Full Te ms & Condi ions o access and use can be ound a
h ps://www. and online.com/ac ion/jou nalIn o ma ion?jou nalCode=ip i20
Scandina ian Jou nal o P ima y Heal h Ca e
ISSN: 0281-3432 (P in ) 1502-7724 (Online) Jou nal homepage: h ps://www. and online.com/loi/ip i20
The use o he B ie Child Abuse Po en ial
In en o y in he gene al popula ion in Finland
Noo a Ellonen, Heidi Ran anen, Sa i Lepis ö, Mika Helminen & Eija
Paa ilainen
To ci e his a icle: Noo a Ellonen, Heidi Ran anen, Sa i Lepis ö, Mika Helminen & Eija
Paa ilainen (2019) The use o he B ie Child Abuse Po en ial In en o y in he gene al
popula ion in Finland, Scandina ian Jou nal o P ima y Heal h Ca e, 37:1, 128-134, DOI:
10.1080/02813432.2019.1571002
To link o his a icle: h ps://doi.o g/10.1080/02813432.2019.1571002
© 2019 Uni e si y o Helsinki. Published
by In o ma UK Limi ed, ading as Taylo &
F ancis G oup.
Published online: 28 Jan 2019.
Submi you a icle o his jou nal
A icle iews: 164
View C ossma k da a
RESEARCH ARTICLE
The use o he B ie Child Abuse Po en ial In en o y in he gene al
popula ion in Finland
Noo a Ellonen
a
, Heidi Ran anen
b,c
, Sa i Lepis €
o
b,c
, Mika Helminen
b,c
and Eija Paa ilainen
b,d
a
Ins i u e o C iminology and Legal Policy, Facul y o Social Sciences, Uni e si y o Helsinki, Helsinki, Finland;
b
Facul y o Social
Sciences, Tampe e Uni e si y, Tampe e, Finland;
c
Tampe e Uni e si y Hospi al, Tampe e, Finland;
d
Sou he n Os obo hnia Hospi al
Dis ic , Sein€
ajoki, Finland
ABSTRACT
Objec i e: The pu pose o his esea ch was o analyze psychome ic in o ma ion in he B ie
Child Abuse Po en ial In en o y (BCAP) in a Finnish gene al popula ion sample.
Design: A sel - epo su ey o pa en s in a p ima y heal h ca e se ing and a hospi al se ing
was used o e alua e he use o he BCAP.
Se ing: The s udy popula ion consis ed o pa en s who we e isi ing one o he ollowing con-
ex s: a p ima y ma e ni y heal h ca e clinic, a child heal h ca e clinic, and he ma e ni y ou -
pa ien clinic, a ious pedia ic ou pa ien clinics, he gene al pedia ic wa d, he pedia ic
su gical wa d, o he neona al in ensi e ca e uni in a hospi al se ing.
Subjec s: The BCAP was gi en o pa en s a he 30–34 h week o p egnancy, when he child
was 5 mon hs old o all pa en s depending on he con ex . The BCAP was deli e ed o 759
pa en s. The inal size o he sample was 453 esponden s.
Main ou come measu e: The BCAP, which consis ed o 25 i ems o sc een child abuse po en ial
and nine i ems o e alua ion o esponden alidi y.
Resul s: The in e nal consis ency o he Abuse Risk Scale was good (.770), and he alidi y scales
wo ked well. The ac o s uc u e mi o s wi h he o iginal ac o s s uc u e.
Conclusion: The psychome ic p ope ies o he BCAP epo ed in he analysis sugges ha he
BCAP could be a alid ins umen o de ec child abuse po en ial in he gene al popula ion in
Finnish heal h ca e se ings. Howe e , among Finnish esponden s he e is e y li le a ia ion in
some pa s o he measu e, which sugges s ha u he esea ch should assess he alidi y o
he ins umen in ep esen a i e samples. Fu he analysis is also needed o e alua e he co ec
classi ica ion a e o he BCAP.
KEY POINTS
Iden i ica ion o amilies a isk o child mal ea men equi es alid ools o ecognize isk wi hin
he gene al popula ion, as pa o child and amily needs and isk assessmen s in amily se ices.
1. The BCAP is alid, eliable, and use ul in b inging pa en al wo ies unde discussion in child
and amily se ices.
2. Resul s o his s udy can be used o a mo e sys ema ic and alid child mal ea men isk
assessmen o iden i ying amilies who need help managing hei e e yday li es.
