HIMSS Europe CIO Summit

farmpaintlickInternet and Web Development

Oct 21, 2013 (3 years and 9 months ago)

71 views

HIMSS Europe CIO Summit

Presented by

HIMSS Analytics Europe

Geneva, Switzerland

20


22 November

Xavier Pastor

CMIO. Hospital Clinic.

University of Barcelona

Barcelona. Catalonia. Spain


Semantic Search of EHRs

Creation of Data Warehouses and
Clinical Registries

Topics to be covered


What we have achieved in EHRs


A second wave of resistance


Present models of IS and new challenges in
healthcare


Proposals for the next future


Implications


Conclusions

What we have

achieved in EHRs


(or we are in the way…)

Registration of the main business
processes in healthcare facilities

Discharge

Patient admission

Antecedents

Actual Disease

Physical exam

Clinical note

Clinical documentation

Service

Request

Appoint
ment

Registry

Results report

Surgery
request

Appoint.

IQ registry

Surgical report

Waiting list

Prescription

Validation

Delivery

Administration

Medical orders

Activity
registry

Planning

Reports

1.
-

Clinical path

2.
-

Complementary examinations path

3.
-

Surgical path

4.
-

Medication

5.
-

Nursing care


Departmental Systems (DIS)


Best solution for extended business
processes


Avoiding corporate processes and
data competition


Strong relationship with technology


Specific reporting


Maximum integration



Clinical Imaging (PACS)


Imaging Database


Standard Format


Full integration with core
business system


Repository for:


Image distribution
through corporate
system


Image sharing with
modalities or
departmental
applications

ERP
-
EPR

DIS
n

RIS

IM

Radiological

modality

(Rx, TC, etc.)

DIS
4

DIS
3

DIS
2

DIS
1


Integration Manager (IM):


HIS to DIS:



Request


(patient, episode,

service, time, health
problem) [Planning]


DIS to HIS:


Result

Activity, Documentation
(report, data, Image, time)


Master Tables


Elements:


Organization


Structure


Services


Business partners


Agendas


Documents


Forms


Patients


Reports


Users


Roles


Categories


Processes


Master data creation and maintenance


Users creation and maintenance


representation of Main business
processes


Permissions


Service requests


Planning


Activity registration


Documentation


Exploitation


ERP
-
EPR
:


Data warehouse:


Extraction


Debugging


Transformation


Summarization


Aggregation


Analysis


Balance Scoreboard

Clinical

Workstation

Management

Administrative

Workstation

Diagnostic or

therapeutic

units

HIMSS EMR adoption model

HCO in Spain

1984: HIS
-

textual interface without structured information


Financial S.

Hospital Clínic: EMR evolution

1995: EPR Clinician
workstation.
Graphical interface
with structured
information

1997: ERP. SAP
R3

2003: ERP + EPR: SAP Health
solution. Graphical interface with
structured and workflow information

2007: Full clinical record

2010: eMedication & CDS

Accomplishment of the Discharge Report
(Jan 1999
-

Dec 2001)

Physician’s adherence to the IS

>95%

EMR assessment on healthcare performance
from the healthcare providers

Time savings

Cost efficiency

Information Quality

Improvement of diagnostic
and therapeutic activities

Risk Management

Effectiveness of processes

Patient Empowerment

Quality of service

Quality systems

Continuity of care

Service
Value



Patient
Empowerment


Efficiency of
care


Quality
Improvement


Accountability

Information management

Clinical Audit

Strategic management

Organizational effectiveness

Efficiency



Effectiveness

Quality of
service

Clinical
Governance

.

Diffusion

Patient workflow

Nasi G, Pastor X, et al.

Bocconi U., Barcelona U.

(submitted for publication, 2011)

Study in 16 HC centers in Barcelona area. 220
questionnaires: 109 Physicians and 111 Nurses

A second wave of resistance








(smarter than first one…)

Some end
-
users comments


All patient data is inside and the
system is strong and powerful
but...



It takes me too much time to gather
data to be informed about patient’s
condition


To many mouse clicks


The system is a black hole. I cannot
extract any information from the
system


I loose the relationship with the
patient

But what about


Immediacy?


Clarity?


Comprehensibility?


Relevance?


Pertinence?

Paper
-
based records

Electronic records

To tell the truth in the majority of institutions
we have changed only the media

“Big” change

Reflexiones...

A formula to keep in mind

integration
simplifying

Added

value

+

=

Professional

implication

Some surprises.

The BI project at Hospital Clinic


Initial project: a pilot with real data
(2004)


Goal: Common and unique balance
scorecard for clinical management.


