scieee Science in your language
[en] (orig)

D3.1 - Interim Data Management Plan for STRONG-AYA

Abstract

This interim Data Management Plan sets out the present vision for decentralized data management. It outlines the decentralised storage of appropriately pseudonymised data in accordance with the principal of data minimization.

Read accessible full text

D3.1 - Interim Data Management Plan for STRONG-AYA

Author: Wee, Leonard; Darlington, Anne-Sophie; Aguado, Maria; Lejeune, Stephane; Stark, Dan; Dekker, Andre; van Beusekom, Bart; Couspel, Norbert; Husson, Olga; Hanebaum, Simone
Publisher: Zenodo
DOI: 10.5281/zenodo.17234740
Source: https://zenodo.org/records/17234740/files/D3.1_DATAMGMTPLAN_STRONGAYA_FINAL.pdf
1
A new, in e disciplina y, mul i-s akeholde Eu opean ne wo k o imp o e
heal hca e se ices, esea ch and ou comes o Adolescen s and Young Adul s.
STRONG-AYA – No. 101057482 – D3.1
Deli e able Repo
In e im Da a Managemen Plan o STRONG-AYA
Deli e able D3.1
Sho i le : In e im Da a Managemen Plan
Due da e o deli e able: 01/04/2023
Ac ual submission da e: 31/03/2023
P ojec :
STRONG-AYA
Lead Con ibu o
Leona d WEE (Clinical Da a Science – Uni e si y Maas ich )
Email
[email p o ec ed]
O he Con ibu o s
[WP1] Anne-Sophie Da ling on ([email p o ec ed].uk) – Uni e si y
Sou hamp on
[WP2] Ma ia Aguado (ma ia.aguado@eo c.o g) - EORTC
[WP2] S ephane Lejeune ([email p o ec ed]) – EORTC
[WP2&3] Daniel S a k ([email p o ec ed]) – Uni e si y Leeds
[WP3] And e Dekke ([email p o ec ed]) – Uni e si y Maas ich
[WP3] Ba an Beusekom ([email p o ec ed] – Ne he lands Comp ehensi e
Cance Cen e
[WP4] No be Couspel ([email p o ec ed]) – Eu opean
Cance O ganiza ion
[WP5] Olga Husson ([email p o ec ed]) – Ne he lands Cance Ins i u e
[WP5] Simone Hanebaum ([email protected]) – Ne he lands Cance Ins i u e
Emails
See a e abo emen ioned pe sons, in pa en heses.
STRONG-AYA – No. 101057482 – D3.1
Due da e
01 04 2023
Deli e y da e
01 04 2023
Deli e able ype
R
Dissemina ion le el
SEN
Desc ip ion o Wo k
Ve sion
Da e
D a ing wo k
V0.1
-
In e im Da a Mgm Plan
V1.0
31/03/2023
Desc ip ion o Deli e able:
An ini ial Da a Managemen Plan will be de eloped and upda ed o he inal deploymen
nea he end o he p ojec . The Da a Managemen Plan(s) will e lec he echnology and
si ua ion-speci ic needs o be add essed by he p inciples o da a p i acy by design and
by de aul . I will de ine da a anonymiza ion s anda ds, minimum sa egua ds and
obliga ions o he da a cus odians. In collabo a ion wi h WP2 i will de ine a STRONG-AYA
in o ma ion secu i y policy.
1. Summa y (max ½ page)
STRONG-AYA is a ision o an in e disciplina y, mul i-s akeholde Eu opean ecosys em u ilising widely
geog aphically dispe sed da a o imp o e heal h se ices, clinical esea ch and pa ien -cen ed ou comes
o 15-39 yea old pe sons wi h cance . This is a special g oup o people in heal hca e, ha expe iences age-
speci ic and li es yle-speci ic issues when diagnosed wi h cance , o en wi h d as ically-al e ed quali y o li e
compa ed o hei cance - ee pee s. The STRONG-AYA p ojec aces his challenge wi h an o ganiza ional
s uc u e and echnological inno a ions ha can b ing oge he heal h da a abou his special g oup o
people, and lea n ac ionable insigh s om his da a, all he while p o ec ing indi idual iden i ies by p inciple
and by design. This in e im Da a Managemen Plan se s ou he p esen ision o decen alized da a
managemen , and may be expec ed o change in ce ain de ails as he STRONG-AYA conso ium de elops
and ma u es. Howe e , he co e concep s con ained he ein a e no expec ed o change signi ican ly.
