scieee Science in your language
[en] (orig)

Violence in migrants and refugees in Europe: determinants and preventable measures

OLIVEIRA, Charlotte Neves de · 2018

Abstract

Introdução A violência sexual e de género (VSG) é um problema global de saúde pública, ao qual refugiados, requerentes de asilo e imigrantes não documentados, estão particularmente vulneráveis. Nos centros de acolhimento Europeus, residentes e profissionais estão predispostos à vitimização e perpetuação de VSG. Objetivos Esta tese tem como objetivo contribuir para melhorar o conhecimento do conceito de VSG, casos reportados, causas, medidas preventivas, e fatores preditivos de vitimização em residentes (refugiados, requerentes de asilo e imigrantes não documentados) e profissionais (prestadores de serviços, e serviços de saúde), em centros de acolhimento Europeus. Métodos Foram utilizados dados recolhidos no âmbito do projeto Europeu "Senperforto", com o objetivo de contribuir para a proteção e promoção da saúde de refugiados, requerentes de asilo e imigrantes não documentados, de forma a prevenir a VSG nos centros de acolhimento. Senperforto incluiu um estudo sobre conhecimentos, atitudes e práticas relativo à VSG, de residentes e profissionais, que vivem e trabalham em centros de acolhimento, em oito países (Bélgica, Grécia, Hungria, Irlanda, Malta, Holanda, Portugal, Espanha). No total foram realizadas 600 entrevistas: 398 a residentes e 202 a profissionais. A análise de dados incluiu uma análise por componentes principais (ACP), testes de associação como o Qui-quadrado ou teste exato de Fisher, e técnicas de machine learning. Resultados O resultado da ACP relativo ao conceito de VSG para o grupo de residentes incluiu 14 dimensões de VSG que representam 83,56% da variância total de dados. No grupo de profissionais resultou em 17 dimensões de VSG correspondendo a 86,92% da variância total de dados. Para ambos, o conceito de VSG diferiu de acordo com o país de acolhimento, sexo, idade e estado civil. Nos residentes, foram encontradas diferenças relacionadas com a duração de residência no país de acolhimento/Europa, e com os tipos de alojamento. Para os profissionais, as diferenças estavam ligadas ao estatuto legal e competências educacionais. Os participantes reportaram 698 casos de VSG (residentes 328, profissionais 370), correspondendo a 1110 atos de vários tipos de violência. As principais causas presumidas foram: frustração e stress (residentes 23,6%, profissionais 37,6%, p0,008) e diferenças relacionadas com aspectos culturais (residentes 19,3%, profissionais 20,3%, p0,884). Os participantes relataram que estes atos de violência poderiam ser evitados melhorando: intervenções preventivas de VSG (residentes 31,5%, profissionais 24,7%, p0,293); condições habitacionais (residentes 21,7%, profissionais 15,3%, p0,232); e comunicação (residentes 16,1%, profissionais 28,2%, p0,042). A maioria dos residentes não tinha conhecimento da existência de medidas preventivas nos centros de acolhimento (58,3%) ou no país deacolhimento (72,4%). As medidas preventivas de VSG propostas pelos participantes incluíram: sensibilização sobre a VSG, melhoria das condições habitacionais e melhoria da comunicação entre residentes e profissionais. Os modelos preditivos de VSG destacaram as condições habitacionais como uma característica importante para prever a vitimização. Assim, instalações sanitárias apropriadas, o tipo de alojamento, o estatuto legal, a idade, o tipo de ocupação e a idade das pessoas com quem as instalações sanitárias são partilhadas, foram variáveis essenciais para prever a vitimização. Ser residente ou profissional provou ter baixa característica preditiva. Conclusão Nos centros de acolhimento Europeus, as estratégias de prevenção primária deverão focalizar-se na harmonização do conceito de VSG, abordando possíveis diferenças relacionadas com características sociodemográficas. Os resultados sugerem que nos centros de acolhimento, tanto os residentes como os profissionais, os homens e as mulheres estão em risco de VSG, reduzindo os estereótipos: masculinos/profissionais - agressores e mulheres/residentes - vítimas. A elevada incidência de VSG apresentada nos nossos resultados sugere que a prevenção secundária deverá incidir numa maior sensibilização para o problema, melhorar as condições habitacionais e de trabalho, melhorar a comunicação, assegurar um procedimento de asilo equitativo e justo, e incluir os residentes e profissionais como participantes ativos no processo de desenvolvimento e implementação destas medidas. Enfatiza-se ainda, a necessidade emergente da criação e implementação de políticas e diretrizes europeias personalizadas que melhorem as condições habitacionais e de trabalho nos centros de acolhimento. Por último, estamos convencidos de que os Estados-Membros poderão beneficiar do desenvolvimento de capacidades e ferramentas para a implementação destas políticas e diretivas.

Read accessible full text

Violence in migrants and refugees in Europe: determinants and preventable measures

Author: OLIVEIRA, Charlotte Neves de
Publisher: Instituto de Higiene e Medicina Tropical
Year: 2018
Source: https://run.unl.pt/bitstream/10362/56922/1/Violence%20migrants%20and%20refugees%20in%20Europe_CharlotteOliveira_PhD%20Thesis_%2001012019.pdf
Uni e sidade No a de Lisboa
Ins i u o de Higiene e Medicina T opical
Violence in mig an s and e ugees in Eu ope:
de e minan s and p e en able measu es
CHARLOTTE NEVES DE OLIVEIRA
DISSERTAÇÃO PARA A OBTENÇÃO DO GRAU DE DOUTOR
EM SAÚDE INTERNACIONAL
ESPECIALIDADE EM POLÍTICAS DE SAÚDE E DESENVOLVIMENTO
Dezemb o, 2018
Uni e sidade No a de Lisboa
Ins i u o
de Higiene e Medicina T opical
Violence in mig an s and e ugees in Eu ope:
de e minan s and p e en able measu es
Au o :
Cha lo e Ne es de Oli ei a
O ien ado :
P o esso a Dou o a Sónia Dias
Co-o ien ado es: P o esso a Dou o a Ma ia do Rosá io Ma ins
P o esso a Dou o a Ines Keygnae
Disse ação ap esen ada pa a cump imen o dos equisi os necessá ios à ob enção do g au de
Dou o em Saúde In e nacional e especialização em Polí icas de Saúde e Desen ol imen o
i
Dedica ion
To my belo ed amily, o blood and hea .

i
ii
Acknowledgemen s
To P o esso Sónia Dias, P o esso Ines Keygnae , and P o esso Ma ia do Rosá io Ma ins
an eno mous hank you o all he suppo along hese yea s. Thank you o pushing me
u he , helping me o keep he ee on he g ound, and belie ing ha I was able o each his
pa h.
To my IHMT colleagues, hank you o you suppo , posi i e and mo i a ing wo ds,
especially in he di icul days, and o always being a ailable when I needed.
To ICRH eam, hank you o you collabo a ion, o ecei ing me a you Ins i u e in such
a wa mly a i ude, and wi h “open a ms”. A special hanks o Saa Bae o he suppo and
o he open doo s a he home.
To all my belo ed iends, a huge hank you. You ha e been an essen ial pilla , wi hou you
cons an p esence, suppo and encou aging wo ds, his wo k could ne e be possible. The
di icul momen s, obs acles and us a ion we e o e come posi i ely and cons uc i ely
wi h all o you by my side. A special hanks o my closes iends Ana Aze edo, Cá ia
Simões, So ia Co a, Joana Camelie and Salomé Fa ia. And a huge hank you o Sebas ião
Ca mo, o all his pa ience and kind suppo .
Abo e all, o my amily, Pai and Mãe, Ri a, So ia and F ancisco, I can only say hank you
om all my hea . You a e my inspi a ion and my eason o “keep mo ing o wa d”. You a e
he eason why my d eams come ue.
Thank you.
Ob igado.
iii
ix
Resumo
In odução
A iolência sexual e de géne o (VSG) é um p oblema global de saúde pública, ao qual
e ugiados, eque en es de asilo e imig an es não documen ados, es ão pa icula men e
ulne á eis. Nos cen os de acolhimen o Eu opeus, esiden es e p o issionais es ão
p edispos os à i imização e pe pe uação de VSG.
Obje i os
Es a ese em como obje i o con ibui pa a melho a o conhecimen o do concei o de VSG,
casos epo ados, causas, medidas p e en i as, e a o es p edi i os de i imização em
esiden es ( e ugiados, eque en es de asilo e imig an es não documen ados) e p o issionais
(p es ado es de se iços, e se iços de saúde), em cen os de acolhimen o Eu opeus.
Mé odos
Fo am u ilizados dados ecolhidos no âmbi o do p oje o Eu opeu "Senpe o o", com o
obje i o de con ibui pa a a p o eção e p omoção da saúde de e ugiados, eque en es de
asilo e imig an es não documen ados, de o ma a p e eni a VSG nos cen os de acolhimen o.
Senpe o o incluiu um es udo sob e conhecimen os, a i udes e p á icas ela i o à VSG, de
esiden es e p o issionais, que i em e abalham em cen os de acolhimen o, em oi o países
(Bélgica, G écia, Hung ia, I landa, Mal a, Holanda, Po ugal, Espanha). No o al o am
ealizadas 600 en e is as: 398 a esiden es e 202 a p o issionais. A análise de dados incluiu
uma análise po componen es p incipais (ACP), es es de associação como o Qui-quad ado
ou es e exa o de Fishe , e écnicas de machine lea ning.
Resul ados
O esul ado da ACP ela i o ao concei o de VSG pa a o g upo de esiden es incluiu 14
dimensões de VSG que ep esen am 83,56% da a iância o al de dados. No g upo de
p o issionais esul ou em 17 dimensões de VSG co espondendo a 86,92% da a iância o al
de dados. Pa a ambos, o concei o de VSG di e iu de aco do com o país de acolhimen o, sexo,
idade e es ado ci il. Nos esiden es, o am encon adas di e enças elacionadas com a
du ação de esidência no país de acolhimen o/Eu opa, e com os ipos de alojamen o. Pa a os
p o issionais, as di e enças es a am ligadas ao es a u o legal e compe ências educacionais.
Os pa icipan es epo a am 698 casos de VSG ( esiden es 328, p o issionais 370),
co espondendo a 1110 a os de á ios ipos de iolência. As p incipais causas p esumidas
o am: us ação e s ess ( esiden es 23,6%, p o issionais 37,6%, p0,008) e di e enças
elacionadas com aspec os cul u ais ( esiden es 19,3%, p o issionais 20,3%, p0,884). Os
pa icipan es ela a am que es es a os de iolência pode iam se e i ados melho ando:
in e enções p e en i as de VSG ( esiden es 31,5%, p o issionais 24,7%, p0,293); condições
habi acionais ( esiden es 21,7%, p o issionais 15,3%, p0,232); e comunicação ( esiden es
16,1%, p o issionais 28,2%, p0,042). A maio ia dos esiden es não inha conhecimen o da
exis ência de medidas p e en i as nos cen os de acolhimen o (58,3%) ou no país de
x i

x ii
Table Index
Table 1: STOP SV – Technical package, speci ic s a egies and associa ed app oaches o
p e en sexual iolence, adap ed om Basile e al (9). ________________________ 42
Table 2: Va iables used o conduc he p incipal componen analysis. ________________ 60
Table 3: Lis o a iables conside ed as inpu s o p edic i e models. ________________ 65
Table 4: Socio-demog aphic cha ac e is ics o Residen s and P o essionals, in EARF. ___ 74
Table 5: P incipal componen analysis o Residen s: ep esen a i e ques ions and ou pu
loading (Va imax a ia ion). ____________________________________________ 75
Table 6: P incipal componen analysis o P o essionals: ep esen a i e ques ions and
ou pu loading (Va imax a ia ion). ______________________________________ 77
Table 7: Residen s – SGBV concep ualisa ion and socio-demog aphic cha ac e is ics (p-
alues: Chi-squa e es and Fishe Exac es ). ______________________________ 82
Table 8: P o essionals – SGBV concep ualisa ion and socio-demog aphic cha ac e is ics (p-
alues: Qui-squa e Tes and Fishe Tes ). __________________________________ 85
Table 9: Li ing condi ions o Residen s a Eu opean asylum ecep ion acili ies. _______ 95
Table 10: SGBV Cases epo ed by esiden s and p o essionals and gende . ___________ 97
Table 11: Numbe o epo ed cases o SGBV by esiden s and p o essionals, pe coun y.
___________________________________________________________________ 98
Table 12: Numbe o ac s o iolence epo ed by esiden s and p o essionals. _________ 98
Table 13: Causes o epo ed cases o SGBV. __________________________________ 100
Table 14: Po en ially p e en able measu es o epo ed cases o SGBV, by esiden s and
p o essionals. _______________________________________________________ 101
Table 15: Exis ence o p e en able measu es a coun y le el and ecep ion asylum acili y.
__________________________________________________________________ 102
x iii
Table 16: Possible p e en able measu es o SGBV desc ibed by esponden s. ________ 103
Table 17: SGBV epo ed cases, 12 mon hs p io o he in e iews, pe coun y o esea ch.
__________________________________________________________________ 114
Table 18: Bes ea u e subse and ea u e impo ance, based on he impo ance o Random
o es algo i hm. _____________________________________________________ 115
xix
Lis o Ac onyms
AS Asylum-seeke (s)
AUC A ea Unde he Cu e
CAB Communi y Ad iso y Boa d
CEAS Common Eu opean Asylum Sys em
CR Communi y Resea che s
EARF Eu opean Asylum Recep ion Facili ies
EP Eu opean Pa liamen
EU Eu opean Union
FGM Female Geni al Mu ila ion
GBV Gende -based Violence
HIV Human Immunode iciency Vi us
HPV Human Papilloma Vi us
IPV In ima e Pa ne Violence
KAP Knowledge, A i ude & P ac ice s udy/su ey
LGBTI Lesbian, Gay, Bisexual, T ansgende and In e se x
ML Machine Lea ning
MS Membe S a es
MSF Médecins Sans F on iè es
MSM man ha ing sex wi h man
NGO Non-Go e nmen al O ganisa ion
PCA P incipal Componen Analysis
xx
PC(s) P incipal componen (s)
PTSD Pos T auma ic S ess Diso de
RAUM Re ugees, asylum-seeke s and undocumen ed mig an s
ROC Recei ing Ope a ing Cha ac e is ics
SGBV Sexual and Gende -Based Violence
STI Sexually T ansmi ed In ec ion
SV Sexual Violence
SVM Suppo Vec o Machine
UNHCR Uni ed Na ions High Commissione o Re ugees
UNFPA Uni ed Na ions Popula ion Fund
VAW Violence Agains Women
WHO Wo ld Heal h O ganisa ion
CHAPTER 1. INTRODUCTION

CHAPTER 1. In oduc ion
2
CHAPTER 1. In oduc ion
3
1. INTRODUCTION
Re ugees, asylum seeke s (AS) and undocumen ed mig an s (RAUM) a e a ulne able
popula ion o sexual and gende -based iolence (SGBV), wi h unique physical, men al,
economic, and social conce ns (1–4). Since he Beijing Decla a ion in 1995 (5), he
in e na ional communi y has assumed he need o a ‘mo e holis ic suppo o e ugee and
displaced women, including hose who ha e su e ed all o ms o abuse, including gende -
speci ic abuse, (…)’ (p.216). Nowadays, we assis o he i s ime since he Wo ld Wa II,
o a massi e mo emen o people seeking in e na ional p o ec ion, leeing con lic ,
pe secu ion and SGBV (6). Many o hese people in ending o each Eu opean coun ies a e
e ugees, AS and undocumen ed mig an s en i led o p o ec ion unde he 1951 Re ugee
Con en ion and i s 1967 P o ocol (7).
SGBV is a majo public heal h p oblem and a h ea o human igh s (4,8,9). In 1996, he
Wo ld Heal h Assembly adop ed a esolu ion decla ing iolence as a leading public heal h
p oblem. In his decla a ion, he WHO u ged membe s a es o assess iolence in hei own
coun y, o ac and o communica e he magni ude o he p oblem. Speci ic eques s ha e
been made o p omo e public heal h in e en ions p e en ing and educing he bu den o
iolence such as:
a) The cha ac e isa ion o all ypes o iolence, hei magni ude, causes and public heal h
consequences, aking in o accoun gende sensi i i y;
b) To assess he ypes and e ec i eness o measu es and p og ammes o p e en iolence
and mi iga e i s e ec s;
c) To p omo e in e en ions aiming o p e en he p oblem wo ldwide;
d) To ensu e he coo dina ed and ac i e pa icipa ion o WHO echnical p og ammes;
e) To ein o ce he collabo a ion be ween di e en s akeholde s wi h go e nmen s, local
au ho i ies and o he o ganiza ions o he Uni ed Na ions sys em in planning,
implemen ing, and moni o ing iolence p e en ion’ p og ammes (10).
CHAPTER 1. In oduc ion
4
Mo eo e , o add ess he complexi y o iolence and i s oo s, a socio-ecological
comp ehensi e and holis ic app oach, ocusing on causes and p e en ion o iolence is
p econized (3).
The la ely p oac i e measu es aken wo ldwide ela es o he 2030 Agenda o Sus ainable
De elopmen (11), whe e iolence p e en ion is highligh ed as he ‘key componen o
de elopmen and o imp o ing quali y o li e in all pa s o he wo ld’ (p.1). Mo eo e , he
2030 Agenda sus ains a op p io i y o all membe S a es o signi ican ly educe all o ms o
iolence and ela ed dea h a es e e ywhe e (11).
Conside ing he ecen mig a ion ends, King & Lulle (12) s a es ha he de elopmen o a
long- e m ision o Eu opean mig a ion policies, e idence-based, and sus ained in eliable
and compa able da a is called o . ‘(…) he EU needs a mo e cohe en mig a ion policy, and
mo e high-quali y esea ch e idence o a la ge-scale compa a i e na u e o help o imp o e
policies di ec ly ela ed o he eali ies o mig a ion in a p agma ic way.’ (p.56). Mo eo e ,
de Haan a i ms ha esea ch add essing iolence should p io i ise he in-dep h
unde s anding o he p oblem (13) in di e en se ings and add essing ulne able g oups,
elabo a ing e ec i e p e en i e measu es and esponses (3). In his sense, ou esea ch
in ends o con ibu e o he de elopmen o e idence-based SGBV p e en ion measu es,
adop ing a public heal h app oach and socio-ecological model.
This doc o al hesis is o ganised in o i e chap e s. Chap e 1 p o ides a li e a u e e iew
syn hesising he s a e o he a ega ding he main subjec s, namely mig a ion in he
Eu opean con ex , concep ualisa ion o iolence and SGBV, and inally SGBV and
mig a ion. Chap e 2 p esen s he gene al and speci ic objec i es o his esea ch. Chap e 3
desc ibes he p ojec on which his hesis is based – Senpe o o P ojec , and he me hods
used o each ou speci ic objec i es. Chap e 4 p esen s he esul s o his esea ch. In chap e
5 we p esen he discussion and he main conclusions o his hesis.
CHAPTER 1. In oduc ion
5
1. Mig a ion in he Eu opean con ex
Mig a ion is conside ed a po en ial de elopmen ool o socie ies when based in humane,
jus and well-go e ned mig a ion policies (14). I can lead people ou o po e y, inc ease
educa ional oppo uni ies and labou demand, os e inno a ion and enable he exchange o
knowledge, skills, and cul u e (14). Howe e , mig a ion can con ibu e o social dispa i ies
and inequali ies in access o se ices, i in e na ional human igh s a e no ensu ed. Fu he ,
he social, cul u al, economic and poli ical con ex in which people’ mo emen s occu s,
migh la gely de e mine whe he mig a ion ansla es in o inc eased well-being and
oppo uni ies, o in o dep i a ion and ulne abili y (14).
Acco ding o he Uni ed Na ions (15) mig a ion can be de ined as he ‘mo emen o people,
ei he wi hin a coun y o ac oss in e na ional bo de s. I includes all kinds o mo emen s,
i espec i e o he d i e s, du a ion and olun a y/in olun a y na u e. I encompasses
economic mig an s, dis ess mig an s, in e nally displaced pe sons (IDPs), e ugees and
asylum seeke s, e u nees and people mo ing o o he pu poses, including o educa ion and
amily euni ica ion’ (p.6).
Wo ldwide, in e na ional mig a ion is a g owing phenomenon. E e y mon h he numbe o
people ying o each s able coun ies is inc easing despi e he condi ions, means, and
consequences. The UNHCR s a ed ha by he end o 2015, 65.3 million o people we e
o ced o displacemen due o di ec o undi ec ed iolence - pe secu ion, con lic ,
gene alized iolence, SGBV o human igh s iola ions (16), accoun ing o he highes
numbe o displaced people since he Wo ld Wa II (16). This numbe has eached a new
eco d in 2016, wi h 65.6 million o people being o cibly displaced wo ldwide. Rega ding
e ugees’ mo emen , a new eco d was eached wi h 22.5 million a he end o 2016 – 1.4
million we e newly displaced (17). Ye , 40.3 million in e nally displaced people and 2.8
million AS we e seeking in e na ional p o ec ion and wai ing o hei legal s a us o be
de ined (17).
CHAPTER 1. In oduc ion
12
De ini ions o iolence end o be ela ed o he use o o ce and agg essi e beha iou di ec ed
o one pe son, o bidden and punishable by law (13,34). This de ini ion is conside ed a ma e
o ‘law and o de ’ and no app oachable by heal h p o essionals (3). Se e al au ho s
ecognised ha iolence could be de ined acco ding o wo di e en app oaches. The i s
de ini ion comp ises a na ow concep o iolence: he in en ional ac o excessi e o
des uc i e o ce ( he Minimalis Concep ion o Violence), he second, de ines iolence as a
iola ion o igh s, suppo ing a b oade concep ion o iolence ( he Comp ehensi e
Concep ion o Violence) (35).
Violence is conside ed mul i ace ed once he e a e di e en o ms o iolence, which a e
conduc ed in di e en con ex s and can be s udied by di e se ac o s’ pe spec i e: (i.e.
pe pe a o , he ic im, hi d pa y, neu al obse e ) (Haan, 2008). The same au ho a i ms
ha iolence is socially cons uc ed because who and wha is conside ed iolence a ies
acco ding o speci ic de e minan s such as socio-cul u al, his o ical condi ions and legal
amewo ks.
De ini ions o iolence should be comp ehensi e in o de o ecognize he mul iple o ms
ha iolence may ake, he di e en con ex s in which iolence may occu , and he
in e ac ion be ween o ms and con ex s (36). E en hough di e en pe spec i es and
de ini ions o iolence can be ound in he li e a u e, concep ualisa ion o iolence wi hin
public heal h is g owing place h ough an en i onmen al and beha iou - ela ed pe spec i e
(3). Conside ing ha , we ha e communicable diseases and iolence epidemics ha we assis
nowadays, we ha e, in ano he side, non-communicable diseases, which esul s om pas
and cumula i e isk ac o s ha can be p e en able and modi iable (37). The au ho s a gued
ha iolence pe o ms in simila ways wi h non-communicable diseases (37). In his sense,
due o he ising bu den o iolence, some au ho s p opose a change in scope o public heal h,
ecognizing he in luence o social jus ice, economics, and globaliza ion as possible oo s o
p ema u e dea h and ill-heal h, whe e iolence could i (37). The WHO in i s i s Wo ld
Repo on Violence and Heal h (3) highligh ed he need o add ess iolence om a b oad
public heal h pe spec i e and in line wi h non-communicable diseases – ‘The ocus is

