ArticlePDF Available

Children with behavioral, non-behavioral, and multiple disabilities, and the risk of out-of-home placement disruption

Authors:

Abstract and Figures

This study examined the relative risk of placement disruption for 3-10 year-old children placed in out-of-home care based on the biological relatedness of the placement caregiver and child disability status: no disability, a non-behavioral disability only, a behavioral disability only, or both a non-behavioral and behavioral disability. Data were used from the baseline and 36 month follow-up of the National Survey of Child and Adolescent Well-Being, a national probability study of children investigated for child abuse and neglect in the United States. Disability status was derived using several different nationally-normed measures of language development, daily-living skills, social skills, and behavioral problems. Around 1 in 4 children placed in out-of-home care experienced a disruption. Placement with kin decreased the likelihood of disruption for a majority of children, and children with different types of disabilities were no more or less likely to disrupt in kinship care compared to children with no disability. Older children with a behavioral disability only or both a non-behavioral and behavioral disability were more likely to disrupt compared to younger regardless of placement. The study findings suggest that maltreated children placed with kin will be afforded the same stability provided to children without a disability.
Content may be subject to copyright.
Child
Abuse
&
Neglect
35 (2011) 956–
964
Contents
lists
available
at
SciVerse
ScienceDirect
Child
Abuse
&
Neglect
Children
with
behavioral,
non-behavioral,
and
multiple
disabilities,
and
the
risk
of
out-of-home
placement
disruption
Jesse
J.
Helton
Children
and
Family
Research
Center,
School
of
Social
Work,
University
of
Illinois
at
Urbana-Champaign,
1010
West
Nevada
Street,
Urbana,
IL
61801,
USA
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
24
September
2010
Received
in
revised
form
10
June
2011
Accepted
14
June
2011
Available online 8 November 2011
Keywords:
Kinship
care
Disability
Placement
disruption
a
b
s
t
r
a
c
t
Objective:
This
study
examined
the
relative
risk
of
placement
disruption
for
3–10
year-old
children
placed
in
out-of-home
care
based
on
the
biological
relatedness
of
the
placement
caregiver
and
child
disability
status:
no
disability,
a
non-behavioral
disability
only,
a
behav-
ioral
disability
only,
or
both
a
non-behavioral
and
behavioral
disability.
Methods:
Data
were
used
from
the
baseline
and
36
month
follow-up
of
the
National
Survey
of
Child
and
Adolescent
Well-Being,
a
national
probability
study
of
children
investigated
for
child
abuse
and
neglect
in
the
United
States.
Disability
status
was
derived
using
several
different
nationally-normed
measures
of
language
development,
daily-living
skills,
social
skills,
and
behavioral
problems.
Results:
Around
1
in
4
children
placed
in
out-of-home
care
experienced
a
disruption.
Place-
ment
with
kin
decreased
the
likelihood
of
disruption
for
a
majority
of
children,
and
children
with
different
types
of
disabilities
were
no
more
or
less
likely
to
disrupt
in
kinship
care
compared
to
children
with
no
disability.
Older
children
with
a
behavioral
disability
only
or
both
a
non-behavioral
and
behavioral
disability
were
more
likely
to
disrupt
compared
to
younger
regardless
of
placement.
Conclusion:
The
study
findings
suggest
that
maltreated
children
placed
with
kin
will
be
afforded
the
same
stability
provided
to
children
without
a
disability.
© 2011 Elsevier Ltd. All rights reserved.
Introduction
Stable
out-of-home
arrangements
are
an
important
component
in
assuring
the
well-being
of
maltreated
children
who
are
without
the
support
of
nurturing
biological
parents.
Although
current
child
welfare
policy
recognizes
kin
placements
as
least
likely
to
disrupt,
little
research
has
been
done
to
determine
the
relative
risk
of
disruption
for
children
with
various
types
of
disabilities
in
kin
and
non-kin
placements.
Thus
the
purpose
of
this
study
is
twofold:
(1)
compare
the
disruption
rates
of
kin
and
non-kin
placements
for
children
with
different
disabilities,
and
(2)
determine
the
child,
caregiver,
and
family
characteristics
associated
with
disruption
for
children
with
different
disabilities.
Four
groups
of
children
will
be
examined:
those
with
no
disability,
those
with
a
non-behavioral
disability
(that
is
a
marked
or
severe
impairment
in
language
development,
daily-living
skills,
or
social
skills),
those
with
an
emotional
or
behavioral
disability,
and
those
with
both
a
non-behavioral
and
behavioral
disability.
Improving
placement
stability
is
a
major
part
of
providing
maltreated
children
with
adequate
care.
When
children
are
placed
in
stable
environments
and
into
stable
relationships
with
caregivers,
they
are
less
prone
to
illness
(Harden,
2004),
less
likely
to
act
out
(Rubin,
O’Reilly,
Luan,
&
Localio,
2007),
less
likely
to
be
delinquent
(Ryan
&
Testa,
2005),
and
more
likely
to
perform
well
in
school
(Zima
et
al.,
2000).
To
increase
the
chance
of
establishing
a
stable
out-of-home
placement,
current
child
welfare
policy
recommends
placement
of
a
maltreated
child
with
the
least-restrictive,
most
family-like
setting
available
(James,
2004).
It
is
widely
accepted
that
children
placed
with
kin
are
more
stable
compared
to
children
placed
in
more
restrictive
levels
of
care,
such
as
with
non-kinship
foster
parents
or
residential
facilities
(James,
2004;
Testa,
2001;
0145-2134/$
see
front
matter ©
2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chiabu.2011.06.004
J.J.
