1 of 32

Can the lessons learned from the Kanta services contribute to the development of openEHR?

Pirkko Kortekangas, chief Specialist UNA Oy

Note: Opinions during the presentation is only Pirkko to blame. Some slides are originally from Marko Jalonen who also has promised to help answering Your questions of Kanta

2 of 32

Agenda

Foreword �Kanta in the big picture�Kanta details that are interesting for the openEHR reference model�A couple of thoughts on the topic of access management in Kanta�How Kanta data can now be used in clinical work: Snapshot UNA

MD, specialist of surgery and neurosurgery

Clinical work 20 years, knowledge management 25 years

Involved in any development or execution concerning management of welfare and health data in Finland.

Earlier Chief Medical Information Officer at Turku University Hospital

UNA Oy, Chief Specialist 2018 -

Co-chair of Hl7 Finland subgroup openEHR Finlans

Pirkko Kortekangas

3 of 32

Why this presention – me or now or here?

I have participated in both Kanta and many of its predecessor like “core data” and regional system and value assessment of data projects.

In addition, I am familiar with the attempts to get Kanta to real use also for professionals.

Now

I try to open up that openEHR has a lot of interoperable elements with Kanta thinking.

I hope You are familiar with openEHR reference model as it may be hard to get my points.

Here

The focus of the Nordic Hachaton is templates – our doings might inspire You?

Kanta represents in a way a “solution of frozen templates”, but it has great value for planning the next generation solution to share data

On the other hand, I will demonstrate that useful basic functionality for the professionals may be built on Kanta-type information

4 of 32

The Finnish social and healthcare

Finland is a nation of 5.6 million inhabitants

21 wellbeing services counties are responsible for providing public social and healthcare services for citizens since the organisational reform 1.1.2024

general and specialized care and social services
rescue services

occupational healthcare is generally available to a large part of the population and is mostly private
also other organizations taking care of social and healthcare services, eg. vocational and higher education, defense forces and prisoners
numerous private social and healthcare providers for self-paying customers and outsourced public services (eg. vouchers, screening outsourcing)

The reform had been planned twice before. It was near to take place in 2018
After the reform we are in the middle of financial challenges. There has been and still is a big push for overhauling professionals’ systems and especially offering citizens’ digital services

Marko Jalonen

13.12.2024

5 of 32

Kanta services in Finland

National social and healthcare data sharing platform provided by the Social Insurance Institution of Finland (Kela) in a close co-operation with other national actors.

The social and healthcare providers are mandated by law to connect to Kanta services.

A massive set of parts and functionalities

6 of 32

Notions of the past and present

Kanta was implemented to healthcare systems in the first half of 2010 decade

It included a huge amount of adaptation and configurations
Only university hospitals had the same EPR product and a shared project to implement the new national demands to the system
Other hospitals and health care centers with their about 5 EHR products and their about 270 local installations had less co-operation leading to somewhat variable ways to interpret the requirements

The systems were and are mostly still from the first generation of systems with relational base

Laboratory, radiology and reservation systems were still mostly old school legacy systems

The first wawe of the Welfare Services County Reform in ICT matters has included in most Counties mainly harmonisation (”consolidation”) of systems

In healthcare it has meant to combine separate installations to one on a county after selecting the one system to be used in the future in the primary and/or secondary care
In social care with less comprehensive systems it has meant more extensive procurement processes and the implementations have just started

Most of the ””new”” systems are the same generation systems as before

7 of 32

Everyone in Finland is somehow involved in Kanta

List of stakeholders

The GDPR and the application of the EHDS will have a major impact on the development the national level

not to underestimate AI and Data Act

8 of 32

From the viewpoints of professionals and citizens

All* in with minor delays

All summaries out

without delay (if not postponed)

Documention on oontinuous medical record

Client documention in social care record

Daily entries in the ward

Social and Health Care

*some exceptions still in social care and some information groups

9 of 32

Key figures

10 of 32

Details of Kanta architecture

Listeners are encouraged to make conclusions of their own in whether the details pointed out may be solved better in the openEHR reference model

