Main Guidelines for a Standardized Electronic Health Record

lovethreewayAI and Robotics

Oct 20, 2013 (3 years and 11 months ago)

83 views

1


Main Guidelines for a
St
andardized Electronic
Health

Record


Mário Macedo

Rua Principal, 10
-
C, Peralva, 2305
-
516 Paialvo, Portugal
,
mario.macedo@mail.telepac.pt

Pedro Isaías

Universidade Aberta

Rua Fernão Lopes, 9, 1
-
Esq., 1000
-
132 Lisbon, Portugal
,


pisaias@univ
-
ab.pt



Abstract
.
The H
ealthcare O
rganizations use information systems with
several different types of data and user interfaces. The lack of
standardization means loss of efficiency and effectiveness. It

limit
s
the
expected quality of healthcare

services
.
S
ome difficulties for this
standardization

are known
. However there are models that can respond to
the complexity of this area of science and evolve with the development
of knowledge.

This paper

present
s the problem of this issue and suggests some methods
that lead to a clinical standard and guidelines in line with existing
standards, but flexible and evolving in accordance with the development
of science and
specificities of each service

of healthcare. Also, the
empowerment of systems to support clinical decision and the use of
workflows for treatment plans may involve even more of an organization
of healthcare will only be possible if they are used standard models and
open

technologies
.


Keywords
:
E
H
R, Medical Guidelines, Healthcare Plan Workflow



1.

Introduction

Information Systems in Health
c
are

knowledge allows, among other things, to
increase the wealth of societies and improve the quality of life of their
members.

According to
Downing

[
1
]
, Healthcare must
be safe, effective, focusing on
patient, just in time, efficient and equitable.

Pat
terson

[
2
]


state that
we move towards

a network of kno
wledge in society.
I
nformation and Communication
Technologies

use
in the areas of Health has

2


been slow due to two types of
factors, some conservatism and the great
complexity of healthcare associated with a constant evolution. It is common to
hear doctors

note,

for example
,

that their profession is a mix of science and art.

Actually the number of
variables to c
onsider the provision of health
care is very
large and difficult to relate. For this reason the automation of workflows, the
definition of data models and devel
opment of knowledge databases are

complex
and difficult to model. Another reason is

that until now it has not been possible
to develop systems entirely appropriate to the requirements of this area.

For example,
it is

clear that there is not a unique

interface to record data

from
an emergency service. Will

a doctor in an emergency
situati
on
use a

keyboard,
a mouse, or even a touch screen? The answer

is negative and the int
erpretation
of voice systems is not

reliable

enough

yet
. Will
it
be
reasonable to ask a
doctor
,

for example,
to get

training in each of the interfaces of
the
various
applications that exist in the market for the same functions? The answer is
negative. Using a computer application should be independent of the prod
uct
and

as intuitive to the user as

driving a car.

The NHS (National Healthcare System), entity responsible
for the English
National Health Service has developed a Framework
with
objects to normalize
with all the interfaces of al
l computer applications. This Framework was made

available to all
the software producers
. This is undoubtedly the way for the
quality,
effectiveness and efficiency of

Information Systems in Healthc
are
.

The main advantage of Information Systems concerning digitized clinical data
is the fact that Health professionals are not interested in looking at data but in
withdrawing knowledge
of

it
.
Many available

systems
on the market just

transform the previous records on paper into electronic records.

The information technologies and the creation of knowledge from them are the
engine of modern society.

In what Health is concerned, information syst
ems can improve the quality of
health services in the following dimensions:



Improving existing resources management;



Supporting diagnosis and care through more information made
available in databases;



Increasing accessibility to clinical data;



Automation

of care processes;



Increasing communication with the user/patient.

Improving existing resources management
:

Improving
existing resources
management will allow to

better plan the needs in
terms of capacity, use and supply. The information systems can also create
3


knowledge based on historical data and support
future planning avoiding flaws
in

answering
requests
.

Supporting diagnosis and care through more information

made available
in databases
:

One of the great difficulties of the doctor is accessing, assimilating and relating
the great amount of information that is produced daily by advances in medicine.

A
ccording to Patterson

[
2
],

the fragmentation and simultaneous explosion of

knowledge in health
care systems are not compatible with traditional care
methods.

