DISABILITY AND REHABILITATION, 2003; VOL. 25, NO.

13, 722–731

Pilot evaluation of a mindfulness-based

intervention to improve quality of life among
individuals who sustained traumatic brain
injuries
MICHEL BÉDARD{{*, MELISSA FELTEAU{,
DWIGHT MAZMANIAN{, KARILYN FEDYK}, RUPERT KLEIN{,
JULIE RICHARDSON}, WILLIAM PARKINSON} and
MARY-BETH MINTHORN-BIGGS{
{ Lakehead Psychiatric Hospital, Ontario, Canada
{ Lakehead University, Ontario, Canada
} Canadian Paraplegic Association, Lakehead University, Ontario, Canada
} McMaster University, Lakehead University, Ontario, Canada

Distress Inventory of the SCL-90R (p = 0.054) approached

Accepted for publication: January 2003 statistical significance.
Conclusions: The intervention was simple, and improved
Abstract quality of life after other treatment avenues for these
participants were exhausted.
Primary objective: To examine the potential efficacy of a
mindfulness-based stress reduction approach to improve
quality of life in individuals who have suffered traumatic
brain injuries. Introduction
Research design: Pre-post design with drop-outs as controls.
Methods and procedures: We recruited individuals with mild to At least 25% of individuals with traumatic brain inju-
moderate brain injuries, at least 1 year post-injury. We ries (TBI) will have a residual impairment;1 this propor-
measured their quality of life, psychological status, and tion climbs to as much as 67% for individuals with
function. Results of 10 participants who completed the major TBI.2 Subtle residual deficits are evident even in
programme were compared to three drop-outs with complete
data. individuals where good recovery was believed to have
Experimental intervention: The intervention was delivered in taken place.3, 4 Residual deficits in executive functions,
12-weekly group sessions. The intervention relied on insight memory and learning are well-documented.5 – 7 At least
meditation, breathing exercises, guided visualization, and 50% of individuals who experienced TBI reported
group discussion. We aimed to encourage a new way of chronic pain, mostly in the form of headaches.8 Among
thinking about disability and life to bring a sense of
acceptance, allowing participants to move beyond limiting a consecutive set of patients referred for neuropsycholo-
beliefs. gical assessment, patients with TBI reported approxi-
Main outcomes and results: The treatment group mean quality mately twice as much pain and sleep problems than
of life (SF-36) improved by 15.40 (SD = 9.08) compared to general neurological patients.9 Reduced independence
7 1.67 (SD = 16.65; p = 0.036) for controls. Improvements on in activities of daily living and poor vocational
the cognitive-affective domain of the Beck Depression
Inventory II (BDI-II) were reported (p = 0.029), while changes outcomes have also been reported.7, 10 – 15
in the overall BDI-II (p = 0.059) and the Positive Symptom Brain injuries may lead to behavioural problems,
alcohol and drug use, and criminal behaviour.16 A great-
* Author for correspondence; Department of Psychology, er proportion of individuals with TBI is found in
Lakehead University, 955 Oliver Road, Thunder Bay, Ontario, psychiatric hospitals compared to general hospitals
P7B 5E1, Canada; e-mail: [email protected] and non-psychiatric medical clinics.17 Individuals with
Disability and Rehabilitation ISSN 0963–8288 print/ISSN 1464–5165 online # 2003 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
DOI: 10.1080/0963828031000090489
 Mindfulness-based intervention in traumatic brain injuries

TBI may be at risk of depression.18, 19 This risk may be self-concept, self-efficacy, and emotional issues are
related to a sense of learned helplessness that develops crucial to rehabilitation and reintegration of individuals
in some individuals.20 Others may have post-traumatic with TBI.31 Improvement in these domains and self-
stress syndrome, and many feel lonely.21 Compared to esteem may allow the injured persons to find more
people with spinal cord injuries, and non-disabled meaning in their lives.21
controls, individuals with TBI reported more unmet Group therapy and peer support groups focusing on
needs,22 and reported changes in,23 and often a loss of these issues are increasing in popularity. Inherent in
‘self’.24 these groups is a focus on healthy interactions with
The ultimate index of a successful rehabilitation is the others and reductions in isolation, the development of
extent to which individuals with TBI resume their pre- self-help and social skills, and reassurance and optimism
injury lifestyle and activities. This goal is often not about the future.32, 33 The ultimate goal is for the indivi-
met, even when considering only individuals with mild dual to attain his/her personal goals.21 Although the
trauma. Two main explanations can be proposed. First, evidence regarding the efficacy of consumer-run peer
recovery has reached a ceiling; individuals with TBI are support groups is recent and preliminary, the feasibility,
not expected to improve beyond the level achieved with and desirability of this type of intervention for a variety
current interventions. Second, individuals with TBI have of illnesses is documented.32, 34 Further, group therapy
a recovery potential that exceeds what current interven- incorporating mindfulness meditation provided substan-
tions can offer. That is, individuals with TBI may benefit tial benefits to individuals experiencing chronic pain,
from the added value of adjunct interventions to current anxiety, dermatological conditions, and cancer.35 – 41
ones. Improvements were evident after follow-ups of up to 3
The first explanation, that recovery has reached a years.37, 39, 40
plateau, is unsatisfactory. Recovery across individuals The primary short-term objective of the study was to
is very variable, with some recovering fully and others determine the potential effectiveness of a mindfulness-
recovering poorly; we have few explanations for this based group-support intervention for individuals who
variability.21 Further, the plateau explanation is defea- suffered mild/moderate TBI. We hypothesized the inter-
tist; it offers little hope and detracts imaginative pursuits vention would improve quality of life, decrease depres-
of novel approaches. On the other hand, the possibility sion symptoms, and increase the sense of control
that further progress can be achieved is theoretically experienced over their lives by individuals with TBI.
plausible. Given that recovery varies across individuals,
and that the human brain is probably more plastic than
traditionally believed,25 it is plausible that better Methods
outcomes are possible for individuals whose recovery
PARTICIPANTS, RECRUITMENT AND ELIGIBILITY
lags behind others with similar conditions.
Rehabilitation following TBI is typically classified The study was conducted in a small urban centre
into one of three stages: acute, sub-acute, and post- (population 125 000) located remotely from medical
acute.26 However, whereas physical, cognitive and beha- academic centres. A convenience sample of 21 partici-
vioural problems are usually addressed in TBI rehabili- pants was recruited from four sources: (1) a commu-
tation, psychological and emotional problems are rarely nity-based rehabilitation programme, (2) referrals from
targeted.21 Yet, these problems may pose a greater threat a local neuropsychologist, (3) the local brain injury asso-
to successful rehabilitation than physical problems.27 ciation, and (4) through media coverage (newspaper
Many individuals with TBI lack awareness into their articles, television, posters, and hand-outs). The study
situation in the initial stages of the post-injury period, was approved by the local Ethics Committee.
effectively reducing motivation.21 Emotional function Inclusion criteria included the ability to speak and
often deteriorates as individuals with TBI become aware read English, aged 18 years or more but less than 65
of their limitations,28 suggesting that improvement of years, and completion of traditional rehabilitation inter-
psychological and emotional function may translate into ventions. Another essential component of a successful
better outcomes. Consistently, others showed associa- intervention was insight into one’s condition.21 Unrealis-
tions between improvement in affect and self-awareness, tic self-appraisal is often found in individuals with TBI21
and goal attainment.29 and individuals with lack of insight were excluded.
Our desire to test a novel intervention stems from the Insight was determined by comparing participants’
realization that rehabilitation is a complex multi-factor- scores on the Patient Competency Rating Scale
ial process.30 Holistic approaches integrating awareness, (PCRS)42 with those obtained by their relatives or signif-

