Content uploaded by Jonathan Davidson
Author content
All content in this area was uploaded by Jonathan Davidson on Dec 29, 2017
Content may be subject to copyright.
Research Article
DEVELOPMENT OF A NEW RESILIENCE SCALE:
THE CONNOR-DAVIDSON RESILIENCE SCALE (CD-RISC)
Kathryn M. Connor, M.D.,
n
and Jonathan R.T. Davidson, M.D.
Resilience may be viewed as a measure of stress coping ability and, as such, could
be an important target of treatment in anxiety, depression, and stress reactions.
We describe a new rating scale to assess resilience. The Connor-Davidson
Resilience scale (CD-RISC) comprises of 25 items, each rated on a 5-point scale
(0–4), with higher scores reflecting greater resilience. The scale was
administered to subjects in the following groups: community sample, primary
care outpatients, general psychiatric outpatients, clinical trial of generalized
anxiety disorder, and two clinical trials of PTSD. The reliability, validity, and
factor analytic structure of the scale were evaluated, and reference scores for
study samples were calculated. Sensitivity to treatment ef fects was examined
in subjects from the PTSD clinical trials. The scale demonstrated good
psychometric properties and factor analysis yielded five factors. A repeated
measures ANOVA showed that an increase in CD-RISC score was associated
with greater improvement during treatment. Improvement in CD-RISC score
was noted in proportion to overall clinical global improvement, with greatest
increase noted in subjects with the highest global improvement and deterioration
in CD-RISC score in those with minimal or no global improvement. The CD-
RISC has sound psychometric properties and distinguishes between those with
greater and lesser resilience. The scale demonstrates that resilience is modifiable
and can improve with treatment, with greater improvement corresponding to
higher levels of global improvement. Depression and Anxiety 18:76–82, 2003.
&2003 Wiley-Liss, Inc.
Key words: resilience; stress coping; wellbeing; posttraumatic stress disorder;
anxiety; depression
INTRODUCTION
Resilience embodies the personal qualities that enable
one to thrive in the face of adversity. Research over the
last 20 years has demonstrated that resilience is a
multidimensional characteristic that varies with con-
text, time, age, gender, and cultural origin, as well as
within an individual subjected to dif ferent life circum-
stances [e.g., Garmezy, 1985; Garmezy and Rutter,
1985; Rutter et al., 1985; Seligman and Csikszentmi-
halyi, 2000; Werner and Smith, 1992]. One theory for
this variability was developed by Richardson and
colleagues, who proposed the following resiliency
model [Richardson et al., 1990; Richardson, 2002].
Beginning at a point of biopsychospiritual balance
(‘‘homeostasis’’), one adapts body, mind, and spirit to
current life circumstances. Internal and external
stressors are ever-present and one’s ability to cope
with these events is influenced by both successful and
unsuccessful adaptations to previous disruptions. In
some situations, such adaptations, or protective
DEPRESSION AND ANXIETY 18:76–82 (2003)
Department of Psychiatry and Behavioral Sciences, Duke
University Medical Center, Durham, North Carolina
Contract grant sponsor: Smith Kline Beecham; Contract grant
sponsor: Pfizer Pharmaceuticals; Contract grant sponsor: Pure
World Botanicals, Inc.; Contract grant sponsor: Organon; Con-
tract grant sponsor: NIH; Contract grant number: R01 MH56656-
01A1
n
Correspondence to: Dr. Connor, Box 3812, DUMC, Durham, NC
27710. E-mail: kathryn.connor@duke.edu
Received for publication 15 September 2002; Accepted 1 April 2003
DOI: 10.1002/da.10113
Published online in Wiley InterScience (www.interscience.wiley.com).
&& 2003 WILEY-LISS, INC.
factors, are ineffective, resulting in disruption of the
biopsychospiritual homeostasis. In time, response to this
disruption is a reintegrative process, leading to one of
four outcomes: (1) the disruption represents an oppor-
tunity for growth and increased resilience, whereby
adaptation to the disruption leads to a new, higher level
of homeostasis; (2) a return to baseline homeostasis, in
an effort to just get past or beyond the disruption; (3)
recovery with loss, establishing a lower level of home-
ostasis; or 4) a dysfunctional state in which maladaptive
strategies (e.g., self-destructive behaviors) are used to
cope with stressors. Resilience may thus also be viewed
as measure of successful stress-coping ability.