With

the advent of the information society
,

it
is po
ssible to develop
systems that search, in
a

permanent and automatic

way
,
available databases

on
the internet, (cal
led
crawlers),

generating local knowledge through the use of
datamining.

These systems are foundations of knowledge and may be associated with neural
networks and they can integrate expert systems to support the dec
ision.

According to Bergman

[
3
]
, the information available on the Web and not
accessed by ordinary search engines is between four hundred to five hundred
times higher. There are also about one hundred thousand sites in the area of

the
"Deep Web" that are not searched.

Increasing accessibility to clinical data
:

The information technologies allow the integration and digital processing of all
the data that were traditionally produced in other media.

In order to integrate
and transmit clin
ical data, several protocols were developed, such as HL7,
version 3 that is the most widely used today.

One of the advantages of electronic data is the standardization of clinical terms
in the field of clinical diagnosis, equipments and procedures. This
standardization also decreases the var
iance of services among Healthc
are

Organizations
. Patterson

[
2
]
stated that it is common in hospitals, for example,
the expression
heart attack

to be used simultaneously with
myocardial
infarction
.

Another advantage o
f the electronic clinical data is the automation of the
communication of infectious diseases cases to public health agencies that study
epidemiology.

Increased access to clinical data translates into an increase in the provided
services efficiency and it
influences a crucial element of treatment, time.

Automation of care processes
:

Care processes automation allows us to optimize the management of resources,
reduce human error and improve the overall quality of services.

Due to the capa
city of information
systems, workflows

check some data
4


automatically and be able to generate alerts when certain events exceed
standards or rules,
it
is
possible
to increase the efficiency of Health
care.

Increasing communication with the user/patient
:

Nowadays
t
he portals for users support are an important communication tool.

These portals also provide advice and information.

They are also an important
means of communication betw
een the citizen and the Healthc
are technician
enabling the prevention of health prob
lems and patient personal care. We are in
the field of CRM, (Custom Relationship Management), namely the provision of
adequate services to the needs of each citizen.

With the concept of Web
2.0 and collaborative work,

the concept of Health 2.0

arose.

The
Web 2.0 provides dynamic
, social, collaborative

platforms
, with different
integrated services without the use of static pages. We are in the field of Rich
Internet Applications
,

that is
,

the applications
in which

the user is both
consumer and producer of content. The keyword is interactivity with the
environment.

The concept of Health 2.0 appears associated wit
h Web 2.0 but it is focused on
H
ealthcare.

According to Shreeve
[
4
]
, Health
2.0 is a new concept in Heal
thc
are in which all
stakeholders, (patients, doctors, suppliers and donors),

collaboratively
contribute to the services provided and their costs using the means available to
improve safety, efficiency and care quality.

Al
so according to this author,

Health

2.0 implies the observation of the
principles of sta
ndardization and connectivity in the

cooperation,

exchange of
information and knowledge transfer

platforms
. The main vector of this concept
is the value added services and allowing the free choice and as
sessment of the
value delivered to the user.

Within the H
ealthcare

O
rganizations there

are, however, great difficulties of
integrating systems. Due to technological advances and the advent of
information systems integrated with medical equipment and developed by each
manufacturer, there is a state of disintegration and redundancy of data.
H
owever, this integration is possible provided that there are interfaces and
protocols appropriate in each computer application.
Some of the most important

advances in this area are
objects
oriented languages and the protocols of
standard messages for commu
nication between

systems in the field of
Healthc
are
.

The protocols of standard messages still have the characteristic of
be
ing flexible in their

structure and include different types of clinical data that
ca
n even be images of medical examinations
.

In conc
lusion, according
to
Pa
t
t
erson
[
2
]
, the new paradigm of Health
c
are is
:



Transition of hierarchical systems to network systems
;

5




Transition of Systems guided by functions to systems orientated by

proceedings;



Systems with rigid structures of medical

records
change in
to flexible
records, focused on tasks.

The needs of permanent increase of knowledge on the clinical data enhanced
the use
of Data Mining tools in Healthc
are.