723
 M. Be´dard et al.

icant others (PCRS-R) as done by Prigatano and collea- into ‘Mental Health’ and ‘Physical Health’ components.
gues.43 We also excluded participants if they were chemi- Depression symptoms were measured with the Beck
cally dependent or had major concurrent mental illness Depression Inventory (BDI-II),48 which can be divided
(DSM-IV). Finally, participants were screened for suici- into a cognitive-affective factor and a somatic factor.49
dal ideation. The decision to include or exclude partici- The BDI-II has high internal consistency (alpha = 0.92)
pants was based on a consensus determined by the study in this population.50 Other psychological domains were
clinical psychologists, in addition to scores on the Symp- measured with the SCL-90-R, the Perceived Stress Scale
tom Checklist 90-Revised (SCL-90-R)44 and on Beck’s (PSS),51 and the Multidimensional Health Locus of
scale for suicidal ideation (BSS).45 Control Scale (MHLC).52 Two composite scores for
the SCL-90-R are presented. The Global Severity Index
(GSI) is a single score that indicates the current number
DESIGN
and intensity of symptoms. The Positive Symptom
We used a pre-post design with control group. Parti- Distress Index (PSDI) represents the symptom intensity
cipants were initially assessed for eligibility. Those who only for those symptoms that were endorsed. The locus
met criteria completed the remaining instruments and of control scale is divided into three domains: internal,
received the intervention. Following the intervention powerful others, and chance. Function was assessed
all questionnaires were administered again. Drop-outs with the home, social, and productive domains of the
were contacted to complete follow-up questionnaires Community Integration Questionnaire (CIQ).53 The
and were used as controls. scales have well-documented validity and reliability.
All data collection sessions were conducted at the
home of the participants. Baseline data were collected
INTERVENTION
using a two-stage process. Following informed consent,
The intervention consisted of a weekly, 12-week participants completed the basic demographics, SF-36,
group intervention, based on Kabat-Zinn’s mindful- SCL-90-R, BSS, and PCRS. Caregivers completed the
ness-based stress reduction programme41 and Kolb’s PCRS-R. For participants who continued with the
experiential learning cycle.46 To ensure the consistent study, a second meeting was arranged by the Research
application of these principles a manual was developed Assistant to complete the BDI-II, the MHLC scale,
and followed throughout the intervention. The and the CIQ. Post-intervention data were collected
approach emphasizes the importance of psychological approximately 4 months after baseline data, also in
well-being as an essential component of health and qual- two sessions at the home of the participants. Feedback
ity of life through present moment awareness and accep- about the content and format of intervention was
tance. Using insight meditation, breathing exercises, obtained by asking participants to complete a question-
guided visualization, and group discussion as the naire with seven questions using a four-point Likert
primary techniques, the facilitators encouraged partici- scale. The information was summarized to guide future
pants to use self-exploration as a tool to harness the programme improvement and is briefly presented in the
transformative power present in themselves, empower- results.
ing participants to exert control over their situations.
Through the exploration of various themes the interven-
STATISTICAL ANALYSES
tion ultimately sought to encourage a new way of think-
ing about disability and how to approach life to bring a To examine efficacy issues, pre-post changes for all
sense of acceptance, allowing participants to move participants who completed the intervention were exam-
beyond limiting beliefs. ined first with paired t-tests for the treatment group
alone. However, this approach did not control for
non-specific effects, hence, it was considered statistically
DATA COLLECTION
liberal (results are presented in tables but not discussed).
Basic demographics collected included age, gender, A more conservative approach used a two-way analyses
marital status, employment status, education, and medi- of variance (ANOVA) with drop-outs as controls for all
cation use. Three types of outcome measures were measures. The independent variables were: (1) time (pre-
collected: (1) quality of life, (2) psychological processes, and post-intervention), and (2) group (intervention or
and (3) function. To measure quality of life we used the control). Given that the hypotheses called for an
Short Form Health Survey (SF-36),47 a generic health- improvement among the intervention group but not
related quality of life instrument. The SF-36 is divided controls, the time by group interaction was the critical