The clinical relevance of resilience and related
constructs has been noted previously. Maddi and
Khoshaba theorized that hardiness was an index of
mental health [Maddi and Khoshaba, 1994] and recent
data has supported this hypothesis [Ramanaiah et al.,
1999]. Tsuang [2000] has emphasized the substantial
clinical implications that follow a better understanding
of the forces that mould resilience. With regard to
trauma and posttraumatic stress disorder (PTSD), it
has been shown that hardiness contributes to protec-
tion against developing chronic PTSD after combat
[King et al., 1998; Waysman et al., 2001].
The growing focus on health promotion and well-
being, shifting emphasis away from pathology and
problem-orientation, provides an opportunity to revisit
the role of resilience in health. Yet there is relatively
little awareness about resilience or its importance in
clinical therapeutics. Conventionally, therapeutic trials
have focused more heavily on measuring morbidity,
although quality of life elements are now included in
many trials. A number of scales have been developed to
measure resilience [Bartone et al., 1989; Wagnild and
Young, 1993] or aspects of resilience [e.g., hardiness:
Hull et al., 1987, Kobasa, 1979; perceived stress, Cohen
et al., 1983]. However, these measures have neither
been widely used nor applied to specific populations
[Carlson, 2001; Mosack, 2002] and thereby lack
generalizability. Of striking note, a textbook of psy-
chiatric measures recently published by the American
Psychiatric Association contains not a single resilience
measure [American Psychiatric Association, 2000].
The need for well-validated measures of resilience
that are simple to use is thus evident. While several
scales have been developed, they have not gained wide
acceptance and no one scale has established primacy.
With these considerations in mind, the Connor-
Davidson Resilience Scale (CD-RISC) was developed
as a brief self-rated assessment to help quantify resilience
and as a clinical measure to assess treatment response.
METHODS
SCALE DEVELOPMENT
We recently became interested in the concept of
resilience as being relevant to treatment outcome in
anxiety, depression, and stress reactions. This interest
arose in part from a finding that fluoxetine produced
greater therapeutic benefit on stress coping than
placebo in PTSD [Connor et al., 1999]. Furthermore,
in reviewing the account of Sir Edward Shackleton’s
heroic expedition in the Antarctic in 1912 [Alexander,
1998], it was noted that the expedition’s leader
possessed many personal characteristics compatible
with resilience and that this may perhaps have
contributed to the successful survival of each member
of the expedition in the face of overwhelming odds.
Together, these observations prompted the authors to
undertake the development of a short self-rated
resilience measure.
The content of the scale was drawn from a number
of sources. From Kobasa’s work with the construct of
hardiness [Kobasa, 1979], items reflecting control,
commitment, and change viewed as challenge were
included. The following features were drawn from
Rutter’s work [Rutter, 1985]: developing strategy with a
clear goal or aim, action orientation, strong self-
esteem/confidence, adaptability when coping with
change, social problem solving skills, humor in the
face of stress, strengthening effect of stress, taking on
responsibilities for dealing with stress, secure/stable
affectional bonds, and previous experiences of success
and achievement (these last two features may reflect the
underpinnings of resilience). From Lyons [1991], items
assessing patience and the ability to endure stress or
pain were included. Lastly, from Shackleton’s experi-
ences, it was noted that the role of faith and a belief in
benevolent intervention (‘‘good luck’’) were likely
important factors in the survival of the expedition,
suggesting a spiritual component to resilience. Table 1
summarizes the salient features of resilience.