Helton
/
Child
Abuse
&
Neglect
35 (2011) 956–
964 957
Usher,
Randolph,
&
Gogan,
1999;
Webster,
Barth,
&
Needell,
2000).
The
stable
relationship
between
kin
caregivers
and
foster
children
is
believed
to
have
ties
to
an
evolutionary
psychological
explanation
of
altruism
(Testa,
2001;
Testa
&
Slack,
2002).
To
understand
kin
altruism,
evolutionary
psychologists
look
to
the
innate
sense
of
connection
humans
share
with
genetically
related
individuals
(Watson,
1997).
This
connection
insulates
children
who
are
biologically
related
to
the
adult
from
situations
harmful
to
their
safety
and
well-being.
The
theory
posits
that
a
stronger
feeling
of
empathy
and
duty
towards
biologically
related
children
increases
a
caregiver’s
ability
to
endure
burdensome
situations
that
a
non-kin
caregiver
may
be
unwilling
to
endure,
thus
decreasing
the
chance
of
placement
disruption.
New
research
is
showing
that
these
types
of
advantages
of
kin
arrangements
happen
towards
the
beginning
of
an
out-of-home
placement
and
then
diminish
as
the
duration
of
time
spent
in
care
lengthens
(Koh
&
Testa,
2008;
Testa,
Bruhn,
&
Helton,
2009).
Aside
from
placement,
some
consensus
exists
as
to
what
child
and
caregiver
characteristics
present
the
greatest
risk
for
placement
disruption
(for
review,
see
Oosterman,
Schuengal,
Slot,
Bullens,
&
Doreleijers,
2007).
Placement
disruption
may
be
more
likely
for
male
children
compared
to
female,
older
children
compared
to
younger,
and
Caucasian
children
compared
to
all
other
races
(Heflinger
&
Taylor-Richardson,
2004;
Leathers,
2006;
Smith,
Stormshak,
Chamberlain,
&
Whaley,
2001;
Webster
et
al.,
2000).
Other
correlates
of
disruption
include
family
level
factors,
such
as
the
number
of
children
in
the
household,
financial
support,
and
caregiver
health
(Chamberlain
et
al.,
2006;
Heflinger
&
Taylor-Richardson,
2004;
Terling-
Watt,
2001).
However,
the
largest
empirical
consensus
is
that
children
with
emotional
or
behavioral
problems
(Aaron
et
al.,
2010;
Barth
et
al.,
2007;
Chamberlain
et
al.,
2006;
James,
2004;
Leathers,
2006;
Newton,
Litrownik,
&
Landsverk,
2000;
Oosterman
et
al.,
2007;
Rolock,
Koh,
Cross,
&
Manning,
2009)
are
more
likely
to
disrupt
compared
to
children
without
any
emotional
or
behavioral
problems.
A
study
by
Barth
and
colleagues
(2007)
using
national
data
found
that
children
ages
7–14
with
emotional
and
behavioral
disorders
(EBD)
were
2.5
times
more
likely
than
their
peers
without
EBD
to
experience
more
than
4
placements
in
3
years.
Disruption
is
more
likely
to
occur
for
children
with
an
emotional
or
behavioral
problem
when
caregivers
become
unable
to
control
a
child’s
conduct.
Children
with
these
problems
require
more
caregiver
time
and
energy
to
properly
address
instances
of
irritability,
non-compliance,
and
emotional
regulation.
If
these
occurrences
of
conduct
problems
are
not
properly
addressed,
they
may
result
in
more
negative
caregiver/child
interactions
(Chamberlain,
1996;
Smith,
Sprengelmeyer,
&
Moore,
2004).
These
difficulties
may
prohibit
the
forming
of
positive
relationships
between
out-of-home
caregivers
and
children
with
behavioral
problems,
leading
to
increased
rates
of
disruption.
A
study
by
Rolock
and
colleagues
(2009),
found
that
out-of-home
caregivers
of
children
with
multiple
placement
disruptions
often
reported
that
they
did
not
possess
the
skills
needed
to
manage
the
behavior
of
these
foster
children.
Although
the
evidence
base
is
smaller,
children
with
other
non-behavioral
problems,
such
as
medical
conditions
(Chang
&
Liles,
2007;
Terling-Watt,
2001)
or
developmental
disabilities
(Rosenberg
&
Robinson,
2004),
are
also
more
likely
to
disrupt
compared
to
normally
developing
children.
Rosenberg
and
Robinson
(2004)
examined
a
national
sample
of
foster
children
under
the
age
of
3
in
1999
Adoption
and
Foster
Care
Analysis
and
Reporting
System
(AFCARS)
administrative
data.
They
found
that
children
with
a
developmental-medical
condition
had
more
prior
foster
care
placements
and
longer
durations
of
stay
in
out-of-home
care
compared
to
their
non-disabled
peers.
Another
study
focusing
on
kinship
care
placements
found
that
children
who
had
disrupted
from
care
were
less
healthy
compared
to
children
who
were
reunified
or
currently
in
a
stable
placement
(Chang
&
Liles,
2007).
Contrary
to
the
concept
of
kin
altruism,
it
is
theorized
that
decreased
feelings
of
connection
and
the
failure
to
form
a
secure
attachment
is
a
major
cause
of
disruptions
for
families
that
include
a
child
with
a
disability
(Rycus
&
Hughes,
1998;
Sobsey,
1994).
Children
with
a
major
medical
problem
or
disability
can
be
at
risk
for
these
problems
in
their
relationship
with
caregivers
due
to
frequent
hospitalizations
or
perceptions
that
these
children
are
unresponsive
or
unaffectionate
(Ammerman
&
Patz,
1996;
Sherman,
1988).