Teams:


Technical: IT Dep + External Provider
(EP)


Functional: IT + Organization + Financing
+ key users


Time delay


Planned end: dec. 2004.
Real end: feb. 2006


Cost increase


Initial: x


(EP) + y


(IT).
Final: 2,5*x


(EP) + 4,6*y


(IT)


Main cause:
Internal disagreement

about the meaning of registered data

Decision

Data

Information

Integration

Monitoring,
anamnesis,
clinical exam

Medical

Knowledge

Actions

Diagnostic
Hypothesis

Differential
diagnosis

Medical

Orders

Syndromic

diagnosis

time

The patients
and their
problems

The clinical method

transformation

combination

inference

registration

planning

execution

Where is the knowledge in our EMR?

time

15 years

Continuous changing

in knowledge. New
relationships among
data and information

Classical implementation of CDS in EMRs


Rules firing under special patient conditions


Knowledge organization:


Premise: a Boolean expression IF… THEN… AND
OR NOT


Conclusion: statements

Use essential programming languages in some
cases, the EPR systems analyze some (very
few) patients’ data to give decision support


Simplicity


Limitations in growth to keep consistency
(complexity, comorbidities, non
-
monotonic or diffuse
reasoning,….)

Present models of IS and new
challenges in healthcare







(and clear limitations

of actual architectures of ICT)

Trends in HC & ICTs


Consumer:


“Full integral healthcare”


Return from his/her taxes


Healthcare providers:


Trend towards fragmentation
& specialization


Collaboration in Clinical
Process (W2W)


Outcome measurements


Benchmarking


Quality evaluation and
improving


Financial sustainability of
the HC system


Ageing population


Chronic diseases


Financing “
per capita



P4 Medicine


Predictive


Preventive


Personalized


Participatory



Pervasive Computing
environment


Is the “consolidated model” sustainable?


Is this Information System model suitable to
encompass the technological tide?


How big is the effort to lesser the natural
information entropy?


Do the actual technology architecture
represent the real Clinical Business?


Is a Hospital self
-
sufficient to offer to their
customers a “Continuum of Care”?

Ten years ago:

Primary


Specialized Care


The mean time to establish a lung cancer diagnosis by the SC after the
initial consultation of patient to his/her GP is 50 days


Top
-
down measures from Health Authorities hadn’t be successful to
improve such situations

Houston, we have a problem !!!

Reengineering
relationship between
Primary and
Specialized Care

Outpatient

Specialized

Care center

Family physician

Specialists

Diagnosis / treatment resources

Patient /

citizen

Specialized

Center B

Primary Care

center

Specialized

Center A

Poor communication among HC professionals

Delay in diagnosis and treatment

Destination to Specialized Center by chance

Technical resources always at the Hospital

No update of clinical info from the hospitals

WE WANT THE PATIENT’S DISCHARGE REPORT !!!

Primary care

center

Specialized care

center

Family physician

Specialists

Diagnosis / treatment resources

Patient /

citizen

Patients’ flow regulated by agreements and clinical
protocols based on scientific evidence

Goal:
Improve the patient care with a new approach over the
relationship among family physicians and clinical specialists.

xml files

Clinical msg

HL7 2.5

PC

EMR

SC

EMR

AISBE project: lessons learned in 4 years

Achievements:


Many patient’s benefits because
new organization


Interoperability in a highly
heterogeneous environ
-
ment (9
different providers) thanks to the
technical support of ICTs.


Additional solution to new needs
(Teledermatology)


Better information about activity


Biggest added value:


The PC physicians get the Reports
from the Hospital immediately


The SC physicians get a clear and
proper request for their services

Problems unsolved:



Semantic Interoperability


Clinical process
management: Patient
workflow still is
represented by
administrative processes
(Centers, Services, etc..)

Example:

It’s no possible to compute
the accuracy in the
suspicion diagnosis of skin
disorders among the PC
physicians, loosing the
opportunity to make a
specific training program to
empower them

Proposals for the next future








(being very humble…)

Clinical process: a business

devoted to solve problems


Main reason of clinical
business: to solve or improve,
the health problems


Health problems must be the
main object related to the
patient to be managed by the
healthcare professionals who
must organize the activities
over the patient

ERP
-
EPR: Nice and full administrative view
with clinical documentation

Ep1 Vis1
(Outp. Gyn)

Ep4 (Hosp


prothesis

Ep3 (Emerg.

Femur F.)

Ep6 (Hosp


Surg.)

Pre
-

Registr.

Hernia

Ep2 (Outp.

Surg &

preop)

Ep5 (Outp.


Orthop.)