With the above considerations, the CD-RISC was
constructed, with the following goals in mind: to
develop a valid and reliable measure to quantify
TABLE 1: Characteristics of resilient people
Reference Characteristic
Kobasa, 1979 View change or stress as a challenge/opportunity
Kobasa, 1979 Commitment
Kobasa, 1979 Recognition of limits to control
Rutter, 1985 Engaging the support of others
Rutter, 1985 Close, secure attachment to others
Rutter, 1985 Personal or collective goals
Rutter, 1985 Self-eff icacy
Rutter, 1985 Strengthening effect of stress
Rutter, 1985 Past successes
Rutter, 1985 Realistic sense of control/having choices
Rutter, 1985 Sense of humor
Rutter, 1985 Action oriented approach
Lyons, 1991 Patience
Lyons, 1991 Tolerance of negative affect
Rutter, 1985 Adaptability to change
Current Optimism
Current Faith
Research Article: Resilience Scale: (CD-RISC) 77
resilience, to establish reference values for resilience in
the general population and in clinical samples, and to
assess the modifiability of resilience in response to
pharmacologic treatment in a clinical population.
The CD-RISC contains 25 items, all of which carry
a 5-point range of responses, as follows: not true at all
(0), rarely true (1), sometimes true (2), often true (3),
and true nearly all of the time (4). The scale is rated
based on how the subject has felt over the past month.
The total score ranges from 0–100, with higher scores
reflecting greater resilience. The individual items
comprising the scale are listed in Table 2.
STUDY SAMPLE
Subjects were drawn from the following study
samples: a random-digit dial based general population
sample [i.e., non help-seeking (Group 1, n¼577;
included subjects with complete data only); primary
care outpatients (Group 2, n¼139); psychiatric out-
patients in private practice (Group 3, n¼43); subjects
in a study of generalized anxiety disorder (GAD;
Group 4, n¼25); and subjects in two clinical trials of
PTSD (Group 5, n¼22; Group 6, n¼22)]. Of note,
subjects in Group 6 are included only for between-
group diagnostic comparisons and in the assessment of
pre- to post-treatment change. Each study protocol
was approved by the Duke University Medical Center
Institutional Review Board and all subjects provided
informed consent.
Demographic characteristics of Groups 1–5 (n¼
806) were as follows: female 65% (n¼510), male
35% (n¼274); white 77% (n¼588), non-white 23%
(n¼181); and mean (sd) age 43.8 (15.3) years (n¼763).
Some missing data occurred for all of these compar-
isons, which explains why the figures do not total 806
in the various comparisons (e.g., data were not always
available for gender, ethnic status, etc.).
DATA ANALYSIS
The data were analyzed with the following objec-
tives: (1) to establish reference scores for the CD-RISC
and to assess whether scores were af fected by clinical
category or demographic factors, (2) to assess the
reliability and validity of the scale, (3) to assess the
factor composition of the CD-RISC in the general
population, and (4) to assess the extent to which CD-
RISC scores can change with clinical improvement
with treatment and over time.
Given that several of the samples were not normally
distributed, median CD-RISC scores were calculated
for each group and pairwise comparisons were per-
formed using the Wilcoxon Rank Sum test, with
Po.05 being regarded as significant. A Bonferroni
correction was used for multiple comparisons to derive
the zscore. Of note, mean CD-RISC scores are also
presented for clinical reference. A Kruskal-Wallis test
was used for multiple group comparisons, with the
expectation that degrees of resilience would be lower in
psychiatric outpatients than in the general population
or primary care patients.
Descriptive statistics were used to characterize
CD-RISC scores in the full sample by gender,
ethnicity, and age. Analysis of variance was used
to analyze categorical variables (e.g., gender and
ethnicity) and correlation with the continuous measure
of age.
The reliability and validity of the scale were assessed
as follows. Test retest reliability was examined in
subjects from Groups 4 and 5 in whom no clinical
change was noted between two consecutive visits.
Internal consistency was evaluated by using Cronbach’s
alpha for the total and item-total scores in subjects
from Group 1. Convergent validity was assessed in
various groups by correlating the CD-RISC with
measures of hardiness [Kobasa Hardiness Scale; Koba-
sa et al., 1979], perceived stress [Perceived Stress Scale
(PSS-10); Cohen et al., 1983], and stress vulnerability
[Stress Vulnerability Scale (SVS); Sheehan et al., 1990],
as well as measures of disability [Sheehan Disability
Scale(SDS); Sheehan et al., 1983] and social support
[Sheehan Social Support Scale (SSSS); Sheehan, 1990].