This documen ou lines he decen alised s o age o app op ia ely pseudonymised da a in acco dance wi h
he p incipal o da a minimiza ion. This da a shall be made FAIR, bu no open due o sensi i e na u e o
pe sonal, medical and li es yle in o ma ion (in spi e o pseudonymiza ion). FAIRness o he p i a e da a is a
p e equisi e o compu e -con olled Fede a ed Analy ics and Fede a ed Lea ning. Fede a ed he e means
he dissemina ion o summa y s a is ics and epidemiological models, wi hou he exchange o indi idual
pa ien -le el da a. Las ly, he ou line o a secu e sys em in as uc u e o assu e p i acy by p inciple and
STRONG-AYA – No. 101057482 – D3.1
p i acy by design is p esen ed. A inalized Da a Managemen Plan is scheduled o deli e y nea he
end o he STRONG-AYA p ojec .
1 In oduc ion
Gene al in oduc ion o STRONG-AYA p ojec
STRONG-AYA is a ision o an in e disciplina y, mul i-s akeholde Eu opean ede a ed ecosys em o da a
a ailabili y o imp o e heal h se ices, clinical esea ch and pa ien -cen ed ou comes o adolescen s and
young adul s (AYAs) wi h cance (i.e., 15-39 yea s o age a i s cance diagnosis).
AYAs comp ise a demog aphically di e se and clinically he e ogenous g oup ac oss he leng h and b ead h
o Eu ope, due o uniquely age-speci ic and li es yle-speci ic issues, as well as d as ically al e ed quali y-o -
li e compa ed o hei cance - ee pee s.
The STRONG-AYA conso ium belie es ha inno a i e and pa adigm-b eaking ways o accessing da a and
sha ing knowledge ac oss ins i u ional siloes is u gen ly needed o add ess heal hca e needs among his
unique g oup o people.
The key pilla s o he p ojec a e: (1) consensus de elopmen o a Co e Ou comes Se i.e. COS; (2)
in as uc u e o collec ing, managing and enhancing AYA da a among pa icipa ing coun ies; (3)
dissemina ing ou comes and da a analysis ools a he pan-Eu opean le el o imp o e pa e ns o ca e o
AYAs.
A ede a ed ecosys em - o decen alized da a managemen as well as enabling so wa e echnology o
s akeholde s (e.g., clinicians and pa ien s) o be able o pe o m ede a ed da a analysis - was selec ed as he
in ended app oach o add ess da a agmen a ion, bypass adop ion ba ie s associa ed wi h cen alized
pa ien da a eposi o ies, and o be mo e easily scalable o o he cen es a e end o cu en p ojec .
STRONG-AYA – No. 101057482 – D3.1
Da a managemen ha complies wi h FAIR (Findable-Accessible-In e ope able-Reusable) da a p inciples is a
co ne s one o ou echnical in as uc u e. The FAIR p inciples will acco dingly be applied a pa ne ,
na ional and pan-Eu opean le els o da a, and subsequen ly digi al a e ac s gene a ed (e.g., so wa e,
publica ions, epo s) om he p ojec wo k.
A u he echnological co ne s one o he p ojec shall be ede a ed analy ics, i.e. pe o ming s a is ical
analyses and summa y isualiza ions o da a bu all he da a s ays locally wi hin he o iginal ins i u ion. The
emphasis is ha indi idual pa ien -le el da a will no lea e he da a-holding ins i u ion. STRONG-AYA
suppo s ede a ed lea ning such ha p ognos ic and p edic i e (epidemiological) models can be de eloped
and alida ed, also wi hou sha ing indi idual pa ien -le el da a be ween ins i u ions.
Fede a ed analy ics (and ede a ed lea ning) ha use geog aphically dispe sed FAIR da a equi es secu e web
connec i i y in as uc u e ha is upg adeable, main ained and moni o ed agains ex e nal a acks. Use -
acing web in e aces hos ing applica ions app o ed by he STRONG-AYA conso ium will be made a ailable,
so dis inc use s will each ha e di e en ia ed pe missions and ge access o analy ical/lea ning ools.
Scope o Da a Managemen Plan(s)
The Da a Managemen plan ou lines (1) in gene al e ms he da a used in his p ojec , (2) ha da a is
p o ec ed by keeping i wi hin he da a-holding ins i u ions, (3) implemen a ion o he FAIR p inciples o
make ins i u ional da a mu ually in e -ope able ac oss he STRONG-AYA conso ium and (4) a ailabili y o
he da a a e he p ojec ends.
