Development of an expert-based scoring system for early identification of patients with inborn errors of immunity in primary care settings – the PIDCAP project

Vol.:(0123456789)
Jou nal o Clinical Immunology (2025) 45:26
h ps://doi.o g/10.1007/s10875-024-01825-3
RESEARCH
De elopmen o anExpe ‑Based Sco ing Sys em o Ea ly
Iden i ica ion o Pa ien s wi hInbo n E o s o Immuni y inP ima y
Ca e Se ings – hePIDCAP P ojec
JacquesG.Ri iè e1,2,3,4· Ge a dCa o ‑Sans5,6· Jo diPie a‑Jiménez5,6,7· Se gidelaTo e5,6· PIDCAP expe g oup·
Xa ie Cos8,9· Xa ie Se a‑Picamal8· Pe eSole ‑Palacin1,2,3,4
Recei ed: 13 Ap il 2024 / Accep ed: 10 Oc obe 2024
© The Au ho (s) 2024
Abs ac
Ea ly diagnosis o inbo n e o s o immuni y (IEIs) has been shown o educe mo ali y, mo bidi y, and heal hca e cos s. The
need o ea ly diagnosis has led o he de elopmen o compu a ional ools ha igge ea lie clinical suspicion by physicians.
P ima y ca e p o essionals se e as he i s line o imp o ing ea ly diagnosis. To his end, a compu e -based ool (based on
ex ended Je ey Modell Founda ion (JMF) Wa ning Signs) was de eloped o assis physicians wi h diagnosis decisions o
IEIs in he p ima y ca e se ing. Two expe -guided sco ing sys ems (one pedia ic, one adul ) we e de eloped. IEI wa n-
ing signs we e iden i ied and a panel o 36 expe s eached a consensus on which signs o include and how hey should be
weigh ed. The esul ing sco ing sys em was es ed agains a e ospec i e egis y o pa ien s wi h con i med IEI using p i-
ma y ca e EHRs. A pilo s udy o assess he easibili y o implemen a ion in p ima y ca e was conduc ed. The sco ing sys em
includes 27 wa ning signs o pedia ic pa ien s and 24 o adul s, adding addi ional clinically ele an c i e ia es ablished by
expe consensus o he JMF Wa ning Signs. Cy openias, ≥ 2 sys emic in ec ions, ecu en e e and b onchiec asis we e he
leading wa ning signs in child en, as b onchiec asis, au oimmune diseases, cy openias, and > 3 pneumonias we e in adul s.
The PIDCAP (P ima y Immune De iciency “Cen e d’A enció P imà ia” ha s ands o P ima y Ca e Cen e in Ca alan) ool
was implemen ed in he p ima y ca e wo ks a ion in a pilo a ea. The expe -based app oach has he po en ial o lessen unde -
epo ing and minimize diagnos ic delays o IEIs. I can be seamlessly in eg a ed in o clinical p ima y ca e wo ks a ions.
Keywo ds Inbo n e o s o immuni y· p ima y immunode iciencies· Je ey Modell Founda ion· wa ning signs· a e
disease· expe -d i en sco ing sys em· ea ly diagnosis· compu e -assis ed medicine
* Jacques G. Ri iè e
[email p o ec ed]
* Pe e Sole -Palacin
pe e.sole @ allheb on.ca
1 In ec ion andImmuni y inPedia ic Pa ien s Resea ch
G oup, Vall d’Heb on Ins i u de Rece ca (VHIR),
Ba celona, Ca alonia, Spain
2 Pedia ic In ec ious Diseases andImmunode iciencies Uni ,
Hospi al In an il I de La Dona Vall d’Heb on, Vall d’Heb on
Ba celona Hospi al Campus, Ba celona, Ca alonia, Spain
3 P esen Add ess: Uni e si a Au ònoma de Ba celona (UAB),
Ba celona, Ca alonia, Spain
4 Je ey Modell Diagnos ic andResea ch Cen e o P ima y
Immunode iciencies, Ba celona, Ca alonia, Spain
5 Ca alan Heal h Se ice, Ba celona, Ca alonia, Spain
6 Digi aliza ion o  heSus ainabili y o  heHeal hca e Sys em
(DS3) Resea ch G oup, L’Hospi ale deLlob ega , Ca alonia,
Spain
7 Facul y o In o ma ics, Mul imedia andTelecommunica ions,
Uni e si a Obe a de Ca alunya, Ba celona, Spain
8 Ins i u Ca alà de La Salu (ICS), Ba celona, Ca alonia, Spain
9 The Founda ion Uni e si y Ins i u e o P ima y Heal h Ca e
Resea ch Jo di Gol I Gu ina (IDIAPJGol), Ba celona, Spain
Jou nal o Clinical Immunology (2025) 45:26 26 Page 2 o 13
In oduc ion
Inbo n E o s o Immuni y (IEIs), also known as p i-
ma y immunode iciencies (PI), a e a ca ego y o diseases
de ined by he p esence o a comp omised immune sys-
em, mainly as a esul o a single-gene loss-o - unc ion o
gain-o - unc ion mu a ion. IEIs a y in he mechanism o
ac ion, mode o inhe i ance, p e alence, and se e i y. They
ha e adi ionally been cha ac e ized by an inc ease in sus-
cep ibili y o se e e and ecu en in ec ions in he ea ly
yea s o li e, wi h many pa ien s ha ing a poo p ognosis
wi h educed quali y o li e and high a es o mo ali y [1].