724
 Mindfulness-based intervention in traumatic brain injuries

effect. Person r coefficients were used to assess correla- The SF-36 Mental Health score of the intervention
tions. Given the low statistical power of the analyses group improved by 15.40 (SD = 9.08) at follow-up
effect sizes are presented. compared to 7 1.67 (SD = 16.65) for the control group
(F(1,11) = 5.70, p = 0.036). The SF-36 Physical Health
was unchanged (F(1,11) = 1.75, p = 0.213, table 2).
The treatment by time interaction for the overall BDI-
Results
II approached statistical significance (F(1,10) = 4.54,
Of the 39 individuals referred by professionals and p = 0.059; see table 3). Depression symptoms were
self-referred, 21 of them agreed to be interviewed, almost halved in the intervention group (mean reduc-
and 19 met inclusion criteria. Ten participants tion = 8.73, SD = 7.78). The effect size of 0.312 was in
completed the programme; three drop-outs agreed to the medium to large range. When we divided BDI-II
be interviewed as the control group. Demographics scores into cognitive-affective and somatic domains we
are presented in table 1. With the exception of gender found a significant improvement on the cognitive-affec-
most characteristics were similar between the treatment tive domain (F(1,10) = 6.48, p = 0.029). The somatic
and control participants. The mean number of medica- domain did not reveal changes (F(1,10) = 0.87,
tions taken by participants was 3.10 (SD = 2.85) for p = 0.374).
the treatment group and 2.90 (SD = 3.00) for controls. It appeared possible that participants who reported
Anti-depressants were the most frequently used type of improvements in quality of life (Mental Health) were
medication; 50% of participants were taking anti- the same who reported reductions in depression symp-
depressants prior to and after the treatment sessions. toms. To test this possibility we correlated changes in
Pearson correlations between participants’ PCRS SF-36 (Mental Health) with changes in BDI-II scores.
scores and scores from significant others revealed a Data from all participants for whom data on both vari-
moderate relationship (r(8) = 0.57). The respective ables at both time points were available (controls and
means for participants and significant others were intervention) were used. We obtained a correlation coef-
87.22 (SD = 15.12) and 86.67 (SD = 25.67). These data ficient of 0.46 (p = 0.132). This relationship is depicted
confirmed that participants’ insight into their limita- in figure 1.
tions were corroborated by relatives. The Global Severity Index (GSI; SCL-90R) remained
unchanged after the intervention (F(1,11) = 1.30,
p = 0.278). However, the Positive Symptom Distress
Table 1 Demographics Index (PSDI) approached statistical significance
(F(1,11) = 4.63, p = 0.054; see table 4), with a moderate
Treatment Group Control Group
Variables (n = 10) (n = 3) to large effect size. The PSS interaction was not statisti-
cally significant (F(1,10) = 0.34, p = 0.575; table 4).
Women (%) 7 (70%) 0 (0%) Internal HLC approached statistical significance
Mean age (min, max) 43 (24, 55) 39 (26, 46)
Married/partner 3 (30%) 1 (33%) (F(1,10) = 4.79, p = 0.053), but this effect appeared
Eduction 4 high school 6 (60%) 2 (67%) explained by a shift towards external locus of control
Employed/student 3 (30%) 1 (33%) among control participants rather than a shift towards
Lives alone 2 (20%) 0 (0%)
internal locus of control among those who completed

Table 2 SF-36

Effect
Variable Treatment Control ANOVA F p size t (df)* p*

SF-36 Mental Health ( + )

Pre 36.70 (10.44) 40.33 (20.82) Time 3.69 0.081 0.251
Post 52.10 (4.61) 38.67 (9.87) Group 0.77 0.399 0.065 7 5.36 (9) 0.001
Diff 15.40 (9.08) 7 1.67 (16.65) T6G 5.70 0.036 0.341
SF-36 Physical Health ( + )
Pre 42.10 (8.96) 53.00 (9.64) Time 0.54 0.476 0.047
Post 46.80 (11.14) 51.67 (16.56) Group 1.37 0.267 0.110 7 2.20 (9) 0.055
Diff 4.70 (6.73) 7 1.33 (7.77) T6G 1.75 0.213 0.137

Note: SF-36 = Short Form Health Survey; BSS = Beck Suicide Scale; (SD); ( + ) = improvement is based on higher test scores; * t-values (df) and
p-values are for the treatment group only

725
 M. Be´dard et al.

Table 3 Beck Depression Inventory-II

Effect
Variable Treatment Control ANOVA F p Size t (df)* p*

BDI-II ( 7 )
Pre 18.43 (12.18) 12.50 (7.78) Time 0.83 0.384 0.077
Post 9.70 (10.64) 16.00 (9.90) Group 0.00 0.982 0.000 3.55 (9) 0.006
Diff 8.73 (7.78) 7 3.50 (2.12) T6G 4.54 0.059 0.312
Cognitive-Affective ( 7 )
Pre 9.66 (7.54) 3.00 (2.83) Time 0.15 0.705 0.015
Post 4.90 (5.80) 6.50 (4.95) Group 0.28 0.607 0.027 3.45 (9) 0.007
Diff 4.76 (4.36) 7 3.50 (2.12) T6G 6.48 0.029 0.393
Somatic ( 7 )
Pre 5.80 (3.58) 6.00 (2.83) Time 1.95 0.193 0.163
Post 3.30 (3.77) 5.50 (3.54) Group 0.21 0.655 0.021 2.71 (9) 0.024
Diff 2.50 (2.22) 0.50 (0.71) T6G 0.87 0.374 0.080

