Symp oma ic ne wo ks in suicide a emp and
ea emp : Rele ance o psychia ic como bidi y
And es Pemau
1
,Alejand o de la To e-Luque
2,3
,Ca olina Ma in-Ma in
1
,
Ma ina Diaz-Ma sa
2,3,4
,Jo ge And eo-Jo e
5,6
,Wala Ayad-Ahmed
4
,
Ma ia Fe B a o O iz
3,5,6,7
,Ma ia Te esa Bobes-Basca án
3,8,9,10,11
,
Manuel Canal-Ri e o
3,12,13
,I ene Canosa Ga cía
3,14
,Ana Isabel Ceb ià
3,15,16
,
Benedic o C espo-Faco o
3,12,13
,Ma ía Ángeles Bo i
17,18,19
,Ma ilde Elices
14
,
Ana González-Pin o
3,20
,I ia G ande
3,17,18,19
,Luis Jiménez-T e iño
3,8,9,10,11
,
Diego J. Palao
3,15,21
,Angela Palao-Ta e o
5,6,7
,Ca la Pé ez-Gue a
2
,
Na alia Robe o
17,18,19
,Miguel Ruiz Veguilla
3,12,13
,Pila A. Sáiz
3,8,9,10,11
,
The SURVIVE Conso ium and Víc o Pé ez
3,14,22
1
Depa amen o de Pe sonalidad, E aluación y Psicología Clínica, Uni e sidad Complu ense de Mad id, Mad id, Spain;
2
Depa amen o de Medicina Legal, Psiquia ía y Pa ología, Uni e sidad Complu ense de Mad id, Mad id, Spain;
3
Cen e
o Biomedical Resea ch in Men al Heal h (CIBERSAM), Spain;
4
San Ca los Uni e si y Clinic Hospi al, Mad id, Spain;
5
Hospi al La Paz Ins i u e o Heal h Resea ch (IdiPAZ), Mad id, Spain;
6
Depa men o Psychia y, Uni e sidad
Au ónoma de Mad id (UAM), Mad id, Spain;
7
Depa men o Psychia y, Clinical Psychology and Men al Heal h, La
Paz Uni e si y Hospi al, Mad id, Spain;
8
Depa men o Psychology, Uni e sidad de O iedo, O iedo, Spain;
9
Ins i u o de
In es igación Sani a ia del P incipado de As u ias (ISPA), O iedo, Spain;
10
Ins i u o Uni e si a io de Neu ociencias del
P incipado de As u ias (INEUROPA), O iedo, Spain;
11
Se icio de Salud del P incipado de As u ias (SESPA), O iedo,
Spain;
12
Hospi al Vi gen del Rocio, IBIS, Se ille, Spain;
13
Uni e sidad de Se illa, Se ille, Spain;
14
Hospi al del Ma Medical
Resea ch Ins i u e (IMIM), Ba celona, Spain;
15
Men al Heal h Se ice, Hospi al Uni e si a i Pa c Taulí, Uni a Mix a de
Neu ociència T aslacional, Ba celona, Spain;
16
Depa men o Clinical and Heal h Psychology, Facul y o Psychology,
Uni e si a Au ònoma de Ba celona, Spain;
17
Depa amen de Medicina, Facul a de Medicina i Ciències de la Salu ,
Ins i u de Neu ociències, Uni e si a de Ba celona (UB), Ba celona, Spain;
18
Bipola and Dep essi e Diso de s Uni ,
Hospi al Clinic de Ba celona, Ba celona, Spain;
19
Ins i u d’In es igacions Biomèdiques Augus Pi i Sunye (IDIBAPS),
Ba celona, Spain;
20
Depa men o Psychia y, Hospi al Uni e si a io Ala a, BIOARABA, UPV/EHU, Vi o ia, Spain;
21
Depa men o Psychia y and Fo ensic Medicine, Facul y o Medicine, Uni e si a Au ònoma de Ba celona, Spain and
22
Ins i u o de Salud Men al, Hospi al del Ma , Ba celona, Spain
Abs ac
Backg ound. One o he mos ele an isk ac o s o suicide is he p esence o p e ious
a emp s. The symp oma ic p o ile o people who ea emp suicide dese es a en ion. Ne wo k
analysis is a p omising ool o s udy his ield.
Objec i e. To analyze he symp oma ic ne wo k o pa ien s who ha e a emp ed suicide
ecen ly and compa e ne wo ks o people wi h se e al a emp s and people wi h jus one a
baseline.
Me hods. 1043 adul pa icipan s om he Spanish coho “SURVIVE”we e pa o his s udy.
Pa icipan s we e classi ied in o wo g oups: single a emp g oup (n = 390) and ea emp g oup
(n = 653). Di e en ne wo k analyses we e ca ied ou o s udy he ela ionships be ween suicidal
idea ion, beha io , psychia ic symp oms, diagnoses, childhood auma, and impulsi i y. A
gene al ne wo k and one o each subg oup we e es ima ed.
Resul s. People wi h se e al suicide a emp s a baseline sco ed signi ican ly highe ac oss all
clinical scales. The symp oma ic ne wo ks we e equi alen in bo h g oups o pa ien s (p > .05).
Al hough he e we e no o e all di e ences be ween he ne wo ks, some nodes we e mo e
ele an acco ding o g oup belonging.
Conclusions. People wi h a his o y o p e ious a emp s ha e g ea e psychia ic symp om
se e i y bu he ela ionships be ween isk ac o s show he same s uc u e when compa ed
wi h he single a emp g oup. All isk ac o s dese e a en ion ega dless o he numbe
o a emp s, bu assessmen s can be adjus ed o be e moni o he occu ence o ea emp s.
Highligh s
•People who ea emp suicide ha e g ea e se e i y ac oss mos clinical scales applied.
•Como bidi y and non-suicidal sel -ha m a e cen al in symp oma ic ne wo ks.
•The single a emp and ea emp symp oma ic ne wo ks a e equi alen .
•Assessmen s can be adjus ed o be e moni o he occu ence o ea emp s.
Eu opean Psychia y
www.camb idge.o g/epa
Resea ch A icle
Ci e his a icle: Pemau A, de la To e-
Luque A, Ma in-Ma in C, Diaz-Ma sa M,
And eo-Jo e J, Ayad-Ahmed W, B a o
O iz MF, Bobes-Basca án MT, Canal-Ri e o M,
Canosa Ga cía I, Ceb ià AI, C espo-Faco o B,
Bo i MÁ, Elices M, González-Pin o A, G ande I,
Jiménez-T e iño L, Palao DJ, Palao-Ta e o A,
Pé ez-Gue a C, Robe o N, Ruiz Veguilla M,
Sáiz PA, The SURVIVE Conso ium, Pé ez V
(2025). Symp oma ic ne wo ks in suicide
a emp and ea emp : Rele ance o
psychia ic como bidi y. Eu opean Psychia y,
68(1), e4, 1–10
h ps://doi.o g/10.1192/j.eu psy.2024.1807
Recei ed: 13 June 2024
Re ised: 21 Oc obe 2024
Accep ed: 12 No embe 2024
Keywo ds:
impulsi i y; ne wo k analysis; suicide; suicide
ea emp ; auma
Co esponding au ho :
Ma ina Diaz-Ma sa;
Email: [email p o ec ed]
© The Au ho (s), 2025. Published by Camb idge
Uni e si y P ess on behal o Eu opean
Psychia ic Associa ion. This is an Open Access
a icle, dis ibu ed unde he e ms o he
C ea i e Commons A ibu ion-
NonComme cial-Sha eAlike licence (h p://
c ea i ecommons.o g/licenses/by-nc-sa/4.0),
which pe mi s non-comme cial e-use,
dis ibu ion, and ep oduc ion in any medium,
p o ided he same C ea i e Commons licence
is used o dis ibu e he e-used o adap ed
a icle and he o iginal a icle is p ope ly ci ed.
