Concept Matching between OpenEHR and SONHO

Ricardo Cruz-Correia

, Carla Ávila

and Rong Chen

CINTESIS, Al. Prof. Hernani, 4200 Porto, Portugal

Hospital S. João, Al. Prof. Hernani, 4200 Porto, Portugal

Cambio Healthcare Systems, SE-1180 Stockholm, Sweden

Abstract. SONHO is the dominant Information System (IS) in hospitals in Por-

tugal. Currently it is seriously compromised due to its outdated infrastructure,

and hospitals seek a solution enabled data migration to new ISs. OpenEHR is an

open standard that describes the management and storage, retrieval and ex-

change of health data in electronic health records. The aim of this work is to

study how to create an application interface to SONHO implemented according

to the openEHR standard. The first phase of this work was to extract the demo-

graphic concepts existing on SONHO. The second phase was to select the

proper openEHR structure holding the patient demographic information. The

third phase was to match both concepts aiming to identify omissions in

openEHR archetypes. 41 concepts were found in SONHO, and 12 in openEHR

person demographics archetype. From the 41 different SONHO concepts, 14

concepts were mapped to a openEHR concept and 27 are missing. From the 12

concepts found in the openEHR person demographics 3 are missing in SONHO.

We claim that many important SONHO concepts are still missing in the

openEHR person demographics archetype (e.g. patient identification numbers).

To build a useful interface the used archetype must include the missing con-

cepts, which leads to the need of creating a new demographics archetype.

1 Introduction

Currently there is a growing number of Health Information Systems (HIS) operating

on Hospitals. Communication among these HIS aiming at semantic operability is a

hot topic [1]. The main driver behind current standardization efforts is the goal of

consistency and predictability across our data and our system processes, with the

added goal of saving effort through re-use, and more controversially, raising quality

[2]. Different standards, namely CEN 13606 [3], HL7 [4] and openEHR [5], have

emerged aiming to solve interoperability questions. These standards are maintained

by organisations that are very different in their origin. CEN is the European Normali-

zation Committee; HL7 is a all-volunteer, not-for-profit organisation involved in

development of international healthcare standards; and openEHR foundation is a not-

for-profit foundation supporting the open research, development, and implementation

of openEHR open standard (i.e. publicly available). OpenEHR is an open standard

that describes the management and storage, retrieval and exchange of health data in

Cruz-Correia R., Ávila C. and Chen R. (2009).

Concept Matching between OpenEHR and SONHO.

In Proceedings of the 1st International Workshop on Open Source in European Health Care: The Time is Ripe, pages 45-53

DOI: 10.5220/0001828400450053

 SciTePress

electronic health records (EHRs) [6]. The openEHR endeavour is about creating

specifications, open source software and tools in the technical space for such a plat-

form [5].

1.1 The Portuguese Scenario

At the end of the 80s the Health Portuguese Ministry was faced with a major disor-

ganization in the administrative departments of different hospitals. So the idea

emerged to create an Information System (IS) that standardized and organized the

information management. This IS, called “Sistema IntegradO de INformação

Hospitalar” (SONHO), was developed by a national agency called “Instituto de

Gestão Informática e Financeira da Saúde” (IGIF). SONHO is an administrative man-

agement system of patient’s data that apparently met the existing organizational

needs. It has proved to be useful in the management of patients, particularly in admin-

istrative and financial areas. It was a pioneering application in Portugal, with no com-

petitors at that time [7].

SONHO is now a dominant system in hospitals in Portugal and is installed in about

90% of the 86 Portuguese public hospitals. This IS currently assumes a fundamental

role in the registration and management of the administrative information of the pa-

tient.

The overall objective of the SONHO was controlling the flow of patients in the hospi-

tal, namely to know who is in, who leaves, what he/she does and if possible, with

what resources aiming to ensure the standardization of statistical data and billing.

SONHO also aims at facilitating the organization and management of the Health

Ministry by initially considering three patient encounter modules: outpatient, inpa-

tient, and emergency, which represented the three main entry points of patients in a

hospital. Subsequently, new modules were added for surgery operation room and the

day care.

This application allows for the possibility of registration of clinical data (including

history of an inpatient encounter, summary report of emergency, referral letters, list of

diagnosis and procedures).