Note: BDI-II = Beck Depression Inventory 2nd edition; values are means (SD); ( 7 ) = improvement is based on lower test scores; * t-values (df) and
p-values are for the treatment group only

Discussion
One of the primary objectives of this intervention was
to improve the quality of life of individuals who
sustained TBI. Consistently, the Mental Health
summary score of the SF-36 increased from 37 at base-
line to 52 after the intervention. This finding was both
statistically and clinically significant. The follow-up
mean is now equivalent to Canadian normative data
for men and women aged 35 – 44 years, which ranges
from 50.5 to 53.5.54 The overall change recorded for
depression symptoms approached statistical signifi-
cance. The treatment group mean dropped by almost
50% whereas the control group mean increased. The
Figure 1 Depicted is the relationship between changes in effect size was strong, suggesting that a larger trial
depression symptoms (BDI-II change; negative change in- would achieve statistical significance. The overall change
dicates improvement) and changes in quality of life (SF-36 on the BDI-II was mostly the results of changes in the
change, Mental Health; positive change indicates improve- cognitive domain scores (which achieved statistical
ment). The closed squares represent actual data points, the line significance). These findings are consistent with the
is the best fitting regression line.
conceptual focus of the intervention. While improve-
ments were noted on psychological/emotional domains,
we did not find changes on the Physical Health summary
the programme. Both the powerful others and chance of the SF-36 or on the somatic domain of the BDI-II.
locus of control domains were not significantly different Results on the SCL-90-R did not reveal an improve-
between completers and controls (F(1,10) = 1.16, ment but must be examined in light of the inclusion/
p = 0.307 and F(1,10) = 2.54, p = 0.142; see table 5). exclusion criteria; potential participants who had severe
We did not find changes on the CIQ following the inter- symptoms were excluded from the study, possibly creat-
vention (see table 6). ing a floor effect. Nevertheless, improvements on the
Nine participants completed and returned Positive Symptom Distress Inventory sub-scale
programme evaluations. The mean response for all approached statistical significance with a good effect
seven questions was 3.34 on a 4-point Likert-type size. These results are consistent with those observed
scale suggesting positive evaluations of the interven- with the SF-36 and the BDI-II, and results reported
tion. Participants agreed with statements such as ‘the by others.35 However, using the PSS we found reduc-
facilitators were effective’ and ‘the sessions were well tions in perceived stress equivalent in both the treatment
organized’. and control groups. Therefore, the impact of the inter-

726
 Mindfulness-based intervention in traumatic brain injuries

Table 4 Global Severity Index (GSI) and the Positive Symptom Distress Inventory (PSDI; SCL-90-R) and Perceived Stress Scale (PSS)

Variable Treatment Control ANOVA F p Effect Size t (df)* p*

GSI ( 7 )
Pre 1.45 (0.86) 1.16 (0.96) Time 7.59 0.019 0.408
Post 0.89 (0.66) 0.92 (0.63) Group 0.07 0.798 0.006 3.88 (9) 0.004
Diff 0.55 (0.46) 0.23 (0.34) T6G 1.30 0.278 0.106
PSDI ( 7 )
Pre 2.23 (0.67) 1.93 (0.90) Time 5.92 0.033 0.350
Post 1.63 (0.53) 1.89 (0.94) Group 0.00 0.955 0.000 4.33 (9) 0.002
Diff 0.60 (0.44) 0.04 (0.07) T6G 4.63 0.054 0.296
PSS ( 7 )
Pre 30.98 (8.50) 32.00 (12.73) Time 8.01 0.018 0.445
Post 22.40 (7.12) 19.00 (18.38) Group 0.04 0.842 0.004 2.66 (9) 0.026
zDiff 8.58 (10.20) 13.00 (5.65) T6G 0.34 0.575 0.033

Note: GSI = Global Severity Index; PSDI = Positive Symptom Distress Inventory; PSS = Perceived Stress Scale; values are means (SD);
( 7 ) = improvement is based on lower test scores; * t-values (df) and p-values are for the treatment group only

Table 5 Health locus of control scales

Effect
Variable Treatment Control ANOVA F p Size t (df)* p*

IHLC ( + )
Pre 24.10 (6.49) 22.50 (4.95) Time 3.81 0.079 0.276
Post 24.50 (5.48) 15.50 (0.71) Group 1.62 0.232 0.139 7 0.29 (9) 0.779
Diff 0.40 (4.38) 7 7.00 (4.24) T6G 4.79 0.053 0.324
PHLC ( 7 )
Pre 19.00 (4.83) 19.50 (4.95) Time 1.33 0.277 0.117
Post 19.10 (6.03) 22.50 (6.36) Group 0.23 0.639 0.023 7 0.09 (9) 0.930
Diff 7 0.10 (3.63) 3.00 (1.41) T6G 1.16 0.307 0.104
CHLC ( 7 )
Pre 19.00 (3.83) 21.00 (0.00) Time 0.02 0.888 0.002
Post 22.00 (4.29) 18.50 (3.54) Group 0.09 0.772 0.009 7 2.09 (9) 0.067
Diff 7 3.00 (4.54) 2.50 (3.54) T6G 2.54 0.142 0.203