Addi ionally, he e shall be a numbe o end p oduc s o STRONG-AYA om a da a managemen pe spec i e,
which a e ele an o he sus ainabili y a e unding ends, and o o he simila ede a ed FAIR da a
ecosys em ini ia i es such as he Eu opean Heal h Da a Spaces. The end p oduc s include:
1) An in o ma ion model (i.e. umb ella p o ocol), ha de ines he Co e Ou comes Se (COS) o be
ob ained by pa ne s in he STRONG-AYA communi y.
2) A FAIR implemen a ion p o ile ha desc ibes how he ins i u ion-speci ic da a schema is mapped a
pos e io i o a uni e sal e minology and axonomic model o he o e -a ching COS.
3) Open-sou ce so wa e, enginee ing know-how, and e e ence implemen a ions o he
a o emen ioned, including syn he ic da a which helps u u e pa ne s se up hei own FAIR da a
eposi o ies acco ding o he in o ma ion model (1) and FAIR implemen a ion p o iles (2).
4) Open-sou ce so wa e eposi o y o ede a ed analy ics (FAIR models) and e e ence implemen a ion
amewo ks o new conso ia (FAIR se ices) ha we e used in STRONG-AYA so ha o he use s can
apply he STRONG-AYA in as uc u e o hei own p ojec s.
Complemen a y documen a ion
This documen is he ea e e e ed o as he Da a Managemen Plan, o he DMP in sho .
The DMP is in ended o he use o pa ne s o he STRONG-AYA P ojec ( e e ence numbe 101057482). The
pa ne s a e de ined in he Gene al Ag eemen (execu ed on 18 May 2022, i s amendmen da ed TBC), ha e
execu ed he Conso ium Ag eemen (20 Decembe 2022), and whose asks a e de ailed in he p ojec scope
o wo k in he a o emen ioned Gene al Ag eemen .
The DMP is o be used complemen a ily wi h he STRONG-AYA (go e nance, legal and ope a ional) ecosys em
documen – see Deli e able 2.1.

STRONG-AYA – No. 101057482 – D3.1
The DMP is o be used complemen a ily wi h he STRONG-AYA echnical bluep in documen – see
Deli e able D3.3.
The DMP is o be used complemen a ily wi h he STRONG-AYA eco d o p ocessing ac i i y (ROPA) and da a
p o ec ion impac assessmen (DPIA) – see Deli e able 2.2.
The de ini ions used in he DMP aligns o e minologies in he o hcoming STRONG-AYA join da a
con olle ship and da a p ocessing ag eemen .
2 P ocedu e
The p elimina y COS Da a i ems we e de eloped by consensus among he pa ne s in Wo k Package 1. A
gene al wish-lis o da a domains we e su eyed among pa icipa ing clinical ins i u ions and cance
egis ies. P agma ically, a mo e de ailed lis o da a ields we e de eloped as al eady ou inely collec ed,
and would be easible o collec in u u e. The COS is in ended o g ow and scale in u u e, as pa ne s jou ney
u he in o his p ojec .
Wi h Wo k Package 2, su eys and uns uc u ed in e iews we e conduc ed in o de o unde s and he
exis ing da a lows wi hin pa ne ins i u ions, and po en ial o scaling up in u u e. This in o ms he
e alua ion o p ocessing ac i i ies and da a p o ec ion impac assessmen . This wo k is ongoing.
The p incipal wo k o ecosys em desc ip ion and echnical bluep in de elopmen was ca ied ou by
easibili y su eys and uns uc u ed in e iews among pa ne s, and d awing hea ily om expe ise in
success ul ede a ed lea ning p ojec s led by IKNL and Uni e si y Maas ich . This wo k is ongoing, howe e
he baseline echnical bluep in will be made a ailable online as a an “agile” documen and hus inco po a es
ex a unc ionali y as his p ojec p og esses.
Engagemen wi h a wide ne wo k o cu en and u u e s akeholde s is he wo k o WP4, and is ongoing.
The engagemen s a e essen ial o unde s and he needs o he u u e da a con ibu o s and po en ial da a
use s o he ede a ed ecosys em, hus he echnical bluep in and he DMP will e ol e hand in hand.
This in e im DMP was d a ed based a ound he EU Ho izon2020 DMP empla e and he guidance
documen a ion om he in e na ional GO-FAIR Ini ia i e.
The co-con ibu o s ha e e iewed his DMP documen , and ag ee on his as an in e im DMP, in o de
p oceed u he wi h ecosys em design and echnical de elopmen , leading owa ds a inalized DMP a he
end o he p ojec as p omised.