Divergent validity was assessed by correlating CD-
RISC scores with the Arizona Sexual Experience
Scale [ASEX; McGahuey et al., 2000] in subjects from
Group 4.
TABLE 2: Content of the Connor-Davidson Resilience
Scale
Item no. Description
1 Able to adapt to change
2 Close and secure relationships
3 Sometimes fate or God can help
4 Can deal with whatever comes
5 Past success gives confidence for new challenge
6 See the humorous side of things
7 Coping with stress strengthens
8 Tend to bounce back after illness or hardship
9 Things happen for a reason
10 Best effort no matter what
11 You can achieve your goals
12 When things look hopeless, I don’t give up
13 Know where to turn for help
14 Under pressure, focus and think clearly
15 Prefer to take the lead in problem solving
16 Not easily discouraged by failure
17 Think of self as strong person
18 Make unpopular or diff icult decisions
19 Can handle unpleasant feelings
20 Have to act on a hunch
21 Strong sense of purpose
22 In control of your life
23 I like challenges
24 You work to attain your goals
25 Pride in your achievements
Connor and Davidson78
An exploratory factor analysis using an ORTHO-
MAX rotation was conducted by using data from the
general population sample (Group 1).
The effects of time and treatment on resilience were
assessed by comparing pre- and post-treatment CD-
RISC scores in treatment responders and non-respon-
ders in the clinical trial samples (Groups 4, 5, and 6) by
using a repeated measures analysis of variance (ANO-
VA), with response as the grouping variable and time as
the repeated measure. Response was defined by a
Clinical Global Improvement (CGI-I; Guy; 1976)
score of 1 (very much improved) or 2 (much improved).
RESULTS
CD-RISC SCORES BY CLINICAL CATEGORY
AND DEMOGRAPHIC GROUP
Mean (sd) and median (1st, 4th quartile) CD-RISC
scores were calculated for the full sample (Groups 1–5)
and for the individual study groups (Table 3). Results of
pairwise comparisons are listed in Table 4 and
significant dif ferences were found for the following
groups: general population (Group 1) vs. each of
the other groups, primary care (Group 2) vs. GAD
(Group 4), and primary care vs. PTSD (Groups 5
and 6). Statistical significance was obtained in the
overall multiple comparison model (w
2
¼142.80, df ¼5,
Po.0001).
Mean (sd) scores were also calculated by demo-
graphic grouping, and no dif ferences were observed in
the characteristics evaluated. A gender comparison
revealed a mean score of 77.1 (16.3) for women and
77.2 (14.2) for men (P¼.63). Mean CD-RISC scores by
racial group were as follows: white subjects, 77.4 (14.8)
and non-white subjects, 76.7 (18.1) (P¼.83). The
mean (sd) age of the full sample was 43.8 (15.4) years,
and no correlation was found between age and CD-
RISC score (Pearson r¼.06, n.s.).
RELIABILITY AND VALIDITY
Internal consistency. Cronbach’s afor the full
scale was 0.89 for Group 1 (n¼577) and item-total
correlations ranged from 0.30 to 0.70 (Table 5).
Test–retest reliability. Test–retest reliability was
assessed in 24 subjects from the clinical trials of GAD
(Group 4) and PTSD (Group 5) in whom little or no
clinical change was observed from time 1 to time 2.
The mean (sd) CD-RISC scores at time 1 [52.7 (17.9)]
and time 2 [52.8 (19.9)] demonstrated a high level
of agreement, with an intraclass correlation coef ficient
of 0.87.