ARTICLE HISTORY
Recei ed 17 Augus 2018
Accep ed 17 Decembe 2018
KEYWORDS
Child abuse; child
mal ea men ; child en;
amily p ac ice; Finland;
pa en s; p e en ion; isk
In oduc ion
I is a commonly sha ed unde s anding ha child en
should no be exposed o any ype o mal ea men .
Ne e heless, child en expe ience i also in No dic
coun ies, al hough all kind o child mal ea men
ha e been banned o decades [1]. Based on a Finnish
sel - epo su ey in 2013, h ee pe cen o 9 h
g ade s epo ed being slapped by hei pa en du ing
pas 12 mon hs, and 21% epo ed expe iences o
mild physical iolence by hei pa en s a any poin in
hei li es [2]. Based on sel - epo su ey conduc ed
o pa en s o child en unde 12 yea s, ou pe cen o
Finnish pa en s and h ee pe cen o Swedish pa en s
epo ha ing slapped o hei child [3]. Compa ed o
o he coun ies, such as Canada [4] o England [5]
hese a es a e signi ican ly lowe .
Resea ch has sugges ed se e al isk ac o s o child
mal ea men , which can be di ided in o child-, pa -
en -and amily- ela ed isk ac o s. Known child- ela ed
isk ac o s a e, o example, child’s disabili y,
CONTACT Noo a Ellonen [email p o ec ed] Ins i u e o C iminology and Legal Policy, Facul y o Social Sciences, Uni e si y o Helsinki, P.O.
Box 24 (Unioninka u 40), Helsinki, FI 00014, Finland.
ß2019 Uni e si y o Helsinki. Published by In o ma UK Limi ed, ading as Taylo & F ancis G oup.
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s
un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed.
SCANDINAVIAN JOURNAL OF PRIMARY HEALTH CARE
2019, VOL. 37, NO. 1, 128–134
h ps://doi.o g/10.1080/02813432.2019.1571002
beha io al p oblems o i i abili y o he child. Risk ac-
o s ela ed o he pa en s may include subs ance
abuse, men al heal h p oblems, emo ional imma u i y,
lack o social suppo , pa en al his o y o mal ea men
as a child and un ealis ic expec a ions ega ding he
child. Known amily- ela ed isk ac o s a e, o
example, se e al child en in he amily, low income,
socioeconomic disad an age, his o y o child p o ec -
i e in e en ions and amily pe cep ion o lack o
social suppo . The accumula ion o any kind o isk
ac o s also inc eases he isk o child mal ea men as
well as any kind o mal ea men aking place is a
known indica o o u he mal ea men [6]. Nega i e
consequences o child mal ea men a e also e iden .
Mal ea men is associa ed wi h men al p oblems,
beha io al p oblems, subs ance abuse, obesi y and
some ch onic deceases, inc eased isk o suicidal
beha io and isky sexual beha io [7].
Va ious measu es exis o help p o essionals assess
pa en ing compe ence and sc een o child mal ea -
men [8,9]. One o he mos equen ly used in e -
na ionally and alida ed measu es o de ec ing
po en ial child mal ea men is he Child Abuse
Po en ial In en o y (CAP) c ea ed by Joel Milne in
1986 [10]. The CAP In en o y is a sel - epo measu e
o 160 o ced-choice (ag ee/disag ee) ques ions. O he
160 i ems, 77 cons i u e an Abuse Risk Scale consis ing
o six sub-scales: Dis ess, Rigidi y, Unhappiness,
P oblems wi h Child and Sel , P oblems wi h Family,
and P oblems wi h O he s. The CAP In en o y includes
h ee alidi y scales and h ee alidi y indexes o
e alua e he eliabili y o he esponses: The Lie Scale,
he Random Response Scale, he Inconsis ency Scale,
he Faking-good Index, he Faking-bad Index, and
he Random- esponse Index [10].
High in e nal consis ency and es - es eliabili y o
he o iginal CAP In en o y ha e been epo ed. The
co ec classi ica ion a e o known abuse s and non-
abuse s anges om 80% o 90% [10–12]. I has good
psychome ic p ope ies and subs an ial li e a u e sup-
po ing he measu e’s alidi y.
The CAP In en o y has been ansla ed in o mul-
iple languages, and i has been used a ound he
wo ld. The ansla ed e sions o he Abuse Risk Scale
ha e also been p o en o be eliable and alid.