The w i en pe mission o Camb idge Uni e si y
P ess mus be ob ained p io o any
comme cial use.
h ps://doi.o g/10.1192/j.eu psy.2024.1807 Published online by Camb idge Uni e si y P ess
In oduc ion
Mo e han 720,000 people die annually by suicide a ound he
wo ld [1]. The WHO has u ged o implemen na ional plans o
cu e he inc easing ends o suicide mo ali y obse ed in some
coun ies in ecen yea s [2-4].
Suicide includes a se ies o complex and luc ua ing hough s
and beha io s, om passi e ideas o dea h o suicide a emp s and
ea emp s. Classical s udies ha e in ended o unde s and his
phenomenon ocusing on speci ic isk ac o s o de ec and p e en
suicide [5-7]. Some o he mos s udied isk ac o s a e impulsi i y,
childhood auma, dep essi e symp oms, o he p esence o p e i-
ous suicide a emp s [8-13]. Speci ically, he p esence o p e ious
suicide a emp s is one o he mos c i ical isk ac o s o ea -
emp s. Recen wo k sugges s ha be ween 20 and 30% o people
who a emp ed suicide will do so again [14,15].
Despi e ob aining aluable da a, his app oach has p o en
limi ed. Mo e ecen ly, idea ion-ac ion models ha e gained ele-
ance [16-19]. These models in end o s udy why some people
ansi ion om suicidal idea ion o suicide a emp s aising he
polyhed ic and mul icausal na u e o suicide. The in eg a ed
mo i a ional- oli ional model has gained he mos ele ance wi hin
his app oach [20]. This model p oposes h ee phases in he suicidal
p ocess: p e-mo i a ional, mo i a ional and oli ional. A i s , and
h ough a iables such as de ea and en apmen , suicidal idea ion
would a ise. La e , h ough he ac ion o ce ain mode a o s, his
idea ion could lead o a suicide a emp . Howe e , many gaps
emain unclea abou how hei in e ac ion inc eases he isk o
suicidal beha io [21,22].
Mo ing om single- ac o models o he idea ion-ac ion pe -
spec i e, he e olu ion in he ield o s udy has come hand in hand
wi h new s a is ical analysis. One o he echniques in oduced
wi h p omising esul s is Ne wo k Analysis. Ne wo k analysis
used o s udy men al pa hology a ises om Bo sboom’sp oposal
and goes beyond being a me e s a is ical app oxima ion [23]. In
his wo k, he sugges ed ha men al pa hology should be unde -
s ood as a complex sys em, ea u ed by he cons an in e ac ion
be ween ele an symp oms. Recu en in e ac ions be ween
symp oms can he e o e be e lec ed by ne wo k s uc u es. Ne -
wo k analysis also allows o know which symp oms a e mos
cen al (mo e in e connec ed and he e o e ele an ) o he diag-
nosis s udied. This way, we could be e cha ac e ize he diag-
noses, begin a i s causal app oach o he phenomena and
e en ually de elop be e ea men s [24]. Fu he mo e, he ne -
wo k p oposal escapes he educ ionism o he adi ional diag-
nos ic ision. I de ies he no ion o common causes o symp oms
and ecognizes he ele ance o eedback loops in psychopa h-
ology [25].
Al hough suicide is no a diagnosis, di e en wo ks ha e ied o
b ing his philosophy o analysis close o suicidal beha io [26-
33]. To da e, isk ac o s s udied, popula ions and esul s p esen
high a iabili y [30-33]. In addi ion o his a iabili y, wo ks
ocused on his echnique a e s ill sca ce.
Se e al au ho s aise he eno mous po en ial o hese echniques
o alida e complex models o suicidal beha io and o compa e
g oups o pa ien s by pe sonalizing ea men s [21,27]. Compa ing
g oups o people wi h a single suicide a emp e sus se e al
a emp s is especially p omising, and i could help de ec di e en
p o iles and isk ac o s [34-36].
Some p e ious s udies ha e app oached his opic, eaching
di e en conclusions. Nuñez e al. [30] ound some di e ences in
he ne wo ks o single-a emp and ea emp g oups, al hough no
s a is ically signi ican . De Beu s e al. [28] also ound no signi ican
di e ences when ocusing on suicidal idea ion.
To o e come some o he limi a ions o p e ious wo k, we
sea ched o people who had a emp ed suicide ecen ly (las
10 days). In addi ion, isk ac o s om mul iple domains
(mo i a ional, oli ional, cogni i e, demog aphic, e c.) we e
included. Speci ically, he isk ac o s conside ed we e impulsi i y,
childhood auma, psychia ic symp oms, p e ious suicidal beha -
io s, non-suicidal sel -ha m, subs ance use, sex, age, and acqui ed
capabili y o suicide. The gene al symp om ne wo k was s udied
based on hese isk ac o s, as well as hei cen ali y and s abili y
indices. Subsequen ly, we compa ed whe he he ne wo k o he
single a emp g oup and ha o he ea emp g oup di e ed in
hei s uc u e.
Ou hypo heses a e p esen ed below. Rega ding he gene al
ne wo k, we belie e ha anxie y, dep ession, and idea ion will be
cen al nodes based on p e ious wo k [26,28,30]. Rega ding di e -
ences be ween g oups, we hypo hesize ha he symp om ne wo k
will be mo e s ongly connec ed in he ea emp g oup han in he
single-a emp g oup. Bo sboom [23] sugges s ha symp oms end
up gene a ing s abili y i hey end o occu oge he . We also
belie e ha impulsi i y will be mo e cen al in he ea emp g oup
ne wo k [30]. Also, he a iable o acqui ed capabili y will p esen
g ea e cen ali y in he ea emp g oup. The acqui ed capabili y is
di ec ly ela ed o g ea e pain ole ance and knowledge o suicide
me hods [16].
Me hod
Pa icipan s
Fo he cu en s udy, 1043 pa ien s admi ed a di e en hospi al
eme gency depa men s due o a suicide a emp pa icipa ed. The
sample came om he “Suicide P e en ion and In e en ion S udy
(SURVIVE)”coho . The SURVIVE s udy pu s oge he esea ch
e o s om esea che s o 10 hospi als sp ead ac oss he Spanish
e i o y. The e hical commi ees o all he hospi als in ol ed
app o ed he s udy. The s udy p o ocol is desc ibed in mo e de ail
elsewhe e [37].
Fo he p esen wo k, he inclusion c i e ia we e he ollowing:
(a) people olde han 18 yea s, (b) a emp ca ied ou wi h a leas
some wish o die, and (c) suicide a emp wi hin he 10 days be o e
he e alua ion. Exclusion c i e ia we e he ollowing: (a) di icul ies
in unde s anding he ins uc ions, ei he due o cogni i e impai -
men o language, (b) unclea in en ionali y o he e en ,
(c) medical damage a e he a emp ha makes i impossible o
answe he ques ionnai es, and (d) he pa ien had mo e han 30
o al li e ime a emp s (conside ing comple ed, abo ed and in e -
up ed). All pa icipan s illed ou he co esponding in o med
consen .
