Feedback interventions for improving self-awareness after brain injury: A protocol for a pragmatic randomised controlled trial

Research Article

Feedback interventions for improving self-awareness afterbrain injury: A protocol for a pragmatic randomisedcontrolled trial

Julia Schmidt,1,2 Jennifer Fleming,2,3 Tamara Ownsworth,4 Natasha Lannin5 andAsad Khan2,3

1Royal Rehabilitation Centre Sydney, Ryde, New South Wales, 2School of Health and Rehabilitation Sciences, TheUniversity of Queensland, Brisbane, Queensland, 3Princess Alexandra Hospital, Brisbane, Queensland, 4School ofPsychology, Griffith University, Brisbane, Queensland, and 5Sydney Medical School, Rehabilitation Studies Unit,The University of Sydney, Sydney, New South Wales, Australia

Background: Occupational therapists working in braininjury rehabilitation use functional tasks as a means ofproviding feedback to improve self-awareness of peoplewho have a brain injury and ultimately improve theiroccupational performance.Purpose: To compare the effectiveness of video, verbal andexperiential feedback for improving self-awareness in peo-ple with traumatic brain injury.Methods: A randomised controlled trial will be conductedto compare the efficacy of video and verbal feedback duringoccupational therapy. Fifty-four participants with trau-matic brain injury will be randomly allocated into threeintervention groups: (i) video plus verbal feedback, (ii)verbal feedback and (iii) experiential feedback (control con-dition). Participants will receive the allocated interventionbased on performance of a meal preparation task. Theintervention sessions will occur four times during a two-week period. Blinded assessment will occur at baseline,post-intervention, and two months follow up. The primaryoutcome will be a measure of on-line self-awareness (num-ber of self-corrected and therapist corrected errors). Sec-ondary outcomes to be assessed include levels ofintellectual self-awareness, emotional distress, and accep-tance of disability. Data will be analysed using an inten-

tion to treat approach. Linear mixed effects models will beused to investigate the intervention effects.Findings and Implications: Results will contribute toevidence-based guidelines to support therapists to choosethe most effective form of feedback for people withdecreased self-awareness after traumatic brain injury.

KEY WORDS feedback, rehabilitation, traumatic braininjury.

Introduction

Self-awareness is described as an understanding of

one’s strengths and limitations and their impact on

everyday functioning (Schlund, 1999). It is conceptua-

lised as having two components, (i) intellectual aware-

ness or the knowledge and beliefs about one’s abilities,

and (ii) on-line awareness including self-monitoring and

self-correction of errors during task performance (Toglia

& Kirk, 2000). Self-awareness is necessary for successful

participation in occupational roles such as work, study,

safe independent living, and social relationships (Flem-

ing, Strong & Ashton, 1998).

Impaired self-awareness is common after traumatic

brain injury (TBI) with some reports of up to 97% of

individuals showing some level of impairment of self-

awareness of their deficits (Sherer et al., 1998b). Devel-

oping self-awareness in people after a TBI is an essen-

tial aspect of the rehabilitation process. When people

have impaired self-awareness after TBI, they can have

limited motivation to participate in therapy, decreased

incentive to use compensatory strategies, difficulty set-

ting achievable goals and difficulty achieving a produc-

tive and independent lifestyle in the community

(Fleming & Ownsworth, 2006; Ownsworth, Fleming,

Desbois, Strong & Kuipers, 2006; Sherer et al., 1998b).

Julia Schmidt BSc (OT); Occupational Therapist, PhD can-didate. Jennifer Fleming PhD; Associate Professor. TamaraOwnsworth PhD; Senior Lecturer. Natasha Lannin PhD;Senior Lecturer. Asad Khan PhD; Senior Lecturer.

Correspondence: Julia Schmidt, Royal Rehabilitation CentreSydney, PO Box 6, Ryde, NSW 1680, Australia. Email: [email protected]

Accepted for publication 23 January 2012.

© 2012 TheAuthorsAustralianOccupational Therapy Journal© 2012OccupationalTherapyAustralia

Australian Occupational Therapy Journal (2012) 59, 138–146 doi: 10.1111/j.1440-1630.2012.00998.x

However, the development of self-awareness has been

linked to increased emotional distress as the individual

begins to understand the full extent of effects of their

injury (Fleming et al., 1998).

Provision of feedback to improveself-awareness

Feedback interventions, which are a component of occu-

pational-based metacognitive training, are used to

improve self-awareness in people with TBIs (Fleming,

2009). Meta-cognitive training uses real-life settings and

meaningful activities partnered with a supportive and

therapeutic context to facilitate error recognition, realis-

tic goal setting and strategy use (Fleming, 2009; Flem-

ing, Lucas & Lightbody, 2006). The feedback component

involves providing specific information about a person’s

functional (real life) task performance in relation to

observed aspects of the performance such as number

and type of errors, the amount of time taken, and the

task outcomes (Fleming & Ownsworth, 2006). Effective

feedback interventions include specific, timely, consis-

tent and respectful feedback on task performance

(Barco, Crosson, Bolesta, Werts & Stout, 1991; Langer &

Padrone, 1992; Mateer, 1999).