3 In e im Da a Managemen Plan
Co e Ou comes Da a
3.1.1 S a emen o in ended pu pose o collec ion
The s a emen o in ended pu pose has been based on ex om he STRONG-AYA (go e nance, legal and
ope a ional) ecosys em epo – Deli e able 2.1.
Adolescen s and young adul s (i.e., AYAs) wi h cance o m a unique g oup; hey ace age-speci ic issues (e.g.
in e ili y, unemploymen , inancial p oblems) and dec eased quali y o li e due o cance and i s ea men .
Unlike dedica ed heal hca e se ices and clinical ials o paedia ic cance pa ien s, AYA-speci ic heal hca e
se ices emain sca ce and p ac ice is he e ogeneous ac oss Eu ope. AYAs, who also a e a he e y co e o
socie y and he economy, need access o e idence-based age-app op ia e and high-quali y dedica ed
heal hca e.
STRONG-AYA – No. 101057482 – D3.1
AYA ca e and esea ch will bene i om collec ion and pooling o pa ien -cen ed da a and
collabo a ion among a wide spec um o s akeholde s: pa ien s, heal hca e p o essionals, scien is s, and
policymake s. The compa isons and desc ip i e analyses ca ied ou will be used o imp o e cance ca e and
ea men a he policy, o ganisa ional, se ice and in e pe sonal le el.
3.1.2 Types o da a collec ed
The p ojec in ends o make use o wide- anging pa ien da a, ine de ails pending he inal COS epo , in
he ollowing gene al domains. This p elimina y lis has been p o ided as a s a ing poin o he conso ium
o de elop and ex end and scale up; i is no o be aken as he inal and de ini i e decla a ion o a inalized
COS da a se .
De e minan s and case mix ac o s
- Tumou /disease pheno ype (such as s aging, si e and ype o cance )
- T ea men (s) ecei ed (such as su ge y, chemo he apy, adio he apy, immuno he apy o
combina ions he eo )
- Biological sex and sel -de ined gende p onouns
- Age a cance diagnosis
- Social de e minan s (such as socioeconomic g ading, e hnici y/cul u al backg ound, highes
educa ion a ained, employmen ca ego y, domes ic s a us, i he e a e dependen pe sons)
- Body composi ion (such as body mass index)
- P e-exis ing medical condi ions
- Risk s a i ica ion
- Use o diagnos ic imaging o diagnosis and disease quan i ica ion
Physiological and clinical ou comes
Quali y o li e ou comes and impac s on ac i i ies associa ed wi h daily li ing
- Physical unc ioning, Social unc ioning, Role unc ioning, Emo ional unc ioning, Cogni i e
unc ioning
- Global quali y o li e
- Pe cei ed heal h s a us
- Pe cep ions o deli e y o ca e (Sa is ac ion/pa ien p e e ence, Accep abili y and a ailabili y o ca e,
Adhe ence/ compliance, Wi hd awal om ea men , App op ia eness o ea men )
- P ocess, implemen a ion, and se ice ou comes
Heal h sys em esou ce u iliza ion pa e n
- Economic/Financial ha dship
- T ea men - ela ed inju y/ oxici y
- Insu ance s a us
- Hospi aliza ion and pos - ea men heal h ca e u iliza ion (e.g. ea ed as adul o by a paedia ic
specialis )
- Need o u he in e en ion a e cance in e en ion
- Socie al bu den/ca e bu den
- Medical ad e se e en s and a e -e ec s
Mo ali y
- All-cause mo ali y
STRONG-AYA – No. 101057482 – D3.1
3.1.3 Re-use o exis ing da a wi hin his p ojec
In STRONG-AYA, p ima y eliance shall be placed on da a ob ained du ing ou ine cance ca e and s anda d
clinical p ocedu es (i.e., eal-wo ld da a) supplemen ed by da a ex ac ions om p e iously concluded
clinical ials ( ha migh be ei he obse a ional o expe imen al in na u e). This means STRONG-AYA is
p incipally a DATA SECONDARY USE ype o s udy.
App op ia e da a pe aining o be abo emen ioned COS gene al domains shall be sou ced by pa ne
ins i u ions om wi hin hei own clinical en i onmen (such as cha audi s, elec onic heal h eco ds,
quan i a i e ques ionnai es), heal hca e p o ide s in hei ca e ne wo k and om hei adjacen
na ional/ egional/hospi al-based cance egis ies. The quali y o li e domain in he abo emen ioned da a is
in ended o be sou ced ia pa ien - epo ed in o ma ion.