Howe e , pa ien s wi h IEIs a e now known o p esen wi h
a wide ange o p esen a ions and an inc easing numbe o
clinical pheno ypes. In addi ion o an inc eased equency
o in ec ions, he de elopmen o immune dys egula ion
and au oimmuni y, cance , and alle gy a e also associa ed
wi h IEIs [1–4].While indi idual IEIs may be a e, wi h as
ew as one o wo cases epo ed in he li e a u e, hey col-
lec i ely accoun o a signi ican bu den o disease [5, 6].
Ea ly diagnosis o IEIs has been shown o educe a es
o pa ien mo ali y and mo bidi y, heal hca e cos s, and
imp o e quali y o li e o pa ien s [6–10]. While he diag-
nos ic delay o IEI a e he i s p esen a ion o symp-
oms is dec easing, i is s ill ypically be ween 1–4yea s
wo ldwide and 2–5yea s wi hin Eu ope, depending on he
coun y and he IEI [11]. Low awa eness o IEIs and lack
o esou ces a e among he p ima y causes o diagnos ic
delay [12–15]. The need o ea ly diagnosis o IEIs has
led o he de elopmen o ools ha may igge ea lie
clinical suspicion o IEIs by p ima y ca e physicians and
pedia icians, which in u n could lead o ea lie diagnosis
and managemen . The mos well-known o hese ools is
he SPIRIT analyze which is based upon he 10 Wa n-
ing Signs o P ima y Immunode iciency (JMF Wa ning
Signs), c ea ed by he Je ey Modell Founda ion in 1993
and upda ed wice since. The ool is based on a lis o
common wa ning signs aimed o help physicians iden i y
indi iduals wi h a suspec ed IEI, a ailable sepa a ely o
pedia ic and adul pa ien s. The e ec i eness o he JMF
Wa ning Signs has o en been e alua ed ega ding hei
abili y o de ec IEIs ha p esen p ima ily as an inc eased
suscep ibili y o in ec ions bu he e a e limi a ions among
he b oade ange o IEI pheno ypes [16–21]. Seconda y
wa ning sign lis s ha e been p oduced in a emp s o aid
in he diagnosis o unde lying IEIs in a ious indica ions,
including lung disease, oncohema ology, gas oen e ol-
ogy, de ma ology, and in ec ious diseases, among o he s
[2, 4, 22]. P e ious a emp s o e ise he JMF Wa ning
Signs in ligh o newly ecognized wa ning signs ha e
also been made, hough mos lis s appea o be limi ed
o a simila numbe and spec um o o al wa ning signs,
wi h some e o s ocusing on aising awa eness p ima ily
in he p ima y ca e se ing [17]. Mo e ecen ly, a emp s
ha e been made o p oduce medical expe sys ems (com-
pu e ools o aid physicians in diagnosis and ea men
choices) o acili a e he diagnosis o IEI wi hin p ima y
ca e se ings [23–31]. These ools may o e a mo e mod-
e n solu ion o aise heal hca e p o essionals’ awa eness
and clinical suspicion wi hin a p ima y ca e se ing. How-
e e , unlike highly p e alen diseases, which can be mo e
easily in es iga ed using e ospec i e da a om elec onic
heal h eco ds (EHRs), a e diseases such as IEIs a e o en
unde epo ed o a e no p ope ly egis e ed. This ea u e
challenges he de elopmen o p edic i e sco ing sys ems
based on s a is ical o machine-lea ning app oaches, which
equi e la ge amoun s o high-quali y da a on bo h cases
and con ols.
Wi h his in mind, we p oduced an expe -based sco -
ing sys em, de eloped by bo h p ima y ca e physicians and
immunologis s in Spain, based on ex ended wa ning signs o
he o iginal 10 JMF Wa ning Signs, o be applied o p ima y
ca e se ings.
Me hods
S udy O e iew andSe ing
The PIDCAP p ojec aimed o de elop and implemen a
sco ing sys em o he ea ly iden i ica ion o indi iduals
wi h IEIs in he p ima y ca e se ing. The p ojec was led
by a ask o ce consis ing o he ollowing p o iles: clinical
expe s and esea che s om he Child en’s Hospi al a Vall
d’Heb on Ba celona Hospi al Campus (Ca alonia, Spain),
echnical s a om he Ca alan Ins i u e o Heal h (ICS),
p ima y ca e consul o s, and a coo dina o om Innobics—
a i ual esea ch pla o m allowing o he applica ion and
managemen o p ojec s p oposed by ICS p o essionals.