Note: IHLC = Internal Health Locus of Control; PHLC = Powerful Others Health Locus of Control; CHLC = Chance Health Locus of Control;
values are means (SD); ( + ) = improvement is based on higher test scores; ( 7 ) = improvement is based on lower test scores; * t-values (df) and p-
values are for the treatment group only

Table 6 Community integration questionnaire

Effect
Variable Treatment Control ANOVA F p Size t (df)* p*

CIQ Home ( + )
Pre 7.00 (2.98) 4.50 (2.12) Time 1.89 0.199 0.159
Post 7.30 (2.31) 6.00 (5.66) Group 0.80 0.392 0.074 7 0.71 (9) 0.500
Diff 0.30 (1.34) 1.50 (3.54) T6G 0.84 0.381 0.077
CIQ Social ( + )
Pre 5.60 (2.22) 8.00 (2.83) Time 2.59 0.139 0.206
Post 6.45 (2.52) 9.50 (0.71) Group 2.68 0.133 0.211 7 1.48 (9) 0.180
Diff 0.85 (1.86) 1.50 (2.12) T6G 0.20 0.666 0.019
CIQ Productive ( + )
Pre 1.56 (1.24) 1.00 (1.41) Time 2.62 0.140 0.225
Post 1.89 (0.93) 2.00 (0.00) Group 0.09 0.771 0.010 7 1.00 (9) 0.350
Diff 0.33 (1.00) 1.00 (1.41) T6G 0.66 0.439 0.068

Note: CIQ = Community Integration Scale; values are means (SD); ( + ) = improvement is based on higher test scores; * t-values (df) and p-values
are for the treatment group only

727
 M. Be´dard et al.

vention on the stress experienced by participants aged women may have had stronger motivation to enrol
remains difficult to interpret. and complete the programme. The majority of men and
We noted a deterioration of internal locus of control younger participants either did not enrol or dropped out
among control participants but no change among of the programme. Regarding retention, transportation
completers. However, it is impossible to determine if this was a concern for many participants especially in the
effect represents a spontaneous deterioration for winter months. Several participants had mobility
controls, a statistical aberration (given the sample size), problems that made attendance to the weekly sessions
or a genuine protective effect of the intervention. This a challenge. Many participants lived out of reach from
remains to be elucidated. This finding did not support the public transportation system and were dependent
our hypothesized gain in internal locus of control. on others for transportation.
Further work is necessary to unveil the mechanism Despite these shortcomings we are encouraged that
underlying the quality of life improvement reported the intervention improved quality of life and possibly
here. A possible explanation is a relationship between reduced depression in individuals who had exhausted
quality of life and depression symptoms.55, 56 Our corre- all previous therapeutic avenues. These findings are
lation between changes in quality of life and changes in especially encouraging for women. Using a meta-analy-
depression symptoms was in this direction but did not sis of eight studies Farace and Alves58 concluded that
achieve statistical significance. Nonetheless, 20% of women have worse outcomes than men after a TBI on
the variability in quality of life change was explained 17 out of 20 outcome measures. One possible explana-
by the variability in depression score change, and the tion for this finding is that men and women have differ-
relationship would have achieved statistical significance ent needs and current interventions target men’s needs
with seven more participants. more successfully. If such is the case, our intervention
Participants’ functioning, as assessed by the CIQ, may help reduce this inequality. Further research should
revealed no sign of improvement. Others have also investigate gender differences in more details, including
reported little changes in function (using the CIQ) the reasons underlying young men’s reluctance to parti-
beyond immediate post-acute gains.57 Furthermore, cipate in this study.
changes on these outcomes may take place months from Future research should also include follow-ups. The
programme cessation, and may require the sustained positive impact of the intervention on quality of life
improvement observed on quality of life and depression may not be maintained in the long-term. On the other
symptoms. hand, it is possible that improvements in other
There are several limitations to the present study that domains (e.g., return to work) will emerge at a later
must be addressed in future research. Foremost is the time. Larger trials would also improve statistical
sample size. The validity of the statistical analyses may power and provide more precise estimates of the inter-
be questioned, and the statistical power was very low, vention effectiveness.
especially in the context of two-way interactions. Finally, future work should include caregivers as
However, we deliberately chose a conservative analytical intervention targets. Loved ones, relatives, and friends
plan to reduce the risk of Type I errors. Nonetheless, the are often affected, albeit indirectly, by TBI.59 – 64 Care-
findings were consistent with the conceptual focus of the givers of individuals with TBI are at higher risk of
intervention and effect sizes were strong. Another limita- experiencing psychological and emotional problems
tion related to the sample size is our inability to consider than controls.65 – 68 Ultimately, caregiver depression,
the impact of medication use or number of years post- coping, and social supports may have a larger impact
injury on our measures. The current sample was on family functioning than injury severity.69
approximately 3 – 10 years post injury, which leaves a There is reason to believe that multidisciplinary inter-
considerable time between conventional approaches to ventions that can improve the situation of individuals
recovery and this intervention. with TBI will positively affect caregivers and relatives.70
The small sample size can be explained by recruitment The intervention we used may also have had a beneficial
and retention difficulties. These difficulties prevent us effect on caregivers. However, it would be misleading to
from generalizing the present findings to the general assume that addressing the needs of individuals with
population of individuals who suffered TBI. Only 50% TBI is sufficient to fully support caregivers. Caregivers
of all potential participants started the programme and may have unmet needs of their own, such as the need
fewer completed it. We experienced difficulties recruiting for emotional support,71 that need to be addressed sepa-
young men despite the prevalence of TBI in this group. rately. Ultimately, interventions focusing on the whole
Judging from those who completed the study middle- family are required.72

