Rehabilitation Psychology © 2010 American Psychological Association

2010, Vol. 55, No. 1, 12–22 0090-5550/10/$12.00 DOI: 10.1037/a0018415

Resilience in the Face of Coping With a Severe Physical Injury: A Study

of Trajectories of Adjustment in a Rehabilitation Setting
Anette Johansen Quale and Anne-Kristine Schanke
Sunnaas Rehabilitation Hospital and University of Oslo

Objective: Despite the popularity of the concept of resilience, little research has been conducted on
populations in physical rehabilitation settings. Our purpose was to identify three trajectories of psycho-
logical adjustment to an acquired severe physical injury characterized by resilience, recovery, or distress
in a longitudinal design. Participants: Eighty inpatients with a severe injury at a rehabilitation hospital.
The participants had spinal cord injury or multiple traumas. Design: Classification into the three
trajectories was based on symptoms of psychological distress (posttraumatic stress disorder, depression,
anxiety, and negative affect) and participants’ level of positive affect at admission to and discharge from
the rehabilitation hospital. Results: The most common trajectory was the resilience trajectory (54%),
followed by the recovery trajectory (25%) and the distress trajectory (21%). The most interesting
differences between the trajectories were the result of optimism, affect, social support, and pain. Trait
negative and positive affect predicted classification into the trajectories. Conclusions: An adaptation
pattern characterized by resilience was found to be the most common response to an acquired severe
injury, and trait affect predicts the outcome pattern. Interventions based on resilience are discussed.

Keywords: severely injured, spinal cord injuries, resilience, trajectories, positive psychology

Recent research has claimed that most individuals facing differ- Evans, 2001). Because of the extreme adversity these individuals
ent kinds of trauma are resilient (Bonanno et al., 2008; Deshields, experience, there is a need for research that examines the role of
Tibbs, Fan, & Taylor, 2006; Hobfoll et al., 2009), but there seems resilience in this population (Campbell-Sills, Cohan, & Stein,
to be a paucity of literature examining the resilience of individuals 2006; White et al., 2008). Our aims in this study were to estimate
in physical rehabilitation settings. Until now, it has been unclear the prevalence of resilience in a rehabilitation setting and to
whether individuals who acquire a severe physical injury with a identify possible predictors for such a psychologically healthy
sequel of disability (e.g., a spinal cord injury [SCI]) in a life- outcome.
threatening event, such as a motor vehicle accident, have the same
capacity to remain psychologically healthy and to return to previ- Resilience in Physical Rehabilitation
ous levels of emotional functioning as individuals experiencing
other potentially traumatic events after which physical function A widely held cultural assumption is that acquiring a severe
returns to normal (White, Driver, & Warren, 2008). An individual physical injury, with a sequel of functional loss, is a devastating
who has acquired a severe physical injury must face both the event and that a return to normal life is improbable. Individuals
trauma that created the injury and the loss of physical function. with disabilities are often met with strong prejudice and pity in
The greater severity and the necessity to remain in a rehabilitation society (Schanke, 2004). However, those individuals with a severe
hospital for a substantial period posttrauma are likely to influence injury are themselves often surprised by their ability to adjust to
the expression of distress symptoms in this population (Kennedy & the physical change and to learn that happiness and living a
relatively normal life is possible (Dunn, Uswatte, & Elliott, 2009).
This is in accordance with the insider– outsider distinction, and
Anette Johansen Quale and Anne-Kristine Schanke, Department of social psychologists within the field of rehabilitation psychology
Research, Sunnaas Rehabilitation Hospital, Nesoddtangen, Norway, and have emphasized this topic (Dunn, 2009). Insiders (people with
Department of Psychology, University of Oslo. disability) know what disability is like, whereas outsiders (people
This project has been financed by a 3-year grant from the Norwegian without disability) often equate disability with illness and may
Foundation for Health and Rehabilitation through the Norwegian Associ- therefore make erroneous assumptions about the experience of
ation of Disabled. Thanks to our colleagues Ingar Larsen, Anita Kjeverud, disability. An important difference between the two perspectives is
and Toril S. Ødegårdstuen for their contribution to the planning of the that the insiders generally see their situations in more favorable
project and for data collection. Thanks to Kathrine Frey Frøslie for her terms than do outsiders and they take positives into account in their
statistical advice. We also thank Morten Hestnes and Nils Oddvar Skaga at
troubling situations (Wright, 1991).
the Trauma Registry at Ullevaal University Hospital for scoring and
permitting use of the injury-related data from the Trauma Registry.
Instead of letting the insiders’ (those who know what its like)
Correspondence concerning this article should be addressed to Anette perspective inform research and practice, it seems as though the
Johansen Quale, Master of science, cand.psychol, Department of Research, outsiders (who appear to have difficulties with accepting that many
Sunnaas Rehabilitation Hospital, 1450 Nesoddtangen, Norway. E-mail: individuals with disabilities adjust to their circumstances reason-
[email protected] or [email protected] ably well; Dunn et al., 2009) have set the standards of research

12
 RESILIENCE IN PHYSICAL REHABILITATION SETTINGS 13

within rehabilitation psychology. Therefore, there has traditionally severe physical injury. According to Mancini and Bonanno (2006),
been an exaggerated focus on identifying risk factors and vulner- some individuals suffer from chronic distress, recurrent intrusive
abilities. This research has been important for the advancement of memories, or sadness for years after such experiences. Others have
our knowledge of why some individuals cope poorly with such life a recovery process of first suffering more acute reactions and then
events, to identify these individuals at an early stage, and to give gradually returning to former levels of functioning. Still others
the right intervention. Fortunately, the trend is now shifting, and it show a resilient adjustment with surprisingly short-lived reactions
is increased interest in the concept of resilience and the assessment and a relatively rapid return to their previous level of functioning
of positives in terms of human strengths that individuals use to (Mancini & Bonanno, 2006). Being resilient does not mean that a
cope with life challenges (Collard & Kennedy, 2007; White et al., person does not experience difficulties or distress. Some emotional
2008). The trend is now more in line with the insiders’ perspective, pain and sadness is common in individuals who have experienced
to emphasize what individuals with severe injuries can do rather major adversity in their lives. Even though individuals manifest
what they cannot do, to strengthen their positive abilities, and to resilience in their behavior and life patterns, it is argued that
learn from the ones who are coping well. resilience is not a trait that someone either has or does not have
Resilience is a broad conceptual umbrella, and the construct (Luthar, 2003). Resilience involves behaviors, thoughts, and ac-
refers to important psychological skills and to the individual’s tions that anyone can learn and develop (American Psychological
ability to use family, social, and external support to cope better Association Help Center, 2004). Thus, it is crucial that the level of
with stressful events (Campbell-Sills et al., 2006; Friborg, 2005). resilience be detected and evaluated as early as possible in the
Most research, in terms of theoretical and conceptual work within rehabilitation process because there is a greater chance of increas-
the resilience field, is conducted with child populations. Far less is ing resilience and rehabilitation outcomes (Luthar, Cicchetti, &
known about how resilience operates in adulthood. It is crucial to Becker, 2000).
differentiate between resilience in children and resilience in adults One common assessment of human strength is optimism. Ac-
because childhood resilience is typically understood in response to cording to Carver, Scheier, Miller, and Fulford (2009), optimists
enduring challenging environments, whereas, by comparison, expect good things to happen and are persistent in trying to reach
adult resilience is more often a matter of coping with an isolated their goals, even when things are hard. However, pessimists expect
and usually brief stressor event. According to George Bonanno
bad outcomes, try to escape the adversity by wishful thinking, and
(2004), resilience is defined
use temporary distractions that do not help solve the problem.
Martin Seligman (1998)and the field of positive psychology have
as the ability of adults in otherwise normal circumstances who are
exposed to an isolated and potential highly disruptive event such as the defined optimism not only as a trait but as a more active charac-
death of a close relation or a violent or life-threatening situation to teristic and as constructive cognitions about the future and have
maintain relatively stable, healthy levels of psychological and physical emphasized that optimism is a human strength that can be built in
functioning . . . as well as the capacity for generative experiences and therapy and that optimists are exposed to the same disappoint-
positive emotions. (pp. 20 –21)
ments and tragedies as pessimists, but they are handling them
better. Growing evidence in the literature indicates that optimistic
On the basis of Bonnanno’s (2004) definition, we define resil-
individuals respond to adversity in more adaptive ways than do
ience in the rehabilitation setting as the ability of adults who are
facing a severe and potentially disabling physical injury to main- pessimists (Carver et al., 2009). Individuals with an optimistic
tain relatively stable, healthy levels of psychological and social attitude about their rehabilitation are usually the same individuals
functioning and to maintain positive emotions and a positive who ward off the depression and hopelessness that can compro-
perception of self and the future. Resilience is a two-dimensional mise progress in therapy (Seligman & Csikszentmihalyi, 2000).
construct concerning the exposure to adversity and the positive Detecting whether the individual displays an optimistic or a pes-
adjustment outcomes to that adversity. Adversity refers to any risks simistic pattern might help predict rehabilitation outcomes and
associated with negative life events that are related to difficulties secondary complications because optimists appear to take action to
with adjustment. Obtaining a severe physical injury is statistically minimize health risks, know more about risk factors, and make
related to difficulties with adjustment and psychological distress more proactive efforts (Carver et al., 2009).
such as posttraumatic stress disorder (PTSD; Kennedy & Duff, Some studies have argued that positive emotions, or positive
2001; Kennedy & Evans, 2001; Starr et al., 2004), anxiety affect, help resilient individuals to construct the psychological
(Kennedy & Rogers, 2000), and depression (Elliott & Frank, 1996; resources that are necessary for coping successfully with signifi-
Kennedy & Rogers, 2000). Yet, some individuals manage to cant catastrophe, such as the September 11, 2001, terrorist attacks
remain psychologically healthy and show a positive adaptation. In (Tugade & Fredrickson, 2004). As a result, positive emotion
the rehabilitation setting, it is not uncommon to see individuals experienced by resilient individuals functions as a protective factor
who have acquired severely disabling injuries adapting to their to moderate the magnitude of adversity to individuals and assists
new situation in positive ways, by reorienting themselves and them to cope well in the future. Conversely, if an individual
adjusting their goals and ambitions in life, despite the adversity exhibits behaviors associated with low resilience (e.g., anger,
that has led to the personal burden and functional and social isolation, substance abuse), then there is an increased likelihood
limitations. Positive adaptation can be explained as doing substan- that the individual will continue to function in a state of disruption
tially better than what would be expected given the exposure to an and adapt poorly to the situation (White et al., 2008). Individuals
adversity. who do not display a resilient adaptation exhibit the difficulties of
There are marked individual differences in the way in which regulating negative emotions and demonstrate sensitive reactions
people react to and cope with aversive events, such as acquiring a to stressful life events.
14 QUALE AND SCHANKE