SONHO is seriously compromised due to its outdated infrastructure, because it de-

pends on currently discontinued Oracle Database and Oracle Forms versions. Many

hospitals using SONHO are seeking alternatives to it, and thereby facing migration

problems.

1.2 The OpenEHR Initiative

OpenEHR is about creating high-quality, re-usable clinical models of content and

process – known as archetypes – along with formal interfaces to terminology [5]. The

innovation of two-level modelling of Electronic Health Records (EHRs) and arche-

types [8] pioneered by openEHR [9] and standardized by CEN/ISO [3] brings us one

step further towards semantic interoperability of EHRs [10]. Instead of being hard-

coded into proprietary software by software developers, clinical content models are

expressed in the Archetype Definition Language (ADL) [11] and authored by the

clinical professionals themselves. Archetypes are used at runtime by EHR systems to

validate user data entry and query fine-grained data in the EHR. Archetype-based

EHR systems are highly adaptive and can evolve when clinical requirements change

over time since volatile clinical requirements are captured in archetypes while soft-

ware systems are built using only the stable openEHR information model and arche-

type language. Archetypes are expressed in a standardized formal language so they

are machine- interpretable and can be shared between systems. This makes the seman-

tics of EHR data available not only to other EHR systems but also to surrounding

systems [12].

An openEHR archetype is “a computable expression of a domain content model in the

form of structured constraint statements, based on some reference model”. Archetypes

are seen as a means of defining clinical knowledge in an explicit way, separating it

out from the system software that uses it. This has dual benefits of enhancing clinical

ownership and making system development and maintenance easier. An openEHR

template is “a directly, locally usable definition which composes archetypes into a

larger structure logically corresponding to a screen form”. Templates have an impor-

tant role in grouping and refining archetypes for specific local applications [13].

1.3 Aim

The aim of this work is to study how to create an application interface to SONHO

implemented according to the openEHR standard.

2 Methods

The first phase of this work was to extract the demographic concepts existing on

SONHO. The second phase was to select the proper openEHR structure holding the

patient demographic information. The third phase was to match both concepts aiming

to identify omissions in openEHR archetypes.

2.1 Existing Concepts in SONHO

The demographic concepts present in SONHO were retrieved by analyzing the user

interface. Unfortunately, there is no database documentation made available for us. A

SONHO expert user was used to confirm the list of concepts retrieved. The patient

demographic data of SONHO is mainly presented in three forms (Figure 1 to 3). A

description of the concepts can be found in the results section.

2.2 Existing Concepts in OpenEHR

The search of the proper openEHR structures holding the person demographic infor-

mation lead to the selection of the person demographics archetype (openehr-

demographic-PERSON.person.v1) created by Thomas Beale in 2003 [14]. It is still a

draft aimed to be used in a demographic service. Its purpose is to be a general model

of a PARTY with legal name, other identities, and contact details. The existing infor-

mation concepts were extracted considering the term definitions on the ontology sec-

tion of the archetype (e.g. legal name, home address).

Fig. 1. First demographic information form from SONHO.

Fig. 2. Second demographic information form from SONHO.

2.3 Matching Concepts from SONHO and OpenEHR

The process of matching of concepts was accomplished by searching for the nearest

concept in meaning in both SONHO and openEHR. When the concepts of SONHO

were non-existent not similar enough to the ones on openEHR, new concepts are

proposed.

Fig. 3. Third demographic information form from SONHO.

3 Results

3.1 Concepts Found in SONHO

The concepts found (N

SONHO

=41) in the three SONHO demographic forms are pre-

sented in Table 1. The terms in italic are not considered as concepts per-se and work

as group labels.

3.2 Concepts Found in openEHR Person Demographic Archetype

The concepts founding the openEHR person demographic archetype (N

openEHR

=12) are

here presented organised in a tree to improve readability. The term identification

between parentheses is the original identification found in the person demographics

archetype.

Person demographics

Identities

Legal name (at0001)

Other name (at0002)

Contacts

Home address (at0003)

Postal address (at0004)

Home contact (at0005)

Phone number (at0006)

Table 1. Concepts existing in SONHO demographics forms (N

SONHO

=41). The concepts are

divided by the first, second and third forms.