CHAPTER 1. In oduc ion
13
b oadening, wi h inc easing emphasis on p e en ion and add essing he oo causes o
iolence.’ (p.1). To no e, ha public heal h’ app oach o iolence should be conside ed a
complemen a y science o c iminology, jus ice, human igh s p o ec ion (…) i we in end o
ackle down he bu den induced by iolence in ou socie y (3).
Conside ing he de ini ion gi en by he WHO (3) iolence can be de ined as ‘The in en ional
use o physical o ce o powe , h ea ened o ac ual, agains onesel , ano he pe son, o
agains a g oup o communi y, ha ei he esul in o has a high likelihood o esul ing in
inju y, dea h, psychological ha m, malde elopmen o dep i a ion.’ (p.4). This concep is a
ma e o judgmen , a ec ed by cul u al belie s, social no ms and alues. The concep o
iolence can be ca ego ized in o sel -di ec ed iolence (suicide o sel -abuse), in e pe sonal
iolence ( amily/pa ne o communi y) and collec i e iolence (social, poli ical o
economic), acco ding o he pe son who commi s he iolen ac , hen hese h ee ca ego ies
a e subdi ided acco ding o he na u e o iolence – physical, sexual, psychological o
dep i a ion/neglec (3) (see Figu e 2).
Figu e 2: Typology o Violence, adap ed om WHO (3).
CHAPTER 1. In oduc ion
14
2.2 De ining Sexual and Gende -Based Violence
SGBV is conside ed a majo public heal h issue and a h ea o human igh s (3,4,38).
UNHCR (4) s a ed ha ‘This kind o iolence pe pe ua es he s e eo yping o gende oles
ha denies human digni y o he indi idual and s ymies human de elopmen .’ (p.7). The
impo ance o unde s anding, p e en ing and esponding o SGBV is mo e han ano he opic
conce ning public heal h, i is a ma e o iola ion o innume ous human igh s o
in e na ional conce n:
-
‘The igh o li e, libe y and secu i y o he pe son;
-
The igh o he highes a ainable s anda d o physical and men al heal h;
-
The igh o eedom om o u e o c uel, inhuman, o deg ading ea men o
punishmen ;
-
The igh o eedom o mo emen , opinion, exp ession, and associa ion;
-
The igh o en e in o ma iage wi h ee and ull consen and he en i lemen o equal
igh s o ma iage, du ing ma iage and a i s dissolu ion;
-
The igh o educa ion, social secu i y and pe sonal de elopmen ;
-
The igh o cul u al, poli ical and public pa icipa ion, equal access o public se ices,
wo k and equal pay o equal wo k.’ (4: p.8).
In e na ionally, di e en e ms a e used o add ess SGBV. In he ollowing pa ag aphs, we
will add ess he de ini ions o iolence agains women and gi ls, sexual iolence (SV),
gende -based iolence and inally, SGBV. E en hough ca ego isa ion o iolence is
essen ial, esea ch should ake in o conside a ion ha hey a e no exclusi e, bu hey ela e
o each o he (39).
Violence agains women is de ined as any ac o gende -based iolence ha can esul in
physical, sexual o psychological/emo ional ha m o su e ing di ec ed o a gi l/woman. (40).
Mo eo e , in he Decla a ion on he Elimina ion o Violence agains Women (40), iolence
is conside ed a ‘mani es a ion o his o ically unequal powe ela ions be ween men and
women, which ha e led o he domina ion o e and disc imina ion agains women by men
CHAPTER 1. In oduc ion
15
and o he p e en ion o he ull ad ancemen o women.’ (p.1). This ype o iolence, based
on gende disc imina ion encou ages o he o ms o iolence agains women and gi ls and
pe pe ua es he accep ance and in isibili y o his ype o iolence (38). In his sense,
ic ims/su i o s a e a aid and discou aged om speaking ou , as agg esso s a e no held
accoun able o hei ac s (38).
Acco ding o Basile e al (41), SV comp ises ‘a sexual ac ha is commi ed o a emp ed by
ano he pe son wi hou he eely gi en consen o he ic im o agains someone who is
unable o consen o e use.’ (p.11). I includes mo e han ape o sexual assaul , i includes
ac s o physical and psychological abuse, a icking and o he o ms o abuse and sexual
exploi a ion (4,5). Mo eo e , he WHO desc ibed he ollowing SV ac s: ape, sexual
ha assmen , sexual abuse o child en, o ced p os i u ion and sex a icking, child ma iage,
and iolen ac s agains he sexual in eg i y o women, including emale geni al mu ila ion
and obliga o y inspec ions o i gini y (3).
The Beijing Decla a ion (5) s a ed ha GBV ega ds any ac o ‘ba e ing and o he domes ic
iolence, sexual abuse, sexual sla e y and exploi a ion, in e na ional a icking in women
and child en, o ced p os i u ion and sexual ha assmen , as well as iolence agains women
esul ing om cul u al p ejudice, acism and acial disc imina ion, xenophobia, po nog aphy,
e hnic cleansing, a med con lic , o eign occupa ion, eligious and an i- eligious ex emism
and e o ism’ (p.137). Any o hese ac s a e a h ea o human li e and human digni y,
he e o e mus be condemned and elimina ed (5). GBV is de ined as a iolen beha iou
di ec o one pe son o a g oup o pe sons based on hei gende and based on unequal powe
ela ions. I includes ‘ac s ha in lic physical, men al o sexual ha m o su e ing, h ea o
such ac s, coe cion and o he dep i a ions o libe y’ (4: p.10).
The ecen guidelines on GBV p e en ion, p oposed by In e -Agency S anding Commi ee
de ined GBV as any ha m ul ac ha is pe pe a ed agains a pe son’s will, and ha is based
on gende di e ences. These ac s can occu in p i a e o public con ex and go beyond
physical ha m o sexual, men al ha m o su e ing, h ea s, coe cion, and o he dep i a ions
o libe y (38). The e m GBV is also used o add ess and highligh iolence agains men and
CHAPTER 1. In oduc ion
16
boys, used wi h a gende -based pu pose. This ype o iolence ein o ces gende inequi able
no ms o males and emales (e.g. SV commi ed in a med con lic aimed a emascula ing o
eminising he enemy) (38).
Ne e heless, GBV also e e s o iolence pe pe a ed agains lesbian, gay, bisexual,
ansgende and in e sex (LGBTI) pe sons (38). This iolence in en s o punish hose pe sons
as i hey we e de ying gende no ms (42). To no e ha LGBTI is conside ed a ulne able
g oup o SGBV (42).
In he con ex o ( o ced) mig a ion, he UNHCR de ined SGBV as any ac o iolence
‘di ec ly agains a pe son on he basis o gende o sex. I includes ac s ha in lic physical,
men al o sexual ha m o su e ing, h ea s o such ac s, coe cion and o he dep i a ions o
libe y’ (4: p.11). Fi e ypes o iolence can be desc ibed when we speak o iolence owa ds
( o ced) mig an s: physical, psychological, sexual, socio-economic iolence and ha m ul
cul u al p ac ices (4). Acco ding o he de ini ion and ypes o SGBV de ined by UNHCR
(4), di e en ac s o iolence ha e been desc ibed:
-
Physical Violence: bea ing, kicking, bu ning, maiming o killing, wi h o wi hou
weapons; a icking, o ced labou o se ices, sla e y, se i ude o emo al o
o gans;
-
Psychological Violence: abuse/humilia ion o con inemen ;
-
Sexual Violence: ape and ma i al ape; child sexual abuse, de ilemen and inces ;
o ced sodomy/anal ape; a emp ed ape; sexual abuse; sexual exploi a ion; o ced
p os i u ion; sexual ha assmen ; sexual iolence as a weapon o wa and o u e;
-
Socio-economic Violence: disc imina ion and/o denial o oppo uni ies, se ices;
social exclusion/os acism based on sexual o ien a ion; obs uc i e legisla i e
p ac ice;
CHAPTER 1. In oduc ion
17
-
Ha m ul Cul u al P ac ices: emale geni al mu ila ion; ea ly ma iage; o ced
ma iage; honou killing and maiming; in an icide and o neglec ; denial o educa ion
o gi ls and women.
Focusing on sexual iolence i became pe inen o dis inguish be ween sexual ha assmen ,
sexual abuse o ape, which depend on he le el o physical con ac .
Sexual ha assmen s ands o unwelcome ac s, usually epea ed and un ecip oca ed sexual
ad ance, unsolici ed sexual a en ion, demand o sexual a ou s, sexual innuendo, display
o po nog aphic ma e ial as being o ced o wa ch po nog aphic ma e ial o being o ced o
und ess in a sexual con ex (4). I includes ac s wi hou physical con ac .
Sexual abuse is he ‘ac ual o h ea ened physical in usion o a sexual na u e, including
inapp op ia e ouching, by o ce o unde unequal o coe ci e condi ions.’ (4: p.16). Sexual
abuse includes ac s wi h physical con ac bu no pene a ion.
Rape s ands o ‘The in asion o any pa o he body o he ic im o o he pe pe a o wi h
a sexual o gan, o o he anal o geni al opening o he ic im wi h any objec o any o he
pa o he body by o ce, h ea o o ce, coe cion, aking ad an age o a coe ci e
en i onmen , o agains a pe son incapable o gi ing genuine consen .’(4: p.16). I includes
all ac s wi h pene a ion wi h a body pa o an objec .
Aligned wi h his de ini ion, he UNHCR desc ibed six possible p o iles o pe pe a o s o
SGBV: in ima e pa ne s; amily membe s, close ela i es and iends; in luen ial communi y
membe s; secu i y o ces and soldie s, including peacekeepe s; humani a ian aid wo ke s;
ins i u ions. The di e en and mul i p o iles o agg esso s should be aken in o conside a ion
while add essing and de ining p e en i e measu es o SGBV (4).
The de ini ion p esen ed by UNHCR is in e na ionally used o add ess esea ch on iolence
among e ugees, AS and ulne able mig an s (4), once i comp ises a b oad de ini ion o
SGBV, as well as, i e di e en ypes o iolence – physical, psychological, sexual, socio-
economic and ha m ul cul u al p ac ices.

CHAPTER 1. In oduc ion
18
The cu en li e a u e p oposes di e en de ini ions o SGBV. In his sense, i becomes
pe inen o unde s and wha he SGBV concep ualisa ion o ulne able popula ion, such as
e ugees, AS and undocumen ed mig an s is compa ed o he SGBV concep ualisa ion o
p o essionals wo king wi h hese popula ions? Is hei concep ualisa ion aligned wi h
UNHCR SGBV de ini ion?
CHAPTER 1. In oduc ion
19
3. Sexual and Gende -based Violence and Mig a ion
3.1 The s a e o Violence and Sexual and Gende -based Violence in he Eu opean
Union
Violence such as physical, sexual and psychological abuse occu s e e y day in e e y coun y,
inducing consequences ha goes beyond he ini ial abuse o pe manen consequences –
inju ies, men al heal h and ep oduc i e heal h p oblems, sexually ansmi ed diseases (43).
Global and na ional accu a e da a on he magni ude o iolence is sca ce and p obably
unde es ima ed (3,44). Ye , in Eu ope, a coun y-wise dis ibu ion o SGBV, in e ugees, AS
and undocumen ed mig an s, is s ill inexis en (4,45).
3.1.1 Violence
Da a om WHO (2002) s a ed ha mo e han 1.6 million o people los hei li es due o
iolence in 2000: almos hal hese dea hs we e suicide, nea ly a hi d homicides and a i h
we e wa - ela ed. Wo ldwide, each yea , i is es ima ed ha 1.3 million people die due o
iolence, accoun ing o 2.5% o mo ali y (8). Rega ding he na u e o iolence, he global
p e alence o physical and/o sexual in ima e pa ne iolence among all e e -pa ne ed
women was 30.0%, and 25% o Eu opean egion (46). The ecen epo o WHO on
iolence p e en ion demons a ed ha : ¼ o all adul s epo ed physical abuse as child en; 1
in 5 women epo ed sexual abuse as a child; 1 in 3 women has been a ic im o physical o
sexual iolence by an in ima e pa ne (8).
S öckl e al. (2013) es ima ed ha women had six imes mo e p obabili y o isk o homicide
by an in ima e pa ne han men, while IPV was esponsible o 13.5% o homicides o bo h
women and men. The same au ho s ound a p e alence o 38.6% o emale homicide
commi ed by an in ima e pa ne (6.3% o men) (47). Fu he mo e, one in h ee women is
expec ed o expe ience psychological abusi e beha iou , being he (ex)pa ne he
CHAPTER 1. In oduc ion
20
pe pe a o (48). The same s udy ound ha du ing childhood, one in h ee women has been
a ic im o physical o sexual iolence (48).
3.1.2 Sexual and Gende -Based Violence
Add essing da a on SV, among he o e all popula ion, 35.6% o women epo ed being
exposed o SV du ing hei li e ime (46); abou 20% o women and 5-10% o men epo ed
ha ing been sexually abused du ing childhood (44). Recen da a e ealed ha 1 in 20 women
had been aped and 55% unde go sexual ha assmen (48). Rape is equen ly used as a wa
weapon; s a is ics es ima es ha in he con lic in Bosnia and He zego ina 10 000 o 60 000
women we e aped by soldie s (3). Rega ding geni al mu ila ion, app oxima ely 500 000
women and gi ls in he EU ha e been exposed o his ype o iolence (49). Da a ega ding
sexual agg ession among young adul s, in Eu ope, epo ed high incidence a es. In his s udy,
he highes one-yea p e alence a e o emale ic imisa ion o since he age o consen was
ound in he Ne he lands, Ge many, Sweden, Spain, Finland and G eece, in opposi ion wi h
Belgium, Po ugal, Mal a and Hunga y. E en hough his s udy assesses he a ailable
e idence o SGBV and compa es s udies wi h a di e en me hodology, sample composi ion
and sexual agg ession de ini ion, i gi es a clea pic u e o he dimension o SGBV
p oblema ic in Eu ope (50).
3.1.3 Sexual and Gende -based Violence and Mig a ion
Mig a ion can lead o high social cos s and inequi ies, nega i ely a ec ing de elopmen and
inc easing ulne abili y, con ibu ing o he iola ion o mig an igh s (14). In he con ex o
( o ced) mig a ion, mig an s, e ugees and AS a e desc ibed in he li e a u e as a ulne able
popula ion o iolence wi h unique physical, men al, economic, and social cha ac e is ics
(1,2). Recen epo s and esea ch om EU coun ies acknowledge he high ulne abili y o
mig an s o iolence. A s udy conduc ed in Belgium and he Ne he lands conce ning
e ugees, AS and undocumen ed mig an s, ound a high p e alence o di ec o indi ec
CHAPTER 1. In oduc ion
21
iolence: 87/223 esponden s we e pe sonally ic imised, and 79 esponden s knew a leas
o one close pe son who was ic imised upon a i al o Eu ope. To highligh ha he
esponden s desc ibed 389 ac s o SGBV, ega dless he o igin, gende , age and s a us g oup
in e iewed (51). In Hunga y-Se ia bo de , UNHCR and he NGO Médecins sans F on iè es
(MSF) ha e documen ed se e al cases o indi idual iolence, being commi ed by
pe pe a o s in uni o ms (52). The s udy conduc ed in MSF clinics in Se ia included 992
mig an s/ e ugees, which o hose 270 (27%) had expe ienced iolen e en s du ing hei
jou ney. F om hose iolen ac s epo ed, mo e han hal (n=141,52%) we e pe pe a ed by
S a e au ho i ies (53). Da a om a ecen s udy conduc ed in Po ugal s a ed ha
app oxima ely 15% o immig an s li ing in Lisbon dis ic we e ic ims o iolence in he
pas yea (54).
Conside ing he na u e o iolence being SV and/o SGBV, indings o a sys ema ic e iew
and me a-analysis sugges ed ha app oxima ely one in i e e ugees o displaced women in
complex humani a ian se ings expe ienced SV (55). Au ho s also men ion he impo ance
o conside ing ha SV is o en unde - epo ed, due o social s igma, shame and ea o
ep isal. Ano he s udy e e en o A ican mig an s in Ge many, disclosed a p e alence o
16% o women epo ing a his o y o SV and men epo ing 6%, o ocus ha he au ho s
conside ed he possibili y o SV being unde epo ed due o ea and s igma (56). A s udy
conduc ed in UK clinic obse ed a 44.2% p e alence o epo ed SV in a con enience sample
o 43 AS (48.8% emale) compa ed o no epo s by a sample o 43 B i ish pa ien s (age- and
gende -ma ched sample) (57). In he same s udy, esul s disagg ega ed by gende epo ed
76.2% p e alence o SV among emale AS and 13.6% among male AS, sugges ing ha
women AS a e mo e exposed o SV when compa ed o men (57). A ecen s udy s a ed ha
sub-Saha an mig an s in Mo occo as well as du ing hei mig a ion jou ney, we e ulne able
o iolence, wi h 90% o he esponden s ha ing epo ed a leas one ype o iolence, 45 %
o which ega ded sexual iolence (58).
Resea ch has shown ha he es ic ed legal s a us is a cen al de e minan o iolence among
e ugees, AS and undocumen ed mig an s (32). Thei legal s a us impedes hei ac i e
CHAPTER 1. In oduc ion
28
Figu e 3: Ecological Model o unde s anding iolence, adap ed om Wo ld Heal h O ganiza ion (3).
Indi idual le el
A he indi idual le el, he Beijing Decla a ion (5) conside s gende a ac o o ulne abili y
o iolence: ‘Adolescen gi ls a e bo h biologically and psychosocially mo e ulne able han
boys o sexual abuse, iolence and p os i u ion, and o he consequences o unp o ec ed and
p ema u e sexual ela ions.’ (p.57). Since 1994, a he In e na ional Con e ence on
Popula ion and De elopmen (76), ha women’s ulne abili y has been assumed om an
indi idual o socie al pe spec i e; wo ldwide, women a e exposed o di e en h ea s in hei
li es, heal h and well-being, due o he lack o powe , gende inequi ies/powe ela ions,
lowe educa ion and o e load o wo k (76). Gi ls and women a e a majo isk o being
ic ims o sexual and gende -based iolence (77) wi h he male being he p ima y
pe pe a o s (38). Mo e ecen ly, a s udy wi h da a om 44 coun ies a i med ha gende
inequali y a he coun y le el is exposing women o iolence (78). Indeed, esea ch
demons a ed ha SGBV ulne abili y, especially o women and gi ls, can inc ease due o
inadequa e li ing condi ions, o e c owding a ecep ion acili ies, lack o gende -sensi i e in
asylum p ocedu es and a ecep ion acili ies (6,79,80).
E en hough SV is mo e equen ly epo ed by women and gi ls, being his iolence
pe pe a ed by men and boys, ape o men by men, and men o ced in o sex is also a eali y
(3). In gene al, only 30% o ic ims o iolence denounce he mos se ious inciden s o