Convergent validity. CD-RISC scores were posi-
tively correlated with the Kobasa hardiness measure in
psychiatric outpatients (Group 3, n¼30; Pearson
r¼0.83, Po.0001). Compared to the Perceived Stress
Scale (PSS-10), the CD-RISC showed a significant
negative correlation (Group 3, n¼24; Pearson
r¼0.76, Po.001), indicating that higher levels
of resilience corresponded with less perceived
stress. The Sheehan Stress Vulnerability Scale (SVS)
was similarly negatively correlated with the CD-RISC
(Spearman r¼0.32, Po.0001) in 591 subjects
from the combined sample. This result also indicates
that higher levels of resilience correspond to
lower levels of perceived stress vulnerability. As a
measure of disability, the CD-RISC demonstrated a
TABLE 3: Connor-Davidson Resilience Scale scores by
study group
Study group
Group
no. NMean (sd)
Median
(1st, 4th Q)
General population 1 577 80.4 (12.8) 82 (73, 90)
Primary care 2 139 71.8 (18.4) 75 (60, 86)
Psychiatric outpatients 3 43 68.0 (15.3) 69 (57, 79)
GAD patients 4 24 62.4 (10.7) 64.5 (53, 71)
PTSD patients 5 22 47.8 (19.5) 47 (31, 61)
6 22 52.8 (20.4) 56 (39, 61)
GAD ¼generalized anxiety disorder; PTSD ¼posttraumatic stress
disorder.
TABLE 4: Pairwise comparisons of Connor-Davidson
Resilience Scale scores
Group
n
Mean rank
difference
Critical rank
difference
Statistically
significant
difference
nn
Group1 vs.
Group2 114.70 66.36 Yes
Group3 193.80 111.02 Yes
Group4 290.50 146.31 Yes
Group5 362.80 152.56 Yes
Group6 329.70 152.56 Yes
Group2 vs.
Group3 79.10 122.55 No
Group4 175.80 155.24 Yes
Group5 248.10 161.15 Yes
Group6 215.00 161.15 Yes
Group3 vs.
Group4 96.70 178.95 No
Group5 169.00 184.09 No
Group6 135.90 184.09 No
Group4 vs.
Group5 72.30 207.29 No
Group6 39.20 207.29 No
Group5 vs .
Group6 33.10 211.75 No
n
Group 1¼general population; Group 2¼primary care; Group
3¼psychiatric outpatients; Group 4¼GAD clinical trial subjects;
Groups 5 and 6¼PTSD clinical trial subjects.
nn
ao.05; Bonferonni correction used to derive zscore; z¼2.94
GAD ¼generalized anxiety disorder; PTSD ¼posttraumatic stress
disorder.
Research Article: Resilience Scale: (CD-RISC) 79
significant negative correlation with the Sheehan
Disability Scale (SDS) (Pearson r¼0.62,
Po.0001) in psychiatric patients (Groups 3 and 4,
n¼40). Lastly, the Sheehan Social Support Scale
(SSS) correlated significantly with the CD-RISC in
589 subjects (Spearman r¼; 0.36, Po.0001). Thus,
greater resilience, as expected, is associated with less
disability and greater social support.
Discriminant validity. The CD-RISC was not
significantly correlated with the ASEX at baseline
(Group 4, n¼23; r¼0.34, P¼.11) or at endpoint
(n¼19; r¼0.30, P¼.21).
FACTOR ANALYSIS
Analysis of data from subjects in the general
population sample yielded five factors whose eigenva-
lues were, respectively, 7.47, 1.56, 1.38, 1.13, and
1.07. These factors could be broadly interpreted in
the following manner. Factor 1 reflects the notion
of personal competence, high standards, and tenacity.
Factor 2 corresponds to trust in one’s instincts,
tolerance of negative af fect, and strengthening effects
of stress. Factor 3 relates to the positive accept-
ance of change, and secure relationships. Factor 4
was related to control and Factor 5 to spiritual
influences. The factor pattern for the scale is presented
in Table 5.
SENSITIVITY TO THE EFFECTS OF
TREATMENT
In subjects with PTSD (Groups 5 and 6), non-
responders (n¼30) had mean (sd) pre and post
treatment scores of 54.0 (16.5) and 54.9 (18.8),
respectively. Among responders (n¼19), mean
pre- and post-treatment scores were 56.8 (18.4) and
68.9 (19.8), respectively. Significant effects were
observed for time (F¼17.36; df 1, 47; Po.0001)
and for time response category (F¼12.87; df 2, 47;
Po001), indicating that CD-RISC scores increased
significantly with overall clinical improvement.
