena. The aim is to obtain new, important theoretical as well as methodological contributions to the work, quality assurance of the case studies, and to develop strategies for publication, as well as transfer of insights to educational programs. Objective: Research quality, publication, and contribution to educational programs.
- Publication of results in international
journals and other media, both text, and hypermedia, for example through
the poplar science web-portal Forskning.no. For the general audience of
software companies, we will organize seminars and experience forums.
5. Project Organization
The owner of the
project is Well Diagnostics, with the head office in Tromsø and a branch in
For the national
partners, letters of intention are attached to the application. For the local
partners and for the user organisations, we have got either written or oral
acceptance for their contribution. The project comprises the following partners:
Vendors:
|
Organisation |
Role and contact person |
|
Well Diagnostics (project owner) |
Strategies for integrating
systems from heterogeneous vendors Contact person: Rolf Dahl |
|
National
partners/contributors |
|
|
Siemens medical |
Develop business models
and strategies for regional integration Contact
person: Anne-Mai Olsen |
|
Dips ASA |
Develop business models
and integration strategies Contact
person: Tor-Arne Viksjø |
|
TietoEnator |
Strategies for integration
and modularisation Contact person: Sigurd
From |
|
Local
partners/contributors |
|
|
Profdoc |
Develop overall strategies
for integration between specialist- and primary care Contact
person: Øyvind Ødegård |
|
Visma Unique |
Models for integrating
homecare- and nursing care systems with hospital and primary care systems Contact
person: Lars Espejord |
|
Jupiter System Partner as |
Strategies for integrating
their Medical Genetics system with the EPR products from Dips, Siemens and
TietoEnator Contact person: Odd-Halvard Bjørnstad |
User organisations
|
Organisation |
Role and contact person |
|
Universitetssykehuset
Nord-Norge HF |
Methodological guidelines
for management of integration projects Contact
person: Magne Johnsen |
|
Rikshospitalet –
Radiumhospitalet HF |
Develop strategies both in
a technical and organisational perspective to achieve our goal of a complete
electronic patient record. Contact
person: Hallvard Lærum |
|
Helse Bergen – Haukeland
Universitetssykehus |
Integration between
different large-scale EPR system from different vendors Contact person: Eva Møller |
|
Helse Nord IKT |
Methodological guidelines
for management of integration projects Contact
person: Bengt Flygel Nilsfors |
|
St Olavs hospital |
Integration strategies for
virtual electronic medical cards Contact
person: Olav Sletvold |
|
Sentrum legekontor |
Seamless information flow
between specialist- and primary care Contact
person: Dag Nordvåg |
|
Kroken sykehjem |
Access to medication lists
for old and chronic patients Contact
person: Ingebjørg Riise |
Research partners
|
Organisation
|
Role
and contact person |
|
Department of telemedicine,
|
Conduct action research
studies in organisations and supervise PhD candidates Contact
person: Associate professor Gunnar Ellingsen |
|
NTNU, Department of
Computer and Information Science |
Conduct action research
studies in organisations and supervise PhD candidates Contact
person: Professor Eric Monteiro |
Management
The project leader is
Associate professor Gunnar Ellingsen, Dept. of Telemedicine,
The project is to be
managed at two levels, a strategic and an operational. Well Diagnostics as will
be involved together with Ellingsen and Monteiro at the strategic level. The
operational level is associated with the work package research themes, which is
lead by a work package leader and coordinated by a steering group consisting of
representatives from the participating organisations.
Ellingsen has also close
and long-lasting collaborative ties with NTNU/ NSEP in general and the other
principal researcher in the project in particular, Professor
In sum, the
organisation of the project will draw on and contribute to the synergy with the
broader research communities at the two principal nodes, UiTø/TTL and NTNU/
NSEP.
6. International cooperation
|
Institution |
Main focus |
Contact Person |
|
Erasmus Univ. Rotterdam |
Implementation of ICT applications in health care
(the EPR in particular) from a socio-technical perspective |
Professor Marc Berg PhD
Roland Bal |
|
The |
Implementation, and standardisation of information and communication
technologies in organisations |
Professor Robin Williams |
|
|
The development, use and management of
information and communication technologies and systems. |
Professor Geoff Walsham |
|
|
Studies of standardization and evidence-based
medicine and how change occurs in contemporary health care |
Professor Stefan Timmermans |
|
Umeå University |
Standardisation and information infrastructures |
Professor Jonny Holmstrøm |
|
London
School of Economics |
Expanding
organisational involvement of IT and the associated shifts in work and
employment forms |
Professor
Jannis Kallinikos |
|
Simon
Fraser University |
User
Participation in Design of Technological Systems |
Professor
Ellen Balka |
|
Vienna
University of Technology |
CSCW,, information technology and its influence on the practice and
quality of health care within hospitals |
Professor
Ina Wagner |
|
Copenhagen
Business school |
Information
technology in health care, especially electronic patient records and medical
informatics. |
Assistant
professor Signe Vikkelsø |
|
Roskilde
University |
Effects-driven
IT development in health care, method dissemination and CSCW |
Assistant
professor Jesper
Simonsen |