According
to
Kraft et al.
[
5
]
, the
Deci
sion Support Systems in Healthc
are support
ed by Neural networks have
been us
ed in several cases, such as in estimating

needs for blood for
transfusions, myoc
ardial strokes, needs to
supply of medicines and prediction of
risk of coronary heart disease.

The neural networks are used to solve complex
problems that can not be described by algorithms. For the classification of data
,

there are also used Baysean models
,
associations

in clusters
and
fuzy

networks
.

2.

The c
oncept
of
Electronic Health Record (EHR)

According to the
HIMSS

[8]

(
Healthcare
Information and Management Systems
Society
)
, an integrated information system e
nables inter
-
operability among

different information systems and allows
that

Health S
ervices operate with
more efficiency and
at
lower costs.

The advantages, according to that o
rganization, are summed up to the following
factors:



R
educing time to care;



Improving the qu
ality of medical records
;



Increasing the
knowledge for the provision of Health
care;



Reducing the costs and errors of administrative work when entering
data.

Another dimension that enables the integration of systems is the
EBM
(
Evidence Based Medicine
)
, that is, a system of alerts based on a comparison of
data from the information system with established standards. The EBM systems
can also be applied to diagram
s of activities called workflows and lead to
warnings whenever deviations are found.

An in
tegrated information system on H
ealth
care

Services
must meet the
following requirements:



Accessing and linking

medical records from different

sources

and
Healthcare Organizations;



Ensuring privacy and data security;



Having

an open and scalable technology platform;



Being
remotely accessible
;



Having heuristic
characteristics of ac
cessibility to the clinical staff
;



Having
research

features
, alerts and r
eminders;

6




Having

components of financial management or interfaces for a
financial system;



Using a
language of the field of Health
care S
ervices;



S
upport
ing

the archive of images
and o
ther multimedia documents
integrated

with other clinical data;

As existing

clinical systems are being developed in different technological
platforms, there are different ways of integrating them. To facilitate this
integration, it was necessary to develop a standard format of the messages and
communication protocols.

The main advantages of each organization concerning
the processing of
H
ealthcare data were the accessibility, communication, flexibility and
adaptation to the requirements of each service. These objectives are found in the
literature associated with the co
ncept of EMR (Electronic Medical Record).

The concept EHR (Electronic Health Record) is wider and is associated with the
treatment and reporting of clini
cal data by entities providing Health
care.
According to Barret
t
o
[
9
]
, there are several definitions for

EHR and various
terms are used as synonyms.

According to Barretto

[
9
]

EHR is a set of clinical data collected electronically
for each subject and
produced by entities providing Health
care. They are likely
to be accessed centrally or distributed through a
network and meet
characteristics of continuity, efficiency and quality.

Also, the
HIMSS

[8]

Health Information Management Systems Soci
ety defines
EHR as a record of h
ealth data for each patient generated by one or more
clinical episodes.

They are included
in this definition data concerning demographics, data on
progress incidents, medication, vital signs, medical history, details of immunity,
laboratory data and radiology.

The concept of EHR also includes the latest standardization of events,
workflow, mult
imedia and
genetic data. For this reason,
E
H
R messages not
only carry static data, but they themselves can also be integrated in dynamic
environments, in transactional environments and with synchronicity between
events. The concept of EHR is a new type of
message that went from a static set
of data to a time dependency.

The structure of messages EHR also allows the communication of protocols of
medicines that inform the doctor of the active substances recommended dosages
of each drug, how to take it and the

expected response time. Known side effects
of each drug can also be included in these protocols.

To ensure the integration of all kinds of messages, some formal models have
been developed known as arc
hetypes that store metadata of
E
H
R messages. The
metada
ta syntax sets the whole structure of the message which contains the
7


elements of sequences of message headers, message elements, message
hierarchies and data values.

An EHR system requires data integration not only within the organization but
also among or
ganizations. An EHR system also implies the existence of a single
reference for each patient with access to all data of that patient. Finally, an EHR
system implies technologies to build knowledge and decision support in
individual levels and in epidemiolo
gical studies.

Many medical equipments have interfaces to integrate data. In most cases they
refer to a standard, known as HL7

[
22
]
. However, there are on the market two:
the 2.X and 3.