The COS domains and da a ields collec ed may be ex ended in scope and scale wi hin he p ojec ; he DMP
shall be modi ied acco dingly in u u e e sions o cap u e such changes in scope and scale o da a ga he ing
by p ojec pa ne s.
One o he in en ions in STRONG-AYA is o suppo he c ea ion, ex ension and sus ainabili y o ins i u ional
wo k lows o collec high quali y COS da a well in o he o eseeable u u e. This means ha u u e da a
collec ion i ems need o become implemen ed in o ou ine clinical ca e as well as in s anda d ollow-up, and
changes also need o be implemen ed in o epo ing p o ocols o cance egis ies.
3.1.4 PROMs as a special ca ego y o da a collec ion
In heal hca e, he collec ion and u iliza ion o Pa ien Repo ed Ou come Measu es i.e. PROMs is an e e
inc easingly impo an sou ce o da a abou quali y o li e, impac s on ac i i ies ela ed o daily li ing and
pe cep ions o he u ili y o ca e. Howe e , sys ema ic collec ion o PROMs among he STRONG-AYA pa ne s
is p esen ly highly he e ogenous, and only some i ems a e being eco ded sys ema ically and in all places.
The collec ion and u iliza ion o PROMs shall be an impo an aspec o STRONG-AYA.
Some pa ne s ha e digi al PROMs acquisi ion sys ems ha con ac he pa ien and solici elec onic
esponses o PROMs-based quan i a i e su eys. Some pa ne s do no ha e any PROMs acquisi ion sys ems
a all, and o he pa ne s ely on pape o ms which hen ha e o be collec ed o digi iza ion.
Fu he mo e, pa ien engagemen will be c ucial, as he AYA pa ien communi y will end up being he d i e s
o he design o PROMs collec ion and PROMs u iliza ion.
The STRONG-AYA p ojec is commi ed o ha moniza ion o elec onic PROMs da a collec ion and i s
in eg a ion wi h as many exis ing sys ems as possible, bu he exac echnical design o such an in as uc u e
hinges upon he selec ion o key u iliza ion cases which hus equi es ex ensi e pa ien engagemen ; his is
he p incipal wo k o Wo k Package 4. The PROMs u iliza ion design shall no iola e he o e all p inciples
laid ou in he STRONG-AYA ision, bu de ailed da a managemen planning pe aining speci ically o PROMs
emains impossible a he p esen ime. A comple e PROMs u iliza ion plan will be included in he de ini i e
e sion o his DMP o be deli e ed a end o p ojec .
3.1.5 Da a sou ces
Re ospec i ely collec ed and eal-wo ld da a
Clinical case mix ac o s, ea men /in e en ion in o ma ion, diagnoses, physiological ou comes, clinical
ollow-up and mo ali y will p edominan ly be sou ced om elec onic medical heal h eco ds main ained by
he pa icipa ing ins i u ion and/o by ex ac ion om al eady-closed clinical ials.
Social de e minan s o ca e, quali y o li e, impac s on ac i i ies o daily li ing and heal h sys em u iliza ion
a e in ended o be ex ac ed om cance egis y da abases and supplemen ed by ex ac ion om al eady-
closed clinical ials.
STRONG-AYA – No. 101057482 – D3.1
P ospec i ely collec ed da a
Fo u u e, expanded and mo e de ini i e e sions o COS da a ields, he abo emen ioned ins i u ional
elec onic heal h eco ds and cance egis ies need o be expanded and upda ed. This is an ope a ional
ques ion ha is impossible o answe a he e y ou se he p ojec , howe e discussions a e ongoing wi h
pa ne ins i u ions o ind ou wha will be easible o u u e da a collec ion.
Gene al and speci ic conside a ions o he da a
The Decla a ion o Helsinki de ines “ ulne able g oups and indi iduals“ as hose ha ing an inc eased
likelihood o being w onged o o incu ing addi ional ha m – in he con ex o medical esea ch. Gene ally
speaking, ulne able indi iduals o popula ions a e hose whose decision making capaci y and abili y o
exe cise and p o ec hei igh s is es ic ed due o hei men al, social, economic o e hnical s a us, age o
heal h condi ion o , in o he wo ds, hose who in one o mo e ways a e a isk o being ha med h ough he
conduc o esea ch.
The a ge popula ion o STRONG-AYA o e laps wi h subjec s ha would be commonly deemed as
“ ulne able indi iduals”, speci ically pe sons wi h incu able diseases, pe sons who a e unemployed o
impo e ished, pe sons om e hnic mino i y g oups, and mino s. The inclusion o such subjec s’ da a equi es
ca e ul conside a ions in e ms o he esea ch ac i i y being ca ied ou as well as p o ec ion o iden i iable
pe sonal cha ac e is ics.