Da a collec ion was pe o med be ween Decembe 2020 and
Ma ch 2023. Pa ien ’s in e iews we e done by specialized men al
heal h pe sonnel on each ec ui men si e.
Pa icipan s we e classi ied in o wo g oups acco ding o he
exis ence o p e ious suicide a emp s: a ea emp g oup, including
people who p esen ed comple ed a emp s p io o he index, and a
single a emp g oup, whose index a emp was he i s .
Ins umen s
Socio-demog aphics, clinical da a, and cha ac e is ics o he sui-
cidal beha io we e collec ed using a clinical in e iew.
2 Pemau e al.
h ps://doi.o g/10.1192/j.eu psy.2024.1807 Published online by Camb idge Uni e si y P ess
Pa ien s we e e alua ed using a s uc u ed diagnos ic in e iew.
I explo es he main psychia ic diso de s o he DSM-5 [38]. Fo
he analyses, he o al numbe o diagnoses was summed. The
p esence o subs ance abuse, bo h alcohol and d ugs, was also
conside ed in he analysis gi en he ele ance o hese ac o s in
p e ious wo ks [36].
Psychia ic symp oma ology was e alua ed using he B ie
Symp om In en o y (BSI) [39,40]. I is a sel -adminis e ed sc een-
ing scale o psychopa hology. I comp ises 53 i ems di ided in o
di e en subscales: soma iza ion, obsessi e-compulsi e, in e pe -
sonal sensi i i y, dep ession, anxie y, hos ili y, phobia, pa anoia,
and psycho icism. The C onbach’s alpha eliabili y coe icien o
he subscales in he Spanish e sion anges be ween .72 < α< .84.
The Spanish alida ion s udy ound he same nine ac o s uc u e
as he o iginal wo k (using con i ma o y ac o analysis).
Impulsi i y was e alua ed using he Ba a Impulsi i y Scale
(BIS-11) [41,42]. This is a sel -adminis e ed scale o 28 ques ions. I
allows ob aining a global impulsi i y sco e as well as h ee sub-
scales: cogni i e, mo o , and unplanned impulsi i y. In he Spanish
e sion, he in e nal consis ency is a ound .8. The es – e es eli-
abili y a e 2 mon hs is .89. The alidi y pa ame e s ( ac o ial
s uc u e) ob ained we e accep able.
Va iables ela ed o he cu en suicide a emp we e assessed
wi h he Columbia Suicide Ra ing Scale (C-SSRS) [43,44]. The
C-SSRS is a clinician-adminis e ed scale ha e alua es di e en
aspec s o suicidal idea ion and beha io . I includes aspec s such as
in ensi y o suicidal idea ion, ypes o suicidal beha io (comple ed,
abo ed, and in e up ed a emp s), and le hali y o said a emp s.
I ems e e ing o he se e i y o idea ion we e included in he
ne wo k (mos se e e idea ion, equency, du a ion, con ollabil-
i y). These domains we e conside ed because ecen wo k poin s
ou he impo ance o adequa ely cha ac e izing suicidal idea ion
and ecognizing di e en aspec s o i [45]. Howe e , he easons
and de e en s o idea ion we e no conside ed in he ne wo k as
hey a e eminen ly quali a i e [44]. I also inqui es abou he
p esence o non-suicidal sel -ha m. The Spanish adap a ion p e-
sen s adequa e con e gen and di e gen alidi y. In his e sion,
C onbach’s alpha was calcula ed only o he idea ion scale, ob ain-
ing a alue o .53.
Childhood mal ea men and abuse- ela ed in o ma ion was
collec ed by using he Childhood T auma Ques ionnai e (CTQ-
SF) [46,47]. This sel -adminis e ed ques ionnai e consis s o
28 i ems. I includes i e subscales: sexual abuse, physical abuse,
emo ional abuse, physical neglec , and emo ional neglec . The
C onbach’s alpha o he subscales in he Spanish sample is be ween
.66 < α< .94 ( he lowes being physical neglec ). The Spanish
adap a ion showed good i o he i e- ac o s uc u e.
Acqui ed capabili y o suicide was assessed using he Acqui ed
Capabili y o Suicide Scale Fea lessness Abou Dea h (ACSS-FAD)
[48]. This is a 7-i em sel -adminis e ed scale, ocused speci ically on
he lack o ea o dea h. The scale p esen ed adequa e con e gen
and disc iminan alidi y.
Da a analysis
Fi s , desc ip i e analyses we e pe o med. Subsequen ly, χ
2
es s
we e pe o med o compa e quali a i e a iables be ween g oups
(single a emp and ea emp ). E ec sizes we e ob ained using
C ame ’s . Fo quan i a i e a iables, S uden ’s - es s o inde-
penden samples o Mann-Whi ney’su- es s we e used (in case o
highly asymme ic dis ibu ions). Hedges’sgwas used as a measu e
o he e ec size in he i s case and Pea son’s in he second case.
A e his, he gene al ne wo k was es ima ed using all pa ien s.
The ne wo k analysis app oach was used o s udy he complex
pa e ns o in e ac ions be ween isk ac o s o suicide. Th ee
ne wo ks we e es ima ed: one o he comple e sample, one o
he single-a emp g oup, and one o he ea emp g oup. In he
ne wo k, nodes ep esen isk ac o s, bo h demog aphic and clin-
ical: age, sex, psychia ic symp oma ology and diagnoses, impul-
si i y, suicidal idea ion, childhood auma, and acqui ed capabili y
o suicide; and he edges joining he nodes ep esen he ela ion-
ship be ween hem once he o he ela ionships a e conside ed.
Mixed G aphical Modeling (MGM) was used o ne wo k es ima-
ion. Ne wo ks we e weigh ed and egula ized by he Leas Absolu e
Sh inkage and Selec ion Ope a o (LASSO).
The in e p e a ion o ne wo ks should no be based on isual
ep esen a ion alone. This can lead o a misunde s anding o he
ele ance and ela ionship o he nodes. Fo his eason, di e en
cen ali y measu es a e included [49]. Th ee cen ali y es ima es a e
p esen ed o desc ibe he ele ance o he di e en symp oms:
s eng h, closeness, and be weenness. S eng h exp esses he sum
o he edges o a gi en node. Closeness is a measu e o he a e age
sho es dis ance om nodes. Be weenness indica es he numbe o
imes a node is on he sho es pa h be ween wo o he nodes. A
highe sco e in any o he h ee indices indica es g ea e cen ali y in
he ne wo k. All measu es a e p esen ed as s anda dized. The
p edic abili y index was also calcula ed. This index ells us how
well we can p edic a ce ain node based on he o he s. Ge s alues
be ween 1 (comple ely de e mined node) and 0 (independen o he
o he s) [50].
Finally, ne wo k obus ness was es ed using boo s apping
me hods [51]. We will conside accep able s abili y o be abo e .5
[51]. Each o he h ee ne wo ks is accompanied by i s co espond-
ing cen ali y and obus ness alues. To es o signi ican di e -
ences in ne wo k s eng h and s uc u e be ween he single-a emp
g oup and he ea emp one, we used he Ne wo k Compa ison
Tes (NCT). I is a pe mu a ion-based hypo hesis es , ha can
assess he di e ence be ween wo ne wo ks [52]. Abou 1000
i e a ions we e conside ed o he gene al compa isons. In he case
o compa isons be ween edges, we wo ked wi h 500 i e a ions.