CHAPTER 1. In oduc ion
29
au ho i ies (48). I we conside in e pe sonal iolence, men bea he bu den o homicide, wi h
82% o all homicide ic ims being men (8).
Child en and adolescen a e mo e ulne able o iolence. Acco ding o he WHO (2002),
child en and e ugees a e mo e suscep ible o collec i e iolence du ing con lic se ing.
Sexual abuse du ing childhood can be associa ed wi h majo social and men al heal h
p oblems, con lic i e pa en al p ac ices and wi h in e gene a ional ansmission o iolence
(81). Fu he mo e, a pe son ha has been abused du ing childhood is mo e likely o engage
in agg essi e and an isocial beha io s, in a la e s age o he /his li e (3,8,81–83). O , in
ano he hand, a pe son wi h inadequa e pa en ing is mo e likely o su e child mal ea men ,
you h iolence and/o in ima e pa ne and sexual iolence agains women (8). Finally,
e idence exis s ha child mal ea men is ela ed o u u e isk o sexual agg ession (84).
Alcohol and subs ance abuse ha e been associa ed wi h sexually agg essi e beha iou
(51,85–87). A sys ema ic e iew examining he ela ionship be ween alcohol consump ion
and men's sexual agg ession pe pe a ion concluded ha men who d ink hea ily o
compulsi e d ink, and d ink be o e da es a e mo e p one o commi sexual agg ession
compa ed wi h men who d ink in small quan i ies (85).
In e pe sonal le el
A he in e pe sonal le el, an impo an de e minan o SGBV is ela ed wi h in e pe sonal
ela ionships, namely exposu e o pa en al IPV in childhood (84). Wi nessing physical abuse
be ween pa en s/gua dians du ing childhood is conside ed a de e minan o la e sexual
agg ession. The s udy conduc ed by Su on & Simons (88) ound ha exposu e o physical
and emo ional iolen beha iou be ween pa en s accoun ed o sexual assaul pe pe a ion
by men and ic imisa ion among women. E en hough, SGBV can occu in public con ex ,
he close amily con ex , pee s, closed communi y, and socie y a e he main se ing whe e
ic imisa ion akes place (4).
CHAPTER 1. In oduc ion
30
Pe pe a o s o SGBV a e mo e likely o be known by he ic ims (46). Speci ic g oups o
popula ion, as mig an s and impo e ished people, end o be mo e ulne able o SGBV and
a e equally ic imised by s ange s, pe sons in au ho i y and/o people assigned o hei
p o ec ion (51).
Communi y and o ganisa ional le el
Coun y income le el is also conside ed a de e minan o ulne abili y o iolence,
speci ically o iolen dea h. A change in socio-economic s a us, due o mig a ion p ocess
can con ibu e o iolence (89). People wi h lowe socioeconomic s a us and people in low
and middle-income coun ies a e mo e han wice a isk o iolen dea h han hose li ing in
high-income coun ies (3). Also, an adjus men in social suppo and ne wo k, linked wi h
economic cons ain s leads o s ess, con lic and iolence (89). Speci ically o women,
economic empowe men is a po en ial de e minan o inc ease o dec ease he isk o being
abused (90). In a s udy compa ing na i e women and immig an exposed o IPV, indings
sugges ha ha ing paid wo k is conside ed a p o ec i e ac o o IPV o na i e women, bu
no o immig an women (91). Mo eo e , ano he s udy a i ms ha women who ha e access
o be e paid-jobs, inc eases he isks o being ic ims o IPV, due o a scena io whe e
husbands powe and au ho i y dec eases (89).
Widesp ead gende disc imina ion and inequali ies o en con ibu e o women and gi ls
ulne abili y o di e en o ms o GBV (38). Mo eo e , adi ional oo o socie ies wi h
igid gende oles and hos ile masculini y a e consis en ly ela ed o sexual assaul
pe pe a ion isk among men (84). A s udy conduc ed wi h an online communi y sample o
young men concluded ha hos ile masculini y, impe sonal sexual beha iou and a i udes,
and subs ance use a iables we e impo an p edic o s o sexual assaul pe pe a ion (92). In
his sense, ‘g owing up in a iolen o b oken home, subs ance abuse, social isola ion, igid
gende oles, po e y and income inequali y, as well as pe sonal cha ac e is ics such as poo
beha iou al con ol and low sel -es eem’ should be conside ed as social de e minan s o
iolence (3: p.19). Mo eo e , a pe cei ed app o al o sexual pe pe a ion among male pee s
CHAPTER 1. In oduc ion
31
and social ne wo ks is also e e ed as a de e minan o SV (84). E idence has shown ha
pe cei ing ha pee s and social ne wo k app o es o ced sex and/o sexual assaul is a isk
ac o o sexual agg ession among young college men (93).
Socie al le el
An impo an and po en ial p edic o o iolence is he mig a ion p ocess. Du ing he pe iod
o mig a ion i sel – bo de -c ossing – iolence is being highly epo ed (89). Fu he mo e,
p e-exposu e o iolence, auma and wa , is conside ed a p edic o o u u e iolen
beha iou (89). Rega ding women, he mig a ion p ocess is conside ed a de e minan o IPV,
due o possible changes in powe dynamics in hei in ima e ela ionship (94,95). In his
sense, p edic o s a e ela ed wi h economic di icul ies, low educa ional le els, being olde
and being sepa a ed o di o ced (91,96). In he con ex o pos -mig a ion, Gu uge (95)
iden i ied di e en social and economic obs acles which con ibu e o in ima e pa ne
iolence, such as social isola ion, poo access o employmen and ai wages o women (and
hei husbands), linguis ic ba ie s, di icul y accessing sa e housing, social and geog aphical
adjus men s, wel a e su eillance, and sys emic acism embedded wi hin heal h and social
se ices’(p.45). Mo eo e , Gu uge e al (89) conside s ha ‘(a) expe iences o iolence in
he p e-mig a ion con ex and du ing bo de c ossing; (b) gende inequi y in he ma i al
ins i u ion; (c) changes in social ne wo ks and suppo s; and (d) changes in socioeconomic
s a us and p i ilege’ (p.103) a e p edic o s o iolence, speci ically, in ima e male pa ne
iolence.
The mig a ion s a us
as such
, such as e ugee, AS o undocumen ed, is also conside ed a
de e minan o iolence. Vulne able communi ies – mig an s, e ugees, asylum seeke s and
undocumen ed mig an s, can ha e li e e en s ha induce suscep ibili y o hei heal h (22).
Since he 1951 Con en ion ela ing o he S a us o Re ugees and i s 1967 P o ocol, ha 148
coun ies ecognised ha e ugees a e among he mos ulne able pe sons in he wo ld,
su e ing a iola ion o di e en human igh s (7). Violence is no neu al, and esea ch shows
ha people wi h lowe socioeconomic s a us, such as mig an s, e ugees, IDP’s and
CHAPTER 1. In oduc ion
32
undocumen ed mig an s a e a majo isk o iolence (3,4). Rela ed o he mig a ion s a us,
he p ocess o accul u a ion is also conside ed a de e minan o iolence. Acco ding o Casey
& Mas e s (84) accul u a ion is de ined as he ‘p ocess in which wo o mo e cul u es in e ac ,
and membe s o each cul u e mus con end wi h he deg ee o which hey e ain hei own
cul u al p ac ices and/o adop aspec s o o he cul u al belie s and p ac ices’ (p.19). Du ing
his p ocess, indings sugges ha mig an s can ha e a highe o lowe ulne abili y o
iolence (84,91,97). Thus, accul u a ion can be conside ed a isk o p o ec i e ac o o SV
(84).
A e iew add essing isk and p o ec i e ac o s adolescen da ing iolence pe pe a ion,
including SV (87) concluded ha empa hy, social suppo , and school connec edness o
academic achie emen wo ks as a p o ec i e ac o iolence pe pe a ion.
Ne e heless, o he social de e minan s can be associa ed wi h iolence. The WHO epo on
iolence p e en ion (8) iden i ied ‘weak go e nance, poo ule o law; cul u al, social and
gende no ms; unemploymen ; income and gende inequali y; apid social change; and
limi ed educa ional oppo uni ies’ (p.33) as ac o s ha can be ela ed wi h iolen beha iou s
(8).
Se e al de e minan s can be associa ed wi h SGBV in ulne able mig an s, such as indi idual
and in e pe sonal de e minan s, socio-demog aphic and socio-economic ac o s (46,51).
Figu e 4 ep esen s an adap a ion o he ecological model o sexual iolence in e ugees, AS
and undocumen ed mig an s (2).
CHAPTER 1. In oduc ion
33
Figu e 4: The ecological model adap ed o sexual iolence in e ugees, AS and undocumen ed mig an s (2).
Violence is mul i ace ed, he e a e di e en ypes o iolence, which a e pe pe ua ed in
di e en con ex s, a any ime, can be di ec o one pe son, a g oup o can be used as a weapon
o wa , and can be s udied by di e se ac o ’s pe spec i es (i.e. pe pe a o , ic im, hi d pa y,
neu al obse e ). Once iolence is socially cons uc ed, who and wha is conside ed a iolen
ac a ies acco ding o speci ic de e minan s such as socio-cul u al and his o ical condi ions
(8,13). In sum, o unde s and iolence and iolence p e en ion, he e is a need o iden i ying
and add essing i s dynamics and in e ela ed social de e minan s (8,84).

CHAPTER 1. In oduc ion
34
3.4 Consequences o Sexual and Gende -based Violence
The impo ance o iolence in public heal h science has g own h ough cos ly consequences
in heal h (3,98). The consequences o iolence in he heal h s a us o a popula ion can be
unde s ood by he impac in human losses, inju ies and less ob ious consequences such as
psychological ha m, dep i a ion and malde elopmen ha will in luence he well-being o
indi iduals, amilies and communi ies (3,8). Mo e ecen ly, in 2030 Agenda o Sus ainable
De elopmen (99) i has been accep ed ha ‘ iolence des oys heal h in bo h he na ow sense
o causing dea h and disabili y—bu also mo e b oadly in holding back c ea i i y, economic
g ow h, and gene a ion o well-being.’ (p.1). Ye , au ho s sugges ed ha iolence has a di ec
impac on heal h, and induces addi ional bu den o heal hca e sys ems, no only due o ini ial
auma bu due o u he heal h consequences (100) and po en ially iolen beha io (a e
exposu e o iolence) (3,11).
Despi e he acknowledgemen o consequences o SGBV, he e a e no in e na ional
guidelines ega ding classi ica ion. Di e en classi ica ions can be ound in he li e a u e,
ega ding he iming o consequences (sho - e m consequences o long- e m consequences),
he na u e o consequences (physical, emo ional o psychological, ep oduc i e and sexual
o socio-economic), o speci ic ulne able g oups (child en, gi ls and women, MSM, LGBTI,
sex wo ke s, e ugees, AS, non-documen ed pe son, …). A desc ip ion o exis en da a will
ollow, g ouped acco ding o he na u e o he consequences: physical, psychological and
social.
Physical consequences
Rega ding physical consequences o iolence da a om WHO (2002) s a ed ha mo e han
1.6 million o people lose hei li e due o iolence in 2000. These dea hs included suicides,
homicides, wa - ela ed dea hs, ma e nal o in an mo ali y and IST ela ed mo ali y (3,4).
The WHO (2002) es ima ed ha 191 million o people – mo e han hal we e ci ilian – los
hei li e due o con lic - ela ed iolence. Speci ically o in ima e pa ne iolence (IPV),
CHAPTER 1. In oduc ion
35
da a ob ained o 66 coun ies epo ed an o e all 13.5% o homicides (47). Rega ding
speci ic inju ies esul ed om physical iolence by an in ima e pa ne , he p e alence anged
om 19% o 55% in e e -abused women (101). O he physical heal h consequences inducing
non- a al ou comes include abdominal and ho acic inju ies, b ain inju ies, bu ns/scalds,
ecchymosis, lace a ions, ac u es and disabili y (4,8). Apa om di ec physical
consequences o iolence, indi ec consequences o iolence such as hea disease, s oke,
cance , diabe es, kidney and li e p oblems and HIV/AIDS, a e he esul o beha iou s ha
ic ims migh adop (smoking, alcohol and d ug abuse, high- isk sexual beha iou ) a e
being exposed o ic imisa ion, and inducing a high bu den in ic ims li e (8).
Nega i e and di ec heal h impac s o SGBV a e majo and can induce ep oduc i e and
sexual heal h consequences, including sexually ansmi ed in ec ions and HIV in ec ion
(102); Abused women ha e an inc ease in 60% o heal h p oblems compa ing wi h non-
abused women, and epo ed mo e gynecological, ch onic s ess- ela ed and cen al ne ous
sys em p oblems (103). The WHO Mul i-coun y S udy on Women’s Heal h and Domes ic
Violence s a ed ha among e e -p egnan women, 25% o 20% o he women who we e
physically abused we e kicked and punched in he abdomen (101). Mo eo e , ep oduc i e
heal h consequences o ic imisa ion can include unwan ed p egnancy, misca iage,
p egnancy complica ions, unsa e abo ion, STIs, including HPV and HIV/AIDS, mens ual
diso de s, complex pain synd omes, ch onic pel ic pain, ecu en aginal in ec ions,
dec eased sexual desi e, gynecological diso de s and sexual diso de s (4,8,103).
Psychological consequences
The bu den o iolence goes beyond physical inju ies o psychological consequences. I
includes dep ession, anxie y, pos - auma ic s ess diso de , hype ac i i y, subs ance use and
dependence, ea ing and sleep diso de s, high- isk sexual beha iou s, suicidal hough s and
beha iou (3,8,95,101,104). In a me a-analysis s udy conduc ed by Golding (105), esul s
included a s ong associa ion be ween IPV and men al diso de s: ‘p e alence o men al heal h
p oblems among ba e ed women was 47.6% in 18 s udies o dep ession, 17.9% in 13 s udies
CHAPTER 1. In oduc ion
36
o suicidali y, 63.8% in 11 s udies o pos auma ic s ess diso de (PTSD), 18.5% in 10
s udies o alcohol abuse, and 8.9% in ou s udies o d ug abuse’ (p.99). Ano he po en ial
consequence is he ac ha ic ims a e usually insecu e and a isk o u he iolence, due
o ea , h ea s and lack o p o ec ion (4). Among child en, psychia ic mo bidi y migh be
linked wi h se e al amilial, pe sonal and en i onmen al ci cums ances, anging om cul u e
con lic , job insecu i y, eg e s abou lea ing home, amily dis up ion o unce ain u u e
oppo uni ies (21).
Social and economic consequences
Social consequences induced by sexual ic imisa ion, comp ises a high le el o social
ejec ion, s igma, sel -ha e and/o dep ession, possibly ela ed wi h he ac ha socie y ends
o blame he ic im (4). Long- e m e ec s a social le el include he deg ada ion o ma i al
ela ionships and amily ins abili y (23). Gene al e ec s as poo heal h s a us, poo quali y
o li e, and high use o heal h se ices a e equen long- e m consequences expe ienced by
SGBV ic ims (106). A he economic le el, a loss o a ole in he socie y can be a
consequence o ic imisa ion, eminisa ion o po e y and inc eased gende inequali ies (4).
A he legal le el, he inapp op ia e o lacking legal suppo o ic ims can lead o low
epo ing o cases, as well as epea ed agg essions due o no consequences o he pe pe a o
(4).
The impo ance o iden i ying and unde s anding he consequences o SGBV is ela ed o he
need o de ining e ec i e p og ams o add ess his p oblema ic (4). In he ollowing sub-
heading we will desc ibe he exis en p e en i e measu es add essing SGBV.
3.5 P e en able Measu es o Sexual and Gende -Based Violence
Violence and mo e speci ically SGBV is a p e en able global phenomenon (8,9,107,108).
Conside ing ha iolence is a lea ned beha iou and s a ing ha i is nowadays a wo ldwide
CHAPTER 1. In oduc ion
37
and majo public heal h p oblem, i could be al e ed by p e en i e s a egies and
in e en ions (8). The impo ance o p e en ing and mi iga ing SGBV is di ec ly ela ed o
he de elopmen o equi able and sus ainable socie ies and economies by imp o ing he
quali y o li e (99,108).
A comp ehensi e app oach o SV based on he ecological model is conside ed a powe ul
amewo k o unde s and and p e en SV (3,84,109,110). This app oach p econizes ha o
p e en SV, in e en ions and s a egies should add ess he social de e minan s a ou le els:
indi idual, in e pe sonal, communi ies, and social con ex (8,9,109,110). A s udy conduc ed
in Belgium and he Ne he lands concluded ha p e en i e measu es o SGBV add essing
e ugees, AS and undocumen ed mig an s should ake in o accoun he dynamics o he ou
le els socio-ecological model (51). Fu he mo e, he public heal h app oach o iolence
should also be used o de ine and implemen p e en i e measu es (37).
Taking in o accoun he ecological model and he public heal h app oach o iolence, se e al
guidelines, ecommenda ions and echnical packages a e a ailable add essing SGBV
p e en ion. In he ollowing poin s, we will gi e an ou line o he main global p e en ion
s a egies ha ha e al eady been de eloped.
The ision o WHO
The WHO epo on iolence p e en ion (8) p esen s he wo ldwide e o s ha ha e been
made on iolence p e en ion a e he epo and ecommenda ions eleased in 2002 – Wo ld
epo on Violence and Heal h (3). WHO and pa ne s (8) ha e iden i ied se en main
s a egies aiming o educe all o ms o iolence:
1.
‘de eloping sa e, s able and nu u ing ela ionships be ween child en and hei
pa en s and ca egi e s;
2.
de eloping li e skills in child en and adolescen s;
3.
educing he a ailabili y and ha m ul use o alcohol;
4.
educing access o guns and kni es;
CHAPTER 1. In oduc ion
44
ha hese p og ams we e local and popula ion speci ic and we e no eplica ed o o he
ulne able g oups.
E en hough SGBV p e en ion is possible, solu ions o he magni ude o he p oblem won’
be easy o each (108). The e is a need o mo ing o wa d he acknowledgemen o he
p oblem in socie ies, o design and implemen plans o ac ions, add essing he speci ici ies
o each coun y. Mo eo e , he need o alloca ing budge will enhance poli ical commi men
(108).
Conside ing he magni ude o SGBV p oblema ic among and owa ds e ugees, AS and
undocumen ed mig an s in he Eu opean con ex , and he lack o e idence-based e ec i e
p e en i e measu es, we a e commi ed o con ibu ing o he de elopmen o e idence-based
SGBV p e en ion public heal h in e en ions and policies.

CHAPTER 2. OBJECTIVES
CHAPTER 2. Objec i es
46
CHAPTER 2. Objec i es
47
2. OBJECTIVES
The aim o his hesis is o con ibu e o expand he knowledge on SGBV concep ualisa ion,
epo ed cases and causes o SGBV, p e en i e measu es and p edic i e ac o s o SGBV in
esiden s ( e ugees, AS and undocumen ed mig an s) and p o essionals (se ices and heal h
ca e p o ide s), li ing and wo king in Eu opean asylum ecep ion acili ies (EARF).
To achie e he men ioned aim o his hesis, speci ic objec i es we e de ined:
1. To unde s and SGBV concep ualisa ion, in a ulne able popula ion o e ugees, AS and
undocumen ed mig an s on he one hand, and in he p o essionals wo king wi h hese
communi ies in EARF, on he o he . To iden i y socio-demog aphic cha ac e is ics o bo h
g oups ha can be associa ed wi h SGBV concep ualisa ion.
2. To explo e epo ed cases o SGBV, causes and p e en able measu es desc ibed by
esiden s – e ugees, AS and undocumen ed mig an s – and p o essionals, li ing and wo king
in EARF om Eu opean hos ing coun ies. Fu he mo e, we in end o analyse po en ially
p e en able measu es o SGBV desc ibed by bo h g oups.
3. To explo e whe he a pa e n/ end o SGBV ic imisa ion in Eu opean asylum con ex
can be iden i ied in o de o de e mine u u e ou comes. Ye , o iden i y socio-demog aphic
cha ac e is ics ha migh p edic SGBV ic imisa ion in esiden s and p o essionals.
In gene al, his esea ch in ends o con ibu e o he de ini ion o e idence-based SGBV
p e en i e measu es in he EU, aking in o accoun iden i ied ac o s ha can induce SGBV
ulne abili y in esiden s and p o essionals, li ing and wo king in EARF.
CHAPTER 2. Objec i es
48
CHAPTER 3. METHODS