There are various forms of feedback interventions

such as verbal feedback, videotaped feedback or task

performance without feedback. Providing verbal feed-

back is the most common form of a feedback interven-

tion, which involves a therapist verbally explaining to a

person with a TBI about his or her strengths and limita-

tions, about the results of formal or informal testing,

or about the observations of functional performance

(Mateer, 1999; Ownsworth, Fleming, Shum, Kuipers &

Strong, 2008). Verbal feedback also involves having a

therapist rate the performance of a functional task of a

person with a TBI, and then comparing this with the

person’s self-ratings of the same performance. The ther-

apist and the person would discuss discrepancies in rat-

ings of the performance, which would then enable the

person to make gains in their self-awareness (Owns-

worth et al., 2008).Videotaped feedback is another form of feedback

intervention and has been shown to be effective in sin-

gle case designs, particularly for improving awareness

of behavioural, sensation and communication problems

(Ownsworth et al., 2006; Tham & Tegner, 1997). Video

feedback interventions involve video taping a person

with impaired self-awareness completing a functional

task. The person later watches the video of the task,

while identifying the errors in the performance, observ-

ing compensatory strategies that were utilised, and

noticing areas of strength. It is suggested that this form

of feedback is most effectively completed with a thera-

pist present while the person with impaired self-aware-

ness is watching the video, as they can provide verbal

feedback and therapeutic support (Ownsworth, Quinn,

Fleming, Kendall & Shum, 2010).

Another form of feedback is experiential feedback,

when a person with impaired self-awareness after TBI

completes a functional task with some therapist input.

For example, the therapist may prompt the person to

improve their performance by utilising strategies or

may point out the persons’ errors in order to identify

solutions while the person during task performance, as

per routine occupational therapy practice. However, this

form of feedback does not necessarily involve any ver-

bal feedback or debriefing with the therapist after the

task is completed.

Despite the well-recognised clinical benefits and com-

mon use of feedback, there is currently limited high

level evidence on the type of feedback that is most

effective to improve impaired self-awareness in people

after a TBI (Fleming & Ownsworth, 2006). A recent sys-

tematic review of awareness interventions with a feed-

back component found only three randomised

controlled trials (RCTs) and reported a moderate effect

on increased self-awareness following feedback inter-

ventions when the data were pooled (Hedges adjusted

g = 0.64, 95% confidence interval 0.11–1.16) (Schmidt,

Lannin, Fleming & Ownsworth, 2011). Schmidt et al.concluded that further high quality research is required

to determine the effects of integrating feedback inter-

ventions into rehabilitation programmes and the impact

of this on functional outcome.

Anecdotally, videotaped feedback is thought to be a

powerful means of engaging people who lack self-

awareness in rehabilitation, but no systematic research

has examined the effectiveness of this approach, or

compared it to other forms of feedback such as verbal

feedback or to task performance without feedback. It is

therefore important for occupational therapy practice to

evaluate different feedback formats as part of a treat-

ment approach for impaired self-awareness in order to

promote better outcomes.

Methods

Purpose

Primary aim

To evaluate the effectiveness of feedback interventions

for improving self-awareness in persons with TBI by

comparing three types of feedback interventions. Specif-

ically, it aims to compare the effectiveness of: (i) video-

taped and verbal feedback, (ii) verbal feedback, and

(iii) experiential feedback for improving on-line aware-

ness in persons with TBI. It is hypothesised that video-

taped feedback in conjunction with verbal feedback

from the therapist will be the most effective type of

feedback provision.

Secondary aim

To compare the effects of the three types of feedback on

intellectual awareness, emotional status and adjustment

© 2012 The AuthorsAustralian Occupational Therapy Journal © 2012 Occupational Therapy Australia

FEEDBACK INTERVENTIONS FOR IMPROVING SELF-AWARENESS 139

to injury. Due to the exploratory nature of these compo-

nents, hypotheses have not been generated.

Study design and setting

An assessor and participant blinded pragmatic RCT will

be used to compare two intervention groups with a con-

trol condition (Altman & Bland, 1999). A pragmatic trial

design was selected, instead of an explanatory trial, to

evaluate the effectiveness of the intervention in routine

clinical practice, as opposed to under ideal conditions

such as in a research clinic (Roland & Torgerson, 1998).

The results of this research will therefore have greater

applicability to clinical settings as the intervention could

be easily replicated under the same conditions.

The study will occur at a single centre, the Royal

Rehabilitation Centre, a specialist rehabilitation centre

with inpatient and community services in Sydney, Aus-

tralia. The trial is registered with the Australian New

Zealand Clinical Trials Registry, identifier number ACT-

RN12609000378224. Ethical approval has been obtained

from the Royal Rehabilitation Centre H/09/RRCS2 (09/

02) and from the University of Queensland Behavioural

and Social Sciences Ethical Review Committee

(Approval Number 2009005991).