The analyses we e ca ied ou using SPSS 28.0.1.1 and R
so wa e e sion 4.2.2 (packages dply , boo ne , ne wo k ools, Ne -
wo kCompa isonTes , and qg aph).
Resul s
Table 1 shows desc ip i e da a on he sociodemog aphic and
heal h- ela ed a iables. Da a a e p esen ed o all pa icipan s
(n = 1043) as well as o subg oups based on numbe o suicide
a emp s: single a emp g oup (n = 390) s ea emp g oup
(n = 653).
Di e ences we e only ound in wo sociodemog aphic a iables:
ma i al s a us and employmen s a us (p< .01). Rega ding clinical
scales, signi ican di e ences we e ound in all cases (p< .01) excep
in Non-Planning Impulsi i y. The e ec sizes o di e ences we e
small o medium ac oss all ac o s, excep o he numbe o suicidal
beha io s being la ge (p< .01; = .75) [53]. The ea emp g oup
p esen ed g ea e se e i y in all cases.
Table 2 shows an analysis ela ed o suicidal idea ion om he
C-SSRS. In summa y, signi ican di e ences we e ound in all cases.
G ea e se e i y was mo e p esen in he ea emp g oup (p< .01;
C ame ’s = .13–.23). E ec sizes we e small o mode a e.
Eu opean Psychia y 3
h ps://doi.o g/10.1192/j.eu psy.2024.1807 Published online by Camb idge Uni e si y P ess
Table 1. Compa isons o sociodemog aphic and clinical da a be ween he single a emp g oup and he ea emp g oup (N = 1043)
Va iable Full sample Single a emp g oup (n = 390) Rea emp g oup (n = 653) χ
2
/u/ E ec size
Age 40.29(15.73) 41.43(17.25) 39.59(14.72) 1.77 .12
Gende
Male 293(28.1%) 123(31.5%) 170(26%) 3.66 .06
Female 750(71.9%) 267(68.5%) 483(74%)
Na ionali y
Spanish 782(75%) 289(74.1%) 493(75.5%) .253 .02
O he 261(25%) 101(25.9%) 160(24.5%)
Ma i al s a us
Single 384(36.8%) 132(33.8%) 252(38.6%) 18.17** .13
Ma ied 231(22.1%) 110(28.2%) 121(18.5%)
In a ela ionship 191(18.3%) 68(17.4%) 123(18.8%)
Di o ced 213(20.4%) 67(17.2%) 146(22.4%)
Widowe 24(2.3%) 13(3.3%) 11(1.7%)
Highes educa ional le el comple ed
No o mal educa ion 20(1.9%) 7(1.8%) 13(2%) 6.62 .08
P ima y educa ion 157(15.1%) 46(11.8%) 111(17%)
Seconda y educa ion 542(52%) 203(52.1%) 339(51.9%)
Uni e si y 324(31.1%) 134(34.4%) 190(29.1%)
Employmen s a us
Unemployed 286(27.6%) 89(22.8%) 197(30.4%) 19.40** .14
Employed 423(40.8%) 189(48.5%) 234(36.1%)
S uden 138(13.3%) 49(12.6%) 89(13.7%)
Re i ed 97(9.3%) 28(9.7%) 59(9.1%)
Unable o wo k 94(9.1%) 25(6.4%) 69(10.6%)
No. o suicidal beha io s 4.57(5.06) 1.54(1.11) 6.37(5.61) 15131** .75
No. o diagnoses 4.27(2.39) 3.55(2.14) 4.70(2.44) 8.05** .50
Subs ance abuse
Yes 221(21.2%) 61(15.6%) 160(24.5%) 11.56** .11
No 821(78.7%) 329(84.4%) 492(75.5%)
ACSS-FAD 18.67(6.47) 17.50(6.58) 19.38(6.58) 4.56** .29
BIS–11
A en ional 21.19(3.75) 20.39(3.76) 21.67(3.67) 5.40** .35
Mo o 24.06(5.82) 22.62(5.48) 24.92(5.85) 6.28** .40
Nonplanning 26.95(5.88) 26.58(5.68) 27.17(5.99) 1.57 .10
BSI
Soma iza ion 1.50(.98) 1.30(.97) 1.62(.97) 102045** .16
OCD 2.19(1.05) 1.99(1.06) 2.30(1.02) 105110.5** .14
Sensi i i y 1.99(1.16) 1.75(1.14) 2.14(1.15) 102026.5** .16
Dep ession 2.71(1.09) 2.45(1.16) 2.87(1.01) 99301.5** .18
Anxie y 1.86(1.00) 1.66(.99) 1.99(0.99) 102520.5** .16
Hos ili y 1.31(1.06) 1.05(.98) 1.46(1.07) 96808** .20
Phobic 1.33(1.12) 1.12(1.05) 1.45(1.14) 104728** .14
Pa anoid 1.65(1.02) 1.45(1.02) 1.77(1.00) 103245.5** .15
Psycho icism 1.74(.96) 1.54(.98) 1.87(.93) 101564** .17
Con inued
4 Pemau e al.
h ps://doi.o g/10.1192/j.eu psy.2024.1807 Published online by Camb idge Uni e si y P ess
Table 1. Con inued
Va iable Full sample Single a emp g oup (n = 390) Rea emp g oup (n = 653) χ
2
/u/ E ec size
CTQ-SF
Emo ional abuse 13.27(6.62) 11.63(6.11) 14.27(6.72) 98421** .19
Physical abuse 9.58(5.90) 8.24(5.11) 10.41(6.21) 99995.5** .19
Sexual abuse 9.46(6.71) 7.86(5.38) 10.43(7.24) 102269.5** .18
Emo ional negligence 13.28(5.69) 12.36(5.61) 13.86(5.67) 107744** .13
Physical negligence 8.58(3.95) 8.13(3.66) 8.86(4.09) 114123** .09
Non-suicidal sel -ha m
Yes 440(42.2%) 115(29.5%) 325(49.8%) 41.19** .20
No 603(57.8%) 275(70.95%) 328(50.2%)
No e: Sco es a e p esen ed as mean (s anda d de ia ion) o con inuous a iables and numbe (pe cen age) o ca ego ical ones. χ
2
is p esen ed o ca ego ical a iables; uis p esen ed o
numbe o suicidal beha io s, BSI sub sco es and CTQ-SF sub sco es. is p esen ed in he es o he a iables. (*) s a is ically signi ican di e ences a p < .05. (**) s a is ically signi ican
di e ences a p < .01.
ACSS-FAD, Acqui ed Capabili y o Suicide Scale Fea lessness Abou Dea h; BIS-11, Ba a impulsi i y scale; BSI, B ie symp oms in en o y; CTQ-SF, Childhood T auma Ques ionnai e- sho o m.
Gende was ca ego ized as 0 = emale and 1 = male. Nºo suicidal beha io s accoun s o all a emp s, whe he comple ed o o he wise.