728
 Mindfulness-based intervention in traumatic brain injuries

Overall, the objectives of this pilot study were to 9 Beetar JT, Guilmette TJ, Sparadeo FR. Sleep and pain complaints
determine the potential effectiveness of a holistic mind- in symptomatic traumatic brain injury and neurologic populations.
Archives of Physical Medicine and Rehabilitation 1996; 77: 1298 –
fulness-based intervention as an adjunct to conven- 1302.
tional medical and rehabilitation interventions. Data 10 Greenspan AI, Wrigley JM, Kresnow M, Branche-Dorsey CM,
on the SF-36 (Mental Health) are very encouraging Fine PR. Factors influencing failure to return to work due to
traumatic brain injury. Brain Injury 1996; 10: 207 – 218.
and warrant further research. Furthermore, while the 11 Mazaux JM, Masson F, Levin HS, Alaoui P, Maurette P, Barat M.
availability of conventional rehabilitation programmes Long-term neuropsychological outcome and loss of social auton-
is sometimes limited because of geographical or finan- omy after traumatic brain injury. Archives of Physical Medicine and
Rehabilitation 1997; 78: 1316 – 1320.
cial reasons,73 and this applies to the remote region 12 Godfrey HP, Bishara SN, Partridge FM, Knight RG. Neuropsy-
where this study was conducted, we are encouraged chological impairment and return to work following severe closed
that a simple intervention, delivered by individuals head injury: Implications for clinical management. New Zealand
Medical Journal 1993; 106: 301 – 303.
who do not have advanced medical training, may have 13 Ruff RM, Marshall LF, Crouch J, Klauber MR, Levin HS, Barth
the potential to improve quality of life. Future research J, Kreutzer J, Blunt BA, Foulkes MA, Eisenberg HM. Predictors of
is required to determine if this intervention could be outcome following severe head trauma: follow-up data from the
Traumatic Coma Data Bank. Brain Injury 1993; 7: 101 – 111.
the building block for an adjunct rehabilitation 14 Vogenthaler DR, Smith KRJ, Goldfader P. Head injury, an
programme. empirical study: Describing long-term productivity and indepen-
dent living outcome. Brain Injury 1989; 3: 355 – 368.
15 Dombovy ML, Olek AC. Recovery and rehabilitation following
traumatic brain injury. Brain Injury 1997; 11: 305 – 318.
Acknowledgements 16 Ommaya AK, Salazar AM, Dannenberg AL, Chervinsky AB,
Schwab K. Outcome after traumatic brain injury in the US military
We extend our sincerest thanks to Ms. Alice Bellavance and Dr. medical system. The Journal of Trauma 1996; 41: 972 – 975.
M.A. Mountain for their invaluable help in the recruitment process 17 McGuire LM, Burright RG, Williams R, Donovick PJ. Prevalence
and their participation in the Advisory Committee. We also thank of traumatic brain injury in psychiatric and non-psychiatric
other members of the Advisory Committee: Ms. Michele Meehan, Mr. subjects. Brain Injury 1998; 12: 207 – 214.
David Shannon, Ms. Erin May, Ms. Nancy Debruyne, and Ms. Janice 18 McCleary C, Satz P, Forney D, Light R, Zaucha K, Asarnow R,
Cerra for their co-operation towards the successful completion of this Namerow N. Depression after traumatic brain injury as a function
of Glasgow Outcome Score. Journal of Clinical and Experimental
study. This study was supported by a research grant from the Ontario
Neuropsychology 1998; 20: 270 – 279.
Neurotrauma Foundation. The results and conclusions are those of the
19 Satz P, Forney DL, Zaucha K, Asarnow RR, Light R, McCleary
authors. C, Levin H, Kelly D, Bergsneider M, Hovda D, Martin N,
Namerow N, Becker D. Depression, cognition, and functional
correlates of recovery outcome after traumatic brain injury. Brain
Injury 1998; 12: 537 – 553.
References 20 Moore AD, Stambrook M. Cognitive moderators of outcome
following traumatic brain injury: A conceptual model and
1 Thurman DJ, Jeppson L, Burnett CL, Beaudoin DE, Rheinberger implications for rehabilitation. Brain Injury 1995; 9: 109 – 130.
MM, Sniezek JE. Surveillance of traumatic brain injuries in Utah. 21 Ponsford J. Traumatic Brain Injury: Rehabilitation for Everyday
Western Journal of Medicine 1996; 165: 192 – 196. Adaptive Living. Hove, UK: Lawrence Erlbaum Assoc, 1995.
2 van Balen HG, Mulder T, Keyser A. Towards a disability-oriented 22 Brown M, Vandergoot D. Quality of life for individuals with
epidemiology of traumatic brain injury. Disability and Rehabilita- traumatic brain injury: Comparison with other living in the
tion 1996; 18: 181 – 190. community. Journal of Head Trauma Rehabilitation 1998; 13: 1 –
3 Conzen M, Ebel H, Swart E, Skreczek W, Dette M, Oppel F. Long- 23.
term neuropsychological outcome after severe head injury with 23 Tyerman A, Humphrey M. Changes in self-concept following
good recovery. Brain Injury 1992; 6: 45 – 52. severe head injury. International Journal of Rehabilitation Research
4 Stuss DT, Ely P, Hugenholtz H, Richard MT, LaRochelle S, 1984; 7: 11 – 23.
Poirier CA, Bell I. Subtle neuropsychological deficits in patients 24 Nochi M. ‘‘Loss of self’’ in the narratives of people with traumatic
with good recovery after closed head injury. Neurosurgery 1985; 17: brain injury: A qualitative analysis. Social Science and Medicine
41 – 47. 1998; 46: 869 – 878.
5 Leon-Carrion J, Alarcon JC, Revuelta M, Murillo-Cabezas F, 25 Wilson BA. Recovery of cognitive functions following nonpro-
Dominguez-Roldan JM, Dominguez-Morales MR, Machuca- gressive brain injury. Current Opinions in Neurobiology 1998; 8:
Murga F, Forastero P. Executive functioning as outcome in 281 – 287.
patients after traumatic brain injury. International Journal of 26 Mazaux JM, Richer E. Rehabilitation after traumatic brain injury
Neuroscience 1998; 94: 75 – 83. in adults. Disability and Rehabilitation 1998; 20: 435 – 447.
6 Woischneck D, Firsching R, Ruckert N, Hussein S, Heissler H, 27 Brazil K. Assessing the consequences of traumatic brain injury.
Aumuller E, Dietz H. Clinical predictors of the psychosocial long- International Journal of Rehabilitation Research 1992; 15: 93 – 101.
term outcome after brain injury. Neurological Research 1997; 19: 28 Fordyce D, Roueche JR, Prigatano GP. Enhanced emotional
305 – 310. reactions in chronic head trauma patients. Journal of Neurology,
7 Ponsford JL, Olver HJ, Curran C. A profile of outcome: 2 years Neurosurgery, and Psychiatry 1983; 46: 620 – 624.
after traumatic brain injury. Brain Injury 1995; 9: 1 – 10. 29 Prigatano GP, Wong JL. Cognitive and affective improvement in
8 Lahz S, Bryant RA. Incidence of chronic pain following traumatic brain dysfunctional patients who achieve inpatient rehabilitation
brain injury. Archives of Physical Medicine and Rehabilitation 1996; goals. Archives of Physical Medicine and Rehabilitation 1999; 80:
77: 889 – 891. 77 – 84.