11 of 32

Kanta information model is based on documents – in both administrative and technical senses

Main title

Section with structured title

Stuctured information content

SERVICE EVENT DOCUMENT

SERVICE ENCOUNTER DOCUMENT

ENTRIES IN AN ENCOUNTER DOCUMENT

relates to

Form a document

Relevant metadata t

Out or ward encounter
Period
Responsible organisation

Relevant metadata

Timelabels
View
Provider
Responsible professional

Relevant metadata

Timelabels
Author incl. executive organisation

Administrative documents

Technical documents

12 of 32

with the exception of the open/ward classification of the Service Event

** Service Event is a ”service package due to a single decision making encounter” trying to express why i.e. some investigations or therapies were ordered

Concept applied iin EHR systems in a very heterogeneous way
Despite of the confusion the dreams to replace it with something else will be heavy because it is so deep in the structures

Yet, some usefull metadata is missing at least if the purpose is to understand the joint ability of the services to produce added value to the citizen

About encounters

☹

☺

The nature/type of an encounter can be found* only in the local “patient administration system modules”

There is no other rigid binding an encounter to other encounters except Servie Event**

a process may bee seen from many viewpoints and by any metadata e.g. client, diagnosis, provider

13 of 32

What happens as we are implementing the new document type ?

ordered

scheduled

started

ended

postponed

cancelled

all knwledge of a single encounter incl. the type of the encounter

Reservation document

Status of the encounter changes as the service realizes

The connection between an encounter and patient documentation is still based on time labels

Correct timelabels are based on the ability of a system to ”force” professionals to choose correct encounters to bind there notes to

☹ Slightly invalid concept by birth: Status ”planned” is excluded related to the management of Service Event – only the same register keeper may edit a document linked to a service event

14 of 32

Data is linked to an entry; entry to a document; a document to a service event, a service event to an organisation that also has a role of registerkeeper

Using data stored by somebody else

How to reuse Kanta data?

If you store data by copying it from an other source, set id points original origin

set id = oid of the original entry

Background to take into consideration

The professional is obliged to document the grounds for care decisions

Many people think that the easiest way to prove this is to store again other party's information that has been viewed – data viewing log entry is not considered to be evidence enough

The vision is that you would allways get the last version regardless of the recorder.

Becomes the fact after some years in the list of medicins
Versioning of contact information is partly taken care by National Population System.
Others like risk factors (alarm information) has no plan

Versionable data

the newest version is the one to use

Cumulative data

every documented note contributes to present and future understanding

15 of 32

Code sets

For example

Registers of register keepers, providers, working units
National labels (titels for sections)

Local labels are allowed but all must be mapped to a national label

Views

For specialities, services, certificate forms, data administrative

Method of getting the servive
Process stages
Laboratory tests
Radiology
Value sets for expressing a finding or alternatives of answer etc.

Common medical classifications as ICD-10, ICPC-2, procedures, Loinc

localized versions

Kodtjänsten

We have code sets for many purposes

Maintenance requires a lot of resources and transparent, functional processes

�Maintenance of language and concepts are part of the operation

16 of 32

Have code sets made life easier for professionals?

☺

☹

People tend to choose incorrectly

too many alternatives
unfamiliar / adminstrative alternatives
unclear definitions or cathegories

The need for new code values or sets may outperform the ability of a centralized manual maintenance process

Coded information hinders correct documentation

An underutilized clinical resource

One word/code can sometimes replace unnecessary chatter

Well adapted in strictly targeted usecases, like ortopedic implant surgery

Supports AI assisted storage of free text into structured data
Parallel encoding by different codesets increases the value of the source data

Unrealistic manual task
Local and national administrative needs are not dumped to professionals

AI could help maintain code sets based on real life data instead of imagination at a desk

strucutured documentation

17 of 32

Stuctured content definitions

Web sites of national content definitions (under development concerning roles)