Table 2. Compa isons o C-SSRS idea ion sco es be ween he single a emp and ea emp g oups
Va iable Full sample Single a emp g oup (n = 390) Rea emp g oup (n = 653) χ
2
E ec size
Mos se e e idea ion
No idea ion 107(10.3%) 50(12.8%) 57(8.7%) 53.81** .23
Wish o be dead 54(5.2%) 35(9%) 19(2.9%)
Nonspeci ic ac i e suicidal hough s 102(9.8%) 57(14.6%) 45(6.9%)
Ac i e idea ion wi hou in en 205(19.7%) 76(19.5%) 129(19.8%)
Ac i e idea ion wi h in en , no plan 240(23%) 83(21.3%) 157(24%)
Ac i e idea ion wi h plan and in en 335(32.1%) 89(22.8%) 246(37.7%)
F equency
No idea ion 107(10.3%) 50(12.8%) 57(8.7%) 42.85** .20
Less han once a week 194(18.6%) 103(26.4%) 91(13.9%)
Once a week 58(5.6%) 23(5.9%) 35(5.4%)
2–5 imes 204(19.6%) 72(18.5%) 132(20.2%)
Daily 249(23.9%) 86(22.1%) 163(25%)
Many imes, each day 231(22.1%) 56(14.4%) 175(26.8%)
Du a ion
No idea ion 107(10.3%) 50(12.8%) 57(8.7%) 28.61** .17
Flee ing 203(19.5%) 97(24.9%) 106(16.2%)
Less han 1 hou /some o he ime 209(20%) 85(21.8%) 124(19%)
1–4 hou s/a lo o ime 217(20.8%) 71(18.2%) 146(22.4%)
4–8 hou s/mos o he day 123(11.8%) 40(10.3%) 83(12.7%)
Mo e han 8 hou s/pe sis en 184(17.6%) 47(12.1%) 137(21%)
Con ollabili y o suicidal hough s
No a emp 254(24.4%) 110(28.2%) 144(22.1%) 21.81** .15
Easily 89(8.5%) 43(11%) 46(7%)
Li le di icul y 73(7%) 34(8.7%) 39(6%)
Some di icul y 185(17.7%) 71(18.2%) 114(17.5%)
Lo o di icul y 202(19.4%) 59(15.1%) 143(21.9%)
Unable 240(23%) 73(18.7%) 167(25.6%)
Con inued
Eu opean Psychia y 5
h ps://doi.o g/10.1192/j.eu psy.2024.1807 Published online by Camb idge Uni e si y P ess
The symp oma ic ne wo k o he en i e sample can be seen in
Figu e 1(a). The cen ali y indices a e p esen ed below
(Figu e 1(b)). Conside ing s eng h, he mos ele an nodes we e
he numbe o diagnoses as well as anxious symp oms and emo-
ional abuse ( ollowed by in e pe sonal sensi i i y and psycho ic
symp oms). Closeness and be weenness poin ed ou he ele ance
o diagnoses, in addi ion o non-suicidal sel -ha m. The ne wo k
had adequa e edge s abili y (CS = .75) and s eng h alues
(CS = .67). The exac p edic abili y alues can be seen in Table S1
o he Supplemen a y Ma e ials. They ange be ween 0 ( o he
ACSS) and .41 (psycho ic symp oms).
The ea emp g oup ne wo k (n = 653) (Figu e 2(a)) showed a
simila con igu a ion o ha o he global ne wo k. The nodes wi h
he highes s eng h we e anxious and obsessi e-compulsi e
Table 2. Con inued
Va iable Full sample Single a emp g oup (n = 390) Rea emp g oup (n = 653) χ
2
E ec size
De e en s
De ini ely s opped you 215(20.6%) 93(23.8%) 122(18.7%) 23.27** .15
P obably s opped you 161(15.4%) 54(13.8%) 107(16.4%)
Unce ain 85(8.1%) 29(7.4%) 56(8.6%)
Mos likely did no s op you 77(7.4%) 22(5.6%) 55(8.4%)
De ini ely did no s op you 362(34.7%) 118(30.3%) 244(37.4%)
Does no apply 143(13.7%) 74(19%) 69(10.6%)
Reasons
Comple ely o ge a en ion 10(1%) 1(.3%) 9(1.4%) 17.83** .13
Mos ly o ge a en ion 14(1.3%) 8(2.1%) 6(.9%)
Equally o ge a en ion and o end he pain 81(7.8%) 30(7.7%) 51(7.8%)
Mos ly o end he pain 202(19.4%) 75(19.2%) 127(19.4%)
Comple ely o end he pain 602(57.7%) 208(53.3%) 394(60.3%)
Does no apply 134(12.8%) 68(17.4%) 66(10.1%)
No e: Sco es a e p esen ed as numbe (pe cen age) o ca ego ical ones (*) s a is ically signi ican di e ences a p < .05. (**) s a is ically signi ican di e ences a p < .01.
Figu e 1. (a) Ne wo k displaying he ela ionship be ween Symp oms in he ull sample. (b) Cen ali y indices o Symp oms. Edges in blue indica e posi i e ela ionship. Edges in ed
indica e nega i e ela ionship. Thicke edges ep esen s onge associa ions. The colo s o he nodes g oup he sco es o he CTQ, he BIS, he BSI, suicide- ela ed beha io s, and
o he co a ia es. The g ay bo de on he nodes e lec s p edic abili y. Subs_abuse = Does he pa ien ha e subs ance abuse; N_Diagnosis = numbe o diagnoses; N_beha -
io s = To al numbe o suicidal beha io s (comple ed, in e up ed, and abo ed a emp s). CTQ (Childhood auma Ques ionnai e): sex_ab = sexual abuse; phys_neg = physical
negligence; phys_ab = physical abuse; emo _neg = emo ional negligence; emo _ab = emo ional abuse. CSS (Columbia suicide se e i y a ing scale): SH = sel -ha m; In ense = mos
in ense idea ion; F eq = idea ion equency; Du = Du a ion o idea ion; Con ol = con ollabili y o suicidal hough s. BSI (B ie Symp oms In en o y): Soma = soma iza ion;
Sens = in e pe sonal sensi i i y; Psy = psycho icism; Phob = phobias; Pa = pa anoia; OCD = obsessi e-compulsi e; Hos = hos ili y; Dep = dep ession; Anx = anxie y. BIS (Ba a
impulsi i y scale): Nplan = unplanned impulsi i y; Mo = mo o impulsi i y; A = a en ional impulsi i y. ACSS (Acqui ed Capabili y o Suicide Scale Fea lessness Abou Dea h).
6 Pemau e al.
h ps://doi.o g/10.1192/j.eu psy.2024.1807 Published online by Camb idge Uni e si y P ess
symp oms. The closeness measu e shows he ele ance o he
numbe o diagnoses, anxie y, and phobic symp oms. Be weenness
p esen ed as ele an o he numbe o diagnoses, in ensi y o
idea ion, and dep essi e symp oms. The ne wo k p esen ed an edge
s abili y coe icien o .75 and a s eng h coe icien o .59, bo h
being adequa e. P edic abili y anged om .1 ( o he ACSS) o .38
( o psycho ic symp oms).
Finally, he ne wo k o people wi h one a emp (n = 390)
(Figu e 2(b)) showed some di e ences in i s cen ali y indices.
Based on s eng h, he mos cen al nodes we e emo ional abuse
and anxious symp oms. Rega ding closeness, in ensi y and e-
quency o idea ion as well as dep essi e symp oms we e he mos
ele an nodes. Looking a be weenness, in ensi y o idea ion,
dep essi e symp oms, and numbe o diagnoses we e he mos
ele an nodes. This ne wo k also had adequa e edge s abili y
(CS = .75) and s eng h indices (CS = .59). P edic abili y anged
om .0 ( o he ACSS and he numbe o beha io s) o .36 (anxious
symp oms and emo ional abuse).
In all ne wo ks, subscales belonging o he same cons uc s
ended o be in e connec ed. The symp oms p esen ed g ea e
densi y in hei connec ions in he global ne wo k and in he
ea emp g oup. In gene al e ms, he auma and acqui ed cap-
abili y sco es we e qui e sepa a ed om he es .