Primary Care

Hospital

cr: 1

Menopausal woman, osteoporosis and inguinal hernia

Ep1 Vis2
(Outp. Gyn)

Ep2 Vis2

(Outp.Surg)

Medical practice: Ideal view for
clinical processes

Act1

(Outp Gyn)
Primary C

Act3 (Hosp

Prothesis)

Hosp.

Act2 (Emerg

Femur F.)

Hosp.

Act3 (Hosp


y CIR)

Hosp.

Act1

Pre
-

Registr.

Hernia

Act2 (Outp.

Surg & preop)

Hosp.

Act4 (Outp


Orthop.)

Hosp.

hc: 1

Menopausal

woman

Act5 (Outp
Gyn)

Primary C.

Act4 (Outp.
Surg)

Hosp.

Clinical Episode: Osteoporosis

Clinical Episode: Inguinal Hernia


New EMR: three layers

time

relations

relations

Level 1
(actual)

Transactional history:

Episodes, encounters, visits,
services, forms, appointments,...

Level 3

Medical knowledge:

Protocols, clinical guidelines,
clinical assay, prototypical
clinical cases,

Level 2

EMR by patient problems
:

Clinical History

Problems, diagnoses, risky factors,
chronic conditions, documentation

1
st

Goal:

EPR

敐位O
=
2
nd

Goal:

ePOMR

+ Knowledge

Knowledge representation

Usage of standard structures to formalize biomedical knowledge

Biomedical Ontologies: they exist!


FMA
: Foundational Model of Anatomy


All Human Anatomy


Available to interoperate with applications


GO
: Gene Ontology. Related to:


Cell components


Biological processes


Molecular functionality


National Cancer Institute


Repositories


NCBO BioPortal

(National Center for Biomedical Ontology)


The OBO Foundry
: The Open Biological and Biomedical Ontologies


Semantic Web (3.0)

Our first approach.

OntoCRF
©
: a test of concept for research

OWL

DATABASE

Web Forms

OntoDDB
®

Next step: OntoCRF towards OntoCR


Building a Clinical Repository about Health
Problems of patients in our Regional HIS


Inclusion of knowledge about Health Problems
(until some extent)


Capability to managed the Health Problems by
physicians (PC and SC) in real time


Recoverable by any EPR in a standard format (EN
-
ISO
-
13606)

A new generation of EHR

Knowledge

Patient’s Clinical

Information Management

Patient management into

business enterprise

ERP1

(High tech Hosp.)

ERP2

(PC center)

ERP3

(Comm.

Hosp.)

ERP4

(Diagnostic

Offices)

Health problems

Clinical pathway

Problems server

Protocol server

Drug knowledge

Services and

procedures

Web Services

Clinical

repository

Medical terminology

Master clinical

Data (archetypes)

Patient history

Organizations

Customizable

Clinical portal

Social

networks

Professional

networks

Semantic
Search on EHRs
really available

Semantic
Interoperability

available

Expected benefits of Semantic Search


“Smart” collection of patient data before
clinical contact using complex queries


Relevance, pertinence, adequacy, appropriateness, …


Clever advise at point of care


Better registration of clinical data


Contextual user interface and interaction
capabilities


Knowledge discovery

Implications






(It means involvement…)

The role of SW engineers


Design SW structures capable to represent
explicitly the knowledge in an scalable and
semantically interoperable way


Design Interfaces suitable for the use of HC
professionals


Design new IS architectures that allow the
progressive transformation from
“consolidated” models to newer ones

The role of HC professionals


New professional positions to be covered related
with clinical data and documentation in EMRs


Collaborate to formalize Biomedical knowledge


Introduction of such new concepts and procedures in
the curriculum of Health Sciences


Learn to manage healthcare processes with non
-
human assistants

The role of HC managers


Promote and finish the transition from paper
PR toward a “suitable” EPR.


Explore and promote with other organizations
new ways to deliver Healthcare taking profit of
ICTs


Stimulate teams of HC professionals to work
around innovation


Play the role of governance in Information and
Communication Systems

The role of the patients


Indeed they will be very involved




It’s unknown how they are going to be
integrated as part of the clinical process

Conclusions


Healthcare Professionals need to get more value from the
Information and Communication Systems


It’s possible to represent biomedical knowledge and
knowledge about clinical data in the Information Systems and
interoperate with it.


Systems must be more than registration machines. They must
evolve to become true assistants in clinical business,
managing knowledge, information and data by themselves.


“Health problem” is the “key stone” for the next generation of
EHRs systems


In order to achieve practical results a big effort must to be
developed among Engineers, Biomedical Scientists and
Healthcare Professionals

Thank you very much for your attention!!!


xpastor@clinic.ub.es