Caregivers
of
children
with
a
disability
have
reported
viewing
their
child
as
“different”
and
unlike
any
other
child,
leading
to
feelings
of
anger
and
embarrassment
(Rycus
&
Hughes,
1998).
Biological
parents
have
reported
that
family
stressors
related
to
the
care
of
a
child
with
a
disability,
such
as
caregiving
demands,
ambiguity
of
parental
roles,
and
the
lack
of
financial
resources
may
also
lead
to
family
disruption
(Hodapp,
2002;
Jackson,
2004
as
cited
in
Roper
&
Jackson,
2007,
p.
147).
Although
Barth
and
colleagues
(2007)
also
examined
children
with
special
needs,
as
reported
by
the
caseworker,
they
did
not
separate
behavioral
from
non-behavioral
disabilities.
No
other
study
to
date
has
attempted
to
parse
out
the
individual
effect
of
emotional
or
behavioral
problems
and
non-behavioral
problems
on
out-of-home
placement
disruption.
While
children
may
be
most
stable
with
out-of-home
kin
placements,
there
are
grounds
to
be
concerned
with
quality
of
the
home
environment
and
the
numerous
challenges—challenges
that
may
make
it
more
difficult
to
care
for
a
child
with
a
disability—kin
caregivers
face
that
non-kin
caregivers
do
not.
For
instance,
Ehrle
and
Geen
(2002)
found
that
39%
of
out-of-home
kin
caregivers
had
incomes
below
the
federal
poverty
line,
which
is
3
times
the
rate
of
out-of-home
non-kin
caregivers
living
in
poverty
(13%).
Kin
caregivers
are
also
more
likely
to
live
in
disadvantaged
neighborhoods
and
public
housing
complexes
and
be
in
poorer
health
compared
to
non-kin
caregivers
(Berrick,
1997;
Gebel,
1996).
Kin
caregivers
also
report
receiving
irregular
or
limited
services
from
child
protective
services
due
to
ineligibility
or
caregiver
refusal
(Gibson,
2002).
If
it
is
true
that
a
child’s
disability
and
medical
problems
may
be
straining
foster
families
to
the
point
of
disruption
(Rosenberg
&
Robinson,
2004),
then
it
stands
to
reason
that
disruption
might
be
more
likely
in
disadvantaged
homes
with
fewer
resources.
Complicating
the
examination
of
placement
disruption
for
children
with
both
emotional
or
behavioral
and
non-behavioral
disabilities
in
out-of-home
care
is
the
possibility
of
the
discretionary
initial
placement
of
difficult
children
away
from
kinship
placements
and
into
more
restrictive,
less
family-like
settings.
Children
with
very
complicated
medical
or
developmental
958 J.J.
Helton
/
Child
Abuse
&
Neglect
35 (2011) 956–
964
conditions
and
difficult
behaviors
may
be
in
need
of
more
intensive
services
with
caregivers
specifically
trained
to
address
those
issues,
preempting
kinship
placement
considerations.
Studies
have
shown
that
children
placed
with
kin
may
be
less
likely
to
have
health
problems
(Grogan-Kaylor,
2000),
behavioral
issues
(Benedict,
Zuravin,
Somerfeld,
&
Brandt,
1996),
or
disabilities
(Beeman,
Kim,
&
Bullerdick,
2000)
compared
to
children
placed
in
more
restrictive,
less
family-like
settings.
Further,
child
welfare
caseworkers
have
reported
being
reluctant
to
utilize
available
kin
due
to
difficulties
supervising
placements,
dysfunctional
family
traditions,
and
fears
of
poor
parenting
skills
and
outdated
beliefs
about
discipline
(Beeman
&
Boisen,
1999;
Peters,
2004).
If
caseworkers
are
viewing
children
with
a
disability
as
more
vulnerable
than
a
normally
developing
child,
they
may
be
reluctant
to
place
these
children
with
families
they
fear
would
be
detrimental
to
their
well-
being
and
more
likely
to
disrupt.
However,
no
evidence
to
date
has
supported
these
assumptions.
Research
questions
While
research
has
shown
that
children
in
out-of-home
care
are
most
stable
with
kin
caregivers,
it
is
unknown
to
what
extent
the
added
stability
afforded
by
kinship
care
extends
to
children
with
different
disability
profiles:
those
with
no
disability,
a
non-behavioral
disability
only,
an
emotional
or
behavioral
disability
only
(referred
to
as
“behavioral
disability”
for
the
remainder
of
the
paper),
or
both
a
non-behavioral
and
behavioral
disability.
With
the
current
policy
recommendations
advocating
placement
with
kin
as
a
favorable
placement
option,
it
is
important
to
ask
whether
kin
placements
are
stable
for
all
children.
It
is
also
important
to
take
into
account
that
kin
caregivers
have
financial
and
health
challenges
that
other
caregivers
do
not—challenges
that
may
make
it
more
difficult
to
care
for
children
with
any
type
of
disability.
Methods
Sample
This
study
used
data
from
the
Child
Protective
Service
sample
of
the
National
Survey
of
Child
and
Adolescent
Well-Being
(NSCAW),
a
national
probability
panel
study
of
5,501
children
who
were
part
of
a
child
maltreatment
investigation
within
a
fifteen-month
period
that
began
in
October
of
1999.
The
NSCAW
subjects
ranged
in
age
from
newborn
to
14
years
old
at
the
time
of
the
first
stage
sampling
and
were
selected
from
92
Primary
Sampling
Units
(PSUs),
which
is
proportionate
to
size
in
97
counties
nationwide.
The
second
stage
involved
the
sampling
of
children
from
lists
of
cases
for
which
an
investigation
had
been
completed
in
the
previous
month.