The PIDCAP sco ing sys em was in ended o be embedded
in o he clinical wo ks a ion o he Ca alan Heal h Se ice’s p i-
ma y ca e se ings. The Ca alan Heal h Se ice p o ides public,
uni e sal ca e o he en i e popula ion o Ca alonia (8 million
inhabi an s) h ough a ne wo k o 64 gene al hospi als, 27 psy-
chia y hospi als, 375 p ima y ca e cen e s, 91 skilled nu sing
acili ies o in e media e ca e, and 130 ou pa ien men al heal h
acili ies. All p ima y ca e cen e s o he Ca alan Heal h Se ice
sha e a single clinical wo ks a ion (i.e., he eCAP pla o m) and
s o e all clinical in o ma ion in a single clinical da a eposi o y.
The ul ima e goal o he PIDCAP sco ing sys em was o inco -
po a e a buil -in ale sys em in he eCAP o iden i y pa ien s a
high isk o ha ing an IEI and o subsequen ly igge a e e al
o expe s o u he IEI in es iga ion.
Jou nal o Clinical Immunology (2025) 45:26 Page 3 o 13 26
All da a gene a ed and used in his s udy we e han-
dled acco ding o he Gene al Da a P o ec ion Regula ion
2016/679 on da a p o ec ion and p i acy o all indi iduals
wi hin he Eu opean Union and he local egula o y ame-
wo k ega ding da a p o ec ion. P ocedu es we e app o ed
by he esea ch e hics commi ee o he coo dina ing cen e
PR(AMI)339/2017. By he ime o admission o specialized
assessmen , all his o ical clinical da a o he pa ien s, i e-
spec i e o he cen e hey egula ly isi ed, was eques ed
and included in he EHR o he admi ing hospi al.
Sco ing Sys em De elopmen
Two sco ing sys ems we e concei ed: one o he pedia ic
popula ion (i.e., younge han 14yea s, as de ined by he
Ca alan Heal h Se ice) and one o he adul popula ion
(i.e. 14yea s o age and olde ). We used a quali a i e Delphi
me hodology o de eloping he sco ing sys ems ha would
be u ilized by he PIDCAP ool. The en i e p ocess consis ed
o h ee phases: (1) iden i ica ion o wa ning signs o IEI and
p elimina y signi icance weigh ing, (2) expe consensus on
he inclusion and weigh ing o each sign wi hin he sco ing
sys ems, and (3) assignmen o s uc u ed disease codes o
he iden i ied wa ning signs.
Po en ial wa ning signs known o sugges he p esence
o an IEI in bo h adul and pedia ic pa ien s we e de e -
mined ia an in-dep h li e a u e e iew. PubMed and Google
Schola da abases we e sc eened o English-language a i-
cles epo ing s udies o iden i y isk ac o s o IEI, includ-
ing hose de eloping and/o alida ing sco ing sys em o his
end. The sea ch combined key e ms (including a ian s wi h
he same oo e ms) ega ding p ima y immunode iciencies
and wa ning signs o sc eening s a egies. Sea ches we e
es ic ed om 1993 onwa d based on he i s published
lis o JMF wa ning signs, wi h no u he es ic ions. The
sea ch s a egy and esul a e desc ibed in he Supplemen-
a y Ma e ial. The ask o ce membe s e iewed all a icles
and ex ac ed he isk ac o s, hen collabo a ed wi h 8 local
expe s (consis ing o 3 pedia ic immunologis s, 1 p ima y
ca e pedia ician, 3 adul immunologis s, and 1 p ima y ca e
gene al p ac i ione ) o come up wi h an ini ial lis o wa n-
ing signs and hei weigh ing wi hin he sco ing sys em, by
means o a sco e be ween 10 and 75. Cu -o alues we e
a bi a ily es ablished: high isk (sco e ≥ 75), mode a e isk
(35–70), and low isk (< 35).
Fo he second phase, we se up a panel o expe s o a e
he ele ance o each o he p e-iden i ied wa ning signs
and i s ela i e con ibu ion (i.e., weigh ). We selec ed 36
expe s om ac oss e e ence cen e s in Ca alonia and Spain
a ending o pedia ic and/o adul pa ien s wi h IEIs, and
p ima y ca e physicians o hei expe ise and ealis ic iew
o p ima y ca e pi alls. Candida es we e in i ed o pa ici-
pa e ia email ques ionnai e wi h a esponse deadline se
5weeks ollowing in i a ion. Accompanying he ques ion-
nai e, he en olled expe s also ecei ed a summa y o he
chosen li e a u e, as well as in o ma ion on he a ionale and
backg ound o he PIDCAP p ojec o con ex . Expe s we e
asked o a e he inclusion and he sugges ed weigh ing o
each wa ning sign on a 1- o-4 scale based on hei ela i e
pe cei ed signi icance in he clinical suspicion o an unde -
lying IEI. Expe s could also sugges al e na i e weigh ings.