CHAPTER 3. Me hods
50
CHAPTER 3. Me hods
51
3. METHODS
In o de o answe ou speci ic objec i es, we used da a om he Eu opean p ojec –
Senpe o o. In his chap e , we p esen a desc ip ion o he Senpe o o p ojec and
concep ual amewo k, ollowed by ou esea ch design, and a desc ip ion o he
me hodology used, including pa icipan s, sampling, da a collec ion and da a analysis
me hods.
3.1 Senpe o o P ojec
The in e na ional esea ch p ojec Senpe o o - F ame o Re e ence in SGBV P e en ion
Agains and Among Young Re ugees, Asylum Seeke s and Unaccompanied Mino s in he
Eu opean Recep ion & Asylum Sec o - was a Eu opean p ojec de eloped in eigh coun ies
- Belgium, G eece, Hunga y, I eland, Mal a, he Ne he lands, Po ugal, Spain, and unded by
EC Daphne Fund. Senpe o o s ands o no mo e iolence, wi hou iolence, in Espe an o.
The o e all objec i e o Senpe o o was o con ibu e o he heal h p o ec ion and p omo ion
o young e ugees, AS and undocumen ed mig an s by p e en ing SGBV in Eu opean asylum
ecep ion acili ies. Mo eo e , Senpe o o aimed o in es iga e wha knowledge, a i ude,
p ac ice (KAP), and needs o p o essionals and hei clien s in he Eu opean ecep ion and
asylum sec o we e o de elop a need-, igh s- and e idence-based, pa icipa o y and gende -
balanced Eu opean F ame o e e ence. The concep ual amewo k was he socio-ecological
model on heal h and iolence, including desi able p e en ion and a Communi y Based
Pa icipa o y Resea ch (CBPR). CBPR is conside ed he adap ed esea ch me hod i we wan
o imp o e heal h h ough ac ion and social change (114).
Senpe o o p ojec included h ee phases. Fi s , a desk s udy was de eloped o gene a e an
o e iew o exis en good p ac ices in his ield. Second, a KAP s udy on SGBV and
p e en ion we e conduc ed among esiden s li ing, and p o essionals wo king, in asylum
ecep ion acili ies in he eigh Eu opean coun ies al eady men ioned. Thi d, he
CHAPTER 3. Me hods
52
de elopmen o a F ame o Re e ence in p e en ion o SGBV (115) – Code o conduc , a
S anda d Ope a ing P ocedu e, and a aining manual: Make i Wo k (116). The S anda d
Ope a ing P ocedu es consis ed on a se o p ac ical ins umen s o help ecep ion acili ies
in ‘de eloping comp ehensi e p ocedu es o p e en ion o SGBV wi hin he cen e, o
assis ing ic ims and o e e ing pe pe a o s’ (2: p.49). The Code o Conduc s ands o a
p ac ical guide o s a membe s and esiden s. ‘I de ines he ou lines and con en o hei
commi men in a i udes and beha iou o p e en ing, comba ing and esponding o e e y
o m o SGBV.’ (2: p.49). The Sensi iza ion Ki is a cul u ally compe en sensi isa ion
ins umen especially add essing AS and EARF p o essionals, bu ha could be adap ed o
any public. The aining manual - Make i Wo k! is a ‘p ac ical hands-on manual wi h an
engaging and non-judgmen al app oach o sensi i e issues such as sexual and ep oduc i e
heal h and SGBV.’ (2: p.49). Senpe o o was implemen ed om Decembe 2008 o
Decembe 2010.
Fo he implemen a ion o Senpe o o, s akeholde s om he eigh Eu opean coun ies we e
included in communi y ad iso y boa ds (CAB) om he eigh pa icipa ing coun ies. These
CAB consis ed o asylum seeke s and e ugees, asylum ecep ion p o essionals,
policymake s, in e media y o ganisa ions, ci il socie y and esea che s engaged in he
asylum and ecep ion sec o . The CAB we e in e media ies who had a c i ical and dis inc i e
impac on he p ocess and esul s, and pa icipa ed in e e y decisi e phase o he p ojec .
The s udy p o ocol applied he WHO and UNHCR e hical and sa e y guidelines in
esea ching iolence, complied wi h he local e hical equi emen s and ecei ed e hical
app o al om he Ghen Uni e si y Hospi al E hical Commi ee [B67020096667].
3.2 Concep ual F amewo k
The concep ual amewo k conside ed as a baseline o he subjec unde e iew comp ises
he socio-ecological model on heal h and iolence, which conside s ha li e-e en s in highe -
o de should in luence he human de elopmen h ough hei impac on e en s in lowe -o de
CHAPTER 3. Me hods
53
social ecosys ems (117). The socio-ecological model ecognizes he complexi y o SGBV
p oblema ic (3,4,75) and he mul i ac o ial causes ha accoun s o ic imisa ion and/o
pe pe a ion (4,8,84,118). I inco po a es ou -le el ac o s –indi idual, ela ionship,
communi y and socie y – o unde s and, mi iga e and p e en iolence (8,9,118). The
in e play o indi idual, ela ional, communi y and socie y- ela ed ac o s, de e mines he
ulne abili y o e ugees, AS and mig an s o ic imisa ion and/o pe pe a ion o SGBV
(2,8,51).
In 1996 he 49 h Wo ld Heal h Assembly ecognised iolence as a leading public heal h
p oblem. In his sense, a public heal h app oach o unde s anding and p e en ing iolence
has been de ined upon ou undamen al s eps: (a) o de ine he p oblem, (b) o iden i y he
ac o s ha inc ease he isk o iolence, (c) o de elop and o es p e en ion s a egies, and
(d) o dissemina e and implemen b oadly (8).
Mo eo e , and aking in o conside a ion he di e en e ms in he li e a u e used o add ess
SGBV and he di e en de ini ions o SGBV, o he pu pose o his esea ch hesis, we adop
he de ini ion o SGBV gi en by UNHCR, which conside s ha SGBV is ‘(…) iolence ha
is di ec ed agains a pe son on he basis o gende o sex. I includes ac s ha in lic physical,
men al o sexual ha m o su e ing, h ea s o such ac s, coe cion and o he dep i a ions o
libe y….’(4: p.11). This de ini ion comp ises i e ypes o SGBV, acco ding o he
consequences i migh induce, being physical, psychological, sexual, socio-economic and
ha m ul cul u al p ac ices (4).
3.3 Resea ch design
The g owing phenomenon o mig a ion and he ulne abili y o ic imisa ion and/o
pe pe a ion o e ugees, AS and undocumen ed mig an s equi e an u gen call o ac ion
(62). Ye , hese popula ions a e in eg a ed in o asylum ecep ion acili ies wi h speci ic
cha ac e is ics ha migh induce p o ec i e o isk ac o s o ic imisa ion. Also, wi hin hese
CHAPTER 3. Me hods
60
in ol es a o a ion o he PCs. Fo ou da a, we ha e applied a Va imax o a ion, being one
o he mos usual o a ion. Va imax o a ion has been de eloped by Kaise (120) and he
au ho e e s ha each componen has a small numbe o la ge loadings, and a la ge numbe
o ze o (o small) loadings. Ye , each o iginal a iable ends o be associa ed wi h one (o a
small numbe ) o he PC, and each PC ep esen s a small numbe o a iables (119).
To assess he speci ic objec i e, we used a ac o analysis app oach using PCA o a ac o
ex ac ion and Va imax o a ion wi h he main goal o ex ac ing he mos ele an
in o ma ion om ou da a, educing he olume o he da a, simpli ying he numbe o
obse a ions and inally o analyse he da a. We conduc ed a mul i a ia e analysis o 82
a iables ega ding SGBV knowledge. These 82 a iables consis ed o 82 ques ions
add essing SGBV knowledge ( able 2), and conside ed he ac ha an ac o iolence was
conduc ed o a gi l/woman o boy/man. We ook in o accoun he p inciples o PCA e e ing
ha PCs a e independen in o conside a ion, and labelling esul ed om he in e p e a ion o
he da a.
Table 2: Va iables used o conduc he p incipal componen analysis.
Unwelcome and unwan ed sexual commen s o in i a ions o gi ls/women (
saying you look sexy,
sugges o do some hing sexual,…)
And i his happens o boys/men?
Made o wa ch somebody und ess as a gi l/woman?
And i his happens o boys/men?
Made o wa ch pho os o naked pe sons as a gi l/woman?
And i his happens o boys/men?
Made o wa ch po n as a gi l/woman?
And i his happens o boys/men?
Ha ing o und ess in on o o he people wa ching as a gi l/woman?
And i his happens o boys/men?
Th ea ening gi ls/women o sexual ac s (saying ha you will be aped, o he/she
will do sexual hings o
you i you don’ do some hing o him/he ,…)
And i his happens o boys/men?

CHAPTER 3. Me hods
61
As a gi l/woman o be isola ed, con ined and/o dep i ed o libe y o mo emen
(no allowed o lea e
he house, no allowed o speak you mo he ongue, no allowed o ha e con ac wi h o he s, locking
up…)
And i his happens o boys/men?
Unwelcome ema ks and commen s om nonsexual na u e o gi ls/women
(cu se, swea , call names,
blame, accuse un ai ly...)
And i his happens o boys/men?
Th ea ening o gi ls/women wi h unwelcome no sexual ac s
(make you eel sca ed, en e in ou p i a e
space, des oy objec s...)
And i his happens o boys/men?
Teasing, showing no espec , acis o disc imina ing commen s o a gi l/woman?
And i his happens o boys/men?
Someone denying a gi l/woman o be oge he wi h hei pa ne in p i a e?
And i his happens o boys/men?
Someone denying a gi l/woman o be oge he wi h he pa en s o child en in p i a e?
And i his happens o boys/men?
Someone denying a gi l/woman o ul il he ole as a mo he (
no money o ood, clo hes, housing,…)
And i his happens o boys/men?
Physical assaul wi h no pe manen consequences
(e.g. hi ing, kicking, pulling you hai , d ag you,…)
And i his happens o boys/men?
Physical assaul wi h pe manen consequences
(e.g. bu ning, s abbing, maiming, mu ila ing, killing,…)
And i his happens o boys/men?
Unwelcome ouching o b eas s, geni als and o he p i a e body pa s o gi ls/women.
And i his happens o boys/men?
Unwelcome kissing and ca essing o gi ls/women?
And i his happens o boys/men?
Unwelcome a emp ed pene a ion by an o gan o an objec in any body opening o gi ls/women.
And i his happens o boys/men?
Unwelcome pene a ion o he mou h by an o gan o by an o gan o by an objec o gi ls/women
And i his happens o boys/men?
Unwelcome pene a ion o he agina and/o anus by an o gan o by an objec o gi ls/women.
And i his happens o boys/men?
Unwan ed sex wi hin a ela ionship and/o ma iage o a gi l/woman?
And i his happens o boys/men?
CHAPTER 3. Me hods
62
Fo ced p os i u ion o gi ls/women?
And i his happens o boys/men?
Sex wi h a gi l/woman in exchange o su i al, ood o he child en, shel e , money, pape s, o he
a ou s.
And i his happens o boys/men?
Sexual sla e y/ a icking o gi ls/women?
And i his happens o boys/men?
Rape o gi ls/women as a weapon o wa ?
And i his happens o boys/men?
Ci cumcision o gi l/woman?
And i his happens o boys/men?
Child ma iage o a gi l/woman?
And i his happens o boys/men?
A anged ma iage agains he will o he gi l/woman?
And i his happens o boys/men?
Neglec ing emale child en, denial om educa ion o emale child en?
And i his happens o boys/men?
Inju ing a gi l/woman in he name o amily honou ?
And i his happens o boys/men?
Killing a gi l/woman in he name o amily honou ?
And i his happens o boys/men?
T a icking o people o hei o gans?
And i his happens o boys/men?
T a icking o labou ?
And i his happens o boys/men?
Being ea ed di e en ly by o he people because o being a gi l/woman?
And i his happens o boys/men?
Being ea ed di e en ly by o he people because o he sexual o ien a ion o a gi l/woman.
And i his happens o boys/men?
Being ea ed di e en ly by o he people because o he e hnic backg ound o a gi l/woman
And i his happens o boys/men?
Being ea ed di e en ly by o he people because o he esidence s a us o a gi l/woman.
And i his happens o boys/men?
Denial o access o educa ion, heal h assis ance o emune a ed employmen because o
CHAPTER 3. Me hods
63
being a gi l/woman.
And i his happens o boys/men?
Denial o access o educa ion, heal h assis ance o emune a ed employmen because o
being a gi l/woman.
And i his happens o boys/men?
Denial o access o educa ion, heal h assis ance o emune a ed employmen because o
being a gi l/woman.
And i his happens o boys/men?
Denial o access o educa ion, heal h assis ance o emune a ed employmen because o
being a gi l/woman.
And i his happens o boys/men?
To answe o he second speci ic objec i e:
o explo e epo ed cases o SGBV, causes and
p e en able measu es desc ibed by esiden s and p o essionals, li ing and wo king in EARF
om Eu opean hos ing coun ies. Fu he mo e, we in end o analyse po en ially p e en able
measu es o SGBV desc ibed by bo h g oups;
Desc ip i e analysis was conduc ed using
SPSS. Associa ion es s as he Chi-squa e and Fishe ’s exac es we e conduc ed o
unde s and i signi ican associa ion exis ed.
Finally, o answe ou hi d speci ic objec i e:
o explo e whe he a pa e n/ end o SGBV
ic imisa ion in Eu opean asylum con ex can be iden i ied in o de o de e mine u u e
ou comes. Ye , o iden i y socio-demog aphic cha ac e is ics ha migh p edic SGBV
ic imisa ion in esiden s and p o essionals;
we ha e used machine lea ning echniques,
which we will desc ibe in he ollowing pa ag aphs.
Machine Lea ning (ML) a e p edic i e modelling, in ending o u n da a in o in o ma ion,
and in o ma ion in o knowledge, by ex ac ing in o ma ion om he aw da a om da abases
(121). ML aims o p edic and o ecas a sys em o a p ocess, based on da a pa e ns ex ac ed
by s a is ical me hods, o algo i hms. The e o e, ML modelling is based on supe ised
lea ning algo i hms elying upon he expe imen al na u e o lea ning p ocesses which can be
simpli ied as compa isons be ween inpu s and ou pu s, o be ween ac ions and i s
consequences in a supe iso y lea ning con ex (122,123). Da a-d i en modelling equi es a
CHAPTER 3. Me hods
64
ep esen a i e da abase which accommoda es concise examples o ou pu esponses o
a ious inpu s a ia ions (124). Shmueli (122) summa ises he s eps o da a-d i en
modelling as ollows:
1.
De ine Goal;
2.
Collec da a;
3.
P epa e da a & Explo a o y da a analysis;
4.
Choose a iables;
5.
Choose me hods;
6.
E alua e, Valida e & Selec models;
7.
Use models.
Acco ding o Smueli (122) ‘S a is ical modeling is a powe ul ool o de eloping and es ing
heo ies by way o causal explana ion, p edic ion, and desc ip ion’ (p. 289). Acco dingly, we
ha e applied he p e iously men ioned s eps o answe ou hi d speci ic objec i e.
Ou goal was de ine as p edic ing he a iable/ou pu “we e you a ic im?”. This ques ion
was ela ed o he desc ip ion o an SGBV expe ience, wi hin 12 mon hs be o e he
ques ionnai e.
The da a collec ion was made h ough ques ionnai es o esiden s and p o essionals o
Eu opean ecep ion and asylum acili ies (de ailed in o ma ion is included in he poin 3.5 o
his Chap e ). Rega ding ou da a, we decided o exclude he Spanish in e iews o he open
ques ions pa ega ding iolence expe iences, a i udes and p e en ion and esponse
measu es, as we doub ed whe he alidi y could be gua an eed (45). This b ough he o al o
analysed in e iews down o 562 in e iews: 375 esiden s and 187 p o essionals. Fu he ,
we ha e iden i ied in o al 30 independen a iables, ela ed wi h socio-demog aphic
cha ac e is ics. In p edic i e modeling o he selec ion o a iable he e is no need o go
deepe in o he exac ole o each a iable in e ms o an unde lying causal s uc u e, ins ead
he selec ion o he a iables should be based on da a quali y (122). The lis o a iables
conside ed a i s o ou p edic i e models is p esen ed in able 3.
CHAPTER 3. Me hods
65
Table 3: Lis o a iables conside ed as inpu s o p edic i e models.
Desc ip ion
Residen o p o essional
Wha is you sex?
Wha is you age in yea s?
Wha is you ac ual ma i al s a us?
A e you cu en ly li ing wi h a pa ne ?
A e you li ing wi h a pa ne in he same ecep ion acili y?
Do you ha e any child en?
I you ha e child en, a e hey li ing wi h you?
Wha is you ac ual s a us acco ding o immig a ion law?
In wha ype o ecep ion/asylum acili y a e you li ing?
Can you speci y which kind o accommoda ion you a e li ing in?
I o he accommoda ion, please speci y:
Is he e a common a ea in his accommoda ion?
Do you ha e a place o sleep in his accommoda ion?
I you ha e a place o sleep, wha is he size o his space?
Wi h how many adul s (equal o olde han 18) do you sha e his space?
Wha is hei ela ionship o you (adul s wi h whom you sha e he place)?
Wha is hei sex (adul s wi h whom you sha e he place)?
Wi h how many child en (younge han 18) do you sha e his space?
Wha is hei ela ionship o you (child en wi h whom you sha e he place)?
Wha is hei sex (child en wi h whom you sha e he place)?
Wha kind o sani a y acili ies do you ha e in his accommoda ion?
Wi h how many pe sons do you ha e o sha e hese sani a y acili ies?
Wha is hei sex (pe sons wi h whom you sha e he sani a y acili ies)?
Wha is hei app oxima e age (pe sons wi h whom you sha e he sani a y
acili ies)?
Since when a e you wo king he e (yea )?
Wha is you cu en occupa ion in his acili y?
When did you lea e he acili y?
Wha was he eason o you depa u e?
Fo ou modelling, we ha e excluded he a iables ha we e only answe ed by one o he
g oups (e.g. main daily ac i i y in you coun y o o igin – esiden s; o e.g. ype o
ecep ion/asylum acili y a e you wo king – p o essionals). The da a was p epa ed by
handling he no esponses/missing alues as accep able esponse choices. E.g. o a yes/no
ques ion he hi d op ion ‘no esponse’ is handled as he hi d op ion.

CHAPTER 3. Me hods
66
The nex s ep consis ed o analysing a iable by a iable ( om he lis p esen ed in Table 3)
and i s impo ance o po en ially p edic SGBV ic imisa ion. Finally, and aking in o
conside a ion he exis en li e a u e and expe s’ opinion we ha e excluded he ollowing
a iables: coun y whe e esiden s we e bo n, coun y o esea ch, ime o a i al o Eu ope
o he hos coun y, and eligion. In ou modelling, we ha e ollowed wo app oaches o
ea u e selec ion, combining expe s’ knowledge wi h ea u e selec ion algo i hms. These
wo app oach me hod o ea u e selec ion is belie ed o deli e mo e obus models (125).
Se e al ML echniques a e a ailable in he li e a u e. The selec ion o which machine
lea ning algo i hms – eg ession, decision ees, andom o es s, suppo ec o machines,
(…) - a e mo e sui able o simula ing how he human b ain wo ks, o how we p edic
beha iou ha e been he subjec o se e al esea che s, and i s ill emains a s uggling
ad ance (126,127). Rega ding he selec ion o models o p edic SGBV ic imisa ion, we
op ed o he ollowing h ee: Random Fo es s, he Suppo Vec o Machines (SVM) and
Logis ic Reg ession. A desc ip ion o each modelling echnique can be ound in annex I. To
analyse he i ing o ou models we will use he a ea unde he cu e (AUC) o he ecei e
ope a ing cha ac e is ic (ROC) cu e.
In ou hi d s udy, we applied machine lea ning algo i hms. We analysed di e en po en ial
p edic i e a iables ela ed o socio-demog aphic cha ac e is ics o bo h g oups - esiden s
and p o essionals ha migh ha e a p edic i e impo ance o SGBV ic imisa ion. P edic i e
modelling can sugges imp o emen s o exis ing explana o y models, by cap u ing
unde lying complex pa e ns and ela ionships be ween a iables (122).
CHAPTER 4. RESULTS
CHAPTER 4. Resul s
68
CHAPTER 4. Resul s
69
4.1 S udy 1 – Concep ualising sexual and gende -based iolence in Eu opean asylum
ecep ion acili ies: esul s om a p incipal componen analysis
In oduc ion
Sexual and Gende -based Violence (SGBV) is a majo public heal h p oblem and a
iola ion o human igh s (3,4). Conside ing he global challenge o ( o ced) mig a ion
(23), UNHCR (4) de ines SGBV as ‘(…) iolence ha is di ec ed agains a pe son on he
basis o gende o sex. I includes ac s ha in lic physical, men al o sexual ha m o
su e ing, h ea s o such ac s, coe cion and o he dep i a ions o libe y (….)’ (p.11).
SGBV hampe s gende -s e eo yped ac s o iolence, based on unequal powe ela ions
and denying human digni y, igh s, and de elopmen (4,38), which can be ca ego ised
in o physical, psychological, sexual, socio-economic iolence and ha m ul cul u al
p ac ices (4). SGBV concep ualisa ion is a ma e o judgemen , a ec ed by cul u al
belie s, social no ms, and alues (44). Mo eo e , iolence has been de ined as a socially
cons uc ed concep once who and wha is conside ed iolen beha iou changes
acco ding o speci ic de e minan s such as socio-cul u al and his o ical condi ions (13).
Resea ch has demons a ed ha emale, male and ansgende e ugees, asylum seeke s
(AS) and mig an s a e ulne able o SGBV wi h unique physical, men al, economic, and
social cha ac e is ics (1,2,23). Findings o a sys ema ic e iew and me a-analysis sugges
ha app oxima ely one in i e e ugee o displaced women in complex humani a ian
se ings, expe ience sexual iolence (55). In a s udy o SGBV among male and emale
e ugees, AS and undocumen ed mig an s in Eu opean asylum ecep ion cen es (EARF)
mul iple ypes o SGBV we e epo ed, ca ego ised as physical (42.18%), emo ional
(40.54%), socio-economic iolence (11.2%) and sexual (5.98%) (45). A s udy conduc ed
in Belgium and he Ne he lands ound a high p e alence o di ec o indi ec SGBV
exposu e among e ugees, AS and undocumen ed mig an s: 87/223 esponden s had been
pe sonally ic imised, and 79 esponden s knew a leas one close pe son who was
ic imised since hei a i al in Eu ope. Responden s desc ibed 332 ac s o SGBV,
ega dless he o igin, gende , age, and s a us g oup o he in e iewed. The majo i y o
pe pe a o s we e male (74.0%), and 69.3% o ic ims we e emale (male ic ims we e
28.6%) (51). Fu he mo e, asylum- ela ed p o essionals we e ound o be he assailan s
in one- i h o he epo ed cases (51). In Ge many, a communi y o A ican mig an s
CHAPTER 4. Resul s
76
PC9 – Ma i al Rape
Unwan ed sex wi hin a ela ionship and/o ma iage o a gi l/woman? 0.754
And i his happens o boys/men? 0.793
PC1 – Abuse, ape and a icking
Unwelcome pene a ion o he agina and/o anus by an o gan o by an objec o
gi l/woman. 0.831
And i his happens o boys/men? 0.831
Fo ced p os i u ion o gi ls/women? 0.817
And i his happens o boys/men? 0.819
Sexual sla e y/ a icking o gi ls/women? 0.749
And i his happens o boys/men? 0.802
Rape o gi ls/women as a weapon o wa ? 0.791
And i his happens o boys/men? 0.789
PSYCHOLOGICAL VIOLENCE
PC3 – Humilia ion
Unwelcome ema ks and commen s om nonsexual na u e o gi ls/women. 0.767
And i his happens o boys/men? 0.787
Teasing, showing no espec , acis o disc imina ing commen s o a gi l/woman? 0.716
And i his happens o boys/men? 0.728
PC6 – Con inemen
Someone denying a gi l/woman o be oge he wi h hei pa ne in p i a e? 0.751
And i his happens o boys/men? 0.700
HARMFUL CULTURAL PRACTICES
PC10 – Denial o educa ion o gi ls and women
Neglec ing emale child en, denial om educa ion o emale child en? 0.623
And i his happens o boys/men? 0.653
PC14 – Geni al mu ila ion
Ci cumcision o gi l/woman? 0.450
PC11 – Ea ly ma iage
Child ma iage o a gi l/woman? 0.803
And i his happens o boys/men? 0.817
PC5 - Hono killing and Maiming
Killing a gi l/woman in he name o amily honou ? 0.751
And i his happens o boys/men? 0.745
SOCIO-ECONOMIC VIOLENCE
PC13 – Disc imina ion
Being ea ed di e en ly by o he people because o being a gi l/woman? 0.569
And i his happens o boys/men? 0.603