Severe Physical Injuries ience and recovery because the term recovery connotes a trajectory
in which normal functioning temporarily gives way to threshold or
Multiple trauma (MT) and SCI are viewed as severe physical subthreshold psychopathology (e.g., symptoms of depression or
injuries because in most cases they lead to functional loss and PTSD), usually for a period of several months, and then gradually
disability. Individuals with MT or SCI may perceive trauma as a returns to pre-event levels. By contrast, resilience reflects the
threat to their physical existence and as a violation of their social ability to maintain a stable equilibrium. The third trajectory,
and personal integrity. This results in feelings of stress and vul- named distress, was characterized by high levels of distress
nerability as they confront the possibility of their own mortality. throughout the rehabilitation process. To separate recovery trajec-
Events including the incident, ambulance journey, surgical proce- tories from resilient and distress trajectories may be especially
dures, or transfer to an intensive care unit are sudden, unexpected, important in a physical rehabilitation setting because the medical
unfamiliar, and frightening and occur at a time when the individ- trauma is extensive and physical improvements that may affect the
ual’s abilities to comprehend and adapt have been severely com- psychological reactions are expected. In studies similar to the
promised (Mohta, Sethi, Tyagi, & Mohta, 2003). In this study, we present one (Bonanno et al., 2008; Hobfoll et al., 2009), a fourth
operationalized resilience in the face of coping with a severe trajectory has been identified, the delayed onset trajectory. Be-
physical injury as a response pattern consisting of major adversity cause other studies have found that delayed onset of reactions
because of the severe physical injury (with the loss of physical
occurs in only 2%–3% of the total sample (for a review, see
functioning) in the context of minor disruption of psychosocial
Bonanno & Field, 2001), we decided not to take this perspective
functioning and low emotional burden (such as no symptoms of
into account here.
anxiety, depression, or PTSD).
Previously, resilience has been defined according to PTSD
symptoms (Bonanno et al., 2007) and according to depression
Trajectories of Resilience symptoms (Bonanno et al., 2002; Deshields et al., 2006). However,
recently it has been argued that resilience is more than just the
The recent literature has proposed that resilience may be best absence of psychopathology (Bonanno, 2004) and that it, in addi-
understood and measured as one member of a set of trajectories tion, includes the capacity for positive affect (Bonanno et al.,
that may follow adversity and potentially traumatic events 2005). Studies with patients with rheumatoid arthritis (Strand et
(Bonanno et al., 2008; Deshields et al., 2006; Hobfoll et al., al., 2006) and with chronic pain (Zautra, Johnson, & Davis, 2005)
2009). Given that passage of time is an important factor in have concluded that positive affect appears to be an important
adapting to a severe physical injury, it seems appropriate to resilience factor within these populations. Positive affect has also
study different trajectories and examine the process of adapta- been shown to be a meaningful predictor of long-term bereave-
tion over time. By studying trajectories, which describe change ment outcome (Bonanno et al., 2005). Our operationalization is
over time, with data from both the beginning and the end of based both on symptoms of PTSD, anxiety, depression, and state
inpatient rehabilitation, one can look at the process of adjust- negative affect and on state positive affect, thus incorporating both
ment rather than at an endpoint. The absence or presence of the most frequently reported symptoms of psychological distress in
psychopathology at one point in time does not ensure that it will this population and a more specific concept of resilience, namely
or will not occur later. positive affect. Although it has not previously been used, we
George Bonanno and his colleagues have identified trajecto-
wanted to investigate whether this operationalization is meaningful
ries of psychological functioning or prototypical outcome pat-
in detecting severely injured individuals’ pattern of psychological
terns among bereaved individuals (Bonanno, Moskowitz, Papa,
adaptation.
& Folkman, 2005), among individuals who have experienced a
potentially traumatic event (Bonanno, Galea, Bucciarelli, &
Vlahov, 2007), and among hospitalized survivors of the severe Aims of the Study
acute respiratory syndrome epidemic (Bonanno et al., 2008).
Others have studied trajectories among a sample under an
The main aim of this study was to estimate the prevalence of
ongoing threat of mass causalities (Hobfoll et al., 2009) and
resilience, recovery, and distress trajectories in individuals with a
among breast cancer survivors after the end of treatment
severe injury during inpatient rehabilitation. Relatively little is
(Deshields et al., 2006). Yet another study examined trajecto-
known about human strengths in rehabilitation and how resilient
ries by personality prototypes in a sample of individuals with
SCIs (Berry, Elliott, & Rivera, 2007). individuals manage to cope so well. Our clinical impression is that
Building on the work of Bonanno and his colleagues, we out- many of our patients in the rehabilitation hospital show a resilient
lined three trajectories in a longitudinal design. The first trajectory, adaptation and that these patients are more likely to exhibit posi-
named the resilience trajectory, was characterized by no symp- tive adaptive behaviors to the injury and the event (e.g., setting
toms of major distress over time. Resilient individuals generally goals, accepting support from others). Moreover, we wanted to
exhibit a stable trajectory of healthy functioning across time, as examine differences between the three trajectories and try to
well as the capacity for generative experiences and positive emo- identify possible predictors of the trajectories. We see it as impor-
tions (Bonanno, 2004). The second trajectory, named the recovery tant to establish methods to detect resilience and to find out how
trajectory, was characterized by initial symptoms of distress with individuals displaying a resilient adaptation are different from
a significant decrease in symptoms scores later in the rehabilitation those who are not. Finally, we discuss implications for resilience-
process. Bonanno (2004) emphasized a distinction between resil- based interventions in rehabilitation settings.
 RESILIENCE IN PHYSICAL REHABILITATION SETTINGS 15