Consensus was conside ed when a gi en sign/weigh
sco ed 3 o highe by a leas hal o he pa icipa ing expe s.
In cases in which consensus o he sugges ed weigh ing o a
wa ning sign was no eached, hey we e ee alua ed by he
local ask o ce based on expe eedback. I ems no eaching
consensus could be ei he emo ed o ede ined. The inal
adul and pedia ic sco ing sys ems we e sen o he expe s,
along wi h he esul s o he consensus p ocess; he expe s
we e o e ed he oppo uni y o sugges inal amendmen s i
deemed necessa y.
Finally, he local ask o ce ansla ed he lis o wa ning
signs ( ypically epo ed in he li e a u e in na u al language)
in o codes o he In e na ional Classi ica ion o Diseases
e sion 10, Clinical Modi ica ion (ICD-10-CM).
Re ospec i e Tes ing in heIEI Coho
The esul ing sco ing sys em was es ed agains a e ospec-
i e egis y o pa ien s wi h con i med IEI a he coo dina -
ing cen e , using da a om hei p ima y ca e EHR. The
Vall d’Heb on Ba celona Hospi al Campus is ecognized
as a e e ence cen e egionally and na ionally and i is one
o he 3 ecognized Eu opean Re e ence Ne wo k o Ra e
Immunological Diso de s (ERN-RITA) cen e s o IEI in
Spain. The hospi al se es mo e han 950,000 pa ien s pe
yea e e ed om local p ima y heal h ca e (wi h a ca ch-
men popula ion o mo e han 500,000 indi iduals), bu also
e e ed om all o e Spain (especially in a e diseases).
Pedia ic and adul pa ien s diagnosed wi h an IEI and
egis e ed in he hospi al da abase be ween Ap il 2005 and
Janua y 2023 we e included in he analysis da ase . Da a on
p e ious diagnoses we e c oss- abula ed wi h he heal hca e
egis y o diagnoses o he Ca alan Minis y o Heal h,
which collec s all diagnoses epo ed o he SISAP cen al
egis y o p ima y ca e. The ollowing indi iduals we e
excluded: pedia ic pa ien s wi h selec i e IgA de iciency,
usually asymp oma ic; adul indi iduals wi h less han 5
diagnosis en ies in he p ima y ca e egis y (co esponding
o p10 o he numbe o adul diagnosis dis ibu ion) we e
conside ed non- ep esen a i e on he assump ion ha less
han 5 diagnosis en ies mean ha he e we e no ollowed
wi hin he public, uni e sal p ima y ca e sys em bu a he
ollowed by ei he a p i a e heal hca e sys em o ou side o
Ca alonia.
Jou nal o Clinical Immunology (2025) 45:26 26 Page 4 o 13
Jou nal o Clinical Immunology (2025) 45:26 Page 5 o 13 26
All pa ien s in he analysis da ase we e e alua ed using
he sco ing sys em, based on hei his o y o diagnoses
eco ded in he p ima y ca e egis y be o e IEI diagnosis.
Fo he e ospec i e es , he wa ning sign o ‘con i med
IEI’, in oduced o he sco ing sys em o ensu e adequa e
ollow-up o indi iduals wi h p e ious IEI diagnosis, was
emo ed as all indi iduals in he egis y al eady me he
c i e ia. We es ima ed he numbe and pe cen age o indi-
iduals alloca ed by he sco ing sys em in each o he isk
ca ego ies and he equency and pe cen age o each o he
wa ning signs. The analysis consis ed o desc ip i e s a is-
ics, and no hypo hesis es ing was conduc ed.
Pilo Implemen a ion inRou ine Ca e
In Ap il 2018, a pilo s udy was ini ia ed o es he ea-
sibili y o using he PIDCAP sco ing sys em in a clinical
wo ks a ion o a p ima y ca e eam. The sco ing sys em was
made a ailable o all heal hca e p o essionals o he El Ca -
mel, Ba celona, p ima y ca e eam, which p o ides ca e o
a ca chmen popula ion o 19,391 indi iduals: 16,794 adul s
and 2,597 child en. The sco ing sys em ac i a ed an ala m
along wi h a e e al ecommenda ion o an expe e alua-
ion o all pa ien s iden i ied a high isk o IEI. All e e als
we e eco ded and communica ed o he PIDCAP ask o ce.