Rega ding he compa ison be ween he ne wo ks, he ne wo k
in a iance es was no signi ican (p= .88). The global s eng h
in a iance es did no ind signi ican di e ences (p= .34). The e-
o e, no di e ences we e ound be ween he ne wo ks in ei he
s uc u e o s eng h. Al hough no di e ences we e ound be ween
Figu e 2. (a) Ne wo k displaying he ela ionship be ween symp oms in he ea emp g oup. (b) Ne wo k displaying he ela ionship be ween symp oms in he single a emp
g oup. (c) Cen ali y indices o Symp oms. Edges in blue indica e posi i e ela ionship. Edges in ed indica e nega i e ela ionship. Thicke edges ep esen s onge associa ions.
The colo s o he nodes g oup he sco es o he CTQ, he BIS, he BSI, suicide- ela ed beha io s and o he co a ia es. The g ay bo de on he nodes e lec s p edic abili y.
Subs_abuse = Does he pa ien ha e subs ance abuse; N_Diagnosis = numbe o diagnoses; N_beha io s = To al numbe o suicidal beha io s (comple ed, in e up ed, and abo ed
a emp s). CTQ (Childhood auma Ques ionnai e): sex_ab = sexual abuse; phys_neg = physical negligence; phys_ab = physical abuse; emo _neg = emo ional negligence;
emo _ab = emo ional abuse. CSS (Columbia suicide se e i y a ing scale): SH = sel -ha m; In ense = mos in ense idea ion; F eq = idea ion equency; Du = Du a ion o idea ion;
Con ol = con ollabili y o suicidal hough s. BSI (B ie Symp oms In en o y): Soma = soma iza ion; Sens = in e pe sonal sensi i i y; Psy = psycho icism; Phob = phobias;
Pa = pa anoia; OCD = obsessi e-compulsi e; Hos = hos ili y; Dep = dep ession; Anx = anxie y. BIS (Ba a impulsi i y scale): Nplan = unplanned impulsi i y; Mo = mo o impulsi i y;
A = a en ional impulsi i y. ACSS = Acqui ed Capabili y o Suicide Scale Fea lessness Abou Dea h.
Eu opean Psychia y 7
h ps://doi.o g/10.1192/j.eu psy.2024.1807 Published online by Camb idge Uni e si y P ess
bo h ne wo ks in global e ms, di e ences be ween speci ic edges
we e s udied. The edges be ween he ollowing a iables di e ed
depending on he g oup: gende and age; age and emo ional abuse;
soma ic and anxie y symp oms; emo ional abuse and physical
neglec ; hos ili y and soma ic symp oms; gende and obsessi e
symp oms; physical abuse and physical neglec and inally, e-
quency and con ol o idea ion (p anging om .001 o .049). The
di e ences should no be o e in e p e ed, gi en he high numbe o
compa isons made.
Figu es S1,S3, and S5 (see Supplemen a y Ma e ial) show he
boo s apped con idence in e als o he edge weigh s o each o
he ne wo ks. Some con idence in e als a e conside ably wide
(e en o e lapping), so i would be ad isable o in e p e he o de
o he edges ca e ully. Figu es S2,S4, and S6 (see Supplemen a y
Ma e ial) show he a e age co ela ions o s eng h measu e sam-
pled wi h pe sons d opped and he o iginal sample. They show
gene ally good s abili y o node s eng h.
Discussion
The p esen wo k has applied he pe spec i e o symp oma ic ne -
wo ks o s udy a wide ange o isk ac o s ele an o suicidal
beha io . P e ious wo k has al eady applied his analysis o suicide
ou comes, bu always ocused on a smalle numbe o isk ac o s
[27–30,32,33]. In addi ion, ou s udy included a wide sample o he
Spanish popula ion wi h a ecen a emp . The aim was o imp o e
he unde s anding o he complex ela ionships be ween isk ac o s
in his g oup o pa ien s. Also, we sough o compa e ne wo ks
be ween people wi h a single suicide a emp e sus se e al a emp s.
People wi h mo e han one suicide a emp ha e g ea e se e i y
ac oss mos clinical scales applied. Ou hypo heses abou he
gene al ne wo k ha e been pa ially ul illed. Al hough anxie y
and dep ession a e ele an nodes, in he p esen wo k, he mos
cen al node is he numbe o diagnoses, acco ding o se e al
indices. We could unde s and his as an indica o o g ea e se e -
i y, and i has al eady been add essed in p e ious wo ks ela ed o
suicide isk [54,55]. The o he mos ele an nodes we e non-
suicidal sel -ha m, anxious symp oms, and emo ional abuse. These
esul s a e in line wi h wha was ound by a ecen me a-analysis
[15]. The p esence o non-suicidal sel -ha m has been pos ula ed as
a ele an isk ac o , among o he hings, because i is unde s ood
as a way o losing he ea o pain and dea h [56]. Besides, he
p esence o non-suicidal sel -ha m is a way o egula ing a deep
discom o ha may ha e o do wi h psychia ic como bidi y,
impulsi i y, and a his o y o auma in pa ien s wi h mul iple
a emp s [57]. Bo h anxious symp oms and emo ional abuse ha e
also demons a ed hei ele ance p e iously [15,58].
Rega ding subg oups’ne wo ks, di e ences we e expec ed
be ween people wi h a single a emp and se e al a emp s. The
ea emp g oup ne wo k p esen s some equi alen indices o he
global one, bu obsessi e-compulsi e, phobic symp oms and in en-
si y o idea ion also appea ele an . Mo e symp oma ic nodes a e
cen al, which could once again indica e he ele ance o como -
bidi y. The single-a emp g oup ne wo k showed some di e ences
in e ms o cen ali y indices, wi h idea ion being mo e cen al
(acco ding o closeness and be weenness). This may be consis en
wi h he idea ion- o-ac ion models as people could ha e made he
ansi ion om idea ion o he i s ime [17]. Addi ionally, he
ea emp g oup’s ne wo k seems mo e in e connec ed. Howe e ,
he ne wo ks we e no signi ican ly di e en ei he in hei s uc-
u e o in hei o e all s eng h. This inding is consis en wi h
p e ious wo k [28,30]. Bo h a gued ha he lack o di e ences
could be because he en i e sample has al eady a emp ed suicide,
which could limi he a iabili y o he esul s. In ou wo k, he
single a emp sample was ela i ely smalle and could a ec he
a iabili y o he esul s. The in eg a ed mo i a ional oli ional
model does no p opose a di e ence be ween a iables in he
epe i ion o he suicide a emp bu a he a as e ansi ion
be ween phases [20].
Gi en he impo ance o he numbe o diagnoses in he ne -
wo k, and he di e ences in all scales, i could be a gued ha people
wi h one o se e al a emp s di e mainly in he se e i y o pa h-
ology. Thei ne wo ks a e no ela ed o di e en in ensi ies o
s uc u es, bu he symp oms a e mo e se ious in he case o people
who make se e al a emp s. Simila da a has been ound in some
p e ious wo ks [15,34,36].
Resul s ela ed o he ACSS-FAD es we e unexpec ed. O’Con-
no ’s model aises i s ele ance in he ansi ion om idea ion o
suicide a emp [19]. Howe e , i has u ned ou o be he leas
cen al node in all he ne wo ks. This goes agains ou ini ial
hypo heses. Howe e , he p esence o non-suicidal sel -ha m has
been ele an [56].