Method Table 1
Descriptions of the Classification in the Three Trajectories
Participants Group Admission Discharge

Resilience Below threshold on the symptoms The same as at

The study was conducted at Sunnaas Rehabilitation Hospital scales for posttraumatic stress admission
(SRH), the most specialized rehabilitation hospital in Norway. The symptoms (PTSS), anxiety,
study population consisted of inpatients, between 16 and 68 years depression and state negative
affect and above threshold for
old, admitted to SRH during a 2-year period from February 2003
state positive affect
to January 2005. People with MT, SCI, or both were consecutively Recovery Above threshold on at least one Below threshold on at
included. Exclusion criteria were severe cognitive deficits or prac- of the symptoms scales for least one scale that
tical problems with carrying out the interview because of severe PTSS, anxiety, depression, and indicated the
state negative affect or below individual was
medical complications, psychosis, or other major psychiatric ill- threshold for state positive above threshold at
ness or foreigners who had insufficient understanding of the Nor- affect admission
wegian language. We excluded 14 patients on the basis of these Distress Above threshold on at least one The same as at
exclusion criteria. Twenty-three patients admitted during this pe- of the symptoms scales for admission
PTSS, anxiety, depression, and
riod were not included because of vacations and turnover among low state negative affect or
the psychologists or for unknown reasons. This left us with 110 below threshold for high state
patients eligible for participation. Of these, nine refused to partic- positive affect
ipate and 21 missed the follow-up at discharge. We then ended up
with a total of 80 participants and a response rate of 73%.
A MT was defined by having a New Injury Severity Score
(NISS) higher than 15 and at least two injuries classified in the Assessments
Abbreviated Injury Scale (Association for the Advancement of
Automotive Medicine, 1998). A NISS score higher than 15 is Psychological interview. The semistructured psychological
considered a serious injury. SCI was defined as damage to the interview was developed to assess a holistic view of the partici-
spinal cord resulting in a loss of function such as loss of mobility pants’ situation, in addition to demographics such as age, gender,
or sensation. All individuals with SCI who came to SRH were marital status, occupational status, and level of education. The
included. This included participants with both traumatically ac- interview was developed by a research group of five experienced
quired SCI (i.e. accidents) and non–traumatically acquired SCI psychologists and was based on clinical experiences and relevant
(i.e., caused by infections and vascular lesions). literature.
Satisfaction with social support was measured on a visual ana-
log scale ranging from 0 to 10. International Classification of
Design Functioning, Disability and Health criteria were followed to diag-
nose PTSD (World Health Organization, 1992). To explore par-
Semistructured psychological interviews and standardized ques- ticipants’ subjective experience of change in their mental health,
tionnaires were administered 1–3 weeks after admittance to the we used one question from the Late Effects of Accidental Injury
rehabilitation hospital or when the participants’ somatic condition Questionnaire (Malt, 1988): “How do you think your psycholog-
allowed it (median days since injury ⫽ 38.5 days; quartiles ⫽ ical health has been after the injury compared with before the
24 – 61) and within a week before discharge (median days since injury?” There were three response alternatives: worsened, un-
injury ⫽ 121 days; quartiles ⫽ 93–175). Norwegian-language changed, or better. In the interview, special attention was paid to
premorbid and comorbid stressors. We asked for potential stressors
versions of all scales were used, and the interviews were per-
consisting of psychological topics often reported by patients such
formed by clinical psychologists. Written consent was obtained
as comorbid or premorbid (a) serious illness, (b) psychiatric illness
from the participants, and participation involved no changes in the
(in need of therapy or medication), (c) serious illness or death in
psychological treatment during the stay at SRH. The study was
close family, (d) marital problems, (e) substance abuse, (f) feeling
approved by the Data Inspectorate and by the Regional Committee
isolated or lonely, and (g) other stressors (such as relational prob-
for Medical Research Ethics, South-East Norway. lems, being refugees, difficulties in childhood, emotional prob-
The participants were classified into one of the three trajectories lems, and social problems including criminal behavior). We also
(resilience, recovery, or distress) on the basis of their level of asked for potentially event-related stressors such as (a) fear of
psychological distress (PTSD, depression, anxiety, and state neg- dying, (b) whether they felt safe when they got help, (c) whether
ative affect) and their level of state positive affect (see Table 1). they got enough support in the acute phase, and (d) feelings of guilt
The resilience trajectory was operationalized as having low dis- associated with the event.
tress and high state positive affect at both admittance and dis- Injury severity. The severity of the injuries was assessed with
charge. The recovery trajectory was operationalized as having the Abbreviated Injury Scale (Association for the Advancement of
improvements (from below to above a set threshold) on at least one Automotive Medicine, 1998). Ratings ranging from 1 (minor) to 6
of symptom scales. The distress trajectory was operationalized as (lethal) are made for injuries in each of six body areas. The NISS
having high distress, high state negative affect, or low state posi- is the sum of squares of the three highest Abbreviated Injury Scale
tive affect at both admittance and discharge. scores, regardless of body region (Osler, Baker, & Long, 1997).
16 QUALE AND SCHANKE