Fig. 1 Ag eemen e e s o he deg ee o ag eemen o disag eemen
wi h he p oposed i ems and hei weigh . Resul s om 16 su ey
esponden s, 11 o whom a e iden i ied as pedia ic immunologis s, 1
as an adul /gene al immunologis and 4 as p ima y ca e pedia icians.
HSV: He pes Simplex Vi us, IEI: inbo n e o s o immuni y
◂
Table 1 Wa ning signs included in he inal pedia ic sco ing sys em
a Wa ning sign c i e ia amended ollowing he esul s o he su ey based on addi ional eedback p o ided by he su eyed expe s, and o align
wi h eal-wo ld implemen a ion ia coding
b Wa ning sign added based on ecommenda ion om su eyed expe , la e e iewed and implemen ed by local ask o ce
Wa ning sign o ‘Food in ole ance/alle gy’ emo ed ollowing he esul o he su ey. Wa ning signs o ‘2 o mo e mon hs o an ibio ic ea -
men ’ and ‘3 o mo e hospi al admissions/yea ’ emo ed due o limi a ions in ICD-10-CM coding wi hin he elec onic heal h eco d sys em
Wa ning sign Weigh in sco -
ing sys em
Wa ning sign Weigh in sco ing
sys em
Pedia ic pa ien s
 ≥ 10 acu e o i is mediaa20 Sys emic au oimmune diseases, no including
au oimmune cy openia (celiac disease, a h i is,
e c.)
30
 ≥ 3 sinusi is o o bi al celluli isa20 Endoc inopa hology: Hypo hy oidism, hype pa -
a hy oidism, diabe es, e c. (No desc ibed as
au oimmune)
30
 ≥ 3 pneumoniaa40 Hema ological malignancy 30
 Failu e o h i e 20 Solid o gan neoplasia (only hose ha ha e been
associa ed wi h inbo n e o s o immuni y in
pedia ics: hy oid)
30
 Deep abscesses (in o gans) 75 O al (den al/pala al) anomalies 20
 ≥ 3 ecu en skin abscess 20 Ch onic dia hea; o ≥ 10 episodes o acu e
dia hea)a
30
 Mucocu aneous candidiasis (o opha ynx, cu ane-
ous, excluded aginal) in pa ien s ≥ 12 mon hs o
age: ≥ 2 episodesa
30 Ch onic i al skin in ec ion; o ≥ 20 acu e episodes 10
 ≥ 2 sys emic in ec ions (including sepsis) 75 Ch onic eczema o o he de ma ological mani es-
a ions ela ed o inbo n e o s o immuni y
10
 ≥ 1 se ious in ec ions ha alone indica e IEI s udy
(meningi is caused by HSV, e c.)
75 Recu en e e 75
 Family his o y o inbo n e o s o immuni ya50 In lamma o y bowel disease in pa ien s ≥ 2yea s
o age
30
 Consanguini y o o he amily his o y compa ible
wi h mani es a ions o inbo n e o s o immuni y
(lymphomas, e c.)a
30 In lamma o y bowel disease in pa ien s < 2yea s
o age
75
 Cy openia (no speci ied as au oimmune) 20 B onchiec asis wi hou cys ic ib osis 75
 Au oimmune cy openiaa40 Vaccine eac ionb20
 P esence o 2 o mo e wa ning signs 10

Jou nal o Clinical Immunology (2025) 45:26 26 Page 6 o 13
Jou nal o Clinical Immunology (2025) 45:26 Page 7 o 13 26
In addi ion o he PIDCAP sco ing sys em, a se ies o educa-
ional sessions in p ima y and e ia y ca e cen e s was ca ied
ou be ween Oc obe 2017 and Janua y 2018 o p epa e o he
loca ions ecei ing he PIDCAP sco ing sys em and o aise
awa eness and clinical suspicion o IEIs wi hin hese loca ions.
Also, a se ies o in og aphics we e p oduced o highligh he
po en ial wa ning signs o IEIs in di e en indica ions. Dissem-
ina ion o hese esou ces was success ully achie ed h oughou
he a ge a ea in line wi h he goals o he p ojec .
Resul s
Sco ing Sys em De elopmen
Wa ning signs o conside a ion we e ex ac ed om he li -
e a u e sea ch epo ing wa ning signs and isk ac o s o
p ima y immunode iciency, and om medical expe ience
o he ask o ce eam. The li e a u e e iew, ollowed by
collabo a ion wi h local expe s, yielded 28 wa ning signs
o he pedia ic lis and 22 o he adul lis .
O he su eyed panel o 36 expe s, 22 (61%) answe ed
he online su ey: 16 he ques ionnai e o pedia ics and
10 o adul s. The gene al cha ac e is ics o he expe s a e
summa ized in TableS1.