CHAPTER 4. Resul s
77
PC2 - Denial o oppo uni ies and se ices
Denial o access o educa ion, heal h assis ance o emune a ed employmen because
o he esidence s a us o a gi l/woman. 0.792
And i his happens o boys/men? 0.790
Denial o access o educa ion, heal h assis ance o emune a ed employmen because
o being a gi l/woman. 0.774
And i his happens o boys/men? 0.763
PC8 – Denial o access o exe cise ci il, social, economic igh s
As a gi l/woman o be isola ed, con ined and/o dep i ed o libe y o mo emen 0.644
And i his happens o boys/men? 0.637
PC7 – Social exclusion/os acism based on sexual o ien a ion
Being ea ed di e en ly by o he people because o he sexual o ien a ion o
gi l/woman. 0.853
And i his happens o boys/men? 0.855
P o essionals
The mul i a ia e analysis o PC’s o he g oup o p o essionals esul ed in 17 new
a iables ep esen ing 86.92% o he o al a iance o collec ed da a. These new a iables
we e analysed and labelled dimensions o SGBV (4), and desc ibe p o essionals SGBV
concep ualisa ion. In his sense, o p o essionals he concep o SGBV includes he
ollowing dimensions: abuse, ape and a icking; denial o oppo uni ies and se ices;
isual sexual ha assmen ; social exclusion/os acism; physical assaul wi h pe manen
consequences; ela ional iolence; ea ly ma iage; physical assaul wi hou pe manen
consequences; e bal iolence; sexual exploi a ion; hono killing and maiming; h ea and
humilia ion; con inemen – indi idual le el; Denudemen ; sexual innuendo; pa en al
ela ional iolence; geni al mu ila ion. The ep esen a i e ques ions o each dimension o
SGBV a e desc ibed in Table 6.
Table 6: P incipal componen analysis o P o essionals: ep esen a i e ques ions and ou pu loading
(Va imax a ia ion).
PROFESSIONALS
Dimensions o SGBV PCA Loading ou pu
SEXUAL VIOLENCE
PC15 – Sexual innuendo
CHAPTER 4. Resul s
78
Unwelcome and unwan ed sexual commen s o in i a ions o gi ls/women. 0.619
And i his happens o boys/men? 0.600
PC3 – Visual sexual ha assmen
Made o wa ch somebody und ess as a gi l/woman? 0.858
And i his happens o boys/men? 0.879
Made o wa ch pho os o naked pe sons as a gi l/woman? 0.887
And i his happens o boys/men? 0.817
PC14 – Denudemen
Ha ing o und ess in on o o he people wa ching as a gi l/woman? 0.698
And i his happens o boys/men? 0.799
PC1 –Abuse, Rape and T a icking
Unwelcome pene a ion o he agina and/o anus by an o gan o by an objec o gi ls/women?
0.930
And i his happens o boys/men? 0.930
T a icking o people o hei o gans? 0.833
And i his happens o boys/men? 0.833
PC10 – Sexual exploi a ion
Sex wi h a gi l/woman in exchange o su i al, ood o he child en, shel e , money, pape s,
o he a ou s. 0.916
And i his happens o boys/men? 0.916
PHYSICAL VIOLENCE
PC8 – Physical assaul wi hou pe manen consequences
Physical assaul wi h no pe manen consequences (e.g. hi ing, kicking, pulling) 0.773
And i his happens o boys/men? 0.790
PC5 – Physical assaul wi h pe manen consequences
Physical assaul wi h pe manen consequences (e.g. bu ning, s abbing, maiming) 0.877
And i his happens o boys/men? 0.877
Killing a gi l/woman in he name o amily honou ? 0.798
And i his happens o boys/men? 0.843
PSYCHOLOGICAL VIOLENCE
PC 12 – Th ea and humilia ion
Th ea ening o gi ls/women wi h unwelcome no sexual ac s (make you eel sca ed.…) 0.548
And i his happens o boys/men? 0.549
Teasing, showing no espec , acis o disc imina ing commen s o a gi l/woman? 0.546
And i his happens o boys/men? 0.546
PC9 – Ve bal iolence
Unwelcome ema ks and commen s om nonsexual na u e o gi ls/women. 0.759
And i his happens o boys/men? 0.759
PC13 – Con inemen , indi idual le el
CHAPTER 4. Resul s
79
As a gi l/woman o be isola ed, con ined and/o dep i ed o libe y o mo emen 0.720
And i his happens o boys/men? 0.642
PC6 – Rela ional iolence
Someone denying a gi l/woman o be oge he wi h his o he pa ne in p i a e? 0.863
And i his happens o boys/men? 0.862
Someone denying a gi l/woman o be oge he wi h he pa en s o child en in p i a e. 0.812
And i his happens o boys/men? 0.812
PC16 – Pa en al ela ional iolence
Someone denying a gi l/woman o ul il he ole as a mo he (no money o ood) 0.669
And i his happens o boys/men? 0.669
HARMFUL CULTURAL PRACTICES
PC17 – Geni al mu ila ion
Ci cumcision o gi l/woman? 0.417
And i his happens o boys/men? 0.632
PC7 – Ea ly ma iage
Child ma iage o a gi l/woman? 0.882
And i his happens o boys/men? 0.862
PC 11 – Hono killing and maiming
Inju ing a gi l/woman in he name o amily honou ? 0.853
And i his happens o boys/men? 0.853
SOCIO-ECONOMIC VIOLENCE
PC2 – Denial o oppo uni ies and se ices
Denial o access o educa ion, heal h assis ance o emune a ed employmen because o he
e hnic backg ound o a gi l/woman 0.874
And i his happens o boys/men? 0.874
Denial o access o educa ion, heal h assis ance o emune a ed employmen because o he
esidence s a us o a gi l/woman. 0.846
And i his happens o boys/men? 0.846
PC4 – Social exclusion/os acism
Being ea ed di e en ly by o he people because o he sexual o ien a ion o a gi l/woman? 0.794
And i his happens o boys/men? 0.806
Being ea ed di e en ly by o he people because o he e hnic backg ound o a gi l/woman? 0.780
And i his happens o boys/men? 0.780
SGBV concep ualisa ion and socio-demog aphic cha ac e is ics
The associa ion be ween each dimension o SGBV concep ualisa ion and esiden s’
socio-demog aphic cha ac e is ics and p o essionals’ cha ac e is ics a e p esen ed in
CHAPTER 4. Resul s
80
Table 7 and 8, espec i ely. Ou esul s show ha wha is conside ed a speci ic beha iou
o a sexual ac as iolence is di e en acco ding o speci ic socio-demog aphic
cha ac e is ics. We will now desc ibe he signi ican indings, i s o esiden s and
subsequen ly o p o essionals, g ouped in o he i e ypes o SGBV p econized by
UNHCR (4).
Residen s
Sexual iolence: Residen s sexual innuendo’ concep ualisa ion was associa ed wi h he
hos coun y (p=0.010), he kind o accommoda ion (p=0.026), he le el o educa ion o
esiden s (p=0.016) o daily ac i i y in he hos coun y (p=0.037). This inding means
ha esiden s li ing in Belgium and I eland, in a con aine , s udio o oom, wi h an
educa ion (p ima y, seconda y o highe ), o he ones ha do no ha e a job in he hos
coun y end o disag ee wi h sexual innuendo being a ype o iolence. Ma i al ape was
associa ed wi h he age o esiden s (p=0.001), and he kind o accommoda ion whe e
hey we e li ing in (p=0.001). You h and adul s’ esiden s (0-39 yea s old) li ing in
con aine s, oom o s udio end o disag ee ha ma i al ape is a o m o iolence. Abuse,
ape and a icking we e associa ed wi h hos coun y (p=0.001). Residen s ha end o
disag ee we e hos ed in Po ugal and Spain.
Psychological iolence: The concep o con inemen was signi ican ly associa ed wi h
age (p= 0.032), meaning ha he esiden s aged un il 18 yea s old ended o disag ee wi h
con inemen as a o m o iolence.
Ha m ul cul u al p ac ices: Denial o educa ion o gi ls as iolence was associa ed wi h
ma i al s a us, meaning ha single esiden s ended o ully ag ee wi h his as a o m o
iolence (p=0.033). The concep ualisa ion o geni al mu ila ion as a o m o iolence was
associa ed wi h a ained educa ion (p=0.033). Hono killing and maiming
concep ualisa ion we e associa ed wi h he coun y o esea ch (p=0.001), sex (male o
emale) (p=0.004) and age (p=0.042) o esiden s. Residen s hos ed in Belgium and
G eece, male and aged om 19 o 39 yea s old end o disag ee wi h his concep ualisa ion
as a o m o iolence.
CHAPTER 4. Resul s
81
Socio-economic iolence: Social exclusion based on sexual o ien a ion concep was
associa ed wi h he ime o a i al o Eu ope o hos coun y (p=0.018 and 0.007), and
daily ac i i y in he coun y o o igin (p=0.046). Residen s ha a i ed ecen ly a he hos
coun y o Eu ope (less han 5 yea s) and used o ha e a job in he coun y o o igin ended
o ully disag ee ha social exclusion based on sexual o ien a ion is a o m o iolence.

CHAPTER 4. Resul s
82
Table 7: Residen s – SGBV concep ualisa ion and socio-demog aphic cha ac e is ics (p- alues: Chi-squa e es and Fishe Exac es ).
Dimensions o SGBV Concep
Socio-demog aphic cha ac e is ics
o Residen s Sexual Psychological Ha m ul Cul u al P ac ices Socio-economic
PC 12 PC 4 PC 9 PC 1 PC 3 PC 6 PC 10 PC 14 PC 11 PC 5 PC 13 PC 2 PC 8 PC 7
Hos coun y
0.010 0.167 0.127 0.001 0.266 0.183 0.155 0.571 0.482 0.001 0.678 0.842 0.078 0.086
Sex
0.268 0.650 0.580 0.056 0.829 1.000 0.886 0.897 0.070 0.004 0.374 1.000 0.852 0.305
Age
0.185 0.625 0.001 0.212 0.806 0.032 0.059 0.545 0.470 0.042 0.616 1.000 0.105 0.174
Ma i al s a us
0.842 0.273 0.754 0.281 0.362 0.204 0.033 0.363 0.189 0.580 0.253 0.565 0.911 0.716
Ha ing Child en
0.104 1.000 0.243 0.288 0.289 0.125 0.502 0.530 0.295 0.874 0.498 1.000 1.000 0.801
S a us immig a ion Law
0.195 0.087 1.000 0.321 0.626 0.798 1.000 0.124 0.328 1.000 0.161 1.000 0.458 0.064
Yea o a i al Eu ope
0.708 1.000 0.544 0.281 1.000 0.773 0.679 0.484 0.603 0.679 0.079 1.000 0.340 0.018
Yea o a i al o hos coun y
0.458 1.000 0.513 0.075 1.000 0.737 0.618 0.420 0.543 1.000 0.295 1.000 0.280 0.007
Type o ecep ion acili y li ing
0.394 0.646 0.792 0.062 1.000 0.332 1.000 1.000 1.000 0.471 0.436 1.000 1.000 1.000
Kind o accommoda ion
0.026 0.439 0.001 0.388 0.934 0.056 0.951 0.520 1.000 0.123 0.728 0.645 0.207 0.603
A ained educa ion
0.016 0.668 0.647 0.415 0.899 0.560 0.447 0.033 0.558 0.180 0.074 0.611 0.940 0.390
Daily ac i i y coun y o o igin
0.587 0.215 1.000 0.232 0.065 1.000 0.453 0.902 0.308 0.412 0.288 1.000 0.659 0.046
Daily ac i i y hos coun y
0.037 1.000 0.233 0.412 0.502 0.650 0.834 0.467 0.176 0.278 0.758 0.308 0.070 0.744
Signi ican p- alue p<0.05 bolded.
PC 12: Sexual innuendo; PC 4: Visual sexual ha assmen ; PC 9: Ma i al ape; PC 1: Abuse, ape and a icking; PC 3: Humilia ion; PC 6: Con inemen ; PC 10: Denial o educa ion o gi ls and women; PC 14: Geni al
mu ila ion; PC 11: Ea ly ma iage; PC 5: Hono killing and maiming; PC 13: Disc imina ion; PC 2: Denial o oppo uni ies and se ices; PC 8: Denial o access o exe cise ci il, social and economic igh s; PC 7: Social
exclusion/os acism based on sexual o ien a ion.
CHAPTER 4. Resul s
83
P o essionals
Sexual iolence: Fo p o essionals, sexual innuendo concep ualisa ion was associa ed
wi h language skills (p=0.012). P o essionals wi h good language skills (a leas 2 EU
languages) ended o ully disag ee. Visual sexual ha assmen concep ualisa ion was
associa ed wi h language skills (p=0.038) and s a us immig a ion law (p=0.037). The
endency o disag ee was ound in p o essionals wi hou he na ional ci izenship o wi h
basic language skills (1 EU language). Denudemen concep ualisa ion was associa ed
wi h he hos ing coun y and language skills (p=0.030, p=0.000, espec i ely).
P o essionals om Po ugal o wi h basic language skills (1 EU language) we e mo e
likely o ully disag ee. Abuse, ape and a icking concep ualisa ion was di e en
acco ding o he age o p o essionals (p=0.021). Olde p o essionals (> han 40 yea s old)
ended o ully disag ee. Fu he , sexual exploi a ion concep ualisa ion as a o m o
iolence was associa ed wi h hos ing coun y, ma i al s a us and language skills (p=0.002,
p=0.014 and p=0.031). The endency o ully disag ee was ound in p o essionals om
Mal a, he Ne he lands and Po ugal, ma ied o wi h good language skills (1 EU and 1
non-EU language).
Physical iolence: The concep o physical assaul wi hou pe manen consequences as a
o m o iolence was signi ican ly associa ed wi h hos ing coun y (p=0.015).
P o essionals wo king in Hunga y ended o ully disag ee.
Psychological iolence: Ve bal iolence was associa ed wi h ma i al s a us (p=0.042),
wi h single p o essionals disag eeing mo e han he a e age o esponden s. Con inemen
(indi idual le el) as a o m o iolence was associa ed wi h hos coun y (p=0.004), he
s a us o immig a ion (p=0.001), language skills (p=0.040) and he ac o being wo king
(p=0.005). P o essionals ha ended o ully disag ee we e om Belgium and he
Ne he lands, o wi hou he na ional ci izenship, wi h good language skills (2 EU
languages) o wi h a cu en job a he ime o he ques ionnai e.
Ha m ul cul u al p ac ices: Geni al mu ila ion concep ualisa ion was associa ed wi h
p o essionals’ sex (p=0.043), meaning ha male a he han emale p o essionals ended
o ully disag ee wi h i as an ac o iolence. Ea ly ma iage concep ualisa ion as a o m
o iolence was di e en acco ding o he hos ing coun y, ype o ecep ion acili y,
CHAPTER 4. Resul s
84
language skills o he ac o being wo king (p= 0.001, p=0.027, p=0.047 and p=0.031).
P o essionals wo king in Belgium, in open ecep ion acili ies o wi h good language
skills ended o ully disag ee.
Socio-economic iolence: Denial o oppo uni ies and se ices as a o m o SGBV we e
associa ed wi h sex (p=0.049), and emale p o essionals we e mo e likely o ully disag ee
ha i ep esen ed a kind o iolence.
CHAPTER 4. Resul s
85
Table 8: P o essionals – SGBV concep ualisa ion and socio-demog aphic cha ac e is ics (p- alues: Qui-squa e Tes and Fishe Tes ).
Dimensions o SGBV Concep
Socio-demog aphic
cha ac e is ics Sexual iolence Physical
Violence Psychological iolence Ha m Cul u al P ac ices
Socio-economic
PC 15 PC 3 PC 14 PC 1 PC 10 PC 8 PC 5 PC 12 PC 9 PC 13
PC 6 PC 16
PC 17 PC 7 PC 11 PC 2 PC 4
Hos coun y
0.363 0.142
0.030 0.516
0.002 0.015 0.687 0.388 0.180 0.004 0.391
0.556 0.594 0.001 0.725 0.081 0.473
Sex
0.451 0.072
0.736 0.078
0.360 0.256 0.503 0.509 0.333 0.932 1.000
0.502 0.043 0.884 0.345 0.049 0.498
Age
0.618 1.000
0.106 0.021
0.647 0.443 0.937 0.126 0.068 0.801 1.000
0.488 0.618 0.871 1.000 0.441 0.483
Ma i al s a us
0.753 0.133
0.734 0.451
0.014 0.469 0.512 0.381 0.042 0.053 0.533
0.773 0.609 0.189 0.489 0.616 0.500
S a us immig a ion Law
0.146 0.037
0.680 0.125
0.784 0.440 1.000 0.234 0.450 0.001 1.000
0.433 0.851 0.607 0.783 0.639 0.301
Type o ecep ion Cen e
wo king
0.851 0.551
0.397 0.497
0.073 0.487 0.829 0.189 0.833 0.282 0.630
0.763 0.734 0.027 0.373 0.229 0.300
Numbe o languages speaking
0.624 0.617
0.381 0.969
0.782 0.267 0.541 0.651 0.587 0.660 0.185
0.233 0.197 0.308 0.997 0.440 1.000
Numbe o languages needed a
wo k
0.012 0.038
0.000 0.131
0.031 0.377 0.470 0.434 0.387 0.040 0.078
0.970 0.706 0.047 0.454 0.615 0.519
Ac ually wo king in a
Recep ion cen e
0.192 1.000
0.125 0.425
0.817 0.649 0.762 0.064 1.000 0.005 1.000
0.674 0.192 0.031 0.443 0.568 0.685
Cu en ocupa ion
0.720 0.593
0.063 0.696
0.213 0.193 0.747 0.235 0.079 0.460 0.833
0.469 0.836 0.353 0.528 0.930 0.819
Signi ican p- alue p<0.05 bolded.
PC 15: Sexual innuendo; PC 3: Visual sexual ha assmen ; PC 14: Denudemen ; PC 1: abuse, ape and a icking; PC 10: Sexual exploi a ion; PC 8: Physical assaul wi hou pe manen consequences; PC 5: Physical
assaul wi h pe manen consequences; ; PC 12: Th ea and humilia ion; PC 9: Ve bal iolence; PC 13: Con inemen , indi idual le el; PC 6: Rela ional iolence; PC 16: Pa en al ela ional iolence; PC 17: Geni al
mu ila ion; PC 7: Ea ly ma iage; PC 11: Hono killing and maiming; PC 2: Denial o oppo uni ies and se ices; PC 4: Social exclusion and os acism.
CHAPTER 4. Resul s
92
Conside ing he Eu opean con ex o asylum ecep ion acili ies, he isk o SGBV is cons an
and he incidence high (6,45,80). Indeed, esea ch shows ha SGBV ulne abili y, especially
o women and gi ls, can inc ease due o inadequa e li ing condi ions, o e c owding a
ecep ion acili ies, lack o gende -sensi i e in asylum p ocedu es and a ecep ion acili ies
(6,79,80). In Eu ope, a coun y-wise dis ibu ion o SGBV incidence, in e ugees, AS and
undocumen ed mig an s, is s ill inexis en (4,45). Howe e , da a exis s on sexual agg ession
among young adul s, in Eu ope. In his s udy, he highes one-yea p e alence a e o emale
ic imisa ion o since he age o consen was ound in he Ne he lands, Ge many, Sweden,
Spain, Finland and G eece, in opposi ion wi h Belgium, Po ugal, Mal a and Hunga y. E en
hough his s udy assesses he a ailable e idence o SGBV and compa es s udies wi h a
di e en me hodology, sample composi ion and sexual agg ession de ini ion, i gi es a clea
pic u e o he dimension o SGBV p oblema ic in Eu ope (50).
Al hough e idence exis s ha SGBV can be p e en ed (9,107,134) e ec i e in e en ions
a e s ill no clea ly iden i ied (140). A sys ema ic e iew o e alua ions o p ima y p e en ion
s a egies o sexual iolence has concluded ha only h ee p og ams ha e p o ed o ha e a
posi i e impac on p e en ion (111). Indeed, i is clea ha p e en ion o SGBV should be
oo ed in a public heal h app oach (37,109,110,134). Al eady in 2003, he Eu opean Council
Di ec i e 2003/9/CE s a ed ha ic ims o ape and sexual iolence should ecei e speci ic
ea men , and ecep ion acili ies should be p epa ed o add ess hem (29,32). The ecas o
2013 (Eu opean Council Di ec i e 2013/33/EU) laying down s anda ds o he ecep ion o
applican s o in e na ional p o ec ion s a ed ha ecep ion and asylum acili ies should
implemen app op ia e measu es o p e en gende -based iolence including sexual assaul
and ha assmen (70). Recen ly, he Cen e o Disease Con ol has launched a echnical
package on p e en i e s a egies add essed o communi ies and s a es o educe he incidence
and consequences o sexual iolence (9). E en hough achie emen s ha e been made, a lack
o esea ch add essing he speci ic con ex o asylum ecep ion acili ies exis s (2,45).