The injury-related data were collected from medical records at the Henry, 2004), we calculated cutoff scores for PA and NA by
participants’ acute care hospital, primarily from the Trauma Reg- adding or subtracting 1 standard deviation from the mean in the
istry at Ulleval University Hospital (Ulleval Trauma Registry, nonclinical sample. For NA, we were interested in those showing
2003). Injury-related data were only collected for the traumatically high negative affect, so we took the norm mean (16) plus the
injured participants. Maximum pain experience during the past standard deviation (5.5), which resulted in a cutoff of 21. For PA,
week was measured, both at admission and at discharge, on a we were interested in those showing low positive affect, so we
visual analog scale ranging from 0 to 10. took the norm mean (31.31) minus the standard deviation (7.65),
Questionnaires. To measure symptoms of PTSD, we used the which resulted in a cutoff of 24.
Impact of Event Scale—Revised (Weiss & Marmar, 1997). The We measured optimism and pessimism using the Life Orienta-
Impact of Event Scale—Revised is a 22-item scale that measures tion Test—Revised (LOT–R; Scheier, Carver, & Bridges, 1994).
all three core phenomena of PTSD: intrusion, avoidance, and The LOT–R consists of 10 items; three items assess optimism,
hyperarousal. However, in our analysis we used only the combined three items assess pessimism, and four items are filler items.
subscale score of intrusion and avoidance (15 items), and we Respondents indicate the extent to which they agree with each item
weighted the items on a 4-point Likert scale ranging from 0 to 5, on a 5-point Likert scale ranging from strongly agree (0) to
according to the original weighting for the Impact of Event Scale strongly disagree (4). Higher scores are indicative of optimism.
by Horowitz, Wilner, and Alvarez (1979). We used a cutoff of 36 The LOT–R exhibits an acceptable level of internal consistency
for the combined score of the Intrusion and Avoidance subscales, with a Cronbach’s alpha of .78 (Scheier et al., 1994). In our
which is considered to be optimal for detecting PTSD with a sample, the Cronbach’s alpha was .80.
sensitivity of .77 and a specificity of .51 (Witteveen, Bramsen,
Hovens, & Van der Ploeg, 2005). Internal consistency for the
Statistical Analyses
Impact of Event Scale—Revised total scale was good in our
sample, with a Cronbach’s alpha of .92.
We used descriptive statistics to characterize the total sample
The Hospital Anxiety and Depression Scale (HADS) was used
and the three trajectories. Bivariate analyses (analysis of variance
to identify symptoms of anxiety (HADS-A) and depression
for continuous variables and cross-tables, with Pearson chi-square
(HADS-D). The two subscales are added separately, and both
tests and Fisher’s exact tests for categorical variables) were used to
HAD-A (Cronbach’s ␣ ⫽ .82) and HAD-D (Cronbach’s ␣ ⫽ .81)
explore the statistical significance of the associations between the
showed adequate psychometric abilities in our sample. The HADS
different trajectories and possible predictor variables. We used
has also shown good psychometric abilities, with a Cronbach’s
correlation analysis to explore the association between the Positive
alpha greater than .60 in populations in both medical and psychi-
and Negative Affect Schedule state and trait versions, as well as
atric settings and in the general population (Bjelland, Dahl, Haug,
associations between the covariates. Regression analysis was used
& Neckelmann, 2002). A review of the literature shows that a
to explore crude and adjusted estimates of the covariates on the
cutoff of 8 or more is most commonly used (Zigmond & Snaith,
three trajectories. Results presented are those from multinomial
1983), and we adapted this to our model.
logistic regression with backwards variable selection. However,
Positive affect and negative affect were measured using the
because of the restricted sample size, both results from parallel
Positive Affect and Negative Affect Schedule (Watson, Clark, &
bivariate logistic regression and ordinal logistic regression were
Tellegen, 1988). The Positive Affect and Negative Affect Sched-
used to support the findings. All statistical analyses were per-
ule consists of two subscales of 10 adjectives, one subscale for
formed using SPSS 15.0 except the Fisher’s exact test for 2 ⫻ 3
positive affect (PA) and one for negative affect (NA). The adjec-
table, which was done in R, version 2.6.2 (R Project for Statistical
tives for PA are interested, strong, inspired, attentive, enthusiastic,
Computing, 2009). We considered p values less than or equal to
proud, alert, lively, active, and determined; for NA, distressed,
.05 significant.
upset, nervous, scared, hostile, irritable, ashamed, jittery, afraid,
and guilty. The participants were asked to indicate on a 5-point
scale (1 ⫽ very slightly/not at all, 2 ⫽ a little, 3 ⫽ moderately, 4 ⫽ Results
quite a bit, and 5 ⫽ very much). The PA score represents the mean
for the 10 PA items, and the NA score represents the mean for the The most common trajectory of adaptation to a severe physical
10 NA items. We used both a state and a trait version. For the state injury was the resilience trajectory (54%), followed by the recov-
version, the participants were asked to indicate to what extent they ery trajectory (25%) and the distress trajectory (21%). All of the
experienced each of the adjectives during the previous week. For participants were classified into one of these trajectories, and no
the trait version, the participants were asked to indicate to what one showed only a delayed response pattern. In Table 2, descrip-
extent they, before they were injured, experienced the same ad- tive statistics of demographics and injury-related variables for the
jectives. Internal consistencies for the PA and NA scales have full sample and for the different trajectories are presented. Of the
shown good reliability among a sample of patients receiving participants, 77% had a traumatic case of injury, and transport
inpatient medical rehabilitation, with Cronbach’s alphas of .85 and accidents were the most common injury mechanism (44%). All of
.90, respectively (Ostir, Smith, Smith, & Ottenbacher, 2005). The the participants who were injured in a sport accident were in the
internal consistency of both the state and the trait version of the PA resilience trajectory, and the two participants with an assault-
and NA scales in this study sample was also good (Cronbach’s related injury were in the distress trajectory. About one third (29%)
␣ ⫽ .85–.92). The Positive and Negative Affect Schedule does not of the participants had MT, and the rest (71%) had SCI. Twenty-
have established cutoff scores, but on the basis of normative data eight percent of those with SCI also had MT. Among the partic-
from 1,003 adults from the general U.K. population (Crawford & ipants with SCI, 56% had paraplegic injuries and 44% had tet-
 RESILIENCE IN PHYSICAL REHABILITATION SETTINGS 17

Table 2
Sample Characteristics of the Resilience, Recovery, and Distress Trajectories
Demographic variables Full sample Resilience trajectory Recovery trajectory Distress trajectory

N (%) 80 (100) 43 (54) 20 (25) 17 (21)

Age (M [SD]) 39.1 (15.6) 39.3 (15.5) 39.2 (16.5) 38.7 (15.6)
Gender (n [%])
Men 59 (73.8) 34 (79.1) 13 (65) 12 (70.6)
Women 21 (26.3) 9 (20.9) 7 (35) 5 (29.4)
Family status (n [%])
Married (or cohabitating) 43 (53.8) 24 (55.8) 11 (55) 8 (47.1)
Single 37 (46.3) 19 (44.2) 9 (45) 9 (52.9)
Education (n [%])
High school or less 60 (75) 29 (48.3) 16 (26.7) 15 (25)
College/university 20 (25) 14 (70) 4 (20) 2 (10)
Work status time of injury (n [%])
In school/occupation 58 (72.5) 35 (81.4) 13 (65) 10 (58.8)
No school/occupation 10 (12.5) 2 (4.7) 4 (20) 4 (23.5)
On sick leave 12 (15) 6 (14) 3 (15) 3 (17.6)
Injury/diagnoses (n [%])
Multiple trauma 23 (28.8) 13 (30.2) 7 (35) 3 (17.6)
Spinal cord injury 35 (43.8) 18 (41.9) 9 (45) 8 (47.1)
Multiple trauma and spinal cord injury 22 (27.5) 12 (27.9) 4 (20) 6 (35.3)
Cause of injury (n [%])
Sports accident 9 (11.3) 9 (20.9) 0 (0) 0 (0)
Assault 2 (2.5) 0 (0) 0 (0) 2 (11.8)
Transport accidenta 34 (42.5) 21 (48.8) 8 (40) 5 (29.4)
Fall accident 14 (17.5) 5 (11.6) 3 (15) 6 (35.3)
Other traumatic cause 4 (5) 2 (4.7) 1 (5) 1 (5.9)
Nontraumatic cause 17 (21.3) 6 (14) 8 (40) 3 (17.6)
Severity of injury (n [%])
New Injury Severity Scale (M [SD]) 31.9 (13.5) 31.1 (12.8) 35.1 (14.8) 31.2 (14.7)
Pain (n [%])
Maximum pain admission (M [SD]) 4.72 (2.83) 4.1 (2.7) 4.5 (2.9) 6.6 (2.3)
Maximum pain discharge (M [SD]) 3.51 (2.51) 3.2 (2.5) 3.2 (2.6) 4.7 (2.1)
a
Transport accident included all types of accidents that involved transport behavior, for example, car, motorbike, and biking accidents.

raplegic injuries; 67% had incomplete injuries, whereas 33% had .003, showed significant differences between the resilience and the
complete injuries. The median days spent in the rehabilitation distress trajectories, but feelings of guilt associated with the event did
hospital was 91 (quartiles ⫽ 73–124). not, ␹2(2, N ⫽ 79) ⫽ .805, p ⫽ .734. The Pearson chi-square analyses
for the participants’ subjective experience of change in their mental
Bivariate Analyses health showed significant differences between the three trajectories,
␹2(4, N ⫽ 80) ⫽ 18.319, p ⫽ .001. The participants in the resilience
We conducted bivariate analyses for the continuous psychoso- trajectory most frequently reported no change in their mental health,
cial variables and the categorical comorbid stressor variables (see and the participants in the distress trajectory most frequently reported
Table 3). Significant differences between the resilience trajectory a worsening of their mental health. All the participants diagnosed with
and the distress trajectory were most common in all the analyses, PTSD were in the distress trajectory both at admission and at dis-
but we also found significant differences between the resilience charge. The relationship between the resilience and the distress tra-
and the recovery trajectories on trait positive affect and trait jectories was significant at admission, ␹2(1, N ⫽ 59) ⫽ 10.601, p ⫽
negative affect, comorbid psychiatric illness, serious illness or .005, but not at discharge, ␹2(1, N ⫽ 60) ⫽ 5.233, p ⫽ .077.
death in close family, and feelings of isolation and loneliness at According to the bivariate analyses, there were no significant
admission and on comorbid feelings of isolation and loneliness and differences between the demographics and the different trajecto-
other comorbid stressors at discharge. ries (data not shown). One injury-related variable that showed a
The only premorbid stressors that differed significantly between statistically significant effect was maximum pain experience at ad-
the resilience and the distress trajectories were substance abuse, ␹2(1, mission. Using the Bonferroni post hoc test, we found significant
N ⫽ 60) ⫽ 3.864, p ⫽ .049, and other premorbid stressors (such as differences between the resilience and the distress trajectories (p ⫽
relational problems, being refugees, difficulties in childhood), ␹2(1, .005). There was no significant difference between the resilience and
N ⫽ 60) ⫽ 9.9, p ⫽ .002. Other premorbid stressors also differed the recovery trajectories (p ⫽ 1.0) and the distress and the recovery
significantly between the resilience and the recovery trajectories, ␹2(1, trajectories (p ⫽ .142) on maximum pain experience at admission.
N ⫽ 63) ⫽ 10.268, p ⫽ .001. Among the event-related stressors, fear We found no significant difference between the three trajectories and
of dying, ␹2(2, N ⫽ 79) ⫽ 8.86, p ⫽ .012; whether they felt safe when maximum pain experience at discharge. There were significantly
they got help, ␹2(2, N ⫽ 79) ⫽ 10.375, p ⫽ .005; and whether they more traumatically injured participants in the resilience trajectory than
got enough support in the acute phase, ␹2(2, N ⫽ 79) ⫽ 11.709, p ⫽ in the recovery trajectory (p ⫽ .021). Although not significant, it is
18 QUALE AND SCHANKE