Wi h espec o p edic abili y, i is obse able ha he alues a e
mode a e-low in mos cases. Tha is, each node is no e y p edic -
able based on o he s. This is especially ele an in he case o he
ACSS-FAD.
The s udy has se e al limi a ions, which a e discussed below.
Da a comes om a c oss-sec ional design p e en ing he es ablish-
men o causal ela ionships. Ne wo k analysis does no p o ide
in o ma ion on di ec ionali y o causali y. Howe e , i allows o
he concep ualiza ion o complex in e ela ionships be ween symp-
oms and psychosocial componen s.
The e a e di e ences in some clinical and demog aphic measu es
be ween ec ui ing cen e s. This is some hing o be expec ed gi en
ha a ep esen a i e sample o he Spanish popula ion is sough , and
each egion p esen s di e en socioeconomic cha ac e is ics. Also,
he p oposed ne wo ks a e no cul u ally independen and mus be
unde s ood con ex ually o hei ime and space [25].
All measu es we e sel - epo ed, and some o hem esul ed in
sco es o se e al symp oms collapsed in o a single measu e, which
could educe a iabili y [30]. Fo he analyses, only da a om
pa icipan s who comple ed all he scales we e conside ed; his
could lead o a ce ain deg ee o sel -selec ion in he sample,
limi ing gene alizabili y.
Fu he mo e, we ha e ocused ou wo k on di e en a iables ha
ha e been ele an in pas s udies (speci ically, impulsi i y, childhood
auma, suicidal idea ion, and psychia ic symp oms). The e a e o he
ele an a iables no conside ed ha could also be ele an , hus
conclusions mus be limi ed o he a iables conside ed.
Despi e limi a ions, di e en ele an aspec s can be ex ac ed.
Ne wo k analysis ep esen s a no el and sca cely used way o
app oaching he suicidal phenomenon. This is an in e es ing
app oach gi en he mul icausali y and complexi y o suicide
[21]. I is p oposed ha his may be use ul o clinicians, ocusing
ea men s on he mos ele an nodes o he ne wo k [24].
People wi h se e al suicide a emp s p esen mo e se e e symp-
oms han people wi h jus one. Symp om ne wo ks a e no signi i-
can ly di e en be ween bo h g oups, bu some nodes and edges
di e ed in each case. The lack o di e ences in ne wo ks could
indica e ha i is necessa y o ho oughly e alua e isk ac o s
ega dless o he numbe o p e ious a emp s. Howe e , di -
e ences a he node cen ali y in each ne wo k sugges ha
assessmen s can be adjus ed o be e moni o he occu ence
o ea emp s.
8 Pemau e al.
h ps://doi.o g/10.1192/j.eu psy.2024.1807 Published online by Camb idge Uni e si y P ess
Supplemen a y ma e ial. The supplemen a y ma e ial o his a icle can be
ound a h p://doi.o g/10.1192/j.eu psy.2024.1807.
Acknowledgemen s. The ollowing a e membe s o SURVIVE: Iñigo Albe di,
Ma ga i a Alcami, Na alia Anga i a, Guille mo Cano-Escale a, Fe nando Co -
balán, Pa icia Diaz-Ca acedo, Jenni e Fe nández-Fe nández, Edua do
Fe nández-Jiménez, Ve ónica Fe nández-Rod ígues, Ainoa Ga cía-Fe nández,
Ad iana Ga cia-Ramos, Na halia Ga ido- o es, El i a La a, En ico La Spina,
Cla a Ma ínez-Cao, Ma a Mele o, Pablo Mola, Ma a Na as, Bea iz O gaz
Ál a ez, Waleska Pe ez-A que os, I án Pé ez-Diez, Pablo Regue a, Julia Ride ,
Jose Sanchez-Mo eno, Lola Saiz, Ca los Schmid , Elisa Seijo Zazo, La a Suá ez
López, Alba Toll, Mi eia Vázquez, Emma Vidal Be mejo, Edua d Vie a, Iñaki
Zo illa. The au ho s would like o acknowledge he suppo o he esea ch
pa icipan s, who helped o make his wo k possible.
Au ho con ibu ion. All he au ho s con ibu ed o his pape . Concep ual-
iza ion: AP, AdlTL, CMM, MDM, VP. Me hodology: AP, AdlTL. So wa e:
AP. Fo mal analysis: AP. In es iga ion: JAJ, WAA, MCR, ICG, BCF, MAB, DJP,
APT, NR, PAS, MFBO, MTBB, LJT. Resou ces: VP, MRV, APT, AIC, IG, AdlTL,
MDM, AGP, MFBO. Da a Cu a ion: AP, AdlTL, CPG. W i ing - O iginal D a :
AP, AdlTL, CMM, MDM. W i ing –Re iew & Edi ing: AP, AdlTL, CMM,
MDM, JAJ, WAA, MFBO, MTBB, MCR, ICG, AIC, BCF, MAB, ME, AGP, IG,
LJT, DJP, APT, CPG, NR, MRV, PAS, VP. Visualiza ion: AP. Supe ision:
AdlTL, CMM, MDM. P ojec adminis a ion: ME, VP. Funding acquisi ion:
VP, MRV, APT, AIC, IG, AdlTL, MDM, AGP, MFBO. All au ho s we e in ol ed
and app o ed he la es e sion o he manusc ip .
Financial suppo . This s udy was suppo ed by he Ins i u o de Salud Ca los
III-FIS esea ch g an s (PI16/00187, PI19/00236, PI19/00569, PI19/00685,
PI19/00941, PI19/00954, PI19/01027, PI19/01256, PI19/01484, PI20/00229,
PI23/01277, PI23/00822), co- unded by he Eu opean Regional De elopmen
Fund (ERDF) “A Way o Build Eu ope”in eg a ed in o he Plan Nacional de
I + D + I and co inanced by he ISCIII-Subdi ección Gene al de E aluación y
con inanciado po la Unión Eu opea (FEDER, FSE, Nex Gene a ion EU/Plan de
Recupe ación T ans o mación y Resiliencia_PRTR); he Ins i u o de Salud Ca los
III; he CIBER o Men al Heal h (CIBERSAM); and he Sec e a ia d’Uni e si a s i
Rece ca del Depa amen d’Economia i Coneixemen (2021 SGR 01358), CERCA
P og amme/Gene ali a deCa alunyaaswellas heFundacióClínicpe laRece ca
Biomèdica (Pons Ba an 2022-FRCB_PB1_2022); he Go e nmen o he P in-
cipali y o As u ias (g an e .: PCTI-2018-2022 IDI/2018/235). I was also
suppo ed by an FPU g an (FPU20/01651) om he Spanish Minis y o Uni-
e si ies and a Uni e sidad Complu ense de Mad id P edoc o al con ac o
esea ch s a in aining (CT82/20-CT83/20).