Greater improvement was noted in CD-RISC
score in proportion to the degree of global clinical
improvement. For example, in subjects with a CGI-I
score of 1 (n¼7), there was a mean increase of 19.9
(26.6%) in the CD-RISC score, compared to an
increase of 7.9 (16.2%) for those with a CGI-I score
of 2 (n¼7), and a deterioration of 0.8 (1.3%) in those
with a CGI-I of 3 or more (minimal or no improve-
ment; n¼18) (F¼3.42, df 2, Po.05). Significant
effects for time (F¼14.82; df 2, 29; P¼.006) and for
TABLE 5: Item-total correlations and rotated factor pattern for the Connor-Davidson Resilience Scale
Factor (Eigenvalue)
Item Item-total correlation
n
1 (7.436) 2 (1.563) 3 (1.376) 4 (1.128) 5 (1.073)
24 0.61 0.70870 0.14250 0.04339 0.19253 0.01779
12 0.62 0.63998 0.22255 0.20851 0.05018 0.11083
11 0.62 0.62497 0.11656 0.13206 0.21732 0.06408
25 0.56 0.60385 0.04385 0.14600 0.22531 0.11798
10 0.52 0.59601 0.17001 0.16642 0.03336 0.10776
23 0.59 0.55800 0.32628 0.00758 0.12202 0.04681
17 0.70 0.40381 0.35512 0.12714 0.35236 0.00409
16 0.62 0.39651 0.37804 0.26274 0.18958 0.03547
20 0.40 0.08774 0.67393 0.05234 0.06238 0.23265
18 0.58 0.29395 0.57585 0.01006 0.19034 0.08147
15 0.57 0.29967 0.53047 0.04440 0.23134 0.01552
6 0.58 0.11507 0.52564 0.40443 0.12267 0.03711
7 0.55 0.14586 0.46703 0.30584 0.01699 0.27429
19 0.64 0.17227 0.43428 0.27115 0.39728 0.01199
14 0.64 0.25215 0.42942 0.26572 0.36228 0.10734
1 0.55 0.07334 0.08512 0.75885 0.10762 0.03223
4 0.64 0.07074 0.19156 0.61921 0.40002 0.02811
5 0.69 0.26961 0.37932 0.55332 0.09561 0.08239
2 0.36 0.23482 0.08203 0.53775 0.14060 0.31552
8 0.67 0.34423 0.34073 0.43996 0.16462 0.04038
22 0.63 0.21396 0.12493 0.09219 0.77469 0.02935
13 0.62 0.15177 0.03725 0.20513 0.54772 0.40077
21 0.64 0.36495 0.15438 0.02278 0.53186 0.32889
3 0.30 0.01386 0.01460 0.15972 0.15786 0.77820
9 0.40 0.12061 0.24612 0.00029 0.05145 0.73662
n
Calculated from standardized variables; Chronbach’s a¼0.93.
Connor and Davidson80
time CGI group effect (F¼7.70; df 2, 29; P¼.002)
were noted.
DISCUSSION
The CD-RISC has been tested in the general
population, as well as in clinical samples, and
demonstrates sound psychometric properties, with
good internal consistency and test–retest reliability.
The scale exhibits validity relative to other measures of
stress and hardiness, and reflects different levels of
resilience in populations that are thought to be
differentiated, among other ways, by their degree of
resilience (e.g., general population vs. patients with
anxiety disorders). Clinical improvement with even
short-term pharmacotherapy in patients with PTSD, a
condition with a propensity toward heightened vulner-
ability to the effects of stress, is accompanied by up to
25% or greater increase in resilience, depending upon
level of global improvement. Furthermore, subjects
with PTSD who showed very much improvement
attained CD-RISC scores close to the mean of the
general population. To the authors’ knowledge, this
is the first demonstration that increased resilience,
as operationally defined, can be associated with a
pharmacologic intervention.
Three areas can be identified where the CD-RISC
might be usefully applied. A number of investigators
have considered possible biologic aspects of resilience.