The in-dep h de elopmen o go e nance p inciples o STRONG AYA will be comple ed by he end o he i s
yea o he p ojec . This is one o he p incipal ou pu s esul ing om Wo k Package 2. An in en o y o
ele an exis ing e hical and go e nance codes and guidelines al eady p esen in he a ious pa ne
ins i u ions will esul in he de elopmen o he p ojec ’s e hical and legal guidelines and amewo k, and
his shall align p inciples om he di e en STRONG AYA pa ne s as well as ele an EU go e nance and
e hics p inciples (such as he ‘Eu opean E hical P inciples o Digi al Heal h’ 2022), and wi h he key p inciples
laid ou in he Gene al Da a P o ec ion Regula ions (GDPR) wi h i s ele an amendmen s.
The ollowing is based on ex om he STRONG-AYA (go e nance, legal and ope a ional) ecosys em epo –
Deli e able 2.1. The a o emen ioned p inciples shall include bu a e no s ic ly limi ed o:
- law ulness, ai ness and anspa ency: da a shall be p ocessed law ully, ai ly and in a anspa en
manne in ela ion o he da a subjec ;
- limi a ion o pu pose: da a shall be collec ed o he speci ied, explici and legi ima e pu poses and
no be u he p ocessed in a manne ha is incompa ible wi h hose pu poses; u he p ocessing
o a chi ing pu poses in he public in e es , scien i ic o his o ical esea ch pu poses o s a is ical
pu poses shall, in acco dance wi h A icle 89(1), no be conside ed o be incompa ible wi h he ini ial
pu poses;
- da a minimisa ion: da a shall be adequa e, ele an and limi ed o wha is necessa y in ela ion o
he pu poses o which hey a e p ocessed;
- accu acy: da a shall be accu a e and, whe e necessa y, kep up o da e; e e y easonable s ep mus
be aken o ensu e ha pe sonal da a ha a e inaccu a e, ha ing ega d o he pu poses o which
hey a e p ocessed, a e e ased o ec i ied wi hou delay;
- limi a ion on s o age: da a shall be kep in a o m which pe mi s iden i ica ion o da a subjec s o
no longe han is necessa y o he pu poses o which he pe sonal da a a e p ocessed; pe sonal
da a may be s o ed o longe pe iods inso a as he pe sonal da a will be p ocessed solely o
a chi ing pu poses in he public in e es , scien i ic o his o ical esea ch pu poses o s a is ical
pu poses in acco dance wi h A icle 89(1) subjec o implemen a ion o he app op ia e echnical
and o ganisa ional measu es equi ed by his Regula ion in o de o sa egua d he igh s and
eedoms o he da a subjec ;
STRONG-AYA – No. 101057482 – D3.1
• So wa e ools (e.g., FAIRi ca ion ools, syn he ic da a examples, ede a ed in as uc u e
and ede a ed analysis/lea ning algo i hms) will become an in eg al pa o an es ablished and well-
managed open-sou ce se o p oduc s, so as o maximise e-usabili y.
R1.2. (Me a)da a a e associa ed wi h de ailed p o enance
• All elemen s associa ed wi h a URL o URI a e e sion con olled.
• Da a p o enance is moni o ed by he da a-owning ins i u ions. Inso a as possible, ins i u ional da a
wo k lows a e sc ip ed and au oma ed and documen ed in a u u e pa ne -speci ic da a wo k lows
epo .
• O e all esponsibili y o da a li ecycle managemen (e.g., s o age, back-ups, dele ions and in eg i y
o unde lying da a sou ces) lies wi h he da a p o ide s, and FAIR-speci ic ooling o upda e, add and
dep eca e da a will be p o ided o pa ne s.
R1.3. (Me a)da a mee domain- ele an communi y s anda ds
• Implemen a ion o RDF and Seman ic Web mee s domain-speci ic s anda ds whe e e hese al eady
exis , pa icula ly o da a.
• All easonable s eps a e aken o be as FAIR as possible whosoe e es ablished communi y s anda ds
do no exis .
3.5.5 Da a Loss Mi iga ion
Pa ne ins i u ions ha hos da a, in he ede a ed analysis/ ede a ed lea ning pa adigm, keep hei own
da a “on p emise” i.e., he COS da a i ems a e placed in a secu ed esea ch and compu a ional en i onmen
con olled by he pa ne ins i u ion hemsel es.