O he 28 wa ning signs o iginally included in he su -
ey o pedia ics, 27 (96%) eached he p e-es ablished
consensus h eshold (Fig.1); ood alle gy did no achie e
a minimum o 50% ag eemen o s ong ag eemen . A e -
age sco e anged om 2.8 o 4.0 (TableS2). A e aking
in o accoun he expe e iew, one wa ning sign (‘ ood
alle gy’) was emo ed, and one i em (‘p esence o cy o-
penia’) was u he sepa a ed in o sepa a e wa ning signs
(‘cy openia ( ega dless o au oimmune o no )’ and ‘au o-
immune cy openia’) wi h wo di e en weigh ings. Fu -
he mo e, wo u he i ems we e no used la e due o
limi a ions in p ima y ca e coding (‘ ≥ 3 hospi al admis-
sions/yea ’ and ‘ ≥ 2mon hs o an ibio ic ea men ’). The
weigh ing o en i ems was adjus ed based on low sco e,
ee commen s and coding limi a ions. One i em (‘ accine
eac ion’) was added acco ding o expe ecommenda-
ion, alida ed by he local ask o ce. Table1 shows he
inal sco ing sys em o he pedia ic popula ion, which
includes 27 wa ning signs.
Rega ding he adul su ey, all 22 wa ning signs
eached he consensus h eshold (Fig.2); he a e age
sco e anged om 2.9 o 3.9 (TableS3). One i em was
pa i ioned in o wo (as abo e, ‘p esence o cy openia’).
Likewise, one i em was no used due o coding limi a-
ions (‘ ≥ 3 hospi al admissions/yea ’), he weigh ing o 8
i ems was adjus ed based on low sco e, ee commen s and
coding limi a ions, and wo i ems (‘O al (den al/pala al)
anomalies’ and ‘ch onic eczema o o he de ma ological
mani es a ions ela ed o inbo n e o s o immuni y’) we e
added acco ding o expe ecommenda ion, and la e ali-
da ed. Table2 shows he inal sco ing sys em o adul s,
which includes 24 wa ning signs.
O he o al 68,000 possible ICD-10-CM codes
e iewed ollowing he su ey, 3,387 o pedia ics and
3,497 o adul s we e e ained and assigned o each wa n-
ing sign (See Supplemen a y Ma e ial o he ICD-10-CM
codes). Subsequen ly, he ask o ce adap ed he sco ing
sys em o he speci ica ions o he eCAP clinical compu e
sys em o he Ca alan Heal h Se ice, aking in o accoun
he changes in oduced a e he su ey. Each wa ning sign
was pai ed manually by he ask o ce wi h ICD codes as
speci ied ea lie . In ou ine ca e, each isi o he p ima y
ca e cen e is conside ed an episode by he sys em, and
coding wi h ICD is manda o y. Some WS we e assigned
a numbe o episodes o o e come in insic limi a ions o
he coding sys em (e.g., ch onic dia hea was decided o
be equi alen o 10 acu e episodes).
Re ospec i e Tes ing inanIEI Coho
O e all, 305 pa ien s (184 child en and 121 adul s) wi hin
he Vall d’Heb on Ba celona Hospi al Campus eco ds me
he selec ion c i e ia o IEI and we e, he e o e, included
in he s udy. The main demog aphic cha ac e is ics o he
pa ien s included in he s udy a e summa ized in TableS4
and he diagnosis g oup classi ica ion in TablesS5 andS6.
The sco ing sys em classi ied 60 (32%) pedia ic pa ien s
wi h con i med IEI as high- isk indi iduals (Table3); 47
(78%) had 2 o mo e wa ning signs. Ch onic eczema, cy o-
penias, wo o mo e sys emic in ec ions, ecu en e e , ail-
u e o h i e, and b onchiec asis in absence o cys ic ib osis
we e he leading wa ning signs. Howe e , ch onic eczema
and ailu e o h i e we e also p esen in indi idual child en
iden i ied as low- isk by he sco ing sys em.
The sco ing sys em classi ied 36 (30%) adul pa ien s
wi h con i med IEI as high- isk indi iduals (Table4).
Nea ly all o hem (33/36 (92%)) had 2 o mo e wa ning
signs. The mos equen wa ning sign among adul s iden i-
ied as high isk was b onchiec asi in he absence o cys ic
ib osis, ollowed by sys emic and endoc ine au oimmune
diseases, cy openias, and mo e han 3 pneumonias. O no e,
sys emic and endoc ine au oimmune diseases we e also p e-
sen in indi idual adul s iden i ied as low- isk by he sco ing
sys em.