For example, resilience is characterized by a response
profile to major stress in which low baseline catecho-
laminergic activity is transformed into high catechola-
mine production, along with increased tissue-specific
response (e.g., glucose levels) and an attenuated
cortisol response [Dienstbier, 1991]. Gormley [2000]
has opined that SSRI drugs may facilitate this
process in depressive, obsessive-compulsive, and panic
disorders but provides no actual evidence in support of
his assertion. The authors have shown previously that
fluoxetine has such an effect in PTSD [Connor et al.,
1999]. It is also possible that relationships exist
between resilience and central serotonergic function
[Andrews et al., 1988; Healey and Healey, 1996]. Thus,
the CD-RISC might prove useful in studies of the
biology of resilience.
A second application of the scale could be in clinical
practice with contemporary resiliency interventions.
Such interventions explore resilience qualities with
individuals, identify them, and nurture them [Rak,
2002]. In focusing on strengths and positive attributes,
an individual tends to become engaged in more
adaptive pursuits, and their problems tend to diminish.
The CD-RISC is compatible with such interventions,
as an aid to identifying resilient characteristics but also
in assessing response to the intervention.
A third potential application of the scale might be in
studies designed to investigate adaptive and maladap-
tive strategies for coping with stress, and as a tool to
assist in screening individuals for high-risk, high-stress
activities or occupations. For example, resilience
(hardiness) was identified as a strong predictor of
protection from PTSD in a cohort of Vietnam veterans
[King et al., 2000]. Lyons [1991] noted a strengthening
effect of extreme trauma in many trauma survivors and
a scale such as the CD-RISC might be useful in
studying such individuals.
The authors note several limitations of this
report. The CD-RISC is a wave two resilience
measure, using the scheme outlined by Richardson
[Richardson, 2002], assessing characteristics of resi-
lience, and does not assess the resiliency process or
provide information about the theory of resilience.
While divergent validity was demonstrated, the mea-
sure used to assess divergence (ASEX, a measure of
sexual functioning) was weakly, albeit nonsignificantly,
correlated with CD-RISC and this finding most likely
reflects the heterogeneity of the resilience construct.
The CD-RISC has not been validated against an
objective (i.e., behavioral or third party) measure, or
against biological measures of resilience, such as
neuropeptide Y responses to extreme stress [Morgan
et al., 1999]. The authors also recognize that it is
possible to perform well in one area in the face of
adversity (e.g., work) but to function poorly in another
(i.e., interpersonal relationships). Would such a person
be considered resilient? Furthermore, resilience may
either be a determinant of response or an effect of
exposure to stress. Assessment of such directional
factors was not undertaken in this report. A prospective
study would be able to inform whether resilience pre-
dated exposure to trauma, protected against post-
trauma problems, or, if through circumstances, some
survivors developed further resilience post-trauma.
CONCLUSIONS
The CD-RISC is a brief, self-rated measure of
resilience that has sound psychometric properties. By
using the CD-RISC, the findings of this study
demonstrate the following: resilience is quantifiable
and influenced by health status (i.e., individuals with
mental illness have lower levels of resilience than the
general population); resilience is modifiable and can
improve with treatment; and greater improvement in
resilience corresponds to higher levels of global
improvement. The CD-RISC could have potential
utility in both clinical practice and research.
Acknowledgements We thank Larry Tupler and
Erik Churchill for their statistical support and Dr. George
Parkerson for facilitating access to primary care subjects.
REFERENCES
Alexander C. 1998. The Endurance: Shackleton’s legendary antarctic
expedition. New York: Alfred A. Knopf.
American Psychiatric Association. 2000. Handbook of psychiatric
measures. Washington, DC: American Psychiatric Association.
Research Article: Resilience Scale: (CD-RISC) 81
Andrews W, Parker G, Barrett E. 1998. The SSRI antidepressants:
exploring their ‘‘other’’ possible properties. J Af fect Disorder
49:141–144.
Bartone PT, Ursano R, Wright K, Ingraham L. 1989. The impact of
military air disaster on the health of assistance workers. J Nerv
Mental Dis 177:317–328.
Carlson DJ. 2001. Development and validation of a College
Resilience Questionnaire. Dissertation Abstracts International, A
(Humanities and Social Sciences). vol 62, Jan 2001, 20025.
Cohen S, Kamarck T, Mermelstein R. 1983. A global measure of
perceived stress. J Health Soc Behav 24:386–396.