In e nal consis ency o he Abuse Risk Scale a ied
om .88 o .91 (C onbach’s alpha), and co ec classi i-
ca ion a es based on disc iminan analyses a ied
om 83.0% o 100% [13]. The only signi ican di e en-
ces ha e been epo ed in he ac o s uc u e o he
CAP In en o y and in he Lie Scale o he alidi y sec-
ion [11,13]. The CAP In en o y has also been
alida ed in he Finnish gene al popula ion, and he
scale has shown o be a alid measu e in ha con-
ex [13,14].
Despi e he ad an ages o he CAP In en o y, some
cha ac e is ics, such as he leng h o he ques ionnai e
and he complex sco ing sys em, limi he measu e’s
use ulness, especially in some p ima y heal h ca e se -
ings. Because o hese limi a ions, Onde sma e al. [15]
de eloped a b ie e sion o he measu e. The b ie e -
sion (BCAP) includes 24/25 i ems and 9 alidi y i ems
o he o iginal CAP. I ems we e selec ed o sho en he
CAP bu o e ain as much sha ed a iance wi h he ull
measu e as possible, o e ain a s able ac o s uc u e
and a use ul alidi y scale, and o maximize he BCAP’s
p edic i e alidi y [15]. Based on he Ame ican sample,
he BCAP has se en sub-scales: Dis ess, Family
Con lic , Rigidi y, Happiness, Feelings o Pe secu ion,
Loneliness, and Financial Insecu i y. As alidi y scales,
he BCAP includes he Lie Scale and he Random
Response Scale. Psychome ic p ope ies o he BCAP
ha e been epo ed o be good, and he BCAP o e laps
ex ensi ely wi h he ull CAP Abuse Risk Sco e [15].
The BCAP has no ye been widely used in e -
na ionally and is no well alida ed in gene al popula-
ions. The BCAP has been used in Japan in a gene al
popula ion sample bu no alida ed in i [16]. The
only e alua ion o he use o he BCAP among a gen-
e al popula ion was pe o med in he Uni ed Kingdom
(UK) [17]. In ha analysis, he o e all eliabili y o he
Abuse Risk Scale was good (C onbach’s alpha .816),
and he ac o s uc u e was almos he same as in
he US sample [15]. Only mino cul u e- ela ed di e -
ences we e ound. Howe e , om he UK pe spec i e,
he alidi y scales (Lie Scale and Random Response)
need de elopmen . Bu he analysis sugges ed ha
he BCAP may be a eliable, quick, and use ul ool o
clinical sc eening o inc eased isk o physical child
abuse in pa en s in he UK se ing [17].
The pu pose o his esea ch is o analyze psycho-
me ic in o ma ion in he BCAP in a Finnish gene al
popula ion sample. In addi ion, he ac o s uc u e
and he sui abili y o he alidi y scales a e es ed. By
doing his, he aim is o ind whe he he BCAP is use-
ul in assessing child mal ea men , desc ibed as pa -
en al wo ies by pa en s hemsel es, wi hin he
gene al popula ion.
Ma e ial and me hods
S udy design
To e alua e he use o he BCAP in he Finnish gene al
popula ion. The BCAP was deli e ed as a one-occasion,
SCANDINAVIAN JOURNAL OF PRIMARY HEALTH CARE 129
sel - epo su ey o pa en s in a p ima y heal h ca e
se ing and a hospi al se ing be ween Janua y 2017
and Ma ch 2018.
S udy popula ion and p ocess
The BCAP was deli e ed o 759 pa en s o whom 464
e u ned ques ionnai es, esul ing esponse a e o 61.
The e a e no da a a ailable o hose, who didn’
esponse, so he ep esen a i eness o he da a can’
be e alua ed. F om he eliabili y poin o iew, we
applied he ule used in he CAP In en o y ha i he
esponden has mo e han 10% missing esponses, he
ques ionnai e should be conside ed in alid. The e o e,
11 esponden s who had a leas h ee missing
esponses we e excluded om u he analysis. The
inal size o he sample was 453 esponden s.
The s udy popula ion consis ed o pa en s who is-
i ed one o he ollowing con ex s: a p ima y ma e ni y
heal h ca e clinic (54 ecei ed esponses), a child
heal h ca e clinic (38), and he ma e ni y ou pa ien
clinic (76), a ious pedia ic ou pa ien clinics (53), he
gene al pedia ic wa d (119), he pedia ic su gical
wa d (54), o he neona al in ensi e ca e uni (25) o
which he i s wo belong o p ima y heal h ca e and
es o specialized ca e. The BCAP was gi en o
pa en s a he 30–34 h week o he mo he ’s p eg-
nancy, when he child was 5 mon hs old, o all
pa en s depending on he con ex . Pa en s we e no
in o med o he s udy be o e en e ing o he con ex .
