Action point 18. Inventory and consolidation registries. healthdata data we care for. RTREH 2015, Brussels,

Action point 18 Inventory and consolidation registries RTREH 2015, Brussels, 2015.05.26.

healthdata

data we care for

AP18 Deliverables

Status 2015.05.26

a. Inventory registers

 n > 160: www.healthstat.be (2014(Q4));  + CRUD (2015(Q2)).

b. Variables by profession & insGtuGon

www.healthstat.be :  Meta-‐data by profession & insBtuBon (2014(Q4));  Variables by profession & insGtuGon (2015(Q4)); www.healthdata.be/dcd :  HD-‐projects Wave 1: technical specificaGons variables, validaBon rules, reference lists (2015(Q1)).

c. Governance

 ConvenGon RIZIV-‐WIV (2015): basics of govern. fed. registries + SteerComm HD;  Tasks SteerCom HD: Procedures & criteria registries + evaluaBon & prioriBsaBon;  SteerCom HD in acBon: 2015Q2;  HD: detailed descripBon of BP available (2014(Q2);  BP for each project available in web portal (2015(Q4);

d. StandardisaGon of variables

 80 registries: + 8000 variables (incl. doubles)  HD: 1st mapping towards MZG variables (2014)  HD: POC: mapping of 2 registries towards SNOMED

Overzicht

healthdata.be data we care for

AP18 Deliverable (Cont.)

Status 2015.05.26

d. StandardisaGon of variables (Cont.)

 HD: analysis of Clinical Building Blocks [CBB’s] (Ned. Fed. Uni. Med. Centr. & NICTIZ) as framework: aligned with clinical pracBce, not specialism-‐specific, & technology neutral) (2015(Q1)).  HD-‐NICTIZ POC with 2 registries: most variables can be easily covered with CBB’s (2015(Q2)).  B e l g i a n va l i d a B o n o f C B B ’s : s h o r t te r m implementaGon in HD registries possible (2016(Q1))

Overzicht

e.+f.+h.+i. : Architecture to extract variables  Not feasible: see g). from Vaults & hubs g. S2S-‐Architecture for mulGfunct. registries  HD S2S-‐Architecture for mono-‐ & mulGfunct. registries + support transiBon period with HD4DP;  Approved by: WG Architecture, Sect. Comm. Health, eHealth;  Successful test installaGons HD4DP: UZLeuven, UZAntwerp, UZBrussels (2015(Q1-‐2)), planned: CHU Charleroi, UZGhent, ZNA, Inkendaal (2015(Q2));  Roll out: all hospitals + laboratories (before 2016(Q1)), all GP’s (2017(Q2));


AP18 Deliverable (Cont.)

Overzicht

Status 2015.05.26

g. S2S-‐Architecture for mulGfunct. registries  42 projects (WIV & RIZIV) redesigned in 3 waves by (Cont.) 2017(Q2) (2015: 2 data collecBons stopped: HD will re-‐use other DB). j. Cancer Registry: batch > conGnuous S2S

 SKR: Development 1st drah own XML + XSD for anatomopathologists  1st phase: S2S for anatomopathologists 2015(Q2)  2nd phase: S2S for care programs in oncology  3rd phase: S2S for specific oncology registries

AP18 Issues a. HD: If large administraGve data collecGons (MZG, MPG, Doc PH, …) have higher freq. in collecBon, use (coded) INSZ, and have shorter processing Bme > higher re-‐use by registries; b. HD: lack of naBonal minimal set of stable, structured, specialism independent, technical neutral, and reusable data specificaBons for (hospital) EPD > HD proposal: clinical building blocks; c. HD: need 4 vision, architecture & roadmap for Next GeneraGon Sequencing (NGS) data collecBon; d. HD: AP18 (like other AP’s) has a significant impact on data providers, data users, governemental policy: CHANGE on level of organisaBon, informaBon, applicaBons, infrastructure, naBonal and internaBonal reporBng obligaBons (ECDC, WHO, EMCDDA, EUROSTAT, …). CHANGE MGMT and transiGon policy on all levels needed; e. SKR: standard format for data transfer adopted by all stakeholders; f. SKR: Gming. healthdata.be data we care for

AP18 new acGon points

Overzicht

Target

1. Een publiek toegankelijke webapplicaBe met gedetailleerde inventaris  vanaf 2015_Q3 van alle in België bestaande recurrente beleidsondersteunende wetenschappelijke gegevensverzamelingen in domein van gezondheid en gezondheidszorg, wordt permanent geactualiseerd (hop:// www.healthstat.be). 2. De taak tot aanmelding en actualisaGe van inventaris wordt in  vanaf 2015_Q3 o v e r e e n ko m s t t u s s e n o p d ra c h t g e v e n d e o v e r h e i d e n d e projectverantwoordelijken van de beleidsondersteunende wetenschappelijke gegevensverzamelingen opgenomen. 3. De procedures en criteria voor opstarten van nieuwe en conGnueren  vóór 2016_Q1 van bestaande recurrente beleidsondersteunende wetenschappelijke gegevensverzamelingen, worden vastgelegd en ter beschikking gesteld via publiek toegankelijke webapplicaBe. 4. De goedgekeurde generieke architectuur (met HD4DP-‐sohware) van  vóór 2016_Q2 (a) healthdata-‐plarorm wordt geïmplementeerd in (a) alle Belgische  vóór 2017_Q4 (b) ziekenhuizen en medische laboratoria, en bij (b) alle Belgische huisartsen. 5. De goedgekeurde generieke architectuur (met HD4DP-‐sohware) van  vanaf 2015_Q2 (a); healthdata-‐plarorm wordt toegepast op alle (a) nieuwe en (b)  vanaf 2016_Q1 (b: voor bestaande recurrente beleidsondersteunende wetenschappelijke allen, vlgs. kalender); gegevensverzamelingen.  vóór 2017_Q4 (b: voor healthdata.be 42 prjct. WIV & RIZIV) data we care for