Concept Form label in Portuguese

First form

Hospital patient identification number PROCESSO Nº

Date of last update Última actualização

Date of record creation Registado em

National patient number Nº Utente do S.N.S

Old hospital patient identification number Nº Antigo Processo

Complete name Nome

Gender code and description Sexo (Feminino,Masculino, Hibrido)

Date of birth Data Nascimento

Age in years Idade … Anos

Nationality code and description Nacionalidade

Country País Or. (Portugal, França, Espanha, etc...)

Identification card - e.g. Passport of National

Identification card

Doc. Identificação (Bilhete Identidade,

Passaporte, Cedula, Numero Fiscal)

Number of identification card Nº Documento

Place of birth Naturalidade, Distrito, Concelho and Freguesia

(Porto, Matosinhos, São Mamede Infesta)

Observations Observações

Second form

Address Morada

Street Rua

Local Localidade

Phone number Telefone

District Districto (Porto, Lisboa, Aveiro, etc…)

County Concelho (Porto, Matosinhos, Maia, etc…)

Parish Freguesia (Aldoar,Bonfim,Campanhã, etc…)

Postal code no label (4464-159 São Mamede Infesta, etc…)

Contacts Contactos

Type of contact Tipo Contacto (Pai, Mãe, etc…)

Name for contact Nome Contacto

Third form

Parents Filiação

Father name Pai

Mother name Mãe

Reimbursement systems Subsistemas

Code Código (910005, 935601,etc…)

Description Designação (A.D.S.E. - ASSIST. DOENÇA

SERVID., SERVICO NACIONAL DE SAUDE,

etc…)

Number of Reinbursment System Nº Beneficiário

Valid until Válido até

Other data Outros Dados

Primary care code and description Centro Saúde

General practitioner name Médico Família

Civil state code and description Est. Civil (Solteiro, Divorciado, Casado, Viuvo,

Outro)

Professional status code and description Sit. Face Prof.

Migrant Migrante

Profession code and description Profissão (Apicultor, Jardineiro, Engenheiro

Textil, etc…)

Habilitations Habilitações (MENOS DE 4 ANOS

ESCOLARIDADE, BACHARELATO, etc…)

Chronic patient Doente Crónico

Exemption of payment Isenção

Special medication Medicação Especial

Fax number (at0007)

Email address (at0008)

Work contact (at0009)

Details

Date of birth (at0011)

Place of birth (at0012)

Sex (at0013)

3.3 Concept Matching

Table 2 presents the matching of concepts between SONHO and openEHR person

demographics. From the 41 different SONHO concepts, 14 concepts were mapped to

an openEHR concept and 27 are missing. From the 12 concepts found in the

openEHR person demographics 3 are missing in SONHO.

4 Discussion

The authors feel that the methods used to collect the SONHO concepts, namely by

using the user forms, allowed a correct description of the concepts. Also, some of the

concepts found on SONHO’s demographic forms are not really demographics (e.g.

“Is this a chronic patient”, “Special medication”). Nevertheless, these concepts were

all considered, as they should be part of the data interface to SONHO.

The openEHR person demographics archetype is very simple regarding its concepts.

This is probably the main reason that leads to other efforts to create new demograph-

ics archetypes which are much more complete (e.g. demographics archetype used in

Brazilian project [15], and used by children cancer treatment system project [16]).

From the process of matching the concepts, the authors claim that many important

SONHO concepts are still missing in the openEHR person demographics archetype

(e.g. patient identification numbers). For a useful interface to be built, the used arche-

type must include the missing concepts, which leads to the need of creating new

demographics archetypes and specialized local archetypes.

Despite SONHO and openEHR present different approaches, it seems feasible to

create an interface that meets the standard for openEHR and the demographics data of

SONHO.

5 Future Work

Further work is required to explore the possibility to auto-generate the application

interface based on openEHR templates composed with publicly available archetypes

and the locally tailored archetypes.

Currently the authors are planning the implementation of the interface to SONHO,

and also a change of the considered concepts aiming to extend from patient demo-

graphics to patient summary following one of the key actions items of the European

Community in the field of eHealth [17].