Only in version 3 can be defined archetypes as described above. It is
quite often
made reference to models of integration that use portals among applications or
integrate charts.

To this effect was created a consortium among industrial, academic institutions
and other organizations to establish EHR models

among some EU count
ries
such as the United Kingdom, Britain, Denmark, Norway, Ireland, Sweden,
Holland, France, Switzerland and Italy. This project had some developments
and resulted in an organization called OpenEHR

[
10
]
.

The OpenEHR was created in order to boost the develo
pment of opensource
applications using standard messages. OpenEHR is a nonprofit institution that
seeks to develop its activities with the aim of creating specifications, open
source software and other sources of knowledge that enable the implementation
of

EHR projects.

The architecture of the OpenEHR
[10]


information system is in line with the
ISO 18308 standard in terms of the data structure, processes, communications,
privacy and security, medical law, ethics, culture and development. The
integration of
clinical data and its transmission among different organizations
also need to use standard clinical terminology. For this purpose it was
developed another standard, the
SNOMED
-
CT
[11
]


(
Systema
tized
Nomenclature of Medicine)
.

A variety of standards of
terminology, their encodings and their relationships
among them led the

NLM
[12
]

(
National Libray of Medicine
) in USA
, to
establish a standard, which could integrate all the existing, known as the
Unified Medical Language System. The purpose of this
standard is to facilitate
the development of software capable of communicating with each other and
store knowledge in the areas of biomedicine and health.

The symbiosis of technology and standards of diagnosis codes, procedures and
equipments made it
possible to develop a federation of clinical data, the FHR
(Federation Health Record).

8


Also increasing knowledge translates into all levels of organizations in a better
performance of their professionals with less effort and resources consumption.

An EHR
system should not only be aligned with all the above but also be
rolling within rules of coherence and harmony with these standards.

The main vector of this principle is the co
ncept of archtype. The archtypes are
metadata containing definitions of the variables EHR, templates and archives of
terminology that form the architecture of an EHR system.

An EHR system should thus provide an interface oriented services for
integration with other software used in Health
c
a
re

Organization

for access to
clinical data, demographic and the central register of patients.

For security and privacy of data
there is a

central register containing the
identification of each patient.

An EHR system should also include an editor of arch
types enabling the
doctors
to de
fine new clinical guidelines and in accordance with existing standards. The
archtypes objects define the field and their clinical attributes, (Eg: laboratory
result). It is possible to
join different
archtypes
and obtain gre
ater complexity or
reuse a generic archtype to introduce
new
attributes.

These

archtypes can
be
common to
different Healthcare care Services and Organizations.



The generation of templates is done with the archtypes and should be made
using the tools with graphical interface without programming. Thus it is
possible there are also different versions of the same archtype
and templates
. So
there should be a system
to control
versions

of archtypes
and
their validation
.

The database of medical terminology should be founded on a standard
terminology
and
can be expanded with
other syntax
created by the
local
Healthcare Organization
. Similarly as in archtypes

and

templ
ates
we must
control the

terminology
version
s
.


3.

Conclusion

Th
e standardization of clinical data electronic records and the definition of
procedures for the structures of metadata and medical terminology evolv
ed
across the different Healthc
ar
e Services
, countries and cultures
and they will be
th
e possible answer to improving Health
care quality, effectiveness and
efficiency.

Also, the standardization of medical terminology and their semantics
are fundamental to the success of information systems in healt
h.

Finally, the independence of information systems of technological
obsolescence, market trends of I
nformation
C
omputing Technologies

and
suppliers will be enhanced with

the standardization of clinical data.

The
knowledge development in this area is also
boosted with the use of
statistical analysis tools, datamining and business intelligence whose
feasibility of use also depends on standardization.

9


References

1.

Downing, Gregory,
http://www.personalizedmedicinecoalition.org/PMC_Response_to_HHS_RFI_on_
HIT_final_1_25_071.pdf

2.

Patterson, Neal L.
: The Mission of IT in Health Care: Creating a System That Cares
in Healthcare Information System Management, Sp
ringer, 3

21 (2004)

3.

Bergman, Michael K. : The Deep Web: Surfacing Hidden Value,
http://www.brightplanet.com/pdf/deepwebwhitepaper.pdf

4.