CHAPTER 4. Resul s
93
Objec i es
Acknowledging ha SGBV is a p e en able public heal h p oblem, we aim o explo e
epo ed cases o SGBV desc ibed by esiden s – e ugees, AS and undocumen ed mig an s
– and p o essionals, li ing and wo king in Eu opean Asylum Recep ion Facili ies (EARF),
causes and p e en able measu es. The epo ed cases ake in o accoun he iolence ha was
wi nessed and/o expe ienced, iolence ha was commi ed among esiden s and iolence
commi ed by p o essionals owa ds esiden s o ice e sa. Fu he mo e, we in ended o
analyse po en ially p e en able measu es o SGBV desc ibed by he same popula ion. In his
sense, we a e commi ed o con ibu ing o he de elopmen o e idence-based SGBV
p e en ion measu es, adop ing a public heal h app oach.
Me hods
Concep ual model
Ou concep ual amewo k was ounded on a public heal h app oach o iolence (8) and he
Socio-ecological Model (75) o add ess SGBV. F om a public heal h pe spec i e, p ima y
p e en ion is conside ed he mos e ec i e way o p e en iolence ha ing a popula ion-
le el e ec (110). In his sense, o unde s and and p e en iolence ou undamen al s eps
should be aken in o accoun : (1) de ining he p oblem, (2) iden i ying he ac o s ha
inc ease he isk o iolence, (3) de eloping and es ing p e en ion s a egies, and (4)
dissemina ing and implemen ing b oadly (8). The Socio-ecological model uses a mul i-
ac o ial sys em o unde s and SGBV causes, consequences and subsequen p e en able
measu es (3,4). Indeed, his model ecognises ha e en s in highe social ecosys ems migh
in luence human de elopmen hus hei impac on e en s in lowe social ecosys ems (117).
I inco po a es ou -le el ac o s –indi idual, ela ionship, communi y and socie y – o
unde s and, mi iga e and p e en iolence (8,9,118). In his sense, we conside ha p ima y
p e en ion measu es should ocus on he con ibu ing ac o s o iolence a ou dynamic
le els – indi idual, ela ional, communi y and socie y. Th ough he analysis o epo ed cases
o SGBV, we will apply a ou -le el pe spec i e in o de o enhance p ima y p e en ion
s a egies o SGBV, in his speci ic con ex – EARF.
CHAPTER 4. Resul s
94
Senpe o o p ojec
A c oss-sec ional s udy was de eloped using da a om he KAP ques ionnai e o he
Senpe o o P ojec .
Popula ion, Sampling and Da a collec ion
As p e iously men ioned, ou s udy sample (n=600) comp ises esiden s ( e ugees, AS and
undocumen ed mig an s) and p o essionals (se ice and heal h ca e p o ide s) om EARF
in eigh Eu opean coun ies: Belgium, I eland, Mal a, G eece, Hunga y, Po ugal,
Ne he lands and Spain. A ull desc ip ion o he sampling s a egy can be ead in Chap e 3.
Da a we e collec ed h ough semi-s uc u ed in e iews based on a KAP ques ionnai e and
implemen ed by ained communi y esea che s (CR). The da a on coun y o esea ch, li ing
condi ions o esiden s wi hin EARF and awa eness o esponden s ega ding he exis ence
o p e en able measu es was collec ed h ough closed ques ions. Fo he da a on SGBV
epo ed cases, causes and p e en able measu es we used open ques ions.
Da a analysis
A i s da a cleaning, we excluded all he Spanish in e iews o he open ques ions pa
ega ding iolence expe iences, a i udes and p e en ion and esponse measu es. The no es
and ansc ip ions o he CR we e sca ce and inconsis en , and we doub ed whe he he
alidi y o he da a could be gua an eed (45). This b ough he o al o analysed in e iews
down o 562 in e iews: 375 wi h esiden s and 187 wi h p o essionals. Fo desc ip ions o
iolence exposu e, causes and p e en able measu es, we i s applied a amewo k analysis
echnique o ca ego ise ypes o iolence, pe pe a o s, ic ims and ela ions. Da a was
en e ed in o IBM® SPSS so wa e. Fo his p ocess h ee esea che s we e in ol ed, hey
p e iously ha e ag eed on a se o ca ego ies ha we e hen included in he da abase.
Quan i a i e da a om closed ques ions we e en e ed di ec ly in o IBM® SPSS so wa e
da abase. Da a analysis comp ises s a is ical es s –
Chi-squa e Tes
and
Fishe ’s exac es
–
, o analyse i a signi ican s a is ical associa ion exis s a he 5% signi icance le el. Fishe ’s
exac es was speci ically used o ables wi h expec ed cell equencies less han 5.
CHAPTER 4. Resul s
95
Resul s
Responden s included 562 pe sons: 375 (66.7%) esiden s and 187 (33.3%) p o essionals.
The majo i y o esponden s we e male (56.9%), aged om 19 o 39 yea s (67.3%). Residen s
mos ly had asylum seeke s a us (60.3%), while p o essionals had na ional ci izenship
(87.2%). The majo i y o esiden s li ed in an open ecep ion cen e (74.0%), mo e
speci ically in a oom (45.3%) o house/apa men (41.6%), wi h a common a ea (89.3%)
and a place o sleep (97.9%) o 2-4m
2
(43.0%). Residen s sha ed his place wi h 1 o 2 adul s
(40.5%), wi h whom hey had no ela ion (52.8%) while he majo i y sha ed he space wi h
hei own child en (65.1%). Fu he de ails o esiden s’ li ing condi ions in EARF a e
p esen ed in able 9.
Table 9: Li ing condi ions o Residen s a Eu opean asylum ecep ion acili ies.
Residen s
Li ing condi ions a
EARF
Female
N(%)
Male
N(%)
To al
N(%)
Type o accommoda ion
Room
62 (45.3) 84 (35.4) 146 (39.0)
House o apa men
57 (41.6) 85 (35.9) 142 (38.0)
Shel e
5 (3.6) 31 (13.1) 36 (9.6)
S udio o con aine
11 (8.1) 20 (8.4) 31 (8.3)
Ten o homeless
0 (0.0) 10 (4.2) 10 (2.6)
O he
2 (1.5) 7 (3.0) 9 (2.4)
Missing
- - 1
Place o sleep – size m
2
2-4
58 (43.0) 84 (36.5) 142 (38.9)
6-8
43 (31.9) 56 (24.3) 99 (27.1)
10-12
24 (17.8) 40 (17.4) 64 (17.5)
14-20
6 (4.4) 44 (19.1) 50 (13.7)
mo e 22
4 (3.0) 6 (2.6) 10 (2.7)
Missing
- - -
Wi h how many adul s do you sha e his space?
0
12 (14.3) 2 (1.2) 14 (5.5)
CHAPTER 4. Resul s
96
1 o 2
34 (40.5) 61 (35.9) 95 (37.4)
3 o 5
24 (28.6) 50 (29.4) 74 (29.1)
> 6
14 (16.7) 57 (33.5) 71 (27.9)
Missing
- - 121
Wha is hei ela ionship o you?
No ela ed
38 (52.8) 78 (48.4) 116 (49.8)
(Co-) esiden (s)
19 (26.4) 43 (26.7) 62 (26.6)
Pa ne . Family o
F iend(s)
15 (20.8) 40 (24.8) 55 (23.6)
Missing
- - 142
Wi h how many child en do you sha e his space?
0
19 (24.1) 63 (50.4) 82 (40.2)
1 o 2
44 (55.7) 41 (32.8) 85 (41.7)
3 o 5
12 (15.2) 10 (8.0) 22
10.8)
> 6
4 (5.1) 11 (8.8) 15 (7.3)
Missing
- - 171
Wha is hei ela ionship o you?
Own Child en
41 (65.1) 9 (15.5) 50 (41.3)
No-Rela ionship
17 (27.0) 38 (65.5) 55 (45.5)
Family o iends
2 (3.2) 7 (12.1) 9 (7.5)
(Co-) esiden s
3 (4.8) 4 (6.9) 7 (5.8)
Missing
- - 254
Repo ed cases o SGBV
Responden s we e asked o desc ibe cases o SGBV ha hey ecalled in he yea p io o he
in e iew ( able 10). In o al 698 cases we e desc ibed: esiden s epo ed 328 cases and
p o essionals epo ed 370. Rega ding he dis ibu ion o epo ed cases pe coun y,
esiden s om Belgium (67 cases) and I eland (67 cases) desc ibed he highes numbe o
cases. Fo p o essionals, esponden s om Mal a ha e epo ed he highes numbe o SGBV
cases (99 cases) ( able 11).
CHAPTER 4. Resul s
97
Conside ing he desc ip ion o ac s o iolence pe SGBV epo ed case ( able 12) was as
ollow: o esiden s, 50.1% epo ed one single case o SGBV, 25.6% epo ed wo cases,
9.6% epo ed h ee cases and only 2.1% o esiden s epo ed a ou h case o SGBV.
Residen s’ SGBV epo ing included 207 (40.6%) ac s o physical iolence, 192 (37.6%) ac s
o psychological iolence, 84 (16.5%) ac s o socio-economic iolence and 27 (5.3%) ac s
o sexual iolence. Fo p o essionals, we ound ha mo e han hal o esponden s epo ed
a i s and second case o SGBV when asked, 74.9% and 62.0%, espec i ely (see Table 10).
P o essionals’ SGBV epo ing included 259 (43.2%) ac s o physical iolence, 260 (43.3%)
ac s o psychological iolence, 43 (7.2%) ac s o socio-economic iolence and 38 (6.3%) ac s
o sexual iolence. Nei he o he wo g oups desc ibed ac s o ha m ul cul u al p ac ices. In
sum, om he 698 cases desc ibed, 1110 ac s o mul i- ypes o iolence we e included.
Table 10: SGBV Cases epo ed by esiden s and p o essionals and gende .
Case Repo N (%) NO Case Repo N (%) Case Repo N (%) NO Case Repo N (%)
Residen s P o essionals
Female Male To al Female Male To al Female Male To al Female Male To al
Fi s CASEª 71 (37,8) 117 (62,2) 188 (50.1) 66 (35,3) 121 (64,7) 187 (49.9) 73 (52,5) 66 (47,5)
140 (74.9)+ 32 (68,1) 15 (31,9) 47 (25.1)
Second
CASE* 36 (37,5) 60 (62,5) 96 (25.6) 98 (35,8) 176 (64,2) 274 (73.1) 66 (56,9) 50 (43,1)
116 (62.4) 39 (55,7) 31 (44,3) 70 (37,6))
Thi d
CASE** 14 (38,9) 22 (61,1) 36 (9.6) 123 (36,3) 216 63,7) 339 (90.4) 49 (58,3) 35 (41,7)
84 (44.9) 56 (55,4) 45 (44,6) 101 (54,6)
Fou h CASEª 4 (50,0) 4 (50,0) 8 (2.1) 133 (36,2) 243 (63,8) 367 (97.9) 19 (63,3) 11 (36,7)
30 (16.0) 86 (55,1) 70 (44,9) 156 (83,9))
To al cases
epo ed 328 370
ª No missing alues, o bo h g oups.
*Fo he desc ip ion o he second case, i e missing answe s o esiden s.
** Fo he desc ip ion o he second case, one missing answe o p o essionals.
+ One p o essional did no desc ibe gende .

CHAPTER 4. Resul s
98
Table 11: Numbe o epo ed cases o SGBV by esiden s and p o essionals, pe coun y.
Residen s P o essionals To al
Coun y N Repo ed
cases N Repo ed
cases N To al
cases
Belgium 61 67 32 76 93 143
G eece 36 27 30 60 66 87
Hunga y 68 35 21 66 89 101
I eland 63 67 32 55 95 122
Mal a 61 61 30 99 91 160
The Ne he lands 33 61 5 10 38 71
Po ugal 53 10 37 4 90 14
TOTAL 375 328 187 370 562 698
Table 12: Numbe o ac s o iolence epo ed by esiden s and p o essionals.
To al Ac s Violence
Residen s P o essionals To al Ac s
Female Male Ac s
N=510 (%)
Female Male Ac s
N=600 (%) N= 1110 (%)
Physical Violence 207 (40.6) 259 (43.2) 466 (42.0)
Singula non li e-
h ea hening 53 (25.6) 73 (35.3) 126 (24.7)
72 (27.8) 58 (22.5) 131* (21.8)
257 (23.2)
Mul iple non li e-
h ea hening 11 (5.3) 18 (8.7) 29 (5.7) 24 (9.3) 16 (6.2) 40 (6.7) 69 (6.2)
Singula li e-
h ea hening 13 (6.3) 23 (11.1) 36 (7.1) 34 (13.1) 25 (9.7) 59 (9.8) 95 (8.6)
Mul iple li e-
h ea hening 2 (1.0) 11 (5.3) 13 (2.5) 18 (6.9) 10 (3.9) 28 (4.7) 41 (3.7)
Killing 1 (0.5) 2 (1.0) 3 (0.6) 1 (0.4) 0 (0.0) 1 (0.2) 4 (0.4)
Psychological Violence 192 (37.6) 260 (43.3) 452 (40.7)
Ve bal iolence 34 (17.7) 43 (22.4) 77 (15.1) 46 (17.7) 43 (16.5) 89 (14.8) 166 (15.0)
Humilia ion 23 (12.0) 52 (27.1) 75 (14.7) 23 (8.8) 22 (8.5) 45 (7.5) 120 (10.8)
Th ea ening 9 (4.7) 20 (10.4) 29 (5.7) 46 (17.7) 39 (15.0) 85 (14.2) 114 (10.3)
Con inemen 1 (0.5) 3 (1.6) 4 (0.8) 3 (1.2) 3 (1.2) 6 (1.0) 10 (0.9)
Rela ional iolence 6 (3.1) 1 (0.5) 7 (1.4) 25 (9.6) 9 (3.5) 35* (5.8) 42 (3.8)
Sexual Violence 27 (5.3) 38 (6.3) 65 (5.9)
Sexual ha asmen 6 (22.2) 6 (22.2) 12 (2.4) 11 (28.9) 10 (26.3) 21 (3.5) 33 (3.0)
Sexual abuse 3 (11.1) 3 (11.1) 6 (1.2) 4 (10.5) 2 (5.3) 6 (1.0) 12 (1.1)
CHAPTER 4. Resul s
99
A emp o ape 1 (3.7) 0 (0.0) 1 (0.2) 0 (0.0) 0 (0.0) 0 (0.0) 1 (0.1)
Rape 0 (0.0) 3 (11.1) 3 (0.6) 2 (5.3) 1 (2.6) 4* (0.7) 7 (0.6)
Sexual exploi a ion 2 (7.4) 3 (11.1) 5 (1.0) 5 (13.2) 2 (5.3) 7 (1.2) 12 (1.1)
Ha m ull Cul u al
P ac ices 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
Socio-economic Violence 84 (16.5) 43 (7.2) 127 (11.4)
Disc imina ion 9 (10.7) 21 (25.0) 30 (5.9) 7 (16.3) 7 (16.3) 14 (2.3) 44 (4.0)
Re usal o assis ance 18 (21.4) 27 (32.1) 45 (8.6) 9 (20.9) 16 (37.2) 25 (4.2) 68 (6.1)
Social exclusion 4 (4.8) 3 (3.6) 7 (1.4) 0 (0.0) 2 (4.7) 2 (0.3) 9 (0.8)
Re usal o legal
p o ec ion 0 (0.0) 2 (2.4) 2 (0.4) 1 (2.3) 1 (2.3) 2 (0.3) 6 (0.5)
*One missing alue.
Causes o epo ed SGBV cases
To unde s and he causes ha igge SGBV in EARF, esponden s we e asked abou hei
assump ion o he main causes o epo ed iolence. F om ou o al sample, only 161 (42.9%)
esiden s and 133 p o essionals (71.1%) answe ed he ques ion (n o al=294). Table 13
p esen s he p esumed causes o SGBV as amed by he esponden s. Residen s epo ed as
main causes: us a ion and s ess (23.6%), di e en cul u al, e hnic backg ounds and
p ac ices (19.3%), asylum p ocedu es (13.7%), communica ion p oblems (9.9%) and bad
accommoda ion (8.7%). Fu he , male esiden s we e mo e likely o epo ha “s a
compe ence” was a cause o iolence (
p- alue
0.012). Causes o iolence mos ly men ioned
by p o essionals we e us a ion and s ess (37.6%), di e en cul u al and e hnic
backg ounds and p ac ices (20.3%), communica ion p oblems (11.3%). Fo his g oup, no
signi ican s a is ical associa ions in gende we e ound (
p- alue
0.501, Fishe es ).
CHAPTER 4. Resul s
100
Table 13: Causes o epo ed cases o SGBV.
Residen s P o essionals TOTAL
Causes o epo ed
SGBV
Female N
(%)
Male N
(%) p Female N
(%)
Male N
(%) p Resid N
(%)
P o s N
(%) p
Coping ( us a ion &
s ess managemen ) 14 (8.7) 24 (14.9) 0.709 30 (22.6) 20 (15.0) 0.369 38 (23.6) 50 (37.6) 0.008
Di e en cul u al/e hnic
backg ounds & p ac ices 15 (9.3) 16 (9.9) 0.310 14 (10.5) 13 (9.8) 0.831 31 (19.3) 27 (20.3) 0.884
Communica ion p oblem 8 (5.0) 8 (5.0) 0.426 8 (6.0) 7 (5.3) 1.000 16 (9.9) 15 (11.3) 0.849
Asylum p ocedu e
ela ed 5 (3.1) 17 (10.6) 0.102 1 (0.8) 4 (3.0) 0.179 22 (13.7) 5 (3.8) 0.004
Bad accommoda ion 8 (5.0) 6 (3.7) 0.252 5 (3.8) 1 (0.8) 0.218 14 (8.7) 6 (4.5) 0.170
Mul i ac o ial 1 (0.6) 8 (5.0) 0.088 4 (3.0) 5 (3.8) 0.732 9 (5.6) 9 (6.8) 0.808
Compe ence s a 0 (0.0) 9 (5.6) 0.012 7 (5.3) 5 (3.8) 0.774 9 (5.6) 12 (9.0) 0.363
Alcohol Abuse 1 (0.6) 1 (0.6) 1.000 2 (1.5) 5 (3.8) 0.246 2 (1.2) 7 (5.3) 0.084
Food 3 (1.9) 4 (2.5) 1.000 0 (0.0) 0 (0.0) - 7 (4.3) 0 (0.0) 0.017
I don' know 7 (4.3) 3 (1.9) 0.053 0 (0.0) 0 (0.0) - 10 (6.2) 0 (0.0) 0.002
O he s 2 (1.2) 1 (0.6) 0.563 1 (0.8) 1 (0.8) 1.000 3 (1.9) 2 (1.5) 1.000
Missing
- - - - - - 214 54 -
TOTAL 64 (39.8) 97 (60.2) 0.013 72 (54.1) 61 (45.9) 0.501 375
(66.7)
187
(33.3) 0.000
Bolded signi ican p- alue <0.05.
P e en able Measu es o SGBV
Acco ding o esponden s, 73.6% o epo ed cases o SGBV could be p e en ed (26.4%
answe ed i could no be p e en ed; 225 pe sons did no answe he ques ion). F om he
esponden s ha belie ed ha his iolence could be p e en ed he majo i y we e esiden s
(66.9%,
p- alue
0.000). Table 14 desc ibes po en ially p e en able measu es sugges ed o
hese cases o SGBV. S a is ical es s conduc ed o explo e possible associa ions ega ding
p e en able measu es desc ibed by he g oups o esiden s and p o essionals indica ed no
di e ences (
p- alue
0.226, Fishe es ). Rega ding s a is ical associa ions by gende , o bo h
g oups no di e ences we e ound
(p- alue
0.940,
p- alue
0.944, espec i ely).
CHAPTER 4. Resul s
101
Table 14: Po en ially p e en able measu es o epo ed cases o SGBV, by esiden s and p o essionals.
Residen s P o essionals To al
Female
N (%)
Male
N (%) p
Female
N (%)
Male
N (%) p
Resid
N (%)
P o s
N (%) p
Imp o ed SGBV
/In e en ion measu es 16 (28.6) 29 (33.3) 1.000 13 (26.0) 8 (22.9) 0.803
45 (31.5) 21 (24.7) 0.293
Imp o ed accomoda ion
& li ing condi ions 12 (21.4) 19 (21.8) 1.000 8 (16.0) 5 (14.3) 1.000
31 (21.7) 13 (15.3) 0.232
Imp o ed compe ence
s a / communica ion
wi h esiden s
8 (14.3) 15 (17.2) 0.653 12 (24.0) 12 (34.3) 0.335
23 (16.1) 24 (28.2) 0.042
Coping ( us a ion &
s ess managemen ) 6 (10.7) 9 (10.3) 1.000 6 (12.0) 4 (11.4) 1.000
15 (10.5) 10 (11.8) 0.829
Imp o ed asylum
p ocedu e 7 (12.5) 5 (5.7) 0.217 3 (6.0) 1 (2.9) 0.640
12 (8.4) 4 (4.7) 0.606
In e cul u al espec &
ole ance 4 (7.1) 5 (5.7) 0.737 3 (6.0) 3 (8.6) 0.687
9 (6.3) 6 (7.1) 1.000
Imp o ed
communica ion be ween
esiden s
2 (3.6) 3 (3.4) 1.000 1 (2.0) 1 (2.9) 1.000
5 (3.5) 2 (2.4) 0.714
O he 1 (1.8) 1 (1.1) 1.000 4 (8.0) 1 (2.9) 0.644
2 (1.4) 5 (5.9) 0.106
I don' know 0 (0.0) 1 (1.1) 1.000 0 (0.0) 0 (0.0) - 1 (0.7) 0 (0.0) 1.000
Missing alues - - -- - - - 232 102 -
TOTAL answe s 56 (39.2) 87 (60.8) 0.940 50 (58.8) 35 (41.2) 0.944
375 (66.7) 187 (33.3) 0.226
Bolded signi ican p- alue <0.05.
In addi ion, ou esponden s we e asked abou he exis ence o p e en able measu es in he
asylum ecep ion acili y whe e hey li ed o wo ked, and also in hei hos ing coun y.
Rega ding exis ing p e en able measu es a asylum ecep ion acili ies, he majo i y o
esiden s (58.3%) we e no awa e o exis ing p e en i e measu es. Fo p o essionals, 65.0%
we e awa e o exis ing p e en i e measu es in he asylum acili y. Conside ing exis ing
p e en able measu es a coun y-le el, he majo i y o esiden s we e no awa e o exis ing
p e en able measu es (72.4%). While 68.8% o p o essionals we e awa e o p e en able
measu es in hei coun y ( able 15). Fu he mo e, esiden s who epo ed he exis ence o
p e en able measu es also epo ed ha he measu es we e e ec i e (68.9%). The same esul
CHAPTER 4. Resul s
108
accoun he li ing condi ions o ou esponden s and he iden i ica ion o i as po en ial
p e en i e measu es, we belie e ha speci ic measu es should be conside ed a p io i y. Big
s eps ha e been aken in his ma e , and in 2016, he Eu opean Asylum Suppo O ice
(EASO) eleased he EASO guidance on ecep ion condi ions: ope a ional s anda ds and
indica o s. (143). This guidance b ings clea s anda ds ha should be p esen a asylum
ecep ion acili ies, and co esponding indica o s o e alua e he li ing condi ions. E en
hough, imp o emen s o EU Di ec i es and EASO guidelines ha e been done, we belie e
ha he ecen “ e ugee c isis” has caused a s ain in Eu opean ecep ion acili ies –
inc easing poo li ing condi ions, o e c owding and lack o p i acy – and a cons an and
high isk o SGBV ic imisa ion and/o pe pe a ion s ill exis s (6,80).
Th ough ou esul s, signi ican associa ions eme ged o esiden s o p o essionals and he
desc ip ion o he speci ic causes o epo ed SGBV and po en ially p e en able measu es.
We highligh he impo ance o legal s a us, asylum- ela ed p ocedu es, and unequal powe
ela ions, as isk ac o s o SGBV ic imisa ion and/o pe pe a ion. Fu he mo e, o in ol e
p o essionals and esiden s o EARF as ac i e s akeholde s, when de ining and implemen ing
SGBV p e en ion measu es, is p ima y. Also, aking in o accoun ha a high incidence o
pe pe a o s a e s a , gua ds o olun ee s (45,80), i is u gen o p omo e SGBV awa eness
and educa ion, by p omo ing compulso y aining on p e en ion and esponse policies,
a ge ing all ulne able g oups.
Addi ionally, and ega dless ha e idence has shown ha iolence can be p e en ed
(8,9,134), we s ill ha e esiden s and p o essionals (14/562) ha belie e ha his is no
possible. To ensu e ha communi ies a isk can be p o ec ed om being ic ims and/o
pe pe a o s o SGBV i is essen ial o engage wi h p o essionals and esiden s while de ining
p e en i e measu es.
E en hough signi ican and ele an indings a ise om ou s udy, we highligh he
impo ance o acknowledging limi a ions o ou esea ch. Fi s , we canno exclude ha he
communi y esea che s conduc ing he in e iews could ha e had a di e en
concep ualisa ion o SGBV despi e he s anda dised aining. Secondly, we ha e no epo ed
cases o ha m ul cul u al p ac ices, wha can be ela ed wi h he e idence ha his ype o