The e will be no ans e o pa ien -le el da a (in any o m) om one STRONG-AYA pa ne o ano he , o
he pu pose o he kinds o analyses pe mi ed in he STRONG-AYA amewo k.
The echnical bluep in documen – see Deli e able 3.3 – equi es ha on-p emise COS da a shall be
pseudonymized (no na i e ins i u ional iden i ie s o equi alen s o social secu i y numbe s a e allowed),
minimized (supe luous in o ma ion ha is no needed di ec ly o analysis such as bi hda es and pos codes
a e no allowed) and ou inely backed-up as pe he ins i u ion’s own s anda d ope a ing p ocedu e
pe aining o in o ma ion echnology de ices.
In e en o ca as ophic loss o he COS da a and i s back-up copy, o wha e e eason, i mus be no ed ha
he o iginal da a exis s in he p ima y sou ce sys ems such as elec onic hospi al eco ds, clinical ials
a chi es and he cance egis ies. Al hough ex emely incon enien , he COS da a ex ac ion p ocedu es pu
in place by he pa ne ins i u ions ha ead om he p ima y da a sou ces would be au oma ed o he
maximum ex en possible, he e o e i may be e-done om sc a ch i uly necessa y. The likelihood o all
sys ems ailing a he same ime would be e y low, and would sugges a highe p io i y c isis a he
immedia e hand han he issue o eco e ing p ojec da a.
The COS da a will be p ese ed on-p emise by he ins i u ions, o a leas he du a ion o he STRONG-AYA
p ojec , as well as beyond he e m o he STRONG-AYA p ojec as de ined in he legal, go e nance and
business ope a ions amewo k.
3.5.6 FAIR so wa e
STRONG-AYA will ake all easonable s eps o make he so wa e in he p ojec o be as OPEN and FAIR as
possible. The so wa e esul ing om he execu ion o his p ojec includes :

STRONG-AYA – No. 101057482 – D3.1
- Fu u e e sions o Van age6 in as uc u e o ede a ed analysis and ede a ed lea ning
- Code o pe o ming ede a ed da a analysis and ede a ed epidemiological modelling
- So wa e ools o making s uc u ed clinical da a FAIR
- In e ac i e dashboa ds o isualiza ion o summa y s a is ics
- Web-based applica ions o de eloping and alida ing epidemiological models in a ede a ed
ecosys em
Making his so wa e FAIR shall include (bu no exclusi ely limi ed o) he ollowing :
- P og amming code will be open sou ce and open access ia a widely-used public so wa e eposi o y
such as Gi Hub, along wi h ich desc ip i e ex in o ming he in ended use how he so wa e shall
be ins alled and ope a ed;
- Pe sis en uni e sal esou ce loca o links (URLs) o he so wa e eposi o ies will always be gi en in
publica ions and dissemina ions pe aining o i s use in STRONG-AYA esul s;
- The esea ch esul s ha ci e he so wa e eposi o y URLs will hemsel es ha e pe sis en Digi al
Objec Iden i ie s (DOIs) ha will be publicly sea chable and indable, and he esea ch esul s ( o
he maximum ex en possible) will be published as open access documen s;
- The p og amming code shall p incipally be in a widely-unde s ood compu e language such as
Py hon, Ja a o C, o a ma hema ical sc ip ing language such as R, and u he mo e each eposi o y
shall con ain links o so wa e lib a y dependencies (such as a equi emen s. x ile in Py hon) o
assis ep oducibili y and e-usabili y;
- Re-use o he so wa e will be widely encou aged using licensing condi ions such as C ea i e
Commons By A ibu ion (wi h o wi hou Sha e-Alike op ion), he GNU Public so wa e license, and
o he such simila licenses ha a e widely used by he open-sou ce so wa e communi y.
In addi ion o he abo e s eps, any epidemiological models de eloped and alida ed in he STRONG-AYA
p ojec shall be documen ed (along wi h i s pe sis en iden i ie s and URLs) wi h he AIME (A i icial
In elligence in bioMEdical esea ch) egis y o added anspa ency - h ps://aime- egis y.o g/. The models
esul ing om STRONG-AYA will also be con ibu ed o he newly s a ed FAIRmodels p ojec
(h ps:// ai models.o g/) which shall hold ich me ada a on he ype o da a on which he model has been
ained on, how he model was de eloped and he an icipa ed limi a ions o he model’s scope o use.