Connor KM, Sutherland SM, Tupler LA, Churchill LE,
Malik ML, Davidson JRT. 1999. Fluoxetine in posttraumatic
stress disorder: a randomized, placebo-controlled trial. Br J Psych
175:17–22.
Dienstbier RA. 1991. Behavioral correlates of sympathoadrenal
reactivity: the toughness model. Med Sci Sports Med 23:846–852.
Garmezy N. 1985. Stress resistant children: the search for protective
factors. In: Recent research in developmental psychopathology,
book suppl number 4 to J Child Psychol Psych. Oxford: Pergamon
Press.
Garmezy N, Rutter M. 1985. Acute stress reactions. In: M Rutter, L
Hersob, editors. Child and adolescent psych: modern approaches.
Oxford: Blackwell.
Gormley N. 2000. Is neuroticism a modifiable risk factor for
depression? Ir J Psych Med 17:41–42.
Healey D, Healey H. 1998. The clinical pharmacologic profile of
reboxetine: does it involve the putative neurobiological substrates
of wellbeing? J Affect Disord 51:313–322.
Hull JG, Van Treuren RR, Virnelli S. 1987. Hardiness and health: a
critique and alternative approach. J Personality Soc Psychol
53:518–530.
King LA, King DW, Fairbank JA, Keane TM, Adams GA. 1998.
Resilience-recovery factors in post-traumatic stress disorder among
female and male Vietnam veterans: hardiness, postwar social
support, and additional stressful life events. J Personality Soc
Psychol 74:420–434.
Kobasa SC. 1979. Stressful life events, personality, and health: an
inquiry into hardiness. J Personality Soc Psychol 37:1–11.
Lyons J. 1991. Strategies for assessing the potential for positive
adjustment following trauma. J Traumatic Stress 4:93–111.
Maddi SR, Khoshaba DM. 1994. Hardiness and mental health. J Pers
Assess 63:265–274.
McGahuey CA, Gelenberg AJ, Laukes CA, Moreno FA, Delgado PL.
2000. The Arizona Sexual Experience Scale (ASEX): reliability and
validity. J Sex Marital Therapy 26:25–40.
Morgan CA III, Wang S, Southwick SM, Rasmusson A, Hazlett G,
Hauger RL, Charney DS. 2000. Plasma neuropeptide-Y concen-
trations in humans exposed to military survival training. Biol
Psychiatry 47:902–909.
Mosack KE. 2002. The development and validation of the R-PLA: a
resiliency measure for people living with HIV/AIDS (immune
deficiency). Dissertation Abstracts International: Section B: the
Sciences and Engineering. vol 62, Mar 2002, 3844.
Rak CF. 2002. Heroes in the nursery: three case studies in resilience.
J Clin Psychol 58:247–260.
Ramanaiah NV, Sharpe JP, Byravan A. 1999. Hardiness and major
personality factors. Psychol Rep 84:497–500.
Richardson GE. 2002. The metatheory of resilience and resiliency.
J Clin Psychol 58:307–321.
Richardson GE, Neiger B, Jensen S, Kumpfer K. 1990. The
resiliency model. Health Education 21:33–39.
Rutter M. 1985. Resilience in the face of adversity: protective factors
and resistance to psychiatric disorders. Br J Psych 147:598–611.
Seligman MEP, Csikszentmihalyi M. 2000. Positive psychology. Am
Psychologist 55:5–14.
Sheehan DV. 1983. The Anxiety Disease. New York: Bantam Books.
Sheehan DV, Raj AB, Harnett Sheehan K. 1990. Is buspirone
effective for panic disorder? J Clin Psychopharmacol 10:3–11.
Tsuang MT. 2000. Genes, environment and mental health wellness.
Am J Psychiatry 157:489–491.
Wagnild GM, Young HM. 1993. Development and psychometric
validation of the Resilience Scale. J Nurs Meas 1:165–178.
Waysman M, Schwarzwald J, Solomon Z. 2001. Hardiness: an
examination of its relationship with positive and negative long-
term changes following trauma. J Traumatic Stress 14:531–548.
Werner E, Smith R. 1992. Overcoming odds: high risk children from
birth to adulthood. Ithaca, NY: Cornell University Press.
Connor and Davidson82