The BCAP was illed ou a home be o e he nex
appoin men o du ing he isi . Pa en s we e gi en
he possibili y o ill ou he o m wi h he o he pa -
en , sepa a ely, o alone.
The majo i y o he esponden s we e emale
(82.3%, n¼373), and he mean age was 33.1 yea s
(SD ¼6.9). Mos o he esponden s had 1–2 child en
(56.1%), 31.8% had 3 o mo e child en, and ou pe
cen we e expec ing hei i s -bo n (missing 8.4%).
Measu e
The BCAP, enamed he Family Wel a e Ques ionnai e,
included a co e le e . A s anda dized back- ansla-
ion p ocedu e was pe o med p e iously wi h he
CAP [18] in Finland and used in he BCAP. I consis ed
o 25 i ems as shown in Table 1. To enable quick
checking o he BCAP in busy heal h ca e con ex s,
he Lie and Random Response scale i ems (shown in
Table 2) we e placed a he end o he ques ionnai e.
Demog aphic ques ions consis ed o he pa en ’s sex,
age, numbe o child en, and age o he child en.
S a is ical analysis
C onbach’s alpha eliabili y coe icien s we e used o
desc ibe he in e nal consis ency o he Abuse Risk
Scale. P incipal componen analysis was used o ind
he sui able ac o s uc u e o his Finnish gene al
sample. SPSS s a is ical so wa e was used o
he analysis.
Table 1. The BCAP i ems and equencies o i ems, % (n).
I ems Disag ee Ag ee Missing
I am a happy pe son 0.0 (0) 100 (453) 0.0 (0)
Some imes I eel all alone in he wo ld 92.7 (420) 7.1 (32) 0.2 (1)
E e y hing in a home should always be in i s place 81.2 (368) 18.1 (82) 0.7 (3)
I o en eel lonely inside 89.8 (407) 9.3 (42) 0.9 (4)
Child en should ne e disobey 95.8 (434) 3.1 (14) 1.1 (5)
I some imes wo y ha I will no ha e enough o ea 98.2 (445) 1.8 (8) 0.0 (0)
People ha e caused me a lo o pain 94.0 (426) 5.7 (26) 0.2 (1)
My li e is happy 0.4 (2) 98.9 (448) 0.7 (3)
Child en should be quie and lis en 90.3 (409) 6.6 (30) 3.1 (14)
My amily igh s a lo 97.8 (443) 2.2 (10) 0.0 (0)
My amily has p oblems ge ing along 96.7 (438) 2.6 (12) 0.7 (3)
I o en eel wo hless 96.9 (439) 2.9 (13) 0.2 (1)
O he people ha e made my li e unhappy 96.2 (436) 3.5 (16) 0.2 (1)
I o en eel e y upse 90.9 (412) 8.6 (39) 0.4 (2)
I ha e a happy li e 0.7 (3) 99.3 (450) 0 (0.0)
I am easily upse by my p oblems 90.9 (412) 8.2 (37) 0.9 (4)
I am o en dep essed 97.8 (443) 2.0 (9) 0.2 (1)
I am o en upse 98.2 (445) 1.5 (7) 0.2 (1)
A child needs e y s ic ules 97.6 (442) 1.8 (8) 0.7 (3)
I am o en upse and don’ know why 99.3 (450) 0.7 (3) 0 (0.0)
I o en eel e y alone 98.0 (444) 1.8 (8) 0.2 (1)
I o en eel alone 90.7 (411) 8.8 (40) 0.4 (2)
My amily has many p oblems 95.6 (433) 3.8 (17) 0.7 (3)
O he people ha e made my li e ha d 95.8 (434) 4.0 (18) 0.2 (1)
I some imes wo y ha my needs will no be me 83.9 (380) 15.9 (72) 0.2 (1)
130 N. ELLONEN ET AL.
Resul s
F equencies
In Table 1, he equencies o each i em a e p esen ed.
The able shows ha in mos i ems he numbe o
missing answe s is low. The e is also e y li le a i-
a ion in he esponses o ou o he i ems. Those
i ems a e: I am a happy pe son, my li e is happy, my
li e is good, and I am o en upse and do no know
why. The lack o a ia ion means ha hose i ems can-
no be used in adjus ing he ac o s uc u e.
The e o e, hose i ems we e emo ed om he p inci-
pal componen analysis.