CHAPTER 4. Resul s
109
iolence is a ely disclosed due o sensi i i y and social aboos wi hin an asylum p ocedu e
(141). Ano he ele an bias is ela ed o he disclosu e o iolence. We belie e ha ou
esul s unde es ima e he eali y o SGBV in EARF. E en hough con iden iali y was
gua an eed du ing he in e iew, esiden s migh ha e ea ed i could in luence hei asylum
p ocedu e. Ye , in some big asylum ecep ion acili ies whe e communi ies wi h honou ules
we e esiding, i was epo ed o us la e ha esiden s discou aged o he s om pa icipa ing
in he s udy, men ioning po en ial s igma and/o communi y consequences (45). Fo
p o essionals, hey assumed no o da e o speak openly, e en i hey had supe io consen .
Finally, we belie e ha u he esea ch is needed add essing he speci ic con ex o asylum
ecep ion acili ies, and e alua ing SGBV p e en i e s a egies (9,45,111). The e is an
impe a i e o unde s and he impac o SGBV p e en i e s a egies and wha wo ks bes ,
acco ding o he a ge popula ion and speci ici y o he social con ex . P io esea ch has
al eady sus ained he need o sys ema ic e alua ion esea ch on p e en ion and managemen
o all kinds o gende -based iolence (107,111,140).
Conclusion
Ou esea ch shows he complexi y o add essing SGBV in EARF con ex , wi h a high
epo ing o mul i- ypes o iolence and a mul i-causali y o SGBV. Residen s and
p o essionals ha e iden i ied po en ial causes ha igge SGBV. Bo h g oups e e ha he
majo i y o epo ed cases o SGBV could be p e en ed wi h e ec i e p e en i e measu es
adap ed o EARF.
A e lec ion on cu en p e en i e s a egies, e idence-based on causes o SGBV, is u gen ly
equi ed o educe SGBV incidence. Conside ing he con ex o EARF, we belie e he e is a
window o oppo uni y o implemen in eg a ed p e en i e s a egies o such a complex and
highly ulne able popula ion. We belie e ha esiden s and p o essionals should be
conside ed ac i e s akeholde in de ining SGBV p e en i e measu es. We highligh he
impo ance o gende -sensi i i y and equi y in asylum p ocedu es and adequa e
accommoda ion acili ies o p omo e an SGBV ee en i onmen .
CHAPTER 4. Resul s
110
Taking in o conside a ion he ecen “ e ugee c isis” in Eu ope, we insis on he impo ance
o imp o ing p e en able s a egies and policies o mi iga e SGBV. E en hough, SGBV
incidence da a is lacking, his is due o he su i o s a oiding disclosu e o hei expe ience
unless isible and se e e heal h consequences a ise (6). Re ugees, AS and undocumen ed
mig an s a e ic ims o SGBV, wi h men and women being ulne able (45), and he ac ha
his iolence is commi ed in EARF equi es an u gen call o ac ion (80).
CHAPTER 4. Resul s
111
4.3 S udy 3 – Can we p edic sexual and gende -based iolence in Eu opean asylum
acili ies? Resul s om a machine lea ning app oach.
In oduc ion
In he con ex o ( o ced) mig a ion, sexual and gende -based iolence (SGBV) is conside ed
a majo public heal h p oblem and a iola ion o basic human igh s (4,38). E idence sugges s
ha e ugees, asylum seeke s (AS) and undocumen ed mig an s a e a ulne able g oup o
mul iple ypes and o ms o iolence (4,6,38,44,51,58), including a high ulne abili y o
SGBV (38,45,144). In a s udy conduc ed in Se ia, ou o 992 e ugees o mig an s, 27% had
expe ienced any iolen ac s du ing hei passage including sexual iolence (53). Ano he
s udy conduc ed in UK clinics obse ed a 44.2% p e alence o epo ed sexual iolence in a
con enience sample o 43 asylum seeke s (48.8% emale) compa ed o no epo s by a
sample o 43 B i ish pa ien s (age- and gende -ma ched sample) (57). In he same s udy,
esul s disagg ega ed by gende epo ed 76.2% p e alence o sexual iolence among emale
AS and 13.6% among male AS, sugges ing ha women AS a e mo e exposed o sexual
iolence when compa ed o men AS (57). A sys ema ic e iew and me a-analysis ound ha
app oxima ely one in i e e ugees o displaced women in complex humani a ian se ings
expe ienced sexual iolence (55). Ano he s udy on A ican mig an s in Ge many shows a
p e alence o 16% o women epo ing a his o y o sexual iolence and men epo ing 6%,
o ocus ha he au ho s conside he possibili y o sexual iolence being unde epo ed due
o ea and s igma (56).
While add essing SGBV owa ds e ugees, AS and undocumen ed mig an s li e a u e
appoin s ha pe pe a o s a e o en people in (s a e) au ho i y, agen s, soldie s o assailan s
wi h powe ela ions wi h po en ial ic ims (2,53,58). In he speci ic con ex o Eu opean
asylum ecep ion acili ies (EARF), p o essionals a e bo h a isk o ic imisa ion and/o
pe pe a ion (2). E idence sugges s ha when p o essionals a e ic ims, he commi ed
iolence ends o be emo ional ac s, and when hey a e pe pe a o s hey end o commi
socio-economic iolence, compa ed o asylum-seeke s (45). A ecen s udy ound ha ou o
270 epo s o iolence among mig an s and e ugees, 52% we e commi ed by S a e
au ho i ies (53).
CHAPTER 4. Resul s
112
A public heal h app oach, based on he Socio-ecological model (75) a e explana o y
pe spec i es o bo h SGBV ic imisa ion and/o pe pe a ion (37). Con ibu ing ac o s o
SGBV a e he esul o dynamics and in e ac ions o indi idual, ela ional, communi y and
socie al ac o s. The unde s anding o igge ac o s o SGBV among ulne able popula ions
s ill lacks accu acy (2,45). Howe e , li e a u e iden i ies ha he es ic ed legal s a us
exposes mig an s, e ugees and asylum seeke s o iolence, including SGBV (45,91,96). The
mig a ion p ocess, o ced o olun a y is ano he po en ial isk ac o o iolence. In a s udy
conduc ed among Somali e ugee’s women li ing in Sweden, he majo i y o esponden s
ha e ela ed hei leeing wi h wa - ela ed iolence, and he subsequen ulne abili y o di ec
o ms o iolence and iola ion o hei sexual heal h and igh s (60). The au ho s enhance
he isk o p e-mig a ion ic imisa ion o consequences on heal h, namely sexual and
ep oduc i e heal h. A s udy among sub-Saha an mig an s in Mo occo, du ing hei
mig a ion jou ney, obse ed hei high ulne abili y o iolence: 90% o he esponden s ha e
epo ed mul iple ac s o iolence, ou o hose, 45% epo ed sexual iolence (Keygnae e
al., 2014).
Acknowledging ha SGBV is p e en able (9,107,111), and aking in o conside a ion he
p e ailing Eu opean Di ec i e 2013/33/EU (70), p o ec i e measu es ha e been de ined o
add ess SGBV, in he con ex o asylum ecep ion acili ies. This Di ec i e decla es ha all
Membe S a es should ensu e app op ia e measu es o p e en SGBV wi hin he acili ies,
and ein o ces he need o p o iding access o medical and psychological ca e. E en hough
p og ess has been made o p o ec ulne able popula ions om SGBV, ecen epo s on he
cu en Eu opean e ugee c isis, upholds ha mig an s, e ugees and AS a e in high need o
p o ec ion, and a lack o go e nmen al and humani a ian esponse exis s (6,53,80).
Objec i es
Ou s udy in ends o explo e whe he a pa e n/ end o SGBV ic imisa ion in Eu opean
asylum con ex can be iden i ied in o de o de e mine u u e ou comes. Ye , o iden i y socio-
demog aphic cha ac e is ics ha migh p edic SGBV ic imisa ion in esiden s and
p o essionals.
CHAPTER 4. Resul s
113
Me hods
Concep ual amewo k: A socio-ecological pe spec i e and public heal h app oach
o
p e en ion
SGBV has been ecognised as a complex public heal h issue (3,10) and conside ed a public
heal h epidemic o he cu en imes (37,145). A socio-ecological app oach is desc ibed in
he li e a u e as a comp ehensi e model o SGBV (4,75). This model assumes ha SGBV is
he esul o a dynamic in e ac ion be ween ou -le el heal h de e minan s – indi idual,
ela ional, communi y and socie y (4). The combina ion o hese le els igge s he pa e ns
o SGBV. In 1996, he 49 h Wo ld Heal h Assembly ecognised iolence as a leading public
heal h p oblem. A public heal h app oach o unde s anding and p e en ing iolence has been
de ined upon ou undamen al s eps: (1) de ine he p oblem, (2) o iden i y he ac o s ha
inc ease he isk o iolence, (3) o de elop and o es p e en ion s a egies, and (4) o
dissemina e and implemen b oadly (8). A socio-ecological pe spec i e and public heal h
app oach o p e en ion will be he baseline o iden i y pa e ns o isk ac o s o SGBV among
esiden s and p o essionals om EARF.
S udy Design, Sample and Da a collec ion
Ou esea ch consis s o a c oss-sec ional s udy using da a om he Senpe o o P ojec
de eloped in eigh Eu opean coun ies (Belgium, G eece, Hunga y, I eland, Mal a, The
Ne he lands, Po ugal and Spain). A comple e desc ip ion o Senpe o o p ojec , sampling
and da a collec ion is included in Chap e 3 – Me hods.
Da a-analysis
The da a-d i en modelling sugges ed by Shmueli (122), and al eady desc ibed in poin 3.6
o he Me hods Chap e , was aken in o accoun o he p epa a ion o ou da a, and he
explo a o y analysis. An in-dep h desc ip ion o he da a analysis is p esen ed in he Chap e
3 – Me hods, wi h complemen a y in o ma ion in annex I and II.

CHAPTER 4. Resul s
114
Resul s
A o al o 562 esponden s, 375 esiden s and 187 p o essionals, ha e epo ed 698 SGBV
cases, om which 138 cases included pe sonal ic imisa ion, in he las 12 mon hs p io o
he ques ionnai e. A desc ip ion o he numbe o SGBV epo ed by coun y o esea ch is
p esen ed in able 17.
Table 17: SGBV epo ed cases, 12 mon hs p io o he in e iews, pe coun y o esea ch.
Residen s P o essionals To al
Coun ies
N Repo ed cases N Repo ed cases N To al cases
Belgium 61 67 32 76 93 143
G eece 36 27 30 60 66 87
Hunga y 68 35 21 66 89 101
I eland 63 67 32 55 95 122
Mal a 61 61 30 99 91 160
The Ne he lands 33 61 5 10 38 71
Po ugal 53 10 37 4 90 14
To al 375 328 187 370 562 698
Following he da a-d i en modelling desc ibed by Shmueli (122), and al eady explained in
he Chap e 3 – Me hods, we p esen in able 18 he lis o selec ed independen a iables
acco ding o he impo ance o Random o es algo i hm (see annex II o u he desc ip ion
on he me hod) o p edic he dependen a iable “we e you a ic im”. A highe alue
ep esen s highe p obabili y o p edic ing SGBV ic imisa ion, while a lowe alue
ep esen s a lowe p obabili y.
CHAPTER 4. Resul s
115
Table 18: Bes ea u e subse and ea u e impo ance, based on he impo ance o Random o es algo i hm.
Desc ip ion P edic i e
Impo ance
Wi h how many pe sons do you ha e o sha e hese sani a y acili ies? 1.000000
Can you speci y which kind o accommoda ion you a e li ing in? 0.367218
Wha is you ac ual s a us acco ding o immig a ion law? 0.332777
Wha is you age in yea s? 0.321193
Wha is you cu en occupa ion in his acili y? 0.315818
Wha is hei app oxima e age? 0.298385
Wha is hei sex? 0.265571
Wi h how many adul s (equal o olde han 18) do you sha e his space? 0.249937
Wi h how many child en (younge han 18) do you sha e his space? 0.247091
I yes, a e hey li ing wi h you? 0.242666
Wha is you ac ual ma i al s a us? 0.242473
I yes, wha is he size o his space? 0.236184
Wha kind o sani a y acili ies do you ha e in his accommoda ion? 0.208560
In wha ype o ecep ion/asylum acili y a e you li ing? 0.197513
Wha is hei ela ionship o you? 0.176883
Do you ha e any child en? 0.151167
Wha is hei sex? 0.130415
A e you cu en ly li ing wi h a pa ne ? 0.102410
Figu e 6 p esen s a g aph wi h he same subse o a iables’ impo ance based on Random
Fo es s algo i hm. Ou modelling sugges ha he mos impo an a iables o p edic SGBV
ic imisa ion a e: he numbe o pe sons wi h whom esiden s and p o essionals sha ed
sani a y acili ies; he ype o accommoda ion whe e pa icipan s we e li ing (namely: oom,
house/apa men , shel e , s udio/con aine , en o homeless);
and he s a us acco ding o
immig a ion law.
CHAPTER 4. Resul s
116
Figu e 7 p esen s a g aph wi h he i ing o ou model, using Random Fo es algo i hm. The
AUC alue is 0.82+/- 0.13. This g aph ep esen s he p edic i e powe adjus men o he
a iables desc ibed in Table 18. Ou model has high sensi i i y ( ue posi i e) and high
speci ici y ( alse posi i e).
Figu e 7: Random Fo es pe o mance wi h op imised hype -pa ame e s.
Figu e 6: Bes ea u e subse and ea u e impo ance based on Random Fo es s algo i hm.
CHAPTER 4. Resul s
117
Discussion
Ou esea ch explo es an SGBV ic imisa ion p edic i e model o esiden s and
p o essionals o EARF. We conside ed a se o socio-demog aphic cha ac e is ics, such as
li ing condi ions, ype o ecep ion acili y and speci ic legal s a us, as possible p edic i e
ac o s ha can be associa ed wi h SGBV ulne abili y.
In he con ex o EARF, ou indings highligh he impo ance o li ing condi ions o p edic
SGBV ic imisa ion. Acco dingly, he numbe o people wi h whom sani a y acili ies a e
sha ed and hei age, he kind o accommoda ion, he immig a ion s a us, he kind o
occupa ion wi hin he acili y and he age o pa icipan s we e impo an a iables o p edic
i a esiden o a p o essional we e SGBV ic ims. Conside ing he ac ha SGBV is
p e en able (4,109), he Eu opean Union has de ined speci ic di ec i es o p o ec ulne able
communi ies. The p e ailing Eu opean Di ec i e 2013/33/EU laying down s anda ds o he
ecep ion o applican s o in e na ional p o ec ion ( ecas ) s a ed ha adequa e li ing
condi ions should be ensu ed o ha e a digni ied li e, ensu ing equal ea men amongs all
applican s (70). E en hough he ecas o EU 2013/33/EU is conside ed a s ep o wa d o
imp o ing and ha monising ecep ion condi ion o ulne able communi ies h oughou
Eu opean coun ies (142), i lacks clea and speci ic o ien a ion in wha is de ined as
“adequa e li ing condi ions”. Mo eo e , his Di ec i e s ill lacks consis en implemen a ion
h oughou he coun ies (80,146,147). Conside ing he ac ha he numbe o e ugees, AS
and undocumen ed mig an s a e inc easing in Eu ope (17), we belie e mo e mus be done.
The EU encou ages coun y membe s o go o wa d he di ec i e and adap i o he eali y
o each coun y. Se e al epo s show he need o implemen ing measu es o ensu e e ec i e
p o ec ion (53,62,80,146,148,149).
The dynamics be ween mig an s immig a ion s a us, li ing condi ions and SV is complex
(32). Mo eo e , we specula e ha he in e ac ion be ween li ing condi ions and SGBV can
induce ulne abili y owa ds no only esiden s bu also p o essionals, li ing and wo king in
EARF. This esul can be pe cei ed by ou modelling and he impo ance o he a iables o
p edic ic imisa ion, namely, li ing condi ions and s a us acco ding o immig a ion law.
CHAPTER 5. Discussion and Conclusions
124
we e ound in esiden s and p o essionals o EARF. To conside ha de ini ions o iolence
ha e e ol e h ough mul iple a ia ions acco ding o he ield and he ange o o ms o
iolence encompassed (36). Gi en his, we belie e a common SGBV concep ualisa ion
should be conside ed while add essing p e en i e measu es. The equi emen o de eloping
in o ma ion, educa ion and communica ion (IEC) in e en ions add essing SGBV has
al eady been acknowledged by UNHCR (2003). We belie e ou esul s s and o he u gen
need o IEC in e en ions, add essing wha is, o wha is no an SGBV ac , in he con ex o
EARF.
Fo bo h g oups di e ences in SGBV concep ualisa ion we e ound based on speci ic socio-
demog aphic cha ac e is ics such as he hos coun y, sex, age and ma i al s a us. Fo
esiden s, speci ic di e ences ela ed o he ime o a i al o Eu ope/hos coun y and ype
o accommoda ion could be ound, while o p o essionals hese di e ences we e linked o
he legal s a us o p o essionals and hei educa ion skills.
As o gende , ou esul s e oke no di e ences in SGBV concep ualisa ion. Mo eo e , he
ac ha a iolen ac is di ec ed o a gi l/woman o a boy/man is equally conside ed iolence.
Howe e , mo ing om SGBV concep ualisa ion o speci ic ypes o SGBV di e ences a ise.
When conduc ing associa ion es s be ween ypes o SGBV (PCs) and he gende o ou
esponden s we ound signi ican associa ions. A mo e in-dep h analysis sugges s male
esiden s end o disag ee ha honou killing and maiming is an SGBV ac when compa ed
wi h he mean a e age o ou esponden s. Mo eo e , male p o essionals disag ee wi h
geni al mu ila ion as a o m o SGBV, and emale p o essionals end o disag ee wi h he
denial o oppo uni ies as a o m o SGBV.
Ano he ele an associa ion was ound be ween age and a speci ic o m o SGBV. Resul s
om ou i s s udy, including he eigh coun ies o esea ch ound ha p o essionals aged
abo e 40 ended o disag ee ha “abuse, ape and a icking” is a o m o SGBV. This
associa ion is pa icula ly sc eaming o ac ion, once we assis o p o essionals wo king wi h
pe sons, al eady in a ulne able si ua ion, and assuming a beha iou legally punishable by
law is accep able. Conside ing ha p o essionals play an impo an ole in SGBV p e en ion,
and he ac ha hey a e in a p i ileged posi ion o mi iga e SGBV, we belie e ha ou