Table 2. Concept matching between SONHO and openEHR person demographics concepts.

SONHO concepts

Existing openEHR concepts

Description Term

Hospital patient identification number

Missing

Date of last update

Missing

Date of record creation

Missing

National patient number

Missing

Old hospital patient identification number

Missing

Complete name Name of person as legally recog-

nized by the state

at0001

Missing

Other name at0002

Gender code and description

Gender of person at0013

Date of birth

Date of birth of person at0011

Age in years

Missing

Nationality code and description

Missing

Country of birth

Missing

Identification card - e.g. Passport of National

Identification card

Missing

Number of identification card

Missing

Place of birth

Place of birth of person at0012

Notes

Missing

Address

Street Person's usual home address at0003

Local

Person's usual home address at0003

Phone number

Person's phone number at0006

Missing

Person's fax number at0007

Missing

Person's email address at0008

District

Person's usual home address at0003

County

Person's usual home address at0003

Parish

Person's usual home address at0003

Postal code

person's postal address at0004

Contacts

Type of contact

Person's home contact details at0005

Name for contact

Person's home contact details at0005

Person's work contact details at0009

Parents

Father name

Missing

Mother name

Missing

Reimbursement systems

Code

Missing

Description

Missing

Number

Missing

Valid until

Missing

Other data

Primary care code and description

Missing

General practitioner name

Missing

Civil state code and description

Missing

Professional status code and description

Missing

Migrant

Missing

Profession code and description

Missing

Academic graduation

Missing

Chronic patient

Missing

Exemption of payment

Missing

Special medication

Missing

References

1. Cruz-Correia, R.J., et al., Reviewing the integration of patient data: how systems are evolv-

ing in practice to meet patient needs. BMC Medical Informatics and Decision Making,

2007. 7(1): p. 14.

2. Hoya, D., et al., Collaborative development of clinical templates as a national resource

International Journal of Medical Informatics, 2008. Epub ahead of print.

3. CEN/TC 251, EN 13606:2007, Electronic Health Record Communication. 2007, CEN -

European Committee for Standardization: http://www.centc251.org/.

4. Health Level Seven Inc. What is HL7. [Web site] 2005 [cited 2005 Dec]; Available from:

http://www.hl7.org.

5. openEHR. Introducing openEHR. 2004 [cited.

6. openEHR, in Wikipedia. 2008.

7. Teixeira, A. and A. Brochado, Quando o S.O.N.H.O. se torna realidade ... Avaliação

estatística do impacto das Tecnologias de Informação nos serviços de consulta externa

hospitalar. Revista Portuguesa de Saúde Pública, 2005(1).

8. Beale, T. and S. Heard, Archetype Principles in The openEHR Foundation Release 0.95, T.

Beale and S. Heard, Editors. 2005, The openEHR foundation.

9. Kalra, D. and T. Beale, The openEHR Foundation. Studies in Health Technology and In-

formatics, 2005. 115: p. 153-173.

10. Garde, S., et al., Towards Semantic Interoperability for Electronic Health Records: Domain

Knowledge Governance for openEHR Archetypes. Methods of Information in Medicine,

2007. 46(3): p. 332-343.

11. Beale, T. and S. Heard, Archetype Object Model. 2006, The openEHR Foundation.

12. Rong, C., et al., An archetype-based testing framework. Studies in Health Technology and

Informatics, 2008. 136: p. 401-406.

13. Beale, T. and S. Heard, Archetype Definitions and Principles, revision 0.6, The openEHR

Foundation: London.

14. Beale, T., openehr-demographic-PERSON.person.v1. 2003, Ocean Informatics:

http://www.openehr.org/svn/knowledge/archetypes/dev/adl/openehr/demographic/openehr-

demographic-PERSON.person.v1.adl.

15. Freire, S.M. and R. Dutra, Representação dos dados de uma pessoa - br-demographic-

PERSON.pessoa.v1. 2007, Universidade do Estado do Rio de Janeiro - UERJ.

16. Chen, R., openEHR-demographic-PERSON.child_patient.v1. 2004, Acode HB.

17. Committee of the European Communities, An action plan for a European e-Health. 2004.

p. 356