Fig. 2 Ag eemen e e s o he deg ee o ag eemen o disag eemen
wi h he p oposed i ems and hei weigh ing. Resul s om 10 su ey
esponden s, 9 o whom a e iden i ied as adul immunologis s, and 1
as a p ima y ca e physician. HSV: He pes Simplex Vi us, IEI: Inbo n
e o s o immuni y, TB: Tube culosis
◂
Jou nal o Clinical Immunology (2025) 45:26 26 Page 8 o 13
Pilo Implemen a ion
Du ing he pilo implemen a ion pe iod, he PIDCAP sco -
ing sys em es ed 16,794 adul s and 2,597 child en; o
hem, 286 (1.8%) adul s and 13 (0.5%) child en we e iden-
i ied as high isk o IEI. TableS7 summa izes he main
demog aphic cha ac e is ics o he sou ce popula ion. P i-
ma y ca e physicians we e no i ied o ollow up wi h hese
pa ien s. An ale in each pa ien ’s EHR was displayed o he
p ima y ca e physician. These ale s speci ically ou lined he
ICD-10-CM codes conside ed, he pa ien ’s o e all sco e,
and ex ac ed he mos ecen blood coun along wi h any
eco ded immunoglobulin le els. Wi hin his ale sys em,
heal hca e p o essionals we e p esen ed wi h h ee ac iona-
ble op ions: (1) eques a undamen al immunological wo k-
up (inclusi e o a ull blood coun and immunoglobulin le -
els) and wai o he esul s o de e mine he nex s ep, (2)
i ually e e he pa ien o a specialized e e ence cen e , o
(3) a ange an in-pe son e e al isi o he e e ence cen e .
Addi ionally, p ima y ca e p o essionals had he disc e ion
o dismiss he ale i hey deemed ha e e al unnecessa y
due o an al e na i e explana ion o i he pa ien was al eady
being ollowed up. The wo k low was e y well ecei ed in
p ima y ca e cen e s. A e p ima y ca e assessmen , a o al
o 40 adul and 3 pedia ic pa ien s we e e e ed o u he
immunological e alua ion. Un o una ely, owing o he p i-
o i iza ion c i e ia du ing he global COVID-19 heal hca e
c isis, no u he ollow-up in o ma ion is a ailable.
Discussion
In his s udy, we de eloped an expe -based sco ing sys em o
iden i ying indi iduals wi h IEI based on diagnoses eco ded in
he p ima y ca e se ing. The esul ing sco ing sys em expanded
on he 10 classical wa ning signs conside ed o IEI sc eening
wi h addi ional clinically meaning ul i ems. The sco ing sys em,
which can be e ec i ely implemen ed in p ima y ca e wo ks a-
ions, showed he abili y o iden i y indi iduals a high- isk o
IEI using e ospec i e da a s o ed in p ima y ca e eco ds.
Table 2 Wa ning signs included in he inal adul sco ing sys em
a Wa ning sign c i e ia amended ollowing he esul s o he su ey based on addi ional eedback p o ided by he su eyed expe s, and o align
wi h eal-wo ld implemen a ion ia coding
b Wa ning sign added based on ecommenda ion om su eyed expe s, la e e iewed and implemen ed by local ask o ce. Wa ning signs o ‘3
o mo e hospi al admissions/yea ’ emo ed due o limi a ions in ICD-10-CM coding wi hin he elec onic heal h eco d Sys em
HSV He pes Simplex Vi us
Wa ning sign Weigh in sco ing
sys em
Wa ning sign Weigh in sco -
ing sys em
Adul pa ien s
 ≥ 8 acu e o i is mediaa30 Consanguini y o o he amily his o y compa ible
wi h mani es a ions o inbo n e o s o immuni y
(haema ological neoplasms)a
30
 ≥ 8 sinusi is o ch onic sinusi isa30 P esence o cy openia (wi hou speci ying i au oim-
mune)
20
 ≥ 3 pneumoniaa30 Au oimmune cy openia 40
 Ch onic dia hea 30 P esence o b onchiec asis wi hou cys ic ib osis 75
 Deep abscesses (in o gans and/o ganglia) 50 Sys emic and endoc ine au oimmune diseases (celiac
disease, a h i is, sys emic lupus, hy oidi is, e c.)
30
 Recu en skin abscesses o epe i ion (3 o
mo e)a
20 Haema ological neoplasia (excluding mul iple mye-
loma, ch onic myeloid leukemia, Waldens öm’s
disease, e c.)a
50
 O opha yngeal o cu aneous candidiasis
(excluding aginal candidiasis)
30 Solid o gan neoplasia (only hose ela ed wi h inbo n
e o s o immuni y: skin, s omach, hy oid)a
30
 Recu en i al in ec ions (colds, he pes, wa s,
condylomas, e c.) 25 o mo e episodesa
30 In lamma o y bowel disease 20
 2 o mo e sys emic in ec ions including sepsis 75 Recu en e e 50
 Unique se e e condi ion ha alone equi e s udy
o inbo n e o s o immuni y
75 O al (den al/pala al) anomaliesb20
 A ypical mycobac e ia in ec ion 50 Ch onic eczema o o he de ma ological mani es a-
ions ela ed o inbo n e o s o immuni yb
10
 Family his o y o inbo n e o s o immuni ya50
 P esence o 2 o mo e wa ning signs 10
Jou nal o Clinical Immunology (2025) 45:26 Page 9 o 13 26
The de elopmen o sco ing sys ems o algo i hms o
au oma ic and ea ly iden i ica ion o indi iduals a high
isk o IEI in he p ima y ca e se ing aces wo impo an
challenges o be conside ed. Fi s , he e y low numbe o
diagnoses hinde s he de elopmen o sophis ica ed s a is-
ical models o p edic he p esence o IEI wi h adequa e
accu acy [32]. Second, he sca ci y o sc eenings, including
gene ic in es iga ions on indi iduals wi h clinical ea u es
sugges i e o IEI has also been associa ed wi h high le els
o unde epo ing [15, 33, 34]. Thus, some indi iduals con-
side ed o be con ols in his sco ing sys em de elopmen
migh be undiagnosed cases.