729
 M. Be´dard et al.

30 McColl MA, Carlson P, Johnston J, Minnes P, Shue K, Davies D, 50 Green A, Felmingham K, Baguley IJ, Slewa-Younan S, Simpson S.
Karlovits T. The definition of community integration: Perspectives The clinical utility of the Beck Depression Inventory after
of people with brain injuries. Brain Injury 1998; 12: 15 – 30. traumatic brain injury. Brain Injury 2001; 15: 1021 – 1028.
31 Ben-Yishay Y, Diller L. Cognitive remediation in traumatic brain 51 Cohen S, Kamarck T, Mermelstein R. A global measure of
injury: Update and issues. Archives of Physical Medicine and perceived stress. Journal of Health and Social Behavior 1983; 24:
Rehabilitation 1993; 74: 204 – 213. 385 – 396.
32 Dunn J, Steginga SK, Occhipinti S, Wilson K. Evaluation of a peer 52 Wallston KA, Wallston BS, DeVellis R. Development of the
support program for women with breast cancer – Lessons for multidimensional health locus of control (MHLC) scales. Health
practitioners. Journal of Community and Applied Social Psychology Education Monographs 1978; 6: 160 – 171.
1999; 9: 13 – 22. 53 Sander AM, Fuchs KL, High WMJ, Kreutzer JS, Rosenthal M.
33 Wilson ME, Flanagan S, Rynders C. The FRIENDS program: A The community integration questionnaire revisited: An assessment
peer support group model for individuals with psychiatric of factor structure and validity. Archives of Physical Medicine and
disability. Psychiatric Rehabilitation Journal 1999; 22: 239 – 247. Rehabilitation 1999; 80: 1303 – 1308.
34 Davidson L, Chinman M, Kloos B, Weingarten R, Stayner D, 54 Hopman WM, Towheed T, Anastassiades T, Tenehouse A,
Tebes JK. Peer support among individuals with severe mental Poliquin S, Berger C, Joseph L, Brown JP, Murray TM, Adachi
illness: A review of the evidence. Clinical Psychology: Science and JD, Hanley DA, Papadimitropoulos E. Canadian normative data
Practice 1999; 6: 165 – 187. for the SF-36 health survey. Canadian Medical Association Journal
35 Astin JA. Stress reduction through mindfulness meditation. 2000; 163: 265 – 271.
Psychotherapy and Psychosomatics 1997; 66: 97 – 106. 55 Steadman-Pare D, Colantonio A, Ratcliff G, Chase S, Vernich L.
36 Kabat-Zinn J, Wheeler E, Light T, Skillings A, Scharf MJ, Cropley Factors associated with perceived quality of life many years after
TG, Hosmer D, Bernhard JD. Influence of a mindfulness traumatic brain injury. Journal of Head Trauma Rehabilitation
meditation-based stress reduction intervention on rates of skin 2001; 16: 330 – 342.
clearing in patients with moderate to severe psoriasis undergoing 56 Corrigan JD, Bogner JA, Mysim WJ, Clinchot D, Fugate L. Life
phototherapy (UVB) and photochemotherapy (PUVA). Psychoso- satisfaction after traumatic brain injury. Journal of Head Trauma
matic Medicine 1998; 60: 625 – 632. Rehabilitation 2001; 16: 543 – 555.
37 Miller JJ, Fletcher K, Kabat-Zinn J. Three-year follow-up and 57 Sander AM, Roebuck TM, Struchen MA, Sherer M, High Jr WM.
clinical implications of a mindfulness meditation-based stress Long-term maintenance of gains obtained in postacute rehabilita-
reduction intervention in the treatment of anxiety disorders. tion by persons with traumatic brain injury. Journal of Head
General Hospital Psychiatry 1995; 17: 192 – 200. Trauma Rehabilitation 2001; 16: 356 – 373.
38 Massion AO, Teas J, Hebert JR, Wertheimer MD, Kabat-Zinn J. 58 Farace E, Alves WM. Do women fare worse: A metaanalysis of
Meditation, melatonin and breast/prostate cancer: hypothesis and gender differences in traumatic brain injury outcome. Journal of
preliminary data. Medical Hypotheses 1995; 44: 39 – 46. Neurosurgery 2000; 93: 539 – 545.
39 Kabat-Zinn J, Massion A, Kristeller J, Peterson LG, Fletcher KE, 59 Lezak MD. Brain damage is a family affair. Journal of Clinical and
Pbert L, Lenderking WR, Santorelli SF. Effectiveness of a Experimental Neuropsychology 1988; 10: 111 – 123.
meditation-based stress reduction program in the treatment of 60 Watanabe Y, Shiel A, McLellan DL, Kurihara M, Hayashi K. The
anxiety disorders. American Journal of Psychiatry 1992; 149: 936 – impact of traumatic brain injury on family members living with