Shreeve, Scott:
The Enabling Technologies and Reform Initiatives for Next
Generation Health Care,
http://health20.org/wiki/Shreeve_Health_2.0_Whitepaper

5.

Kraft, Margaret R. et. al:
Data Mining in Health
care Information Systems: Case
Study of a Veterans’ Administration Spinal Cord Injury Population in
Proceedings
of the 36th Hawaii International Conference on System Sciences
,
http://csdl2.computer.org/comp/proceedings/hicss/2003/1874/06/187460159a.pdf

6.


Castells, Manuel: A Sociedade em Rede, Fundação Calouste Gulbenkian, Lisboa,
(2005).

7.

Ro
cha, Cristianne Maria Famer:As redes em saúde: entre limites e possibilidades in
VI Congresso Nacional da Rede Unida,
www.opas.org.br/rh/publicacoes/textos_apoio/redes_cap3.p
df

8.

HIMSS,
http://www.himss.org

9.

Barret
t
o, Sistine Ann: Designing Guideline
-
based Workflow
-
integrated Electronic
Health Records. University of South Australia (2005)
.

10.

open
EHR Architecture,
http://svn.openehr.org/specification/BRANCHES/Release
-
1.0.1
-
candidate/publishing/architecture/overview.pdf

11.

SNOMED
-
CT,
http://www.snomed.org/

12.

NLM,
http://www.nlm.nih.gov

13.

A
n EHR User
-
Satisfaction Survey:Advice From 408 Family Physicians
,
2005
:

American Academy of Family Physicians,


h
ttp://www.aafp.org/fpm/20051000/ehrsurveyappendix.pdf

14.

Castells, Manuel & Ince, Martin
:
Conversations with Martin Ice, Blackwell
Publishing Ltd, Oxford
,(2004)
.

15.

'harmonized' European Standards :
CEN



http://www.cen.eu/cenorm/businessdomains/businessdomains/healthcare/index.asp

16.

Fragata, José & Martins, Luís
:

O Erro em Medicina, Almedina, Coimbra
, (2006)
.

17.

Fragata,

José et al.
:

Risco Clínico, Almedina, Coimbra
,(2006)
.

18.

Gardner, Reed M.
:
Computarized Clinical Decision
-
Support in Respiratory Case
,
Respiratory
C
are

http://www.rcjournal.com/contents/04.04/04.04.0378.pdf#search='reed%20gardner
%20respiratory%20care

19.

GLIF
-

Guideline Interchange Format ,
http://www.glif.org

20.

Hurlen, Petter et al. , “Synapses Project”,


https://www.cs.tcd.ie/synapses/public/deliverables/part4.pdf

21.


Electronic Health Records Workgroup,

10



http://www.hhs.gov
/healthit/ahic/healthrecords

22.

HL7,

http://www.hl7.org

23.

ICD (International
Classification of Diseases),
http://www.who.int/whosis/icd10/

24.

Jablonski, Toward et al
.

:
Toward a Flexible, Process
-
Oriented IT Architecture for
an
Integrated Healthcare Network
,
http://www.acm.org

25.

Kalra, Dipak et al.
:
Software Components developed by UCL for the SynEx Project
& London Demonstrator Site
,

http://www.chime.ucl.ac.uk/HealthI/SynEx/

26.

Kalra, Dipak
:
Clinical Foundations and Information Architecture for the
Implementation of a Federated Health Record Service, U
CL
, United
Kingdom
,(2002)
.

27.

Karla , Dipak et. al.
:

C
lient Interfaces to Synapses, Record

Interfaces Using Corba,
http://www.omg.org/docs/corbamed/98
-
04
-
05.pdf

28.

OASIS
,
http://www.oasis
-
open.org

29.

OMG Health Domain Taskforce
,
http://www.omg.org/

30.

Pember, Marvin et al.
:

Clarion Health: Clinical Excellence Through Quality and
Patient Safety in Healthcare Information System Management, Springer, 53

78

(2004)
.

31.

WHO (World H
ealth Organization),


http://www.searo.who.int/EN/Section898/Section1441.htm

32.


W3C,
World Wide Web Consortium
,
http://www.w3c.org