Reliabili y and Finnish no ms/cu o poin s
The in e nal eliabili y o he measu e (25 i ems) is
good. C onbach’s alpha is .770. I he ou i ems wi h
oo li le a ia ion a e emo ed, C onbach’s alpha is
.781. Finnish no ms o he BCAP (mean sco es and
s anda d de ia ions) we e calcula ed o he Abuse
Risk Scale. The mean o he abuse sco e was 1.14
(SD ¼1.20). In he UK sample (also om he gene al
popula ion), he mean was 5.89 (SD ¼3.60), and in he
ou US samples, he means a ied be ween 8.4
(SD ¼6.4) and 9.4 (SD ¼5.9).
In he BCAP, 9 o 12 ha e been used as a cu o
poin o he Abuse Risk Scale. These numbe s ha e
been epo ed o p esen well o he cu o sco es
(166 o 215) used in he long CAP In en o y [15].
Because he no ms o he Abuse Risk Scale a e signi i-
can ly lowe in he Finnish sample, he use is no ea-
sonable. I would include only he op 0.4% o he
dis ibu ion o he Abuse Risk Scale. The manual o
he CAP In en o y also encou ages he de e mina ion
o locally app op ia e cu o poin s [10].
In he long CAP In en o y, a cu o poin o 100
was used in he Finnish se ing, which is almos 20%
o he maximum sco e (486). Based on he same ea-
soning, he app op ia e cu o poin o he BCAP
would be i e. Acco ding o ha cu o poin , 6% o
he esponden s ha e an ele a ed isk o child
mal ea men .
Validi y o esponses
The Lie Scale includes six i ems and he Random
Response Scale h ee i ems. The equencies o he
i ems a e p esen ed in Table 2. Onde sma e al. [15]
sugges ed ha any sco e o ou o abo e on he Lie
Scale and one o abo e on he Random Response
Scale should be conside ed in alid especially when
conside ed oge he . Acco ding o hese cu o poin s,
66.7% o he esponses would be alid acco ding o
he Lie Scale and 86.5% acco ding o he Random
Response Scale. When he scales a e conside ed
oge he , meaning ha bo h scale sco es a e e al-
ua ed, only 5.9% o he esponses would be in alid.
Fac o s uc u e
All 21 i ems we e subjec ed o a p incipal
componen analysis wi h p omax o a ion o analyze
he ac o s uc u e. In he Finnish sample, a i e- ac o
s uc u e was he mos sui able when eigen alues, a
con ibu ion p o ided by he ac o s o he o e all
explained a iance, and in e p e abili y we e consid-
e ed. Loadings o e .40 we e selec ed o he ac o s
(Table 3). Those ac o s a e: Loneliness and dis ess,
Impac o o he s, Family con lic , Rigidi y and
Financial insecu i y.
Discussion
P incipal indings
In his Finnish gene al popula ion sample, he in e nal
consis ency o he Abuse Risk Scale was good
(C onbach’s Alpha .770).The mean BCAP Abuse Sco e
(M¼1.14, SD ¼1.20) was conside ably lowe han in
any o he s udies applying he measu e. In he UK sam-
ple (also om he gene al popula ion), he mean was
5.89 (SD ¼3.60), and in he ou US samples, he means
a ied be ween 8.4 (SD ¼6.4) and 9.4 (SD ¼5.9).
These di e ences suppo e alua ing he app op ia e
cu o poin acco ding o na ional, no in e na ional,
no ms. Based on he alida ion o he long CAP
In en o y, he cu o poin seemed o be i e.
Table 2. The BCAP alidi y scale i ems and equencies o he alidi y scale i ems.
I ems Disag ee Ag ee Missing
I some imes ac wi hou hinking 55.6 (252) 42.2 (191) 2.2 (10)
I know wha is he igh and w ong way o ac 4.9 (22) 93.8 (425) 1.3 (6)
I some imes lose my empe 30.0 (136) 68.2 (309) 1.8 (8)
I is okay o le a child s ay in di y diape s o a while 87.6 (397) 8.4 (38) 4.0 (18)
Some imes I ha e bad hough s 88.7 (402) 9.7 (44) 1.5 (7)
Child en should no lea n how o swim 99.6 (451) 0 (0.0) 0.4 (2)
I some imes ail o keep all o my p omises 38.0 (172) 60.9 (276) 1.1 (5)
People some imes ake ad an age o me 81.7. (370) 16.6 (75) 1.8 (8)
I some imes say bad wo ds 29.4 (133) 68.2 (309) 2.4 (11)
SCANDINAVIAN JOURNAL OF PRIMARY HEALTH CARE 131

Acco ding o ha , six pe cen o he esponden s had
ele a ed isk o child mal ea men and should be
o e ed suppo o managing hei e e yday li es wi h
child en. F om he p e en i e pe spec i e, in he gen-
e al popula ion all he pa en s’wo y i ems should be
aken se iously, and a amily’s need o suppo should
be discussed genuinely ega dless o he abuse sco e.
The alidi y scales ( he Lie Scale and he Random
Response Scale) seemed o wo k in he Finnish sample
when hey we e conside ed oge he (only 5.9% o
in alid esponses). I also seems ha he measu e was
easy o answe , while he numbe o missing
esponses was gene ally low. This di e s om he
Ame ican and UK samples, in which he numbe o
in alid esponses a ied be ween 27% and 31.9% [12].
I should also be conside ed ha acco ding o he
long CAP In en o y, he cu o sco es o he Lie and
Random Response scales should be se o he 95 h
pe cen ile o he equency dis ibu ion. The e o e, in
he Finnish sample in he Lie Scale cu o poin would
be i e o mo e, no ou o mo e as in he Ame ican
o UK sample. Then he p opo ion o in alid
esponses would be 2.9% i he Lie Scale and he
Random Response Scale we e conside ed oge he .
The ac o s uc u e shows ha he Abuse Risk
Scale includes a ange o dimensions associa ed wi h
physical child abuse. The analysis mi o s o some
ex en he US and UK analyses al hough he analysis
sugges ed i e ac o solu ion. In Ame ican sample, a
se en- ac o s uc u e [15] and in he UK sample [17]a
six- ac o solu ion we e conside ed bes . Happiness,
which was ound in he US and UK samples, was no
iden i ied in he Finnish sample, because o he lack o
a ia ion in all i ems e lec ing happiness. Loneliness
and dis ess we e sepa a e ac o s in he US and UK
samples, bu hey a e in one ac o o he Finnish
sample. Rigidi y and amily con lic we e also ound in
he US and UK samples. Impac o o he s was ound in
he UK sample as a ac o o i s own, bu in he US
sample, hose i ems we e connec ed o he amily
con lic ac o . In he UK sample, he e was no inan-
cial insecu i y ac o , and i ems e lec ing inancial
insecu i y in he US sample we e connec ed o happi-
ness in he UK sample. In he Finnish sample, in he
i h ac o only one i em was loaded: ‘I some imes
wo y ha I will no ha e enough o ea .’The e o e,
his ac o could be seen as a inancial insecu -
i y ac o .
S eng hs and weaknesses o he s udy
The size o he da a (N¼453) was qui e good and big-
ge han he US o UK alida ion analysis, bu because
he a e o he Abuse Risk Scale is so low, he numbe
o esponden s wi h ele a ed isk o child mal ea -
men is qui e low (27). The esponse a e was 61 and
he cha ac e is ics o non- esponden s a e no a ail-
able o analysis. Al hough he esponse a e is qui e
good, he e is no gua an ee o he ep esen a i eness
o he da a. Almos comple e lack o a ia ion in i ems
e lec ing ad e si y and poo men al heal h (happiness
i ems, being upse e c.) sugges s ha esponde s may
be posi i ely selec ed. Da a is also skewed based on
he gende o he esponden s. ( emales 82%).
The e o e, assessing he alidi y o he ins umen in
ep esen a i e samples is a conside able challenge o
u u e esea ch.
Table 3. Fac o s uc u e o he i ems.
Loneliness and dis ess Impac o o he s Family con lic Rigidi y Financial insecu i y
I o en eel alone .785
Some imes I eel all alone in he wo ld .714
I o en eel lonely inside .692
I am o en dep essed .621 .517
I o en eel e y alone .613
I o en eel e y upse .583
I o en eel wo hless .563 .623
I am o en upse .415
I am easily upse by my p oblems .406
People ha e caused me a lo o pain .787
O he people ha e made my li e ha d .76
O he people ha e made my li e unhappy .745
My amily has many p oblems .718
My amily has p oblems ge ing along .794
My amily igh s a lo .78
I some imes wo y ha my needs will no be me .608
Child en should be quie and lis en .774
Child en should ne e disobey .696
E e y hing in a home should always be in i s place .576
I some imes wo y ha I will no ha e enough o ea .627
A child needs e y s ic ules –.517
132 N. ELLONEN ET AL.
The da a do no include in o ma ion o analyze he
co ec classi ica ion a e, which would also be impo -
an in alida ing a sc eening measu e. This is he nex
phase o alida ing he BCAP in Finnish child and
amily se ices. In his s udy, he BCAP was es ed
wi h he gene al popula ion. Acco ding o Milne and
C ouch, he co ec classi ica ion a es wi h he CAP
ha e been lowe han 90% in mo e di e se popula-
ions han known abusi e pa en s [12]. Milne and
C ouch sugges ed ha he CAP mo e likely ails o
de ec abusi e ( alse nega i es) pa en s han misclassi-
ies nonabusi e compa ison pa en s as abusi e ( alse
posi i es). In addi ion, he Abuse Risk Scale speci ici y
should be es ed u he o i s app op ia eness in
medical se ings as sugges ed [12]. Fu he esea ch
in he medical con ex is p omising because mos
pa en s accep ed he BCAP well, and only a ew
nu ses epo ed ha he pa en had conside ed he
BCAP inapp op ia e and annoying.
Findings in ela ion o o he s udies
In he Finnish gene al popula ion sample, he no ms
o he abuse sco e we e di e en om hose epo ed
o US and UK samples. This is in line wi h he ea lie
esea ch indings sugges ing ha a es o child mal-
ea men a e lowe in No dic coun ies compa ed o,
o example, UK. This emphasizes ha also he alid-
a ion o measu es o de ec child mal ea men po en-
ial should be alida ed sepa a ely in each coun y. In
addi ion, he a ia ion in some i ems was smalle ; in
ou i ems, almos ze o, which means ha hose i ems
do no sepa a e Finnish esponden s. These i ems
e lec ed he happiness ac o . The ze o a iance in
he happiness i ems may exis because he analysis
was based on a gene al sample, i may e lec some
cul u al di e ences o i may e lec some skewness o
he da a. Small (o none) a ia ion is some i ems
emphasizes u he esea ch o analyze he alidi y o
he ins umen in samples, o which ep esen a i e
na u e can be assessed.
The Finnish sample conduc ed i e sub-scales when
in he US sample he bes ac o solu ion was se en
ac o s and in he UK sample six ac o s. Despi e his
di e ence, he ac o s uc u e in he Finnish analysis
o e lapped clea ly wi h he ac o s om he US and
UK analyses o he BCAP. In addi ion, ac o s a e no
used o p oduce indi idual ac o s sco es bu o
ensu e he Abuse Risk scale includes a ange o
dimensions associa ed wi h physical child abuse isk
[15]. Tha was shown o be ensu ed in he
Finnish sample.
Validi y sco es c ea ed based on he US sample
seemed o wo k qui e well in he Finnish gene al
popula ion, when he Lie Scale and he Random
Response scale we e conside ed oge he , al hough in
he UK sample he indings we e di e en . The a e o
in alid esponses was s a is ically signi ican ly lowe in
he Finnish sample han in he US and UK samples.
O e all, he analysis o he Finnish sample con-
cluded he same as he US and UK analyses: The BCAP
could be conside ed a eliable, quick, and use ul clin-
ical ool o sc eening po en ial child mal ea men
among pa en s when he o iginal longe e sion o
he CAP In en o y is seen as oo exhaus ing o p ac i-
ione s. This is an impo an inding in p og ess o u -
he e alua ion o he use o he BCAP. Fu he
esea ch is, howe e , needed o analyze he co ec
classi ica ion a e o he measu e and o alida e he
ins umen in a sample, which ep esen a i eness can
be e alua ed.
Implica ions
Assessmen p ac ices o he isk o child mal ea -
men , as well as o he p ac ices [19], need o be
na ionally de eloped, and based on scien i ically and
clinically es ed in e na ional ins umen s. I is always
mo e e hical and cos -e ec i e o p e en child mal-
ea men han only ea child mal ea men inju ies.
Findings o he alida ion o he CAP In en o y in
Finland [13], and clinical es ing o he BCAP, his
ins umen could be conside ed as a alid ins umen .
The e alua ion p ocess will con inue by launching an
elec onic e sion o he BCAP and i s manual, and by
con inua ion o he esea ch especially ela ed o co -
ec classi ica ion a e and he ole o alidi y scales in
clinical wo k.
Acknowledgemen s
The au ho s wan o hank all amilies and s a o pa ici-
pa ing, and he Minis y o Social A ai s and Heal h, Finland,
o unding his clinical es ing phase o he BCAP.
Disclosu e s a emen
Au ho s ha e no con lic o in e es .
E hical app o al
The s udy p o ocol was app o ed by he Resea ch E hics
Commi ee o Pi kanmaa Hospi al Dis ic (R11198H).
SCANDINAVIAN JOURNAL OF PRIMARY HEALTH CARE 133
Funding
The s udy was inancially suppo ed by Minis y o Social
A ai s and Heal h, Finland [201610044].
Re e ences
[1] Du an JE, Smi h AB. Global pa hways o abolishing
physical punishmen : ealizing child en’s igh s. New
Yo k: Rou ledge; 2011.
[2] Fage lund M, Pel ola M, K€
a€
a i€
ainen J, e al.
Las en ja nuo en €
aki al akokemukse 2013:
Lapsiuh i u kimuksen uloksia. [Expe iences o io-
lence among child en and young people 2013: esul s
om he Finnish child ic im su ey]. Tampe e,
Finland: Police Uni e si y College; 2014.
[3] Ellonen N, Tindbe g Y, Lucas S, e al. Pa en s’sel -
epo ed use o co po al punishmen and o he
humilia ing upb inging p ac ices in Finland and
Sweden—A compa a i e s udy. Child Abuse Re .
2017;26:289–304.
[4] Cl
emen ME, Boucha d C. P edic ing he use o single
e sus mul iple ypes o iolence owa ds child en in
a ep esen a i e sample o Quebec amilies. Child
Abuse Neglec . 2005;29:1121–1139.
[5] Rad o d L, Co al S, B adley C, e al. The p e alence
and impac o child mal ea men in he Uni ed
Kingdom. Child Abuse Neglec . 2013;37:801–813.
[6] Paa ilainen E, Flinck A. 2015. E icien me hods o
iden i ying child mal ea men in social and heal h
ca e. Nu sing Resea ch Founda ion: Clinical guidelines
– om e idence o ou comes. h p://www.ho us. i/
sys em/ iles/SUOSITUS_las en_kal oinkoh elu_
ENGLANTI.pd
[7] No man RE, Byambaa M, De R, e al. The long- e m
heal h consequences o child physical abuse, emo-
ional abuse, and neglec : a sys ema ic e iew and
me a-analysis. PLOS Med. 2012;9:11.
[8] Bailhache M, Le oy V, Pille P, e al. Is ea ly de ec ion
o abused child en possible?: a sys ema ic e iew o
he diagnos ic accu acy o he iden i ica ion o
abused child en. BMC Pedia . 2013;13:202.
[9] Heinze MC, G isso T. Re iew o ins umen s assessing
pa en ing compe encies used in child cus ody e alua-
ions. Beha Sci Law. 1996;14:293–313.
[10] Milne J. The Child Abuse Po en ial In en o y manual.
2nd ed. Webs e : Psy ec; 1986.
[11] Milne JS, C ouch JL. Psychome ic cha ac e is ic o
ansla ed e sions o he Child Abuse Po en ial
In en o y. Psychol Violence. 2012;2:239–259.
[12] Milne JS, C ouch JL. Child physical abuse isk assess-
men : pa en and amily e alua ions. In: Campbell JC,
Messing JT, edi o s. Assessing dange ousness –
Domes ic iolence o ende s and child abuse s. 3 d
ed. New Yo k: Sp inge ; 2017. p. 55–88.
[13] Ellonen N, Lepis €
o S, Helminen M, e al. C oss-cul u al
alida ion o he Child Abuse Po en ial In en o y in
Finland: p elimina y indings o he s udy among
pa en s expec ing a baby. J Soc Se Res. 2017;43:
308–318.
[14] Lepis €
o S, Ellonen N, Helminen M, e al. Family heal h,
unc ioning, social suppo and child mal ea men
isk o amilies expec ing a baby. J Clin Nu s. 2017;26:
2439–2451.
[15] Onde sma S, Ca in M, Mullins S, e al. A b ie o m o
he Child Abuse Po en ial In en o y: de elopmen
and alida ion. JCCAP. 2005;34:301–311.
[16] Ono M, Honda S. Associa ion be ween social suppo
and child abuse po en ial among Japanese mo he s.
Child You h Se Re . 2017;73:88–92.
[17] Walke C, Da ies J. A c oss-cul u al alida ion o he
B ie Child Abuse Po en ial In en o y (BCAP). J Fam
Viol. 2012;27:697–705.
[18] Haapasalo J, Aal onen T. Child abuse po en ial. J
In e pe s Violence. 1999;14:571–585.
[19] Ahl o n KC, Ho wi z EB, Osika W. A Swedish e sion o
he Consul a ion and Rela ional Empa hy (CARE)
measu e. Scand J P ima y Heal h Ca e. 2017;35:
286–292.
134 N. ELLONEN ET AL.