CHAPTER 5. Discussion and Conclusions
125
esul s a e o pa icula ele ance. F om one side we assis o p o essionals ha ing a b oade
SGBV concep ualisa ion when compa ed wi h esiden s. Howe e , p o essionals aged abo e
40, do no conside abuse, ape and a icking as a o m o SGBV. In his sense, we belie e
he e is a need o a s ic sc eening when engaging p o essionals o wo k in EARF and
con inuous sensi isa ion and aining on SGBV. Ou esul s a e aligned wi h p e ious
e idence epo ing he equi emen o heal hca e wo ke s’ egula aining (129), in eg a ed
and widesp ead p e en i e and esponse measu es (45). Fu he mo e, p o essionals wo king
wi h mig an s and e ugees ha e been iden i ied as po en ial pe pe a o s o SGBV. A ecen
s udy on mig an s and e ugees a elling h ough Balkan coun ies o No he n Eu ope s a ed
hal o he pe pe a o s o epo ed iolence we e s a e au ho i ies (53). Fu he mo e, in
EARF con ex , p o essionals ha e been iden i ied as po en ial pe pe a o s o SGBV,
especially socio-economic iolence (45).
Speci ic ypes o SGBV no being ecognised as a iolen ac is o majo impo an while
add essing p e en i e measu es in EARF. Residen s and p o essionals mus ha e a comple e
and equal knowledge ega ding di e en ypes o SGBV o a oid being ic ims and/o
agg esso s. Placing SGBV in a public heal h pe spec i e, we can assume SGBV
concep ualisa ion is he baseline o p ima y p e en ion (44). Fu he mo e, signi ican
associa ion wi h socio-demog aphic cha ac e is ics ha e a isen om ou esul s. We call o
an u gen ac ion om he di e en s akeholde s o inc ease he knowledge on SGBV o
esiden s and p o essionals, based on IEC in e en ions, as he baseline o p e en iolence
be o e i occu s.
Pe inen esea ch opic ega ds he po en ial associa ion be ween SGBV concep ualisa ion
and case disclosu e. Mo eo e , i is o u mos impo ance o ha e a clea and in-dep h
unde s anding o p o essionals’ SGBV concep ualisa ion. The ac ha p o essionals migh
pe pe ua e SGBV ac s, and exe cise a highe powe ela ion owa ds esiden s, ep esen s a
call o ac ion. We challenge esea che s o go beyond he unde s anding o p o essionals’
SGBV concep ualisa ion and o conside he in luence o i wi h he po en ial pe pe ua ion o
iolence. Ano he ele an aspec o conside in he u u e ega ds he e alua ion o p ima y
p e en i e measu es, and speci ically he ocus on p omo ing and implemen ing a
CHAPTER 5. Discussion and Conclusions
126
widesp ead SGBV concep ualisa ion among esiden s and p o essionals. I we each a le el
whe e p o essionals and esiden s ha e simila SGBV concep ualisa ion, will we s ill wi ness
high le els o SGBV?
E en hough ele an indings we e desc ibed i is impo an o acknowledge po en ial
limi a ions. The ac ha we use seconda y da a o Senpe o o p ojec is by i sel a limi a ion,
once he sampling me hods and ques ionnai es we e p e iously decided and we we e no
included in he p ocess. By he ac ha an in-dep h s udy o he p ojec was conduc ed we
belie e o ha e educed his limi a ion. Mo eo e , Senpe o o p ojec applied mul i- ypes o
sampling me hods, as andom and ep esen a i e sampling was no possible in all coun ies.
E en hough ou esul s canno be gene alised, we belie e i can be ans e able o simila
popula ions in compa able con ex s, in a sense ha a b oad SGBV concep ualisa ion is
p esen ed in ou esea ch – unde s anding e ugees, AS and undocumen ed mig an s’
pe spec i e and also p o essional’s pe spec i e. Speci ically ela ed wi h SGBV
concep ualisa ion, we canno exclude ha communi y esea che s conduc ing he in e iews
du ing he implemen a ion o Senpe o o p ojec , could ha e had a di e en SGBV
concep ualisa ion. This limi a ion was o e come by he implemen a ion o a s anda dised
aining.
S epping ou o EARF, we belie e i would be pe inen o compa e SGBV concep ualisa ion
be ween mig an s and hos ing popula ion, once public heal h policies should be adap ed o
he cul u al and s uc u al con ex . Mo eo e , i is impo an o conside he challenge o
ha ing e ugees, AS and undocumen ed mig an s wi h di e en SGBV concep ualisa ion
“in eg a ed” in Eu opean coun ies, especially i hey ha e a na ow concep . Acco dingly,
we belie e mig an s migh be exposed o highe ulne abili y o bo h ic imisa ion and
pe pe a ion. Conside ing he ecen mig a ion wa e o Eu opean coun ies, i u ges o
add ess his issue. SGBV concep ualisa ion needs o be add essed equally, no only o
mig an s and p o essionals, bu also o hos ing popula ions. Wha is, o wha is no an SGBV
ac should no di e acco ding o a mig a ion s a us. By no doing i , we belie e Eu opean
coun ies and i s ep esen a i es migh be inc easing mig an s’ ulne abili y and inducing
obs acles o hei in eg a ion.
CHAPTER 5. Discussion and Conclusions
127
Repo ed cases and causes o SGBV
Mo ing om p ima y p e en ion o seconda y and e ia y p e en ion, ou esea ch has
shown ha SGBV is a eali y in EARF, and i u ges o implemen e ec i e p e en i e
measu es, o mi iga e he p oblem. Ou esponden s ha e desc ibed 698 cases o SGBV, wi h
esiden s desc ibing 328 and p o essionals desc ibing 370 SGBV cases. Toge he he cases
comp ise 1110 ac s o mul i- ypes o iolence. Rega ding cases o ha m ul cul u al p ac ices,
no cases we e desc ibed, wha can be ela ed wi h he exis en e idence a i ming his ype
o iolence is a ely disclosed due o sensi i i y and social aboos wi hin an asylum p ocedu e
(Banda & Agyapong, 2016).
Ou indings indica e ha wi hin EARF, we assis o a mul i-causali y o SGBV, wi h
esiden s and p o essionals being able o iden i y po en ial causes. The main assumed causes
o SGBV desc ibed by esiden s and p o essionals, wi hin he epo ed cases we e he lack
o coping skills, us a ion and s ess, and di e ences ela ed o cul u al backg ound.
Mo eo e , esiden s ha e iden i ied asylum p ocedu e as a po en ial cause o epo ed cases
o SGBV. In ecen indings wi hin asylum ecep ion cen es in Ge many, he cu en ime-
consuming asylum p ocedu e leads o o e c owding and inadequa e li ing condi ions, which
inc eases emale esiden s’ ulne abili y o iolence (80). Women and child en ha e been
sexually abused in ecep ion cen es in Ge many, a epo iden i ies li ing condi ions, as
o e c owding and lack o space and p i acy, as inc easing condi ions o pe pe a e iolen
ac s (80). We belie e ou esul s highligh ha asylum p ocedu e should be conside ed as a
isk ac o o SGBV. No wi hs anding, a gende -sensi i e and equi able asylum p ocedu e
is u gen ly needed. The in- igo Di ec i e 2013/32/EU (71) ecognises he need o a gende -
sensi i e asylum p ocedu e, ensu ing s a a e also awa e o gende -speci ic ulne abili ies.
Con inuous aining o he s a and pe sons in au ho i y should be acknowledged, in o de
o ul il hei obliga ions while implemen ing his Di ec i e. P o essionals wo king in EARF
ha e shown a b oad SGBV concep ualisa ion aligned wi h UNHCR de ini ion (4), hey ha e
epo ed a high numbe o SGBV cases happening in EARF, ha e been able o iden i y
po en ial causes o SGBV and po en ially p e en able measu es. Howe e , e idence exis s
ha p o essionals, s a e au ho i ies and pe sons in powe a e s ill iden i ied as SGBV
CHAPTER 5. Discussion and Conclusions
128
pe pe a o s owa ds e ugees, AS and undocumen ed mig an s (45,53,58,62,155). Gi en
his, we ein o ce ha i is essen ial o implemen a s ic p o essionals’ code o conduc ,
beha iou moni o ing, con inuous SGBV sensi isa ion, and p ecise p ima y p e en i e
measu es.
Analyzing ou esul s in a dynamic and in e ac i e pe spec i e, us a ion and s ess as a
cause o iolen ac s can be ela ed o bad accommoda ion. Li ing condi ions p e iously
desc ibed by esiden s a e poo , and sha ing accommoda ion wi h adul s and child en, male
and emale, wi h no ela ionship, should be conside ed a s esso and igge o SGBV
ic imisa ion and/o pe pe a ion. This inding is aligned wi h ecen esea ch iden i ying bad
accommoda ion as an inc easing ac o o e ugee women’s ulne abili y o SGBV (79,80).
As men ioned be o e, cu en li ing condi ions o e ugees - women and gi ls – in G eek
Islands a e s ill desc ibed as being a om s anda ds ha mi iga e SGBV ic imisa ion (6).
Once, no gende associa ion wi h li ing condi ions was ound in he esul s, we specula e
ha inadequa e li ing condi ions can be a igge o SGBV ic imisa ion o bo h women and
men. Fo u u e esea ch, and conside ing we use seconda y da a, we belie e i would be
pe inen o unde s and possible s esso s ha could in luence bo h ic imisa ion and
pe pe a ion.
A ele an bias ela ed o gene al disclosu e o iolence should be conside ed. We belie e
ou esul s unde es ima e he eali y o SGBV in EARF. E en hough con iden iali y was
gua an eed du ing he in e iew, esiden s migh ha e ea ed i could in luence hei asylum
p ocedu e. In some big asylum ecep ion acili ies whe e communi ies wi h honou ules
we e esiding, i was epo ed o us la e ha esiden s discou aged o he s om pa icipa ing
in he s udy, men ioning po en ial s igma and/o communi y consequences (45). Fo
p o essionals, hey assumed no o da e o speak openly, e en i hey had supe io consen .
SGBV p e en i e measu es
Ou esul s indica e ha SGBV is p e en able, which is aligned wi h exis ing esea ch
documen ing ha iolence p e en ion is possible (8,9,107). We belie e he e idence o ou
CHAPTER 5. Discussion and Conclusions
129
esea ch co obo a e he ac ha iolence is desc ibed as a socially cons uc ed concep (13),
in luenced by indi idual, in e pe sonal, communi y and socie al ac o s (3,4), and can be
p e en ed (8,9,110,111). Ou esponden s ha e iden i ied seconda y and e ia y p e en i e
measu es ha could be implemen ed in EARF. The main men ioned in e en ions included
SGBV sensi isa ion and awa eness, imp o ing li ing condi ions and imp o e communica ion
be ween esiden s and p o essionals. To go beyond heo e ical p e en i e policies is
pa amoun , o he eal unde s anding o wha wo ks owa ds iolence p e en ion. In his
sense, ou esul s b ing a clea se o in e en ions ha could wo k o p e en SGBV in
EARF. Residen s and p o essionals can desc ibe wha could wo k in p e en ion, which is o
majo ele ance o achie ing e ec i e, e iciency and sus ainable p e en i e measu es (136).
Add essing he concep o desi able p e en ion, i has al eady been acknowledged in p e ious
esea ch (2,58) he impo ance o in eg ali y, pa icipa ion, inclusi eness, add essing oo
causes and maximizing agency o add ess SGBV p e en ion. Mo eo e , conside ing he
ecological model (75), i s dynamics le el, ulne able mig an s and p o essionals should no
be conside ed as isola ed om each o he . Thus, esiden s and p o essionals a e ac i e
s akeholde s in p e en ion. And should ake ac i e esponsibili y o ensu e p e en ion
measu es exis s in asylum ecep ion cen es.
Responden s belie ed SGBV epo ed cases could be p e en ed by imp o ing li ing
condi ions and communica ions. Re e ing o he Eu opean Di ec i e 2013/33/EU laying
down s anda ds o he ecep ion o applican s o in e na ional p o ec ion ( ecas ) in o ce,
we belie e mo e needs o be done. E en i he Membe S a es ha e di ec i es wi hd awing
minimum s anda ds o li ing condi ions o e ugees, AS and undocumen ed mig an s, li ing
condi ions a e s ill iden i ied as a possible measu e o p e en SGBV. This ac sugges s he
Di ec i es a e no p ope ly implemen ed by MS and/o lack clea ins uc ions o be placed
in o p ac ice.
Also, conside ing he IASC guidelines, layou and accommoda ion a e deemed o be essen ial
in SGBV p e en ion. I is acknowledged ha inadequa e housing o he denial o p ope y
a e con ibu ing ac o s o GBV (38). The IASC guidelines (38) ecognizes ha he ‘Lack o
adequa e housing du ing displacemen and ese lemen – whe he in u ban slums, squa e

CHAPTER 5. Discussion and Conclusions
130
se lemen s, collec i e cen es, e ugee se lemen s o wi h hos amilies – may con ibu e o
sexual assaul and exploi a ion.’ (p.167). We go u he , sugges ing ha asylum ecep ion
acili ies as he i s housing o e ugees, AS and undocumen ed mig an s, should p omo e
an SGBV ee-en i onmen wi h adequa e li ing and wo king condi ions.
The ac ha esiden s and p o essionals we e able o iden i y po en ial causes o SGBV is o
majo ele ance o u he design and implemen e idence-based p e en i e s a egies. E en
wi nessing a high numbe o epo ed cases o SGBV, ou esponden s belie ed hese ac s
could be p e en ed. In his sense, seconda y and e ia y p e en i e measu es epo ed by
bo h g oups we e ela ed o gene al SGBV p e en ion measu es. This is aligned wi h he ac
ha he majo i y o esiden s we e no awa e o he exis ing p e en i e measu es in he
ecep ion acili y whe e hey we e li ing.
Conside ing esponden s’ knowledge o exis en p e en able measu es, he majo i y o
esiden s a e no awa e o exis en p e en able measu es in he hos coun y o asylum
acili y. While he opposi e was ound o p o essionals. Once mo e, we assis o
p o essionals ha ing a b oade knowledge, no only ega ding SGBV concep ualisa ion bu
also ega ding exis ing p e en able measu es. Since he 1951 Gene a Con en ion ela ed o
he s a us o e ugees and i s 1967 P o ocol (4,26), i has been acknowledged he igh o a
“ iolence- ee” en i onmen and he need o p o ec e ugees and mig an s om iolence.
We belie e esiden s o EARF, by ha ing a na ow knowledge o SGBV concep ualisa ion
and e e ing no being awa e o exis ing p e en i e measu es, a e exposed o a highe isk
o ic imisa ion and/o pe pe a ion. Ou esul s also highligh he lack o moni o ing owa ds
p e en i e measu es. Moni o ing o in e en ions should be done con inuously (136) by he
s a wo king in EARF; he ac ha esiden s a e no awa e o exis ing p e en i e measu es
should be conside ed a call o ac ion owa ds mo e in o ma ion and sensi isa ion. The e is a
necessi y o con inuous e alua ion o p e en i e measu es o achie e e ec i e, e icien ,
ele an and sus ainable in e en ions (136). In e na ional policies, e idence-based, on how
o manage public heal h in he con ex o mig a ion a e c ucial and long o e due (21).
Fu he esea ch should add ess he speci ic con ex o asylum ecep ion acili ies and
e alua ing SGBV p e en i e s a egies (9,45,111). To unde s and he impac o SGBV
CHAPTER 5. Discussion and Conclusions
131
p e en i e s a egies and wha wo ks bes , acco ding o he a ge popula ion and speci ici y
o he social con ex is essen ial. P io esea ch has al eady sus ained he need o sys ema ic
e alua ion esea ch on p e en ion and managemen o all kinds o gende -based iolence
(107,111,140).
P edic i e models o SGBV ic imisa ion
Rela ed o he ac ha SGBV is p e en able (8,9,107), we ha e explo ed a pa e n o SGBV
ic imisa ion o bo h g oups. Ou esul s highligh ha being a esiden o a p o essional is
no a p edic i e ac o o ic imisa ion. The ea u e esiden o p o essional did no imp o e
modelling o p edic SGBV ic imisa ion. Which is aligned wi h p e ious esea ch
sugges ing ha wi hin EARF bo h g oups a e ulne able o SGBV (2,45). We belie e ha
wi hin EARF bo h g oups a e exposed and ulne able o iolence. This inding also highligh
he need o inclusi e p e en i e measu es, no only add essing esiden s bu also add essing
p o essionals.
Conside ing socio-demog aphic cha ac e is ics, such as sex and ma i al s a us, i seems hey
do no imp o e ou p edic i e models. In opposi ion, li ing and wo king condi ions in EARF
ha e p o en o be impo an p edic i e cha ac e is ics o SGBV ic imisa ion. Speci ically,
sha ing sani a y acili ies, he numbe o people, sex and age wi h whom hey a e sha ed, he
ype o accommoda ion ( oom, house/apa men , shel e , s udio/con aine , en o homeless),
and immig a ion s a us a e impo an cha ac e is ics o p edic ic imisa ion.
Ano he ele an esul is ela ed o he ea u e immig a ion s a us acco ding o he law. This
ea u e has p o en o be impo an o p edic iolence. Ou inding suppo s p e ious
esea che s, e e ing ha e ugees, AS and undocumen ed mig an s a e in a ulne able
posi ion o SGBV ic imisa ion (45,62,80,146). We belie e MS o he EU has al eady
acknowledged he ulne abili y o mig an s o iolence. The ecen Annual Repo on
Mig a ion and Asylum (147)
desc ibes he mig a ion and asylum si ua ion and achie emen s
a MS le el.
The same epo (147) e e s ha ‘Policy de elopmen s a he na ional le el
a ec ed all aspec s o (Membe ) S a es’ asylum sys ems, om access o he asylum
CHAPTER 5. Discussion and Conclusions
132
p ocedu e o ecep ion condi ions and he ea men o ulne able asylum seeke s.’ (p.2).
The e o e, we belie e achie emen s a e being made in Eu opean coun ies, e en i no in a
balanced implemen a ion ac oss coun ies (147). Howe e , we a e con on ed wi h high
epo ed cases o SGBV (53,62,80), wha makes us wonde : wha a e we s ill missing?
Acco dingly, we belie e ou esea ch con ibu es o answe ing his ques ion. We ound ha
SGBV concep ualisa ion is disc epan be ween g oups, esiden s a e no awa e o exis ing
p e en i e measu es a acili y and coun y le el, and he p edic i e model o SGBV
iden i ies li ing condi ions as o mos impo an p edic o s o ic imisa ion. The e o e, e en
i p og ess has been made, we a e s ill somehow a om achie ing a “ ee SGBV
en i onmen ” wi hin he EARF.
Rega ding he me hodology used in he hi d s udy, we belie e he numbe o pe sonal
ic ims migh be unde epo ed, and hus in luencing he modelling esul s. I should be
conside ed ha modelling could be imp o ed wi h a mo e signi ican sample.
No wi hs anding, we belie e machine lea ning echniques may suppo policymake s in
de ining e idence-based p ima y, seconda y and e ia y p e en ion s a egies o SGBV. We
challenge esea che s o apply hese echniques o explo e pa e ns o SGBV ic imisa ion
and pe pe a ion, in se e al con ex s and di e en popula ions. E en i we ecognise he
p edic i e accu acy o machine lea ning algo i hms, limi a ions should be conside ed. The
p edic i e a iables and i s impo ance a e no causes o he p oblema ic being s udied (154).
Con ibu ing o e idence-based SGBV p e en i e measu es
Acco ding o King & Lulle (12) ‘Mig a ion is pa o he li es o us all in Eu ope, whe he
we ou sel es ha e mo ed o no . The e o e, managing mig a ion mus also include lanking
policies and s a egies which deal wi h ela ed issues, bu in an in eg a ed way’ (p.122). In
his sense, we emphasize he equisi e o including p o essionals wo king wi h e ugees, AS
and undocumen ed mig an s as ac i e s akeholde s in he p ocess o designing and
implemen ing p e en i e s a egies and policies. Fu he mo e, esiden s should also be
acknowledged as ac i e oices o SGBV p e en ion.
CHAPTER 5. Discussion and Conclusions
133
The nega i e, sho and long- e m heal h consequences in mig an s o an SGBV expe ience
has been shown in p e ious esea ch (8,58,140). Ou indings show a high numbe o SGBV
cases in EARF, and low awa eness o exis en p e en able measu es epo ed by esiden s.
We belie e changes u ge o be implemen ed o mi iga e he nega i e impac o SGBV in
esiden s and p o essionals’ well-being. Go e nmen s, ci il socie y, NGO’s, and o he
ele an s akeholde s ha e he esponsibili y o inducing a heal hy and socially p oduc i e
mig a ion p ocess, in o de o ha e equi y on all he popula ion li ing in a coun y (21).
Recen esea ch as also acknowledge he need o a mo e cohe en Eu opean mig a ion
policy, and mo e high-quali y esea ch e idence o imp o e policies ela ed o he “g ound
eali ies” o mig a ion p agma ically (12). Mo eo e , mig a ion is no an isola ed p oblem,
and a single- ocus policy is no he solu ion (12). In his sense, ou esea ch goes beyond he
ocus o s e eo yped ic ims – esiden s – and s e eo yped agg esso s – p o essionals. We
ha e bo h g oups pe spec i es, wi h esul s showing ha being a esiden o a p o essional is
no ele an o p edic ic imisa ion.
Since he Beijing Decla a ion (5) ha ‘Gende -based pe secu ion has been accep ed as a basis
o e ugee s a us in some coun ies.’(p. 216). The EU has se he in igou Di ec i e laying
down s anda ds o he ecep ion o applican s o in e na ional p o ec ion (2013/33/EU)
calling all MS o ake adequa e p e en i e measu es in ecep ion acili ies o mi iga e SV,
including sexual assaul and ha assmen . Mo eo e , 27 Eu opean coun ies ha e a i ied he
Con en ion o Is anbul, going beyond he adop ion o he p e en i e measu es a asylum
acili y le el, o ensu e p o isions o adequa e medical, psychological, legal and o ensic ca e
o all ic ims o SV (66). S a ing ha SGBV is a eali y in EARF, wi h bo h esiden s and
p o essionals being a isk o ic imisa ion, we belie e we a e some way o he Eu opean
legal amewo k and i s o ien a ions. We challenge in e na ional s akeholde s and MS o
commi o cu en Di ec i es and in e na ional SGBV p e en ion guidelines, and o mo e
o wa d i , implemen ing e icien and e idence-based p e en i e s a egies. In speci ic
espec o Di ec i e 2013/33/EU (70), ecommended o ensu e he access o housing and
ood. No wi hs anding, we sugges ha he Di ec i e should be mo e p ecise, by ensu ing
one accommoda ion by amily, wi h one sani a y acili y by amily, espec ing amily