Table 3
Bivariate Analyses for Psychosocial Variables for the Resilience Trajectory Versus the Recovery and the Distress Trajectories
Variables Full sample Resilience trajectory Recovery trajectory Distress trajectory

Psychosocial
N (%) 80 43 (54) 20 (25) 17 (21)
Continuous M (SD) M (SD) M (SD) pa M (SD) pa
Social support
Admission 8.5 (2) 9.14 (1.18) 8.16 (2.59) 0.2 7.19 (2.46) 0.002ⴱ
Discharge 8.4 (2.1) 8.98 (1.47) 8.58 (2.48) 1.0 6.59 (2.09) ⬍0.001ⴱ
Life Orientation Test—Revised: admission 16.6 (5.2) 18.36 (4.2) 16.28 (5.4) 0.41 13.06 (5.56) ⬍0.001ⴱ
Positive affect—Trait: admission 35.7 (8) 38.71 (6.79) 33.75 (8.48) 0.04ⴱ 30.13 (6.95) ⬍0.001ⴱ
Negative affect—Trait: admission 17.9 (6.8) 14.71 (4.31) 19.55 (7.31) 0.01ⴱ 24.13 (6.73) ⬍0.001ⴱ
Categorical variables n (%) n (%) n (%) ␹2 n (%) ␹2
Comorbid stressors at admission
Serious illness 10 (12.7) 4 (5.1) 3 (3.8) 0.41 3 (3.8) 0.76
Psychiatric illness 12 (15.2) 2 (2.5) 5 (6.3) 5.54ⴱ 5 (6.3) 7.03ⴱ
Illness or death in family 11 (13.9) 3 (3.8) 6 (7.6) 5.70ⴱ 2 (2.5) 0.33
Marital problems 9 (11.4) 3 (3.8) 3 (3.8) 0.96 3 (3.8) 1.46
Substance abuse 9 (11.4) 4 (5.1) 0 (0) 2.04 5 (6.3) 3.70
feeling isolated or lonely 11 (13.9) 1 (1.3) 5 (6.3) 7.93ⴱ 5 (6.3) 9.68ⴱ
Other stressors 10 (12.7) 1 (1.3) 3 (3.8) 3.58 6 (7.6) 12.54ⴱ
Comorbid stressors at discharge
Serious illness 4 (5) 1 (1.3) 1 (1.3) 0.32 2 (2.5) 2.29
Psychiatric illness 12 (15) 1 (1.3) 3 (3.8) 3.69 8 (10) 19.12ⴱ
Illness or death in family 11 (13.8) 6 (7.5) 3 (3.8) 0.01 2 (2.5) 0.05
Marital problems 8 (10) 2 (2.5) 3 (3.8) 2 3 (3.8) 2.69
Substance abuse 7 (8.8) 3 (3.8) 1 (1.3) 0.09 3 (3.8) 1.54
Feeling isolated or lonely 9 (11.4) 0 (0) 4 (5.1) 8.98ⴱ 5 (6.3) 13.5ⴱ
Other stressors 14 (17.5) 4 (5.1) 9 (11.3) 10.62ⴱ 1 (1.3) 0.19
a
p values are adjusted according to the Bonferroni correction method.
ⴱ
p ⬍ .05.

interesting to note that the participants in the recovery group had the is greater than 1, which indicates that as the negative affect
highest scores on NISS (see Table 2). increases, so does the odds of the membership in the trajectory.
More specifically, the odds ratios 1.16 and 1.26 indicate a 16% and
Predictors 26% increase, respectively, for recovery versus resilience and
distress versus resilience when negative affect is present. The odds
According to correlation analyses, there was no significant re- ratio for trait positive affect in both conditions is less than 1, which
lationship between PA state and PA trait (r ⫽ .439) and NA state indicates that as the positive affect increases, the odds of mem-
and NA trait (r ⫽ .268) at admission, indicating that PA and NA bership in the trajectory decreases. A comparison between the
trait could be treated as possible predictors in regression analysis, recovery and the distress trajectories was also done according to
even though PA and NA state, at admission, appear in the opera- multinomial logistic regression, but the results were not signifi-
tionalization of the response variable. The variables PA trait, NA cant. Results from parallel bivariate logistic regression and ordinal
trait, LOT–R, and comorbid feelings of isolation or loneliness at logistic regression were similar to the results from the multinomial
admission were included in multinomial logistic regression anal- logistic regression and thus supported the findings.
ysis. However, we excluded LOT–R and comorbid feelings of
isolation or loneliness because of lack of significance after adjust- Respondents Versus Nonrespondents
ment for PA trait and NA trait. PA trait and NA trait were
significant in the final model for both groups (see Table 4). The Because in addition to the exclusion criteria, we excluded the 21
odds ratio for trait negative affect in both conditions, recovery participants who missed the follow-up at discharge, we compared
versus resilience trajectory and distress versus resilience trajectory, those participants with data from the discharge to those without

Table 4
Final Model of Multinomial Logistic Regression Analyses
Recovery vs. resilience trajectory Distress vs. resilience trajectory Recovery vs. distress trajectory
Variable OR 95% CI p OR 95% CI p OR 95% CI p

Trait negative affect 1.16 [1.04, 1.31] .01 1.26 [1.11, 1.43] ⬍.001 1.09 [0.98, 1.20] .10
Trait positive affect 0.92 [0.84, 0.99] .04 0.87 [0.79, 0.97] .01 0.95 [0.87, 1.04] .29

Note. OR ⫽ odds ratio; CI ⫽ confidence interval.