Code server
Sosmeta
Termeta

Examples of contents on different level

Diagnosis

Laboratory result

Tooth condition�Childbirth�Eye health check

Medicin

Medication

The somewhat evident lack of standard terminology reflects in e.g. diffuse classification of Termeta contents:

Document
Form
Entry
Other structure
Component
Piece of a component
Group of component

Templates

Achitypes

Datapoints

18 of 32

If we called all Finnish document-level and other content definitions as templates and the minimum usefull grain size of information as architypes…

We have a lot of national templates

We have only few achitype level definitions

Structured data in a template may be smashed to free text sections when stored to Kanta

However structured data your system had in the beginning, Kanta is unable to manage it without a national ”architype” model

Open market for numerous separate systems for very spesific profesional needs

A lot of integration needs
Many separate silos

Every new template is planned in your system from beginnig

i.e.

Primary use summaries
How to aggregate data for a certain summary purpose like European Patient Summary or certificates
Comparison of treatment from the point of view of quality
Administrative reports and bench marking

What if,,,

19 of 32

Points of data protection and Kanta

20 of 32

National legislation looks the issue of data protection from register keepers right to hand out ��GDPR talks about users right to use

Consent for disclosure via Kanta or from other sources

Prohibition to disclose (all, registerkeeper, service event, mprescription)

Recording a document at the request and on behalf of the client

21 of 32

Use of data for various purposes

Local documentation of services

Re-splittiing of documents for reuse

Kanta

Organisation spesific data lakes for secondary use

Control of disclosure for care

Product and implementation spesific databases

MANUAL ACCESS CONTROL

The world Kanta was built to

22 of 32

Care

Kanta

Other data

Health data

Health related data

The world that I believe we should look to

Documentation of services

My life

INTEROPERABLE ACCESS CONTROL AND CONSENTS

Use of data for various purposes

Findata

Finnish Social and Health Data Permit Authority

Citizens data

Technical location and ownership county or national

Data of the service execution

Technical location and ownership provider

23 of 32

INTEROPERABLE ACCESS CONTROL AND CONSENTS

What and why is included?

Only data protection or even ”view history”

How granular sets of data should be identified as the object of use

By what different kinds of metadata a dataset should be regocnized

What different origins of rights

Variation of transactions

Variations of ownership of data in primary and secondary use

Variation of consent types

…

Interoperable solutions for access control need shared concepts

24 of 32

What has been possible to build on Kanta

Demo of UNA Snapshot

25 of 32

UNA Oy

Missions

EVERYDAY Solutions to facilitate everyday services

CHANGE Expertise to support change

DEVELOPMENT Services to support progressive steps to the future

UNA as a company

Established 15.12.2017

15 employees working all over the country

Home site Tampere

No-profit in-house company of 11 Welfare Dservice Counties

Services

Expert services
Solutions for

Sharing data
Joint modules

26 of 32

Patient summary - Professionals - Kanta.fi

The Kanta Services are Finland’s contact point in the disclosure of patient data to health care organisations in other EU member states

The patient’s consent is required to send a Patient Summary

Access is possible already in some regions of Spain

The transfer of patient data to Finland will be enabled in the near future

27 of 32

Why

Professionals don’t use Kanta as it takes too much time and effort

Professionals are in a multiapplication trap if the patient is digitally active

Professionals / scientist / managers haven’t access to self reported care or -data

28 of 32

SNAPSHOT una

Shows all data of the client summed up on a screen

Kanta documents are splitted and information rearranged under headings (infoboxes) familiar for professionals

You can

drill down deeper into the information down to the details
view the data in the calender context
arrange your screen to your needs
see both healthcare and social care
show self reported data among the patient record data

29 of 32

SNAPSHOT una��is an interface on� �CORE una

CORE una

collects and manages all available data of the client from Kanta and other sources in real time
controls that the present user gets the data he/sha has right to access

due to the context of customership
role at work and
customers will / spesific legal right

30 of 32