Data
were
drawn
from
a
baseline
interview
conducted,
on
average,
four
months
after
the
completion
of
a
child
protective
service
investigation
and
from
follow-up
interviews
at
1
year,
years,
and
3
years
thereafter.
The
purpose
of
NSCAW
is
to
examine
the
experiences
of
children
and
families
with
the
child
welfare
system
(NSCAW
Research
Group,
2002).
From
the
original
baseline
sample
of
5,501
children,
data
were
restricted
to
children
between
the
ages
of
3
and
10
placed
with
either
an
out-of-home
kin
or
out-of-home
non-kin
caregiver
at
the
time
of
baseline
interview.
Children
under
the
age
of
3
and
over
the
age
of
10
were
dropped
because
certain
standardized
measures
of
functionality
were
only
used
with
children
between
3
and
10
years
of
age.
Children
placed
in
group
homes
or
temporary
placements
following
an
investigation
were
also
dropped
due
to
the
high
likelihood
of
placement
moves
(Park
&
Helton,
2010).
In
the
end,
4,204
cases
were
dropped
due
to
placement
restrictions
and
an
additional
982
cases
were
dropped
due
to
age
restrictions.
This
resulted
in
a
final
sample
of
315
cases.
Measures
Disruption.
NSCAW
investigators
were
asked
at
each
wave
if
the
child’s
caregiver
was
the
same
as
the
previous
interview.
Investigators
who
reported
that
a
child’s
current
caregiver
was
the
same
during
each
of
the
interviews
between
baseline
and
3
years
were
considered
living
in
stable
placements
and
coded
as
0,
with
a
new
caregiver
reported
at
anytime
between
baseline
and
3
years
considered
a
disrupted
placement
and
coded
as
1.
New
caregivers
at
any
wave
who
identified
themselves
as
either
a
biological
parent
or
adopted
parent
were
dropped
from
the
analysis.
These
legally
permanent
moves
were
not
considered
a
disruption
in
care
or
stable.
Data
on
the
reason
why
a
child
moved
from
one
caregiver
to
another
were
not
available.
Disability
status.
Disability
encompasses
many
different
types
of
physical,
mental,
psychological,
and
social
impairments.
To
obtain
a
valid
measure
of
disability,
an
effort
was
made
to
use
as
many
independent
assessments
possible
to
distin-
guish
disability
status.
The
NSCAW
is
unique
in
its
potential
contribution
to
this
line
of
inquiry,
as
it
contains
standardized
(that
is
nationally-normed)
measures
that
can
be
used
to
construct
valid
indicators
of
both
non-behavioral
and
behavioral
disabilities.
To
be
classified
as
having
a
non-behavioral
disability,
children
must
show
limitations
in
gross
or
fine
motor
skills,
self-care
activities
of
daily
living,
such
as
eating,
dressing,
cognition
and
learning,
and
social
skills
adequate
to
interact
with
others
(Msall,
Bobis,
&
Field,
2006).
Children
who
showed
limitations
were
identified
in
NSCAW
if
they
demonstrated
serious
deficits
in
language,
daily-living,
or
social
skills
on
four
nationally-normed
measures:
Kaufman
Brief
Intelligence
Test
(K-BIT;
Kaufman
&
Kaufman,
1990);
Preschool
Language
Scale
(PLS-3;
Zimmerman,
Steiner,
&
Pond,
1992);
Vineland
Adaptive
Behavior
Scale
(VABS;
Sparrow,
Carter,
&
Cicchetti,
1993);
and
Social
Skills
Rating
System
(SSRS;
Gresham
&
Elliot,
1990).
Children
were
classified
as
disabled
if
they
were
(a)
more
than
3
standard
deviations
below
the
population
mean
on
J.J.
Helton
/
Child
Abuse
&
Neglect
35 (2011) 956–
964 959
one
of
the
above
measures,
or
(b)
more
than
2
standard
deviations
on
at
least
2
measures.
Children
identified
by
caregivers
as
having
HIV,
total
blindness,
total
deafness,
cerebral
palsy,
Down
syndrome,
a
physical
disability,
orthopedic
disability,
or
traumatic
brain
injury
were
also
classified
as
“disabled”.
This
classification
closely
follows
the
federal
guidelines
for
identifying
children
eligible
for
Supplemental
Security
Income
(SSI)
because
of
an
impairment.
SSI
Regulation
Code
416.926
states
a
child
is
eligible
for
SSI
if
he
or
she
exhibits
“marked”
limitations
in
2
domains
of
functioning
or
an
“extreme”
limitation
in
one
domain
(SSA,
2010).
A
“marked”
limitation
is
1
that
is
more
than
moderate
but
less
than
extreme,
or
is
the
equivalent
of
the
functioning
we
would
expect
to
find
on
standardized
testing
with
scores
that
are
at
least
2,
but
less
than
3,
standard
deviations
below
the
mean.
An
“extreme”
limitation
is
the
equivalent
of
scoring
below
3
standard
deviations
below
the
mean.
For
this
study,
the
criterion
for
“marked”
limitations
was
changed
from
2
standard
deviations
below
the
mean
to
1.5
standard
deviations
below
the
mean
due
to
small
proportion
of
children
who
met
the
more
conservative
parameter.
Behavioral
problems
were
measured
by
the
caregiver
reported
Child
Behavioral
Checklist
(CBCL).
The
CBCL
is
a
widely
used
measure
of
social
competence
and
behavioral
problems
standardized
by
age
and
gender
(Achenbach,
1991).