 RESILIENCE IN PHYSICAL REHABILITATION SETTINGS 19

data from the discharge to look for potential bias. The two groups resilience and the distress trajectories for the psychosocial vari-
did not differ significantly in gender, age, family status, injury ables were most common. Participants in the distress trajectory
severity, length of stay in rehabilitation hospital, posttraumatic were significantly less optimistic and less satisfied with social
stress symptoms, anxiety, depression, and trait positive and nega- support both at admission and at discharge, had premorbid sub-
tive affect. However, the nonrespondents had a significantly higher stance abuse, were more afraid of dying, felt less safe when they
state positive affect (p ⫽ .003) and a significantly lower state got help, and did not get enough support in the acute phase
negative affect (p ⫽ .003) at admission than did the respondents. compared with participants in the resilience trajectory. It is inter-
This means that the nonresponse data may not be missing at esting to emphasize that the individuals in the resilience trajectory
random, but because the two groups did not differ significantly in were significantly more optimistic than the participants in the
any of the other variables, we conclude that it will not cause a distress trajectory. This supports previous studies on the relation-
problem in generalizing our results. ship between individuals characterized by optimism and adjust-
ment. Studies have shown that optimistic individuals experience
Discussion less distress after an adversity, such as a childbirth (Carver &
Gaines, 1987), and retain higher quality of life after treatment of a
The purpose of this study was to identify trajectories of psycho- medical condition (Allison, Guichard, & Gilain, 2000) and that
logical adjustment to an acquired severe physical injury charac- optimism predicts lower likelihood for rehospitalization after by-
terized by resilience, recovery, or distress; to explore differences pass surgery (Scheier et al., 1999).
between these trajectories; and to identify possible predictors. We also attempted to distinguish the participants in the recovery
First, we found that resilience is the most common response to an trajectory from those in the resilience trajectory, and we found that
acquired severe physical injury, with more than half of the partic- the participants in the recovery trajectory had serious illness or
ipants displaying a trajectory characterized by resilience. This death in near family at admission and reported other comorbid
finding was consistent with those of other studies on similar stressors at discharge (such as relational problems and social
populations, such as individuals who have experienced the loss of problems). There was a trend in all the scores in the continuous
a loved one (Bonanno et al., 2002), a potentially traumatic event data that the recovery trajectory lay between the two other trajec-
(Bonanno, Galea, Bucciarelli, & Vlahov, 2006), or breast cancer
tories. According to the analyses, it also seemed as though the
(Deshields et al., 2006). However, comparing our results with
comorbid stressors were more important than the premorbid stres-
Berry et al.’s (2007) study of personality prototypes in a sample
sors in predicting outcomes.
with SCI, we found a higher prevalence of resilience than they did
We found no differences between the trajectories on any of the
(54% compared with 28%, respectively). This discrepancy might
demographic variables, meaning that age, gender, family status,
be the result of the different operationalizations. In Berry et al.’s
and level of education did not predict the classification into the
study, the resilient prototype was characterized by low neuroticism
three trajectories. Our results did not support the findings of
and above average extraversion, openness, agreeableness, and
Bonanno and colleagues, who found that resilience was most
conscientiousness. Even though neuroticism has been associated
prevalent among married, younger individuals, males, and individ-
with negative affect and extraversion with positive affect, our
uals with a high education level in a population of individuals
additional inclusion of the absence of psychopathology in our
operationalization of resilience makes the operationalizations dif- exposed to a potentially traumatic event, namely the September 11,
ferent. Because resilience is believed to be not a trait but rather a 2001, terrorist attack in New York City (Bonanno et al., 2007).
process of adjustment that involves behaviors, thoughts, and ac- The discrepancy might be explained by the dissimilar populations
tions (Luthar, 2003), future studies should further investigate per- and the different adversities. One example is that the New York
sonality and resilience in populations with severe injuries. City population had a higher education level (Galea et al, 2002).
The commonalities of adults displaying a resilient adaptation in The nonsignificant findings from this study may also be the result
our study might be because these individuals are exposed to an of the low statistical power because of a relatively small sample
isolated stressor event. Individuals who are exposed to several (N ⫽ 80). By comparison, in the study by Bonanno and his
stressful events over time (e.g., torture or violence) or endure colleagues, there were 2,752 participants.
challenging environments (e.g., child maltreatment or war) might The only significant injury-related variables were maximum
not have the same possibilities for a resilient adaptation. Because pain experience at admission, with the participants in the distress
our results show data from a rehabilitation process over approxi- trajectory experiencing higher levels of pain than the participants
mately 3 months rather than cross-sectionally at one point in time, in the resilience trajectory, and cause of injury, with the partici-
we assume that the individuals who fell into the resilient trajectory pants in the resilience trajectory more often having a traumatic
during rehabilitation will still be categorized as resilient several cause of injury than the participants in the recovery trajectory.
years after injury. Future studies should investigate resilience Although not significant, the participants in the recovery trajectory
several years after injury to see whether they find the same inci- had the highest injury severity, which might indicate that these
dence and investigate predictors of a long-term resilient outcome. individuals need a longer time for psychological adjustment be-
Second, according to bivariate analyses we found significant cause of the extensive physical trauma. Previous research has
differences between both the resilience and the recovery trajecto- stated that injury severity does not have an impact on posttrau-
ries and the resilience and the distress trajectories on trait positive matic stress symptoms (Quale, Schanke, Froslie, & Roise, 2009),
affect, trait negative affect, comorbid psychiatric illness at admis- but in this study we examined the adaptation process with more
sion, and feelings of isolation and loneliness both at admission and widespread variables such as affect, anxiety, and depression in
at discharge. Not surprisingly, significant differences between the addition to PTSD symptoms.
20 QUALE AND SCHANKE

Third, trait positive and negative affect distinguished the resil- their circumstances reasonably well is an important contribution to
ience trajectory from the recovery trajectory and the resilience the field.
trajectory from the distress trajectory. This means that the individ-
uals’ reports of their affect or emotions preinjury predict the Resilience-Based Interventions
outcome course of the adjustment postinjury. The influence of
positive affect and negative affect on outcome is supported by It is essential that rehabilitation psychologists identify, enhance,
other studies that have found that individuals with high positive and encourage reliance on each severely injured individual’s
affect are more likely to participate in social activities (Ryff & strengths, thereby maximizing psychological and physical recov-
Singer, 1998) and successfully cope with stressful situations (Folk- ery (Dunn & Dougherty, 2005). Individuals are able to learn and
man, 1997). Individuals with high negative affect are found to be demonstrate resilience even during and after extremely traumatic
more likely to be depressed or anxious or to report additional events (Luthar et al., 2000), and both significant others and psy-
health complaints (Watson, Clark, & Carey, 1988a). chologists need to understand this. Resilience-based interventions
prompt helping professionals to search for individual strengths in
clients and to nurture them. Because the goal of the rehabilitation
Limitations and Strengths psychologist is to help individuals learn to cope and adjust after
severe illness or injury, resilience-based interventions seem to
Because of the limited literature on trajectories of resilience in have obvious applications for the field (White et al., 2008).
adults and the limited knowledge of resilience in rehabilitation Resilience-based interventions emphasize that resilience is or-
settings, we met some methodological challenges in the operation- dinary, not extraordinary; that being resilient does not mean that a
alization of the trajectories. Using both presence and absence of person does not experience difficulties or distress; that resilience is
psychopathology such as anxiety and depression, together with not a trait that individuals either have or do not have; and that
state positive affect, complicated the operationalization. Rather anyone can learn and develop resilient behaviors, thoughts, and
than measuring the severity of psychopathology, we only mea- actions (American Psychological Association Help Center, 2004).
sured the presence of each symptom. Our study would be strength- By developing resilience skills (e.g., social skills, self-compe-
ened if we additionally identified those with a few symptoms of tence) through therapy or education, individuals will have a greater
psychological distress but who are functioning well, because this is repertoire of skills to help them adapt to adversity. One way for
also regarded as a sign of resilience (Hobfoll et al., 2009). Another rehabilitation psychologists to develop resilience skills in individ-
point for further discussion is whether resilience can be defined as uals with severe injuries could be by using the framework of the
the absence of psychopathology. Is resilience the opposite of conception of the good life after disability presented by Dunn and
psychopathology, or the opposite of vulnerability? Bonanno Brody (2008). They argued that a resilient adaptation is accessible
(2004) claimed that resilience is more than the simple absence of if individuals living with a severe injury are willing to make
psychopathology, and there are multiple routes to good mental certain choices and take certain actions, both behavioral and cog-
health, of which resilience is just one (Bonanno, 2004). We com- nitive, that may enhance their daily living. Choices and actions are
plemented the operationalization by adding state positive and illustrated within three defining areas: making connections with
negative affect, which in turn resulted in the more complicated others, developing positive traits, and enacting life regulation
operationalization. At this point in time, a “gold standard” for how qualities. This framework could be a nice starting point for intro-
resilience is defined does not exist. This leads to nuances in how ducing resilience-based interventions in rehabilitation settings.
each researcher defines the concept in the field of psychological There seems to be a common tendency to interpret resilient
research and makes resilience a difficult construct to quantify. In adaptations to loss or potentially traumatic events as “denial” or a
future studies, we will use an instrument developed to directly pathological refusal to experience pain (Mancini & Bonanno,
measure the construct of resilience, the Resilience Scale for Adults 2006). If we turn to the insider– outsider distinction, this tendency
(Friborg, Hjemdal, Rosenvinge, & Martinussen, 2003), to see is probably a result of the outsiders’ difficulties with accepting that
whether this gives us an even better grasp of the construct in many individuals are able to cope well with even highly traumatic
relation to individuals with a severe injury. events or tragic losses, so they expect pathological reactions to
Additional work on at-risk populations’ achievement of positive arise later. However, the outsider’s perspective is not supported in
outcomes has been called for (Luthar et al., 2000), also within the research. The now classic work by Wortman and Silver (1989)
field of rehabilitation psychology (White et al., 2008). Thus, this stated that many individuals do remain psychologically healthy
study has expanded our knowledge of resilience in rehabilitation after a traumatic event and that not all individuals have to expe-
settings with empirical data. As far as we know, this is the first rience grief and psychological disturbance to avoid a pathological
longitudinal study of resilience in the field of rehabilitation, and no or a delayed pathological reaction. According to research done by
previous studies have examined trajectories of adaptation, with Shelley Taylor and her colleagues (Taylor, 1983; Taylor, Kemeny,
respect to resilience, during rehabilitation. Cross-sectional studies Reed, Bower, & Gruenewald, 2000), realistic expectations about
do not display the process of resilience and cannot distinguish a the specific challenge and the future are not associated with psy-
resilient adaptation from a recovery trajectory, which might be chological health. However, positive illusions and unrealistic op-
important in a rehabilitation setting because of the major physical timistic beliefs, which may be perceived as and labeled denial by
challenges the individual must overcome. Because the focus in outsiders, are health protective and associated with good adapta-
rehabilitation psychology research has mainly been to identify risk tion to adversity. In our study population, no one showed only
factors and vulnerabilities, this quantification of the reality that delayed reactions. Thus, we conclude, with support from the
most individuals who acquire a severe physical injury adjust to research mentioned earlier, that resilience might be considered as
 RESILIENCE IN PHYSICAL REHABILITATION SETTINGS 21