943. patients: A preliminary study in Japan and the UK. Disability and
40 Kabat-Zinn J, Lipworth L, Burney R. The clinical use of Rehabilitation 2001; 23: 370 – 378.
mindfulness meditation for the self-regulation of chronic pain. 61 Lezak MD. Living with the characterologically altered brain
Journal of Behavioral Medicine 1985; 8: 163 – 190. injured patient. Journal of Clinical Psychiatry 1978; 39: 592 – 598.
41 Kabat-Zinn J. An outpatient program in behavioral medicine for 62 Frosch S, Gruber A, Jones C, Myers S, Noel E, Westerlund A,
chronic pain patients based on the practice of mindfulness Zavisin T. The long term effects of traumatic brain injury on the
meditation: theoretical considerations and preliminary results. roles of caregivers. Brain Injury 1997; 11: 891 – 906.
General Hospital Psychiatry 1982; 4: 33 – 47. 63 Wade SL, Taylor HG, Drotar D, Stancin T, Yeates KO. Family
42 Prigatano GP, Bruna O, Mataro M, Munoz JM, Fernandez S, burden and adaptation during the initial year after traumatic brain
Junque C. Neuropsychological Rehabilitation After Brain Injury. injury in children. Pediatrics 1998; 102: 110 – 116.
Baltimore: Johns Hopkins University Press, 1986. 64 Kreutzer JS, Gervasio AH, Camplair PS. Primary caregivers’
43 Prigatano GP, Bruna O, Mataro M, Munoz JM, Fernandez S, psychological status and family functioning after traumatic brain
Junque C. Initial disturbances of consciousness and resultant injury. Brain Injury 1994; 8: 197 – 210.
impaired awareness in Spanish patients with traumatic brain 65 Gillen R, Tennen H, Affleck G, Steinpreis R. Distress, depressive
injury. Journal of Head Trauma Rehabilitation 1998; 13: 29 – 38. symptoms, and depressive disorder among caregivers of patients
44 Derogatis LR. SCL-90-R Administration, Scoring, and Procedures with brain injury. Journal of Head Trauma Rehabilitation 1998; 13:
Manual. Minneapolis: National Computer Systems, 1994. 31 – 43.
45 Beck AT, Steer RA, Ranieri WF. Beck Scale for Suicidal Ideation. 66 Linn RT, Allen K, Willer BS. Affective symptoms in the chronic
Toronto: The Psychological Corporation, Harcourt Brace, 1988. stage of traumatic brain injury: A study of married couples. Brain
46 Kolb, DA. Experiential Learning: Experience As the Source of Injury 1994; 8: 135 – 147.
Learning and Development. Englewood Cliffs, NJ: Prentice-Hall, 67 Gosling J, Oddy M. Rearranged marriages: Marital relationships
1984. after head injury. Brain Injury 1999; 13: 785 – 796.
47 Ware JE, Kosinski M. SF-36 Physical and Mental Health 68 Pessar LF, Coad ML, Linn RT, Willer BS. The effects of parental
Summary: A Manual for Users of Version 1. Lincoln, RI: traumatic brain injury on the behaviour of parents and children.
QualityMetric, 2001. Brain Injury 1993; 7: 231 – 240.
48 Beck AT, Steer RA, Brown GK. Beck Depression Inventory-II 69 Douglas JM, Spellacy FJ. Indicators of long-term family function-
(BDI-II). Toronto: The Psychological Corporation, Harcourt ing following severe traumatic brain injury in adults. Brain Injury
Brace, 1996. 1996; 10: 819 – 839.
49 Whisman MA, Perez JE, Ramel W. Factor structure of the Beck 70 Semlyen JK, Summers SJ, Barnes MP. Traumatic brain injury:
Depression Inventory - second edition (BDI-II) in a student Efficacy of multidisciplinary rehabilitation. Archives of Physical
sample. Journal of Clinical Psychology 2000; 56: 545 – 551. Medicine and Rehabilitation 1998; 79: 678 – 683.

730
 Mindfulness-based intervention in traumatic brain injuries

71 Sinnkaruppan I, Williams DM. Family carers and the adult head- 73 Delmonico RL, Hanley-Peterson P, Englander J. Group psy-
injured: A critical review of carers’ needs. Brain Injury 2001; 15: chotherapy for persons with traumatic brain injury: Management
653 – 672. of frustration and substance abuse. Journal of Head Trauma
72 Harris JK, Godfrey HP, Partridge FM, Knight RG. Caregiver Rehabilitation 1998; 13: 10 – 22.
depression following traumatic brain injury (TBI): A consequence
of adverse effects on family members?. Brain Injury 2001; 15: 223 –
238.

731