The
CBCL
measures
domains
of
both
internal,
such
as
somatic
complaints,
anxiety,
and
depression,
and
externalizing
behavior
prob-
lems,
such
as
delinquent
and
aggressive
behaviors
and
external
presentations
of
behavioral
problems,
which
were
combined
for
this
study.
Children
scoring
at
least
1.5
standard
deviations
above
the
mean
for
behavioral
problems
were
coded
as
having
a
behavioral
disability.
These
children
were
coded
as
having
a
behavioral
disability
regardless
of
scores
on
non-behavioral
assessments.
Child
and
family
characteristics.
Demographic
characteristics
associated
with
placement
disruption
discussed
in
the
intro-
duction,
including
child
age,
gender,
race,
and
the
primary
caregiver’s
self-reported
health
status
were
controlled
for
in
the
analyses.
Poverty
was
also
examined
and
was
measured
by
calculating
the
family’s
income-to-needs
ratio,
which
was
esti-
mated
by
dividing
family
income
by
its
corresponding
poverty
threshold
in
1999.
The
poverty
threshold
varies
by
family
size
and
is
based
on
the
money
necessary
for
the
minimally
accepted
amounts
of
food
(Bishaw
&
Iceland,
2003).
Income-to-needs
ratio
within
the
sample
ranged
from
.19
to
6.15,
with
1.00
representing
the
overall
poverty
threshold.
Those
families
that
fell
below
the
1.00
threshold
were
coded
as
being
in
poverty.
Analysis
All
analyses
for
the
current
study
were
performed
using
STATA
Statistical
Software
Release
10.
Due
to
the
complex
sam-
pling
design,
STATA’s
survey
commands
were
applied
to
obtain
unbiased
estimates
of
the
population
parameters
(NSCAW
Research
Group,
2002).
All
percentages
are
adjusted,
or
weighted,
for
sample
probabilities.
The
association
of
disruption
with
child
characteristics
(age,
gender,
race,
and
disability
status),
the
health
of
the
primary
caregiver,
and
family
character-
istics
(placement
type
and
poverty
level)
were
examined
with
bivariate
contingency
table
analysis
using
the
Pearson
2test.
To
account
for
the
complex
survey
design,
2statistics
were
converted
to
F-statistics
with
noninteger
degrees
of
freedom
using
a
second-order
Rao
and
Scott
correction
(StataCorp,
2003).
It
is
possible
that
disruption
rates
vary
by
different
child,
caregiver,
and
family
characteristics
that
are
associated
with
the
placements
type.
Therefore,
bivariate
analysis
of
child,
caregiver,
and
placement
characteristics
associated
with
kin
or
non-kin
placement
were
also
examined.
Logistic
regression
analysis
modeled
disruption
as
a
function
of
placement
type
while
controlling
for
other
influential
characteristics
of
the
child,
caregiver,
and
family.
Interaction
effects
between
placement
and
disability
status,
along
with
interactions
between
significant
child,
caregiver,
or
family
characteristics,
were
entered
in
a
second
logistic
regression
model.
Missing
data
were
limited
to
the
family
poverty
measure
and
therefore
did
not
require
adjustments
in
the
analysis:
22
out
of
315
cases
(or
7%)
were
missing.
Results
Table
1
shows
the
demographic
characteristics
of
the
sample
and
the
rate
of
disruption
3
years
following
the
baseline
interview.
Overall,
a
little
over
1
in
4
placements
disrupted
between
baseline
and
the
3
year
follow-up
interview.
A
substan-
tially
greater
proportion
of
placements
with
older
children
compared
to
younger
children
disrupted,
2(2,
n
=
315)
=
6.30,
p
=
.01.
Over
5
times
as
many
children
living
with
a
non-kinship
care
placement
at
baseline
disrupted
within
3
years
com-
pared
to
placements
with
kinship
care,
2(1,
n
=
315)
=
25.01,
p
=
.001.
There
were
no
differences
in
placement
disruption
rates
between
the
four
categories
of
disability.
Table
2
presents
child,
caregiver,
and
family
characteristics
by
type
of
placement.
Only
poverty
status
was
associated
with
placement
type,
with
4
times
more
kinship
families
reporting
being
below
the
federal
poverty
line
compared
to
non-kinship
families,
2(1,
n
=
293)
=
9.24,
p
=
.01.
Also
of
note
was
the
absence
of
a
difference
in
placement
type
by
the
four
category
disability
status.
Table
3
presents
a
logistic
regression
model
predicting
any
placement
disruption
between
baseline
and
3
years.
Living
with
a
kin
caregiver
decreased
the
odds
of
disruption
.07
times
compared
to
living
with
a
non-kin
caregiver
for
children
in
out-
of-home
care.
Type
of
disability
had
no
effect
on
placement
disruption.
However,
having
a
behavioral
disability
approached
significance
(p
=
.07)
and
increased
the
odds
of
disruption
3.8
times
compared
to
no
disability.
Three
models
were
added
to
the
analysis,
alternating
reference
groups
between
the
4
categories
of
disability
to
assess
the
relative
risk
of
disruption:
960 J.J.
Helton
/
Child
Abuse
&
Neglect
35 (2011) 956–
964
Table
1
Baseline
child,
caregiver,
and
family
characteristics
by
placement
disruption
(N
=
315).
Percent
(SE) Total
Stable
Disrupted
Total
74
(4.9)
26
(4.9)
Child
characteristics
Age**
3–5
31 (7.0)
79 (7.8)
21 (7.8)
6–8
40 (7.2)
86 (4.1)
14 (4.1)
9–10
29
(5.8)
50
(10.2)
50
(10.2)
Gender
Male
45
(7.2)
70
(8.0)
30
(8.0)
Female
55
(7.2)
76
(5.7)
24
(5.7)
Race
African-American
41
(7.4)
71
(8.5)
29
(8.5)
Caucasian
36
(6.0)
75
(6.4)
25
(6.4)
Hispanic
16
(6.5)
79
(15.0)
21
(15.0)
Other 7 (2.8)
73 (13.2)
27 (13.2)
Disability
No 42
(7.4)
82
(5.3)
18
(5.3)
Non-behavioral
17
(4.0)
73
(9.7)
27
(9.7)
Behavioral
19
(5.7)
61
(15.2)
39
(15.2)
Multiple
22
(6.6)
68
(11.8)
32
(11.8)
Caregiver
characteristics
In
poor
health
Yes
24
(5.8)
80
(8.2)
20
(8.2)
No
76
(5.8)
72
(6.0)
28
(6.0)
Family
characteristics
Placement
type**
Kinship 71 (5.4)
88
(3.9)
12
(3.9)
Foster
29
(5.4)
38
(8.8)
62
(8.8)
Below
federal
poverty
line
Yes
15
(4.6)
81
(10.9)
19
(10.9)
No
85
(4.6)
73
(5.8)
27
(5.8)
Note.
SE
=
standard
error.
** p
<
.01.
non-behavioral
and
all
others,
behavioral
and
all
others,
and
non-behavioral/behavioral
and
all
others.
No
significant
effects
were
found;
the
results
are
not
reported.
Interaction
effects
between
kin
placement
and
disability
status
and,
because
of
the
significant
bivariate
relationship,
child
age
and
disability
status
were
entered
into
the
logistic
regression
model.
No
interaction
effects
were
found
between
kin
placement,
disability
status,
and
disruption.
However,
for
each
additional
year
of
age,
having
a
behavioral
or
both
a
non-behavioral
and
behavioral
disability
increased
the
odds
of
disruption
3.14
and
2.08
times
for
children
in
out-of-home
care.
Discussion
This
study
was
predicated
on
the
theory
that
kin
caregivers
possess
a
stronger
sense
of
responsibility
and
affinity
for
biologically
related
children
placed
under
their
care
compared
to
other
caregivers
fostering
non-biologically
related
children.
It
is
that
biological
relatedness
that
encourages
caregiving
resiliency
during
demanding
or
difficult
situations,
thus
shielding
children
in
kinship
care
from
disruption.
Like
previous
research
(James,
2004;
Testa,
2001;
Usher,
Randolph,
&
Gogan,
1999;
Webster,
Barth,
&
Needell,
2000),
the
results
for
the
total
sample
supported
this
hypothesis,
showing
that
children
who
lived
with
kin
were
at
a
low
risk
for
disruption
within
the
first
3
years
of
out-of-home
placement.
This
study
further
hypothesized
that
placement
stability
with
kin
may
not
be
as
secure
for
children
with
different
and
more
severe
types
of
burdensome
caregiving
needs.
This
was
not
supported,
as
kinship
placements
fostering
children
with
different
types
of
disabilities
at
baseline
were
not
more
or
less
likely
to
disrupt
compared
to
children
without
a
disability.
In
the
end,
kinship
placements
were
as
stable
as
non-kinship
placements
for
a
majority
of
children,
regardless
of
disability
status.
Placements
with
older
children
with
either
a
behavioral
disability
only
or
a
combination
of
behavioral
and
non-behavioral
disabilities
were
more
likely
to
disrupt
compared
to
younger.
These
findings
mirror
results
from
previous
empirical
work
on
biological
families
of
children
with
behavioral
and
non-behavioral
disabilities.
For
instance,
a
longitudinal
study
of
children
with
different
developmental
disabilities
by
Hauser-Cram,
Warfield,
Shonkoff,
and
Krauss
(2001)
found
that
parental
stress
increased
steadily
for
biological
mothers
and
fathers
alike
as
their
child
with
behavioral
problems
grew
older,
but
parental
stress
was
consistently
lower
over
the
years
for
parents
of
children
with
a
developmental
disability
without
behavioral
problems.
Likewise,
over
a
2
year
period
of
time,
Baker
et
al.
(2003)
found
that
increased
parenting
stress
was
associated
with
increased
maladaptive
child
behavior
for
parents
of
children
with
developmental
disabilities,
but
not
for
parents
of
children
without
a
disability.
The
increased
levels
of
parental
stress
could
be
due
to
the
difficulty
caring
for
the
physical
J.J.
Helton
/
Child
Abuse
&
Neglect
35 (2011) 956–
964 961
Table
2
Baseline
child,
caregiver,
and
family
characteristics
by
placement
type.
Percent
(SE)
Kin
Non-kin
Total
Child
characteristics
Age
3–5
36 (9.0)
18 (4.9)
6–8
41 (9.3)
38 (8.7)
9–10
23
(6.5)
44
(9.0)
Gender
Male
44
(9.5)
47
(9.2)
Female
56
(9.5)
53
(9.2)
Race
African-American
44
(9.3)
33
(9.4)
Caucasian
33
(7.5)
45
(9.2)
Hispanic
17
(8.7)
14
(8.4)
Other 6 (3.5)
8 (4.0)
Disability
No 47
(9.4)
29
(8.5)
Non-behavioral
15
(5.5)
21
(5.3)
Behavioral
18
(7.1)
23
(8.5)
Multiple
20
(8.8)
27
(7.0)
Caregiver
characteristics
In
poor
health
Yes
26
(7.4)
21
(9.5)
No
74
(7.4)
79
(9.5)
Family
characteristics
Below
federal
poverty
line**
Yes 19 (6.5)
4
(1.7)
No
81
(6.5)
96
(1.7)
Note.
SE
=
standard
error.
** p
<
.01.
needs
of
older
children
compared
to
younger
children
with
certain
disabilities
inhibiting
physical
movement
and
daily-
living
tasks.
Add
to
this
difficulty
child
problems
with
externalizing
behaviors,
such
as
defiant
and
impulsive
conduct,
or
internalizing
problems,
such
as
anxiety
and
depression,
and
the
combination
could
increase
parental
frustration
and
strengthen
perceptions
of
unresponsiveness.
Further,
the
lack
of
developmental
success
coupled
with
increased
behavioral
problems
has
been
shown
to
decrease
caregiver
feelings
of
parenting
competence
as
the
child
grows
older
(Hauser-Cram
et
al.,
2001).
A
strength
of
this
study
was
the
expansion
of
disability
into
social,
functional,
and
language
domains.
These
types
of
non-
behavioral
disabilities
have
been
generally
ignored
in
previously
research,
as
authors
tended
to
focus
on
children
with
only
behavioral
problems
(Aaron
et
al.,
2010;
Barth
et
al.,
2007;
Chamberlain
et
al.,
2006;
Leathers,
2006).
To
my
knowledge,
no
previous
study
has
looked
at
the
separate
risk
factors
for
children
with
a
behavioral
compared
to
a
non-behavioral
disability.
Another
strength
was
the
operational
definition
of
placement
moves
as
disruption,
meaning
a
non-permanency
move
in
out-of-home
care.
This
created
an
outcome
measure
that
better
captured
the
type
of
detrimental
bouncing
between
homes
that
harms
so
many
children
in
out-of-home
care
(Rubin
et
al.,
2007).
The
limitations,
on
the
other
hand,
included
the
age
restriction
placed
on
the
sample
selection
due
to
the
standardized
disability
measures.
It
would
have
been
beneficial
to
look
at
children
over
the
age
of
10
with
the
same
design,
as
adolescents
may
be
even
more
likely
to
disrupt
compared
to
younger
children
(Barth
et
al.,
2007).
Furthermore,
the
decrease
in
sample
size
from
5,501
to
315,
due
to
the
standardized
measure
restriction
and
type
of
out-of-home
placement,
limits
the
generalizability
of
the
findings.
Another
limitation
of
this
study
was
the
absence
of
data
on
why
a
placement
disrupted.
It
was
unclear
whether
the
disruption
was
actually
a
planned
or
routine
system
or
policy
related
move,
such
as
placement
with
a
sibling,
move
from
a
temporary
placement,
the
closure
of
a
foster
home,
and
so
forth
(James,
2004).
It
was
also
unknown
why
a
child
was
placed
in
either
a
kinship
or
non-kinship
home
post-investigation.
Because
the
baseline
interview
occurred
4
months,
on
average,
after
the
close
of
the
investigation,
some
children
could
have
already
experienced
a
placement
disruption
prior
to
interview.
Additionally,
baseline
caregivers
may
not
have
been
the
first
choice
of
caseworkers,
but
a
second
option
given
the
child’s
disability
status.
However,
recent
qualitative
research
has
shown
that
the
presence
of
cognitive
and
physical
disabilities
would
not
preclude
foster
parents
from
taking
in
a
child
(Rosenwald
&
Bronstein,
2008).
This
study
has
implications
for
child
welfare
policy
and
practice.
The
results
show
that
a
majority
of
children,
regardless
of
type
of
disability,
will
find
their
kin
placement
as
stable
as
children
without
a
disability
placed
with
kin.
Efforts
to
reassure
caseworkers
that
placing
children
with
any
type
of
disability
with
a
kin
caregiver
are
warranted
given
that
caseworkers
are
more
likely
to
recommend
placement
of
children
with
a
disability
into
out-of-home
care
after
substantiated
allegations
compared
to
children
without
a
disability
(Manders
&
Stoneman,
2009),
but
have
reported
being
reluctant
to
utilize
available
962 J.J.
Helton
/
Child
Abuse
&
Neglect
35 (2011) 956–
964
Table
3
Logistic
regression
model
predicting
placement
disruption
between
baseline
and
3
years.
OR 95%
CI
OR
95%
CI
Child
characteristics
Age
1.02
[.78,
1.33]
1.04
[.81,
1.32]
Male
1.37
[.49,
3.86]
1.05
[.29,
3.76]
Race
African-American 2.36
[.75,
7.44] 2.38
[.80,
7.09]
Caucasian
Hispanic
.80
[.21,
3.00]
.79
[.22,
2.79]
Other
1.35
[.49,
3.71]
1.23
[.42,
3.67]
Disability
No
Non-behavioral
1.60
[.30,
8.45]
1.02
[.02,
57.82]
Behavioral
3.80
[.88,
16.45]
.02
[.01,
9.32]
Multiple
2.37
[.75,
7.53]
.03
[.01,
2.22]
Caregiver
characteristics
In
poor
health
.60
[.11,
3.17]
.64
[.15,
2.81]
Family
characteristics
Placement
Kinship
.07** [.02,
.24]
.12*[.02,
.74]
Foster
Below
poverty
line 1.46
[.20,
10.98] 1.68
[.26,
10.67]
Interactions
Placement
and
disability
Kin
*
Non-Beh
1.21
[.12,
12.61]
Kin
*
Beh
.06
[.01,
1.59]
Kin
*
Mult
.19
[.02,
2.85]
Age
and
disability
status
Age
*
Non-Beh
1.04
[.61,
1.79]
Age
*
Beh 3.14*[1.14,
8.60]
Age
*
Mult
2.08*[1.12,
3.86]
Note.
OR
=
odds
ratio;
CI
=
confidence
interval.
*p
<
.05.