a common and natural response to adjusting to a severe physical Bjelland, I., Dahl, A. A., Haug, T. T., & Neckelmann, D. (2002). The
injury and that the absence of psychopathology and the continued validity of the Hospital Anxiety and Depression Scale: An updated
ability to function adequately does not reflect denial but rather an literature review. Journal of Psychosomatic Research, 52, 69 –77.
inherent and adaptive resilient coping. This information is helpful Bonanno, G. A. (2004). Loss, trauma, and human resilience: Have we
news for individuals facing a severe physical injury, and it should underestimated the human capacity to thrive after extremely aversive
events? American Psychologist, 59, 20 –28.
be included in psychoeducative programs. These individuals
Bonanno, G. A., & Field, N. P. (2001). Examining the delayed grief
should be reassured that resilience as a response is common hypothesis across 5 years of bereavement. American Behavioral Scien-
despite the physical and emotional challenges and that delayed tist, 44, 798 – 816.
reactions are not to be expected. Bonanno, G. A., Galea, S., Bucciarelli, A., & Vlahov, D. (2006). Psycho-
logical resilience after disaster: New York City in the aftermath of the
September 11th terrorist attack. Psychological Science, 17, 181–186.
Positive Psychology and Rehabilitation
Bonanno, G. A., Galea, S., Bucciarelli, A., & Vlahov, D. (2007). What
predicts psychological resilience after disaster? The role of demograph-
Our study has clear roots in the fast-growing field of positive ics, resources, and life stress. Journal of Consulting and Clinical Psy-
psychology because enhancing resilience seems to be an underly- chology, 75, 671– 682.
ing theme in clinical positive psychological work. Positive psy- Bonanno, G. A., Ho, S. M., Chan, J. C., Kwong, R. S., Cheung, C. K.,
chology is the science of understanding human strengths and the Wong, C. P., & Wong, V. C. W. (2008). Psychological resilience and
practice of promoting these strengths to help individuals cope dysfunction among hospitalized survivors of the SARS epidemic in
psychologically and physically (Lopez & Snyder, 2009; Seligman Hong Kong: A latent class approach. Health Psychology, 27, 659 – 667.
& Csikszentmihalyi, 2000). Positive psychology focuses on iden- Bonanno, G. A., Moskowitz, J. T., Papa, A., & Folkman, S. (2005).
tifying the strengths of an individual when faced with adversity Resilience to loss in bereaved spouses, bereaved parents, and bereaved
rather than his or her weaknesses (e.g., depression, anxiety). A gay men. Journal of Personality and Social Psychology, 88, 827– 843.
Bonanno, G. A., Wortman, C. B., Lehman, D. R., Tweed, R. G., Haring,
positive rehabilitation psychologist should do more than focus on
M., Sonnega, J., . . . Nesse, R. M. (2002). Resilience to loss and chronic
treatment issues or adaptation to disability; he or she must capi-
grief: A prospective study from preloss to 18-months postloss. Journal
talize on the individual’s psychosocial strengths to maintain or of Personality and Social Psychology, 83, 1150 –1164.
enhance psychological and physical well-being and to prevent Campbell-Sills, L., Cohan, S. L., & Stein, M. B. (2006). Relationship of
pathology (Dunn & Dougherty, 2005). resilience to personality, coping, and psychiatric symptoms in young
Future longitudinal studies should examine whether the adapta- adults. Behaviour Research and Therapy, 44, 585–599.
tion trajectories change in prevalence when people leave inpatient Carver, C. S., & Gaines, J. G. (1987). Optimism, pessimism, and postpar-
rehabilitation and return to living in the community. The transition tum depression. Cognitive Therapy and Research, 11, 449 – 462.
from the rehabilitation hospital to the home setting is one of the Carver, C. S., Scheier, M. F., Miller, C. J., & Fulford, D. (2009). Optimism.
most challenging periods for the newly injured individual. Future In S. J. Lopez & C. R. Snyder (Eds.), Oxford handbook of positive
studies with this study’s population should aim to examine the psychology (2nd ed., pp. 303–312). New York, NY: Oxford University
relationship between resilience and quality of life and the opera- Press.
Collard, C., & Kennedy, P. (2007). A longitudinal analysis of emotional
tion between personality variables, coping style, and secondary
impact, coping strategies and posttraumatic psychological growth fol-
complications after rehabilitation. lowing spinal cord injury: A 10-year review. British Journal of Health
As rehabilitation researchers, we must aim at a more positive Psychology, 12, 347–362.
focus and learn more about positive adjustment, happiness, well- Crawford, J. R., & Henry, J. D. (2004). The Positive and Negative Affect
being, and positive growth after a person acquires a severe injury. Schedule (PANAS): Construct validity, measurement properties and
We can help both injured and noninjured individuals to see that normative data in a large non-clinical sample. British Journal of Clinical
such life events do not need to be devastating and that the future Psychology, 43, 245–265.
can bring both hope and quality. We might then start to challenge Deshields, T., Tibbs, T., Fan, M. Y., & Taylor, M. (2006). Differences in
the cultural assumptions and the prejudice toward individuals with patterns of depression after treatment for breast cancer. Psychooncology,
disabilities to enhance hope and quality of life. 15, 398 – 406.
Dunn, D. S. (2009). The social psychology of disability. In R. G. Frank, M.
Rosenthal, & B. Caplan (Eds.), Handbook of rehabilitation psychology
References (2nd ed., pp. 379 –390). Washington, DC: American Psychological As-
sociation.
Allison, P. J., Guichard, C., & Gilain, L. (2000). A prospective investiga- Dunn, D. S., & Brody, C. (2008). Defining the good life following acquired
tion of dispositional optimism as a predictor of health-related quality of physical disability. Rehabilitation Psychology, 53, 413– 425.
life in head and neck cancer patients. Quality of Life Research, 9, Dunn, D. S., & Dougherty, S. B. (2005). Prospects for a positive psychol-
951–960. ogy of rehabilitation. Rehabilitation Psychology, 50, 305–311.
American Psychological Association Help Center. (2004). The road to Dunn, D. S., Uswatte, G., & Elliott, T. R. (2009). Happiness, resilience,
resilience. Retrieved from http://www.apahelpcenter.org/featuredtopics/ and positive growth following physical disability: Issues for understand-
feature.php?id⫽6 ing, research, and therapeutic intervention. In S. Lopez & C. R. Snyder
Association for the Advancement of Automotive Medicine. (1998). The (Eds.), Oxford handbook of positive psychology (2nd ed., pp. 651– 664).
Abbreviated Injury Scale—1990 revision, update 98. Des Plaines, IL: New York, NY: Oxford University Press.
Author. Elliott, T. R., & Frank, R. G. (1996). Depression following spinal cord
Berry, J. W., Elliott, T. R., & Rivera, P. (2007). Resilient, undercontrolled, injury. Archives of Physical Medicine and Rehabilitation, 77, 816 – 823.
and overcontrolled personality prototypes among persons with spinal Folkman, S. (1997). Positive psychological states and coping with severe
cord injury. Journal of Personality Assessment, 89, 292–302. stress. Social Science & Medicine, 45, 1207–1221.
22 QUALE AND SCHANKE

