Beck AT, Epstein N, Brown G, Steer RA. An Inventory for Measuring Clinical Anxiety: Psychometric Properties. J Consult Clin Psychol 56: 893-897

Abstract

The development of a 21-item self-report inventory for measuring the severity of anxiety in psychiatric populations is described. The initial item pool of 86 items was drawn from three preexisting scales: the Anxiety Checklist, the Physician’s Desk Reference Checklist, and the Situational Anxiety Checklist. A series of analyses was used to reduce the item pool. The resulting Beck Anxiety Inventory (BAI) is a 21-item scale that showed high internal consistency (α = .92) and test—retest reliability over 1 week, r (81) = .75. The BAI discriminated anxious diagnostic groups (panic disorder, generalized anxiety disorder, etc.) from nonanxious diagnostic groups (major depression, dysthymic disorder, etc). In addition, the BAI was moderately correlated with the revised Hamilton Anxiety Rating Scale, r (150) = .51, and was only mildly correlated with the revised Hamilton Depression Rating Scale, r (153) = .25.

Full text

Journal
of
Consulting
and
Clinical
Psychology
1988,
Vol.
56, No. 6,
893-897
Copyright
1988
by the
American
Psychological
Association,
Inc.
0022-006X/88/$00.75
An
Inventory
for
Measuring Clinical Anxiety:
Psychometric
Properties
Aaron
T.
Beck
Department
of
Psychiatry
University
of
Pennsylvania
Gary
Brown
Department
of
Psychiatry
University
of
Pennsylvania
Norman
Epstein
Department
of
Family
and
Community Development
University
of
Maryland
Robert
A.
Steer
Department
of
Psychiatry,
School
of
Osteopathic
Medicine
University
of
Medicine
and
Dentistry
of New
Jersey
The
development
of
a
21
-item self-report inventory
for
measuring
the
severity
of
anxiety
in
psychiat-
ric
populations
is
described.
The
initial item pool
of 86
items
was
drawn
from
three preexisting
scales:
the
Anxiety Checklist,
the
Physician's Desk
Reference
Checklist,
and the
Situational
Anxiety
Checklist.
A
series
of
analyses
was
used
to
reduce
the
item
pool.
The
resulting Beck
Anxiety
Inven-
tory
(BAI)
is a
21
-item scale that showed high internal consistency
(at
=
.92)
and
test-retest
reliability
over
1
week,
r(81)
=
.75.
The BAI
discriminated anxious diagnostic groups (panic disorder, general-
ized
anxiety disorder,
etc.)
from
nonanxious
diagnostic groups
(major
depression,
dysthymic
disor-
der,
etc).
In
addition,
the BAI was
moderately correlated
with
the
revised Hamilton Anxiety Rating
Scale,
r(150)
=
.51,
and was
only
mildly correlated
with
the
revised Hamilton Depression Rating
Scale,
r(153)
=
.25.
Studies
addressing
the
distinctiveness
of
anxiety
and
depres-
sion depend
on the
availability
of
reliable
and
valid assessment
instruments.
However,
a
number
of
studies
have
reported high
correlations
(r >
.50) between
the
widely
used rating scales
of
anxiety
and
depression (e.g., Dobson, 1985; Mendels,
Weinstein,
&
Cochrane,
1
972;
Moumjoy&
Roth,
1982;Prusoff
&
Klerman,
1974; Riskind,
Beck,
Brown,
&
Steer, 1987;
Ta-
naka-Matsumi
&
Kameoka,
1986). These
findings
raise
the
question,
Are the
high correlations
due to a
genuine shared
symptomatology,
or do
they simply
reflect
a
lack
of
discrimi-
nant validity? Consequently,
to the
extent
that
a
given
study
fails
to
differentiate
anxiety
from
depression,
it is not
possible
to
know
whether anxiety
and
depression
are
truly
indistinguish-
able
or
whether
the
results simply reflect
the
shortcomings
of
the
instruments used
to
measure
the two
syndromes.
A
possible contributing factor
to the
lack
of
discriminant
va-
lidity
is the
inclusion
of
anxiety
and
depression symptoms
on
measures
of
both
syndromes
(Lipman,
1982; Riskind, Beck,
Brown,
&
Steer, 1987). When
emphasis
is
placed
on
theoretical
(e.g.,
Spielberger,
Gorsuch,
&
Lushene,
1970)
and
clinical (e.g.,
Hamilton,
1959, 1960; Zung,
1971)
considerations
in the
early
stages
of
clinical
test
construction,
the
discriminant validity
of
each test item
is
often
overlooked. Discriminant validity
is
fre-
quently addressed
in the
later stages
of
test construction when
attention
has
shifted
to
total
scores
and
away
from
individual
test
items (e.g., Zung,
1
97
1
).
A
post
hoc
approach
to
reducing overlapping
symptomatol-
This study
was
supported
by
National Institute
of
Mental Health
Grant
MH38843
to
Aaron
T.
Beck
and by the
Foundation
for
Cognitive
Therapy.
We
gratefully
acknowledge
the
contribution
of
Paul McDer-
mott.
Correspondence
concerning
this
article should
be
addressed
to
Aaron
T.
Beck, Center
for
Cognitive Therapy, Room
602,
133
South 36th
Street, Philadelphia, Pennsylvania
19104.
ogy
across
measures
of
anxiety
and
depression
has
involved
shifting
items
to the
more relevant
scale
by
using
applicable
external
criteria.
For
example, Riskind, Beck, Brown,
and
Steer
(1987)
found
that
the
Hamilton Rating Scales
for
Anxiety
and
Depression (Hamilton, 1959, 1960)
contained
overlapping
items
and
produced significantly correlated
scores.
When
the
authors revised
the
scales
by
deleting
nondiscriminating
items
and
transferring
other
items
to
more
appropriate
scales,
the new
scales were less correlated
and
discriminated
better
between
pa-
tients
with primary anxiety
and
depression
diagnoses.
On
the
assumption that validity should
be
built
into
the
test
from
the
outset, other test constructors
have
used
a
sequential
or
multistage approach
to
test
construction
(Anastasi,
1986;
Jackson,
1970;
Millon,
1983).
This strategy
was
followed
in the
present study
to
develop
a new
instrument
for
the
measurement
of
clinical anxiety,
the
Beck
Anxiety
Inventory
(BAI).
The BAI
was
developed
to
address
the
need
for an
instrument
that
would
reliably
discriminate anxiety
from
depression
while displaying
convergent validity. Such
an
instrument
would
offer
advantages
for
clinical
and
research purposes over existing
self-report
mea-
sures
of
anxiety, such
as the
State-Trait
Anxiety Inventory
(STAI;
Spielberger
et
al.,
1970)
and the
Self-Rating Anxiety
Scale (SRAS; Zung,
1971),
which have
not
been shown
to
differentiate
anxiety
from
depression adequately (e.g.,
Dobson,
1985;
Tanaka-Matsumi
&
Kameoka, 1986).
Method
Subjects
Three samples
of
psychiatric outpatients were drawn
from
consecu-
tive
routine
evaluations
at the
Center
for
Cognitive Therapy
in
Philadel-
phia,
Pennsylvania,
from
successive time periods beginning
in
early
1980
and
lasting until late
1986.
The
total sample size
was
1,086.
The
patients were either
self-referred
or
referred
by
other professionals.
There
were
456 men
(42%; mean
age
=
36.35,
SD
=
12.41)
and 630
893
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.