** p
<
.01.
kin
(Beeman
&
Boisen,
1999;
Peters,
2004).
Providing
disability-specific
training
to
caseworkers
authorized
to
place
children
into
out-of-home
care
has
been
reported
by
caseworkers
as
needed
(Orelove,
Hollahan,
&
Myles,
2000)
and
may
ease
this
reluctance
to
place
children
with
a
disability
with
kin.
In
fact,
better
knowledge
of
children’s
disabling
conditions
may
help
match
children
with
a
disability
with
more
appropriate
non-kin
caregivers,
possibly
lowering
the
disruption
rates
for
this
group.
Understanding
how
caseworkers
decide
which
placement
is
appropriate
for
a
child
with
a
disability
could
also
help
develop
policy
to
appropriately
match
caregivers
with
children
with
different
needs.
Conclusion
By
utilizing
standardized
measures
of
both
behavioral
and
non-behavioral
disabilities
within
a
large
national
sample
of
children
who
had
been
recently
placed
in
out-of-home
care,
this
study
presents
new
evidence
that
placement
with
kin
represents
an
appropriately
stable
placement
for
children
3–10
years
of
age,
regardless
of
disability
status.
Data
from
this
study
suggest
that
placing
children
with
any
type
of
disability
with
kin
may
be
an
appropriate
placement
decision
based
on
their
low
likelihood
of
disruption.
However,
additional
evidence
on
the
well-being
of
children
with
various
types
of
disabilities
and
levels
of
functionality
living
with
kin
is
needed.
These
children
present
unique
challenges
for
the
foster
care
system,
challenges
that
may
have
direct
impact
on
service
utilization
and
indicators
of
physical,
psychological,
and
emotional
well-being
yet
examined.
References
Aaron,
G.
A.,
James,
S.,
Monn,
A.
R.,
Raghavan,
R.,
Wells,
R.
S.
&
Leslie,
L.
K.
(2010).
Behavior
problems
and
placement
change
in
a
national
child
welfare
sample:
A
prospective
study.
Journal
of
the
American
Academy
of
Child
&
Adolescent
Psychiatry,
49(1),
70–80.
Achenbach,
T.
M.
(1991).
Manual
for
the
Child
Behavior
Checklist.
Burlington,
VT:
University
of
Vermont
Department
of
Psychiatry.
Ammerman,
R.
T.
&
Patz,
R.
J.
(1996).
Determinants
of
child
abuse
potential:
Contribution
of
parent
and
child
factors.
Journal
of
Clinical
Child
Psychiatry,
25,
300–307.
Baker,
B.
L.,
McIntyre,
L.
L.,
Blacher,
J.,
Crnic,
K.,
Edelbrock,
C.
&
Low,
C.
(2003).
Pre-
school
children
with
and
without
developmental
delay:
Behavior
problems
and
parenting
stress
over
time.
Journal
of
Intellectual
Disability
Research,
47,
217–230.
Barth,
R.
P.,
Lloyd,
E.
C.,
Green,
R.
L.,
James,
S.,
Leslie,
L.
K.
&
Landsverk,
J.
(2007).
Predictors
of
placement
moves
among
children
with
and
without
emotional
and
behavioral
disorders.
Journal
of
Emotional
and
Behavioral
Disorders,
15(1),
46–55.
Beeman,
S.
&
Boisen,
L.
(1999).
Child
welfare
professionals’
attitudes
towards
kinship
foster
care.
Child
Welfare,
78(3),
315–337.
J.J.
Helton
/
Child
Abuse
&
Neglect
35 (2011) 956–
964 963
Beeman,
S.
K.,
Kim,
H.
&
Bullerdick,
S.
K.
(2000).
Factors
affecting
placement
of
children
in
kinship
and
non-kinship
foster
care.
Children
and
Youth
Services
Review,
22(1),
37–54.
Benedict,
M.
I.,
Zuravin,
S.,
Somerfield,
M.
&
Brandt,
D.
(1996).
The
reported
health
and
functioning
of
children
maltreated
while
in
family
foster
care.
Child
Abuse
&
Neglect,
20(7),
561–571.
Berrick,
J.
D.
(1997).
Assessing
quality
of
care
in
kinship
and
foster
family
care.
Family
Relations,
46,
273–280.
Bishaw,
A.
&
Iceland,
J.
(2003).
Poverty:
1999.
In
Census
2000
brief.
U.S.
Department
of
Commerce.
Chamberlain,
P.
(1996).
Treatment
foster
care
for
adolescents
with
conduct
disorders
and
delinquency:
The
Oregon
Social
Learning
Center
Monitor
Program.
In
E.
D.
Hibbs,
&
P.
S.
Jensen
(Eds.),
Psychological
treatments
for
child
and
adolescent
disorders
(pp.
475–490).
Washington,
DC:
American
Psychological
Association.
Chamberlain,
P.,
Price,
J.
M.,
Reid,
J.
B.,
Landsverk,
J.,
Fisher,
P.
A.
&
Stoolmiller,
M.
(2006).
Who
disrupts
from
placement
in
foster
and
kinship
care?
Child
Abuse
&
Neglect,
30,
409–424.
Chang,
J.
&
Liles,
R.
(2007).
Characteristics
of
four
kinship
placement
outcome
groups
and
variables
associated
with
these
kinship
placement
outcome
groups.
Child
and
Adolescent
Social
Work
Journal,
24(6),
509–522.
Ehrle,
J.
&
Geen,
R.
(2002).
Kin
and
non-kin
foster
care:
Findings
from
a
national
survey.
Children
and
Youth
Services
Review,
24,
15–35.
Gebel,
T.
J.
(1996).
Kinship