Friborg, O. (2005). Validation of a scale to measure resilience in adults. Scheier, M. F., Matthews, K. A., Owens, J. F., Schulz, R., Bridges, M. W.,
Unpublished manuscript, Department of Psychology, Faculty of Social Magovern, G. J., & Carver, C. S. (1999). Optimism and rehospitalization
Sciences, University of Tromsø, Tromsø, Norway. after coronary artery bypass graft surgery. Archives of Internal Medi-
Friborg, O., Hjemdal, O., Rosenvinge, J. H., & Martinussen, M. (2003). A cine, 159, 829 – 835.
new rating scale for adult resilience: What are the central protective Seligman, M. (1998). Learned optimism: How to change your mind and
resources behind healthy adjustment? International Journal of Methods your life. New York City, NY: Vintage Books.
in Psychiatric Research, 12, 65–76. Seligman, M. E., & Csikszentmihalyi, M. (2000). Positive psychology: An
Galea, S., Resnick, H., Ahern, J., Gold, J., Bucuvalas, M., Kilpatrick, introduction. American Psychologist, 55, 5–14.
D., . . . Vlahov, D. (2002). Posttraumatic stress disorder in Manhattan, Starr, A. J., Smith, W. R., Frawley, W. H., Borer, D. S., Morgan, S. J.,
New York City, after the September 11th terrorist attacks. Journal of Reinert, C. M., & Mendoza-Welch, M. (2004). Symptoms of posttrau-
Urban Health Studies, 79, 340 –353. matic stress disorder after orthopaedic trauma. Journal of Bone and Joint
Hobfoll, S. E., Palmieri, P. A., Johnson, R. J., Canetti-Nisim, D., Hall, Surgery, 86, 1115–1121.
B. J., & Galea, S. (2009). Trajectories of resilience, resistance, and Strand, E. B., Zautra, A. J., Thoresen, M., Odegard, S., Uhlig, T., & Finset,
distress during ongoing terrorism: The case of Jews and Arabs in Israel. A. (2006). Positive affect as a factor of resilience in the pain-negative
Journal of Consulting and Clinical Psychology, 77, 138 –148. affect relationship in patients with rheumatoid arthritis. Journal of Psy-
Horowitz, M., Wilner, N., & Alvarez, W. (1979). Impact of Event Scale: chosomatic Research, 60, 477– 484.
A measure of subjective stress. Psychosomatic Medicine, 41, 209 –218. Taylor, S. E. (1983). Adjustment to threatening events: A theory of
Kennedy, P., & Duff, J. (2001). Post traumatic stress disorder and spinal cognitive adaptation. American Psychologist, 38, 1161–1173.
cord injuries. Spinal Cord, 39, 1–10. Taylor, S. E., Kemeny, M. E., Reed, G. M., Bower, J. E., & Gruenewald,
Kennedy, P., & Evans, M. J. (2001). Evaluation of post traumatic distress T. L. (2000). Psychological resources, positive illusions, and health.
in the first 6 months following SCI. Spinal Cord, 39, 381–386. American Psychologist, 55, 99 –109.
Kennedy, P., & Rogers, B. A. (2000). Anxiety and depression after spinal Tugade, M. M., & Fredrickson, B. L. (2004). Resilient individuals use
cord injury: A longitudinal analysis. Archives of Physical Medicine and positive emotions to bounce back from negative emotional experiences.
Rehabilitation, 81, 932–937. Journal of Personality and Social Psychology, 86, 320 –333.
Lopez, S. J., & Snyder, C. R. (2009). Oxford handbook of positive psy- Ulleval Trauma Registry. (2003). Ulleval University Hospital Trauma
chology (2nd ed.) New York, NY: Oxford University Press. Registry. Oslo, Norway: Ulleval University Hospital.
Luthar, S. S. (2003). Research on resilience: An integrative review. In S. S. Watson, D., Clark, L. A., & Carey, G. (1988). Positive and negative
Luthar (Ed.), Resilience and vulnerability: Adaptation in the context of affectivity and their relation to anxiety and depressive disorders. Journal
childhood adversities (pp. 510 –551). New York, NY: Cambridge Uni- of Abnormal Psychology, 97, 346 –353.
versity Press. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
Luthar, S. S., Cicchetti, D., & Becker, B. (2000). The construct of resil- dation of brief measures of positive and negative affect: The PANAS
ience: A critical evaluation and guidelines for future work. Child De- scales. Journal of Personality and Social Psychology, 54, 1063–1070.
velopment, 71, 543–562. Weiss, D. S., & Marmar, C. R. (1997). The Impact of Event Scale—
Malt, U. (1988). The long-term psychiatric consequences of accidental Revised. In J. P. Wilson & T. M. Keane (Eds.), Assessing psychological
injury: A longitudinal study of 107 adults. British Journal of Psychiatry, trauma and PTSD: A handbook for practitioners (pp. 399 – 411). New
153, 810 – 818. York, NY: Guilford Press.
Mancini, A. D., & Bonanno, G. A. (2006). Resilience in the face of White, B., Driver, S., & Warren, A.-M. (2008). Considering resilience in
potential trauma: Clinical practices and illustrations. Journal of Clinical the rehabilitation of people with traumatic disabilities. Rehabilitation
Psychology, 62, 971–985. Psychology, 53, 9 –17.
Mohta, M., Sethi, A. K., Tyagi, A., & Mohta, A. (2003). Psychological care Witteveen, A. B., Bramsen, I., Hovens, J. E., & Van der Ploeg, H. M.
in trauma patients. Injury, 34, 17–25. (2005). Utility of the Impact of Event Scale in screening for posttrau-
Osler, T., Baker, S. P., & Long, W. (1997). A modification of the injury matic stress disorder. Psychological Reports, 97, 297–308.
severity score that both improves accuracy and simplifies scoring. Jour- World Health Organization. (1992). The ICD-10 classification of mental
nal of Trauma, 43, 922–925. and behavioral disorders: Clinical descriptions and diagnostic guide-
Ostir, G. V., Smith, P. M., Smith, D., & Ottenbacher, K. J. (2005). lines (ICD-10). Geneva, Switzerland: Author.
Reliability of the Positive and Negative Affect Schedule (PANAS) in Wortman, C. B., & Silver, R. C. (1989). The myths of coping with loss.
medical rehabilitation. Clinical Rehabilitation, 19, 767–769. Journal of Consulting and Clinical Psychology, 57, 349 –357.
Quale, A. J., Schanke, A. K., Froslie, K. F., & Roise, O. (2009). Severity Wright, B. A. (1991). Labeling: The need for greater person-environment
of injury does not have any impact on posttraumatic stress symptoms in individuation. In C. R. Snyder & D. R. Forsyth (Eds.), Handbook of
severely injured patients. Injury, 40, 498 –505. social and clinical psychology (pp. 569 – 487). New York, NY: Perga-
R Project for Statistical Computing. (2009). Retrieved from http://www.r- mon Press.
project.org Zautra, A. J., Johnson, L. M., & Davis, M. C. (2005). Positive affect as a
Ryff, C. D., & Singer, B. (1998). The contours of positive human health. source of resilience for women in chronic pain. Journal of Consulting
Psychological Inquiry, 9, 1–28. and Clinical Psychology, 73, 212–220.
Schanke, A. K. (2004). Never second best? A narrative perspective on the Zigmond, A. S., & Snaith, R. P. (1983). The Hospital Anxiety and De-
shaping and reconstruction of identity in twenty polio survivors. Scan- pression Scale. Acta Psychiatrica Scandinavica, 67, 361–370.
dinavian Journal of Disability Research, 6, 89 –110.
Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing
optimism from neuroticism (and trait anxiety, self-mastery, and self- Received July 1, 2009
esteem): A reevaluation of the Life Orientation Test. Journal of Person- Revision received September 29, 2009
ality and Social Psychology, 67, 1063–1078. Accepted November 4, 2009 䡲