Participants

Fifty-four adult participants with TBI will be recruited

from both the inpatient and community services. Each

new admission to the inpatient unit as well as new

referral to the community services will be screened for

eligibility for the study using the following inclusion

criteria: (i) diagnosis of a TBI, (ii) over 16 years old, (iii)

emerged from post-traumatic amnesia to allow for

informed consent and to increase the likelihood of par-

ticipants remembering feedback, (iv) functional in the

English language to maximise comprehension of infor-

mation, (v) difference in scores between the therapist

and the participant on the Awareness Questionnaire

(described below) indicating impaired self-awareness,

and (vi) displays adequate cognitive capacity to provide

informed consent to participate in the study. People

who meet the eligibility criteria will be provided with

detailed written and verbal information about the study

using a consent form, approved by the ethical review

committee. People who choose to participate in the

study will sign the consent form.

The exclusion criteria for participation in the study

are: (i) severe communication disability or behavioural

changes that would compromise ability to participate in

an intervention, (ii) expected length of rehabilitation less

than intervention study period (i.e. two weeks), and (iii)

error-free meal preparation on baseline assessment.

Figure 1 displays the recruitment, randomisation and

intervention process, as well as the process for partici-

pants who are screened and not eligible or elect not to

participate.

A power calculation was conducted to determine

sample size. The number of errors at baseline is esti-

mated to be 17.92, based on a preliminary study by

Ownsworth et al. (2008). The intervention is expected to

reduce errors by 30%, which was considered a clinically

significant change by the authors. With power set at

0.80 and 5% level of significance with no period effect,

16 participants are required in each group (Tango,

2009). Sample size has been inflated to 18 participants

per group to allow for an estimated dropout rate of

10%. Therefore, it is calculated that a sample size of 54

participants is required for the study to be adequately

powered to detect significant differences between the

groups.

Measures

The primary outcome measure is performance on a task

assessing online awareness. It will be operationalised as

the number of errors made by the participant during a

standardised functional task (in this case meal prepara-

tion). Following the guidelines of Ownsworth et al.(2010), errors during task performance will be classified

as self-corrected errors or therapist-corrected errors. To

enable the opportunity for self-correction, the therapist

will delay responding to the error for up to approxi-

mately 5–10 seconds (note: the exception is where

safety is compromised). The number of participant

‘checks’ (i.e. asking the therapist for directions or

confirmation) will also be recorded. A number of

secondary outcome measures will be completed as

described in Table 1.

Additionally, demographic and injury-related data

will be recorded for each participant, including age,

gender, years of education, date of injury, initial Glas-

gow Coma Score (GCS), and MRI/CT results. Neuro-

psychological assessments of memory and executive

function will be used for descriptive purposes and to

indicate the extent of cognitive impairments relating

fronto-temporal brain damage. The measures are

described in Table 2 and are all standardised assess-

ment tools with established psychometric properties.

Intervention

Following pre-intervention assessments (described next),

the participant will select a functional activity from a

choice of three meal preparation tasks (pasta and sauce,

sausages and mashed potatoes, or omelette and toast).

Participant selection of the meal is considered essential

for motivation and engagement in the therapy task

(Ownsworth et al., 2006).These activities have been thoroughly analysed, using

an approach based on the Executive Functional Perfor-

mance Task (EFPT) (Baum et al., 2008). This approach

separates a task into components of initiation, execution

(including organisation, sequencing, judgment and

safety), and completion (Baum et al., 2008). The EFPT


140 J. SCHMIDT ET AL.

rates each component on a six-point scale according to

the amount of assistance required (i.e. 0 = independent

and 5 = do for participant). The EFPT has been adapted

for the functional tasks used in this study, and named

the Meal Independence Rating Scale (MIRS), which will

be used by the therapist and participant to rate the level

of independence in task performance. The MIRS will

not be used as an outcome measure per se, but instead

as part of the intervention to allow self-ratings of per-

formance by participants and stimulate discussion for

the verbal feedback component of the task.

The participant will perform the chosen task four

times during occupational therapy sessions, within a

two to four weeks period. All sessions will be video-

taped by an occupational therapist involved in the ses-

sion, using a hand held digital video recorder. The

therapist will hold the camera at shoulder level to

ensure limited intrusion to the participant performing

the task. Participants may request that their face is not

filmed, and this will be accommodated.

The feedback intervention will be provided immedi-

ately following each meal preparation session, according

to group allocation as per Figure 2. Upon completion of

the four intervention sessions, as part of the post-inter-

vention assessment the participant will complete the

self-report measures (Awareness Questionnaire (AQ),

Depression Anxiety and Stress Scale (DASS) and Self

Perception in Rehabilitation Questionnaire (SPIRQ)) and

the participant’s treating occupational therapist will

complete the significant others’ version of the AQ.

A follow-up assessment will take place 8–10 weeks

after the completion of fourth intervention session by

the therapist. This will assess maintenance of gains and,

if occurring in the home after a participant is dis-

charged from an inpatient unit, generalisation of task

performance. During the follow-up assessment, the par-

ticipant will prepare the chosen meal, the participant

and the therapist will each independently complete the

MIRS and the participant and therapist will complete

the post-intervention assessments.

Admissions screened for eligibility

Not eligible, reasons for exclusion documