INNOVATIVE HEALTH CARE CHANNELS

Towards Declarative Electronic Decision Support Systems Focusing on

Patient Security

Kerstin Ådahl, Jenny Lundberg and Rune Gustavsson

Department of Interaction and Systems Design, Blekinge Institute of Technology, 37225, Ronneby, Sweden

Keywords: Semantics, Health-care support systems, Info sense, Rule-based models, Meta-information.

Abstract: The main contribution in this paper is a structured approach supporting validated quality of information

sharing in Health Care settings. Protocols, at different system levels, are used as a method to design and

implement intelligible information sharing structures. Our approach can preferably be seen as a context

dependant information modelling framework that could be implemented using, e.g., web 2.0 techniques in a

professional context. The main challenge is how to trustworthy convey and analyze the huge amounts of

information available in Health Care contexts. Our innovative information health channel concept provides

an approach to analyze and structure information as well as a contextual support towards increasing patient

security.

1 INTRODUCTION

Health Care must be adequate, safe, secure, and

efficient. Deficiencies in information transferring

and sharing activities, verbal as well as written,

might jeopardize patient security and impair the

quality of care. Quality of information sharing is

important, not least in domiciliary care, where

delegations of tasks often are necessary.

The following Figure 1 captures some main

challenges to be addressed in order to ensure patient

security in future distributed health care.

Health care team

Innovative health information sharing Channel

Requirements

-Trustworthy conveying of representations of informatio

- Support of workflows

- Support of detection, analysis, and repair of break-do

- Support of validation of procedures (protocols)

- Support of learning in teams

Figure 1: Challenges and requirements of Innovative

Health sharing channels.

In the figure we have denoted the information

sharing media as Innovative Health Care Channel

(IHCC). The main purpose of the IHCC is to enable

empowerment of professionals or patients to ensure

increased patient security. Empowerment of patients

might include insights in how to change behaviour

to stay healthy or to support trustworthy

rehabilitation measures. Empowerment in healthcare

teams will increase quality of healthcare teams by,

e.g., supporting learning and debriefing of cases.

From Figure 1 we note that the first challenge of

building and maintaining IHCC concerns trust-

worthy conveying of representations of information

by the channel. To that end we have to address

information and information sharing in contexts, i.e.,

interpretation of representation by teams of human

agents (Section 2.1). The second challenge focus on

support of workflows or transactions between

situations. The third challenge focus on support of

detection, analysis, and repair of breakdowns. The

forth challenge focus on support of validation of

procedures (protocols) and work practises. The fifth

challenge focus on learning in teams. Obviously the

last three challenges are instrumental in improving

patient security but they are critically dependant on

a proper understanding and solutions of the first two

challenges.

In Sweden, it is a duty to report incidents that

are, potentially or directly, risks for patients safety

(Lex Maria). However, Ödegård (2006) proves in a

study that local, as well as national, reporting

systems result in insufficient information about risks

423

Ådahl K., Lundberg J. and Gustavsson R. (2009).

INNOVATIVE HEALTH CARE CHANNELS - Towards Declarative Electronic Decision Support Systems Focusing on Patient Security.

In Proceedings of the 11th International Conference on Enterprise Information Systems - Information Systems Analysis and Speciﬁcation, pages

423-426

DOI: 10.5220/0002156004230426

 SciTePress

in an activity. Accordingly, complementary methods

and additional instruments for the existing reporting

systems are necessary in order to offer secure and

safe domiciliary care: both for care recipients and

care providers.

Personnel from the domiciliary care, working

close to the individual care recipient, is an especially

exposed party of health care since they in practise

work without satisfying supporting- and security

systems. An important function of a supporting

system is to minimize undesirable variations of the

quality of care (NICE 2007) with the aim of

facilitating and securing care.

Foundations for a decision support system might

be common directions, quality registers, decision

bases like national guiding principles, nursing

programs, routines, PM or constitutions. If a

supporting system, aiming at supporting health care

professionals, should contribute to and develop

evidence based care, it must first of all be grounded

in systematically collected, inspected and compound

knowledge. Regarding this requirement, there are

large deficiencies in available supporting systems of

today. To find a solution to this problem, we are

aiming at analyzing and developing rule based

semantic context dependent prototypes for

development and assessment of Declarative

Electronic Decision Support Systems (DDSS) that

are in fact instances of IHCC of Figure 1.

DDSS is an example of Professional Empower-

ment, a complementary direction related to the

worldwide movement of Patient Empowerment.

This paper describes our planning work with such an

innovative Health Care Channel.

The remaining part of the paper is organized as

follows. In Section 2 the concept of IHCC and the

nature of information are discussed in some detail.

Section 3 is focusing on basic requirements of

DDSS and some clarifications of the requirements

and challenges stated in Figure 1. Section 4

concludes this paper and gives some pointer to the

future. Section 5 gives some references.

2 INNOVATIVE HEALTH CARE

CHANNELS

Quality of Health Care is grounded in quality and

trustworthiness of information and information flow.

Break downs or misses in information flow causes

multiple incidents and accidents every year

worldwide. A conservative estimate for the

industrialized world is that between 6-10% of all

patients suffer from care related injuries, sometimes

causing deaths. In Sweden, frequent Lex Maria

Cases and reports to the HSAN (Medical

Responsibility Board) indicate extended deficiencies

of information flow, information interpretation or

other information related faults in care. The

National Board of Health and Welfare concludes, in

a set of reports regarding Patient Security (Social-

styrelsen 2008) that insufficient communication and

information breakdowns are most frequent regarding

causes for injuries in Health Care.

It is quite evident that there is a lack of

robustness in how health related information is

handled. Our suggestion is to provide a framework

for an Innovative Health Care Channel (Figure 1),

supporting trustworthy communication. By such a

channel, systematically collected, inspected and

compound information is available for use in Health

Care situations. In the next section (2.1) we will

clarify how our perspective on Information as a

concept has influenced our proposal.

2.1 A Model of Innovative Health Care

Channels

Situation Theory, originally presented by Barwise &

Perry (1983) and elaborated by Keith Devlin (1991,

2001), constitutes a foundation for our approach.

Situation Theory addresses the contexts of

information to support automatization of information

flow possible taking into account its meaning.

Situation Theory clarifies the important distinction

between information and representation. Information

is representation of information, interpreted by a

human agent and appropriate context.

Representations of information are interpreted by

human, or machines, following appropriate

procedures for encoding and decoding. Procedures

can, and often is, built on experiences and learning

over time. This is valid not at least for medical

information that is quite complicated to understand,

given a representation (texts), and requires a context

to be fully intelligible and used by a human agent or

a team.

Patient and Professional Empowerment are both

focusing on providing necessary context to enable

proper information extraction out of its

representations at hand (being text, video, film,

recordings, and so on). Patient Empowerment,

defined as”the increasing ability of patients to

actively understand, participate in and influence

their health status” (Degoulet et al in Nelson & Ball

2004), is pointing at the importance of co-operation

in health care; between patients and professionals.

ICEIS 2009 - International Conference on Enterprise Information Systems

424

On the other hand, Professional Empowerment is

highly topical. The goal must be to match the

meaning of information between the sender and the

receiver but as every interpretation differs from the

others (Devlin 2001), it is not conceivable to attain

exact understanding, but as close as possible (Ådahl

2007).

Considering the second challenge of Figure 1, we

focus on support of workflows or transactions

between situations. Information occurs in a context

(situations) and should be interpreted dependent on

each specific situation, connected to meta-level types

of situations, to support information transactions.

However, it is important to exhaustively identify

workflow in health care in order to build information

channels for co-operation and secure information

exchange (Lundberg 2007). The third challenge of

Figure 1 is support of detection, analysis, and repair

of breakdowns. To avoid breakdowns in the

transactions between situations and strengthen

patient security, robust channels supporting

information exchange is essential.

3 DECLARATIVE ELECTRONIC

DECISION SUPPORT SYSTEMS

(DDSS)

Rule based systems are a useful model for

implementation of knowledge based systems (KBS)

and also support knowledge management. However,

the proper translation from the high-level, situation

type workflow support design of previous section is

still not clear, since we are in fact abstracting away

the semantic context into a syntactic context

(Lundberg-Gustavsson, 2009).

The logical conclusions drawn from the known

rules cannot capture the complete situation.

However, a shared awareness of the intended

context by the users, e.g., health care professionals,

supported by system implementation guidelines, as

well as complementary documentation such as

active documents and active databases, expressed

using suitable ontologies is a starting point for

designing and implementing a IHCC (Figure 2).

Furthermore, the middle box in Figure 2,

referring to Services, structured information and

meta-information, focusing on those aspects should

be a realistic strategy towards enabling more robust

transactions between situations. We consider the

possibilities to use a rule based approach for the

IHCC in order to develop a DDSS.

Fundamentally, this information is validated by

Rule base

Ontologies

Inference

engine

Services

structured information and meta-information

Active

documents

and

databases

Infrastructure for configurable mobile or stationary

networks for distance bridging work and information

sharing. Access points: TV, Mobile phones, Netbooks,

smartphones etc.

Figure 2: Architecture for Innovative Health Care

Channels.

highlighting the intentions behind the data flowing

(representations) and this must be executed in

relation to context. Medical information is extremely

vital and mistakes by using invalidated and not

quality secured information from unreliable sources

might be lethal. Assigning an administrative person

as information analyst could be very risky and

would mirror a lack of understanding of the

importance that information analysis play in a

medical profession. However using only the most

medical competent persons in this position could

create bottlenecks. Having addressed Challenges 1

and 2 of Figure 1, we now return to Challenges 3, 4

and 5 of the same figure.

The third challenge is a focus on support of

detection, analysis, and repair of breakdowns. The

DDSS should support health care professionals in

work flow, preventing information breakdowns from

happen. Confirmation devices such as

acknowledgement, continued attention etc. can be

used as a starting point to consider further similar

tools appropriate in Health Care context.

The forth challenge focus on support of

validation of procedures (protocols). This challenge

concerns the predefined procedures of how to

perform a certain health activity. The activity

performed with the use of computer support (rule

based) is to be validated according to the logic in the

protocols.

The fifth challenge focus on learning in teams.

For example, the web 2.0 solution concerns team

members exchanging opinions about a certain report,

drug, treatment or a certain activity performed.

Accordingly to the analysis above, a suitable

architecture for IHCC such as DDSS is presented in

Figure 3. The infrastructure for configurable mobile

INNOVATIVE HEALTH CARE CHANNELS - Towards Declarative Electronic Decision Support Systems Focusing on

Patient Security

425

or stationary networks enables services with

structured information and meta-information, to be

accessed by human agents. The rule base with

connected inference engine constitutes the core for

the declarative reasoning. Ontologies connected is a

vital part in the team learning, with reference to the

fifth challenge, since the different roles have

different corresponding terminology, definitions as

well as perspective upon the same issue.

To implement semantics (i.e. meaning) in the

system, information about information is needed as a

“pointer” to situations. In this matter, meta-

information needs to be considered, to transform

data (uninterpreted information) into context

dependent information, to be interpreted by human

minds. One possible way to make a rule-based

system situation dependant is to implement possible

situations, thus following the spirit of Lenats Cyc

project (Lenat, Guha, 1990). Restricting the focus

area and consider different perspectives and

situations and validate them, we can approach a

suitable solution. Adding data from mobile devices

used close to the situation and open up for informal

communication possibilities between team members

is a possible way to consider. Furthermore, we also

consider the accessibility perspective with situation

and domain dependant search structures (Antoniou

and Harmelen, 2004).

4 CONCLUSIONS

In this paper we have presented an approach towards

IHCC. We state that information in health care must

be viewed as situation dependent and that the

representations at hand for the teams of users have to

be augmented by common ground semantics to

enable the intended Professional of Patient

Empowerment. Mistakes concerning patient security

are frequently reported to HSAN (Medical

Responsibility Board) in Sweden every year and, to

a large extent, many depend on information

breakdowns. Accordingly, we have identified five

Challenges to meet the requirements of IHCC

(Figure 1). A rule based solution, DDSS, to enable

development of prototypes for selected scenarios are

the present focus for further investigations.

REFERENCES

Antoniou, G., and van Harmelen, F., 2004, A semantic

Web Primer. The MIT Press

Barwise J, and Perry S, 1983. Situations and Attitudes,

Cambridge, MA: MIT Press.

Degoulet P, Fieschi M, Jaulent M-C, Ménard J, 2004.

Patient Empowerment, Cybermedicine, and Citizen

Education in Nelson R, Ball M J. 2004. Consumer

Informatics – Applications and Strategies in Cyber

Health Care Health Informatics Series Springer-

Verlag, New York, Inc

Devlin Keith, 2001. InfoSense – Turning Information into

Knowledge, W.H. Freeman and Company, New York

Devlin, Keith, 1991, Logic and Information. Cambridge

University Press. ISBN 0521499712

Douglas Lenat and R. V. Guha. 1990. Building Large

Knowledge-Based Systems: Representation and

Inference in the Cyc Project. Addison-Wesley. ISBN

0-201-51752-3

Lundberg, Jenny, 2007 Principles of workflow support in

life critical situations, Blekinge Institute of

Technology, 2007:02. p. ISBN/ISRN:978-91-7295-

105-1

Lundberg, Jenny and Gustavsson, Rune (2009): Robust

approach towards context dependant information

sharing in distributed environments. In Proceedings of

ICEIS 2009.

NICE, 2007. A Step-By-Step guide to developing

Protocols. NICE National Institute for Clinical

Excellence. (Link Available 2008-12-08)

http://www.nodelaysachiever.nhs.uk/NR/rdonlyres/E6182

916-60F3-4DFA-9E5F-

73606513D8BA/0/RG0013StepbyStepguidetoDevelop

ingProtocols.pdf

Socialstyrelsen, 2008. Tema: Patientsäkerhet. Nr 2.

Kommunikation i Vården, Artikelnummer 2008-109-9

(Available 2009-01-07:

http://www.socialstyrelsen.se/Publicerat/2008/10011/2

008-109-9.htm)

Van Harmelen Antoniou G. F., 2004 A semantic web

primer, The MIT Press.

Ådahl, Kerstin, 2007. Transparency of Critical

Information for Patient Empowerment in eHealth

Blekinge Institute of Technology, Licentiate

Dissertation Series No. 2007:03

Ödegård, S., 2006. Säker Vård. Patientskador,

Rapportering och Prevention, (Avhandling) NHV

Nordiska Högskolan för Folkhälsovetenskap.

Göteborg. Rapport 2006:1

ICEIS 2009 - International Conference on Enterprise Information Systems

426