Despi e hese challenges, se e al au ho s ha e wo ked
o expand he 10 JMF Wa ning Signs [6, 18, 34] mos ly by
including non-in ec ious como bidi ies, unde - ep esen ed
in he JMF lis , such as au oimmune diso de s [18], o
hema ological signs [6], among o he s. Likewise, he panel
o expe s who pa icipa ed in he PIDCAP s udy, iden i-
ied non-in ec ious condi ions, such as cy openia, sys emic
au oimmune diseases, and ch onic skin condi ions, which
may aise suspicion o an IEI and may be impo an wa n-
ing signs o IEI. These indings also aligned wi h p e ious
wo k by Dąb owska e al., who highligh ed he impo ance
o o he wa ning signs (hema ooncologics, au oimmuni y,
and eczema) wi h simila indings [35].
The applica ion o ou sco ing sys em should be con-
side ed wi hin i s in ended use, which is au oma ically
assessing he isk o IEI om in o ma ion s o ed in p i-
ma y ca e EHRs. This app oach p ecludes wa ning signs
ha a e no adequa ely epo ed o eliable in p ima y ca e
eco ds. This is he case o amily his o y o IEI, which
has been highligh ed as an impo an p edic o o IEI [16,
Table 3 The sco ing sys em classi ica ions o pedia ic pa ien s wi h con i med IEI
Wa ning signs Risk To al
High Medium Low
Pedia ic IEI coho e ospec i e es
 ≥ 10 acu e o i is media 3 1 0 4
 ≥ 3 sinusi is o o bi al celluli is 0 0 0 0
 ≥ 3 pneumonias 8 3 0 11
 Failu e o h i e 12 5 7 24
 Deep abscesses (in o gans) 0 0 0 0
 ≥ 3 ecu en skin abscesses 11 5 2 18
 Mucocu aneous candidiasis (o opha ynx, cu aneous, excluded aginal) in pa ien s ≥ 12 mon hs o age: ≥ 2
episodes
2 0 0 2
 ≥ 2 sys emic in ec ions (including sepsis) 17 0 0 17
 ≥ 1 se ious in ec ion ha alone indica e IEI s udy (Meningi is caused by HSV, e c.) 10 0 0 10
 Family his o y o inbo n e o s o immuni y 1 1 0 2
Consanguini y o o he amily his o y compa ible wi h mani es a ions o inbo n e o s o immuni y (lymphomas,
e c.)
4 3 0 7
 Cy openia (no speci ied as au oimmune) 24 8 6 38
 Au oimmune cy openia 5 1 0 6
 Sys emic au oimmune diseases, no including au oimmune cy openia (celiac disease, a h i is, e c.) 11 2 2 15
 Endoc inopa hology: Hypo hy oidism, hype pa a hy oidism, diabe es, e c. (No desc ibed as au oimmune) 7 1 0 8
 Hema ological malignancy 2 1 1 4
 Solid o gan neoplasia (only hose ha ha e been associa ed wi h inbo n e o s o immuni y in pedia ics: hy-
oid)
0 0 0 0
 O al (den al/pala al) anomalies 11 1 5 17
 Ch onic dia hea; o ≥ 10 episodes o acu e dia hea) 2 0 1 3
 Ch onic i al skin in ec ion; o ≥ 20 acu e episodes 0 0 0 0
 Ch onic eczema o o he de ma ological mani es a ions ela ed o inbo n e o s o immuni y 22 6 14 42
 Recu en e e 14 0 0 14
 In lamma o y bowel disease in pa ien s ≥ 2 yea s o age 0 0 0 0
 In lamma o y bowel disease in pa ien s < 2 yea s o age 0 0 0 0
 B onchiec asis wi hou cys ic ib osis 12 0 0 12
 Vaccine eac ion 0 0 0 0
 P esence o 2 o mo e i ems o hose desc ibed abo e 47 16 0 63
 Numbe o pa ien s 60 20 104 184