3.5.7 FAIR se ices
In ou case, “se ices” e e o he ope a ional know-how o c ea ing a conso ium a ound a ede a ed da a
ecosys em pa adigm, which includes:
- A o mal conso ium ag eemen ha de ines he gene al igh s, oles and du ies pe aining o
conso ium membe ship;
- A go e nance and ope a ions amewo k ha de ines he wo k o he conso ium, including i s
p ocess o legal e iew and e hical e iew o wha he conso ium may (and may no ) do wi h he
collec ed da a, as well as he humanis ic alues (e.g. p inciples o pa ien empowe men ) guiding i s
conduc ;
- A echnology bluep in de ining he ha dwa e and so wa e equi emen s ha cons i u e he
echnical pa s o he ede a ed in as uc u e;
- A se o legal mechanisms, such as a join da a con olle ship ag eemen and an in as uc u e use
ag eemen wi h hi d-pa y se ice p o ide s, ha a e necessa y o allow ede a ed
lea ning/ ede a ed analysis o be pe o med.
- An in o ma ion model (schema) including a axonomy/lexicon/ e minology de ining o he STRONG-
AYA COS i ems, bu wi hou he da a i sel .
STRONG-AYA – No. 101057482 – D3.1
STRONG-AYA makes he abo e se ices open ( o he maximum ex en possible) and FAIR in he ollowing
manne :
- Whe e app op ia e, he a o emen ioned ins umen s shall ha e names, add esses and ins i u ions
and o he po en ially iden i iable and/o con iden ial in o ma ion edac ed, such ha he esul ing
documen will be made openly accessible (wi h publicly accessible URL) as a empla e o u u e
collabo a ions;
- Each o he abo emen ioned documen s will be egis e ed wi h a pe sis en DOI, and he en i e
package o documen s comp ising he STRONG-AYA se ices will be lis ed on a publicly-sea chable
FAIR objec s ca alogue (such as Zenodo);
- The FAIR ca alogue en y shall include ich desc ip ions abou hese se ices, and hese desc ip ions
shall emain in he FAIR eposi o y e en i he documen s hemsel es a e aken o line;
- Re-use o he se ices as empla es will be co e ed by a pe missi e open access license o be selec ed
by he conso ium a he end o he p ojec .
3.5.8 Pos -P ojec Da a Sus ainabili y
Da a in he COS p ojec is made FAIR as a condi ion o i s use du ing he p ojec , so he e a e no special cos s
o make da a FAIR a e he end o he p ojec . As s a ed abo e, one o he in en ions in STRONG-AYA is o
enable c ea ion, ex ension and sus ainabili y o ins i u ional wo k lows o collec high quali y COS da a well
in o he o eseeable u u e and his necessa ily implies ools o making he da a FAIR on-p emises.
Since he da a s ays wi hin he ins i u ion ha con ibu es he da a, STRONG-AYA me hodology mus
in eg a e wi h exis ing da a managemen p ocedu es as well as da a managemen esponsibili ies ha
al eady exis wi hin pa ne ins i u ions. As he p ojec , STRONG-AYA is esponsibili y o he go e nance,
ope a ion and pa ien igh s p o ec ion wi hin he ede a ion in as uc u e.
Cos s o long e m ope a ion and p ese a ion o he ede a ion in as uc u e is pa o he go e nance and
business ope a ion plan o he en i e conso ium. Cos s o long- e m ope a ion and p ese a ion o he COS
da a wi hin he pa icipa ing ins i u ions will be bo ne di ec ly by ha ins i u ion i sel . Cos s o he ongoing
deli e y o he AI echnology will ha e o be bo ne by he conso ium as a whole, as hese echnologies
pe mea e u he in o ou ine applica ion. Howe e , desc ibing he means by which hese o ganisa ions
migh inance o o he wise sha e cos s wi h he conso ium o he main enance o hei COS da a will be a
alid opic o he go e nance and business ope a ion plan.
4 Conclusion
This deli e able epo consis s o an in e im Da a Managemen Plan ha is adequa ely de ailed o ac as a
empo a y oadmap o u he de elopmen and e olu ion o he STRONG-AYA p ojec . I will be expec ed
ha ongoing consul a ions be ween wo k packages and wi h s akeholde s will p o ide deepe de ail in
ce ain sec ions o his epo , as he p ojec i sel akes inc easingly conc e e shape o e he coming yea s.
This Da a Managemen Plan cap u es he cu en design and implemen a ion plan o he echnology and
si ua ion-speci ic equi emen s o he p ojec , whe esoe e i in e sec s wi h da a p i acy needs o
indi iduals. A inalized Da a Managemen Plan has been p omised as a u he deli e able nea he end o
he p ojec .
Reposi o y o p ima y da a (annex)
No Applicable.