Compe ing in e es . PAS has been a consul an o and/o has ecei ed hon-
o a ia o g an s om Adamed, Al e Medica, Angelini Pha ma, CIBERSAM,
E hypha m Digi al The apy, Eu opean Commission, Go e nmen o he P inci-
pali y o As u ias, Ins i u o de Salud Ca los III, Johnson & Johnson, Lundbeck,
O suka, P ize , Plan Nacional Sob e D ogas and Se ie . AGP has ecei ed g an s
and se ed as a consul an , ad iso , o CME speake o he ollowing en i ies:
Janssen-Cilag, Lundbeck, O suka, Al e , Angelini, No a is, Ro i, Takeda, he
SpanishMinis yo ScienceandInno a ion(CIBERSAM), heMinis y o Science
(Ca los III Ins i u e), he Basque Go e nmen , and he Eu opean F amewo k
P og am o Resea ch. NR con ac is co- unded by he Ins i u o de Salud Ca los
III, wi h ile code CD23/00088, by i ue o Resolu ion o he Di ec ion o he
Ins i u o de Salud Ca los III, O.A., M.P. o Decembe 13, 2023, awa ding he Sa a
Bo ell and “Co- unded by he Eu opean Union”Con ac s. IG has ecei ed g an s
and has se ed as a consul an , ad iso o CME speake o he ollowing en i ies
(un ela ed o he p esen wo k): ADAMED, Angelini, Casen Reco da i, Es e e,
Fe e , Gedeon Rich e , Janssen Cilag, Lundbeck, Lundbeck-O suka, Luye, SEI
Heal hca e, Via is ou side he submi ed wo k. She also ecei es oyal ies om
Ox o d Uni e si y P ess, Else ie , Edi o ial MédicaPaname icana. The emaining
au ho s ha e no con lic s o decla e.
Re e ences
[1] Wo ld Heal h O ganiza ion. Suicide, h ps://www.who.in /news- oom/
ac -shee s/de ail/suicide; 2024 [accessed 01 Oc obe 2024].
[2] de la To e-Luque A, Pe ez-Diez I, Pemau A, Ma inez-Ales G, Bo ges G,
Pe ez-Sola V, Ayuso-Ma eos JL. Suicide mo ali y in Spain du ing he
COVID-19 pandemic: Longi udinal analysis o sociodemog aphic ac o s.
Eu Neu opsychopha macol. 2024;82:29–34. doi: 10.1016/j.eu oneu o.20
24.02.006.
[3] Ma ínez-Alés G, Szmulewicz A, López-Cuad ado T, Mo ison CN, Keyes
KM, Susse ES. Suicide ollowing he COVID-19 pandemic ou b eak:
a ia ion ac oss place, o e ime, and ac oss sociodemog aphic g oups.
A sys ema ic in eg a i e e iew. Cu Psychia y Rep. 2023;25(7):283–300.
doi: 10.1007/s11920-023-01427-7.
[4] Wo ld Heal h O ganiza ion. Suicide p e en ion: a global impe a i e. 2014.
[5] Baumeis e RF. Suicide as escape om sel . Psychol Re . 1990;97(1):90–
113. doi: 10.1037/0033-295X.97.1.90.
[6] Beck AT, S ee RA, Ko acs M, Ga ison B. Hopelessness and e en ual
suicide: A 10-yea p ospec i e s udy o pa ien s hospi alized wi h suicidal
idea ion. Am J Psychia y. 1985;142(5):559–63. doi: 10.1176/ajp.142.5.559.
[7] Linehan MM, Com ois KA, Wa d-Ciesielski EF. Assessing and managing
isk wi h suicidal indi iduals. Cogn Beha P ac . 2012;19(2):218–32. doi:
10.1016/j.cbp a.2010.11.008.
[8] And eo-Jo e J, Ramos JC, Bobes J, B a o-O iz M, Ceb ia AI, C espo-
Faco o B, e al. The media ing ole o e lec i e unc ioning in he
associa ion be ween childhood auma and suicide a emp . J Psychia
Res. 2024;171:30–7. doi: 10.1016/j.jpsychi es.2024.01.005.
[9] Da ay FM, Rojas SM, B idges AJ, Badou CL, G endas L, Rodan e D, e al.
The independen e ec s o child sexual abuse and impulsi i y on li e ime
suicide a emp s among emale pa ien s. Child Abuse Negl. 2016;58:91–8.
doi: 10.1016/j.chiabu.2016.06.011.
[10] Haw on K, Saunde s KEA, O’Conno RC. Sel -ha m and suicide in ado-
lescen s. Lance . 2012;379(9834):2373–82. doi: 10.1016/S0140-6736(12)
60322-5.
[11] Mendez-Bus os P, De Leon-Ma inez V, Mi e M, Baca-Ga cia E, Lopez-
Cas oman J. Suicide ea emp e s: A sys ema ic e iew. Ha Re Psych-
ia y. 2013;21(6):281–95. doi: 10.1097/HRP.0000000000000001.
[12] Ng QX, Yong BZJ, Ho CYX, Lim DY, Yeo WS. Ea ly li e sexual abuse is
associa ed wi h inc eased suicide a emp s: an upda ed me a-analysis.
J Psychia Res. 2018;99:129–41. doi: 10.1016/j.jpsychi es.2018.02.001.
[13] Za i C, Rosa V, Ba os A, Valdi ia L, Calega o VC, F ei as LH, e al.
Childhood auma and suicide a emp : a me a-analysis o longi udinal
s udies om he las decade. Psychia y Res. 2017;256:353–8. doi: 10.1016/
j.psych es.2017.06.082.
[14] Goñi-Sa iés A, Yá noz-Goñi N, López-Goñi JJ. Psychia ic hospi aliza-
ion o a emp ed suicide and ea emp a he one-yea ollow-up.
Psico hema. 2022;34(3):375–82. doi: 10.7334/psico hema2021.455.
[15] de la To e-Luque A, Pemau A, Ayad-Ahmed W, Bo ges G, Fe nandez-
Se illano J, Ga ido-To es N, e al. Risk o suicide a emp epe i ion a e
an index a emp : a sys ema ic e iew and me a-analysis. Gen Hosp
Psychia y. 2023;81:51–6. doi: 10.1016/j.genhosppsych.2023.01.007.
[16] Joine T. Why people die by suicide. Ha a d Uni e si y P ess; 2005.
[17] Klonsky ED, Sa e BY, B yan CJ. Idea ion- o-ac ion heo ies o suicide: a
concep ual and empi ical upda e. Cu Opin Psychol. 2018;22:38–43. doi:
10.1016/j.copsyc.2017.07.020.
[18] Klonsky ED, May AM. The h ee-s ep heo y (3ST): A new heo y o
suicide oo ed in he “idea ion- o-ac ion” amewo k. In J Cogn The .
2015;8(2):114–29. doi: 10.1521/ijc .2015.8.2.114.
[19] O’Conno RC. Towa ds an in eg a ed mo i a ional– oli ional model o
suicidal beha iou . In: O’Conno RC, Pi kis J, edi o s. In e na ional
Handbook o Suicide P e en ion: Resea ch, Policy and P ac ice. John
Wiley & Sons; 2011, pp. 181–98.
[20] O’Conno RC, Ki ley OJ. The in eg a ed mo i a ional– oli ional model o
suicidal beha iou . Philos T ans R Soc B Biol Sci. 2018;373(1754):
20170268. doi: 10.1098/ s b.2017.0268.
[21] Ga cía-Ha o J, González-González M, Fonseca-Ped e o E, Al-Halabí S.
Concep ualización de la conduc a suicida. In: Al-Halabí S, Fonseca-
Ped e o E, edi o s. Manual de psicología de la conduc a suicida. Pi ámide;
2023. pp. 32–68.
[22] Nock M, O’Conno R. The psychology o suicidal beha iou . Lance
Psychia y. 2014;1:73–85. doi: 10.1016/S2215-0366(14)70222-6.
Eu opean Psychia y 9
h ps://doi.o g/10.1192/j.eu psy.2024.1807 Published online by Camb idge Uni e si y P ess