AP18 new acGon points (Cont.)

Overzicht

Target

6. Het gebruik van goedgekeurde generieke architectuur (met HD4DP-‐  vanaf 2015_Q3 sohware) van healthdata-‐plarorm bij (a) nieuwe en (b) bestaande r e c u r r e n t e b e l e i d s o n d e r s t e u n e n d e w e t e n s c h a p p e l i j k e g e g e v e n s v e r z a m e l i n g e n w o r d t i n o v e r e e n ko m s t t u s s e n opdrachtgevende overheid en de verantwoordelijken van de gegevensverzamelingen opgenomen. 7. Afspraken, architectuur, en planning voor beleidsondersteunende  vóór 2016_Q1 gegevensverzamelingen met meervoudige bestemming (MyCarenet, hubs & kluizen, naBonaal implantatenregister, …). 8. Afspraken, architectuur, en planning voor verzameling, beheer en  vóór 2016_Q1 ontsluiten van humane Next GeneraGon Sequencing (NGS) data. 9. Een Belgische adaptaGe wordt uitgevoerd van elke beschikbare  vóór 2016_Q1 specialisme-‐oversBjgende en technisch neutrale NFU-‐NICTIZ Clinical Building Block, en wordt na validaBe in een publiek toegankelijke centrale digitale catalogus gepubliceerd (hop://www.healthdata.be/ dcd/). 10. Alle (a) nieuwe en (b) bestaande recurrente beleidsondersteunende  vanaf 2016_Q1 (a); wetenschappelijke gegevensverzamelingen worden inhoudelijk  vanaf 2016_Q1 (b: voor samengesteld doormiddel van de voor België beschikbare gevalideerde allen, vlgs. kalender); Clinical Building Blocks;  vóór 2017_Q4 (b: voor 42 prjct. WIV & RIZIV)


AP18 new acGon points (Cont.)

Overzicht

Target

11. De waardenlijsten van Clinical Building Blocks in alle (a) nieuwe en  vanaf 2016_Q1 (a) ; (b) bestaande recurrente beleidsondersteunende wetenschappelijke  vanaf 2016_Q1 (b: voor gegevensverzamelingen in domein van gezondheid en gezondheidszorg, allen, vlgs. kalender); worden prioritair met SNOMED-‐CT concepten opgemaakt;  vóór 2017_Q4 (b: voor 42 prjct. WIV & RIZIV).


APPENDIX Inventarisering en consolidatie registers RTREH 2015, Brussels, 2015.05.26.

healthdata

data we care for

The challenge for scienGfic data collecGon Signalitics, typical available in authentic sources Variables needed for scientific research question

Register A

Register B Information needed in context of continuity of care or internal administration

Register C

Register D Clinical building blocks


Information mostly not available in primary systems EPD, HIMS, LIMS, …)

Example Clinical Building block: “breathing”

Values: SNOMED-‐CT healthdata.be data we care for

Care transfer

that can be used in different contexts

Stable, reuseable building blocks

Quality indicators healthdata.be data we care for

Research

PaGent summaries

Registries & Terminology: the Challenge Now: A lot (>150) of data collecBons with even more (>8000) Parameters Maximal reuse from primary systems is our key objecBve Hospitals, laboraBes and general pracBBoners need to be able to provide this data in a structured way… How can we create a mutually understandable “language” between the reasearchers and the data providers? ► Basis for data exchange ► Maximum simplificaBon of the administraBve work at the data provider’s side ► Tangible results within 2 years


ApplicaBve IntegraBon

Messages (e.g. SUMEHR, Recip-‐e PrescripBon)

KMEHR Messages HL7 Messages

N..M Common Understanding

Building Blocks (e.g. PaBent, Concern, …)

Implicit No Broad coverage (yet)

1..N Detailed Terminology


Concepts & Reference Lists (e.g. ID, sex)

KMEHR CodeTables LOINC SNOMED CT

SuggesGon ObjecBve: ► Inventory of building blocks ► Data providers make sure they can provide the building blocks (in KMEHR or other structured standard) ► Researchers primarily use building blocks for their data collecBons Approach: ► Make the exisBng building blocks in KMEHR messages (more) explicit and document them separately at conceptual level ► Liaise with NICTIZ to exchange building block experience and define a common building block roadmap


Action point 18. Inventory and consolidation registries. healthdata data we care for. RTREH 2015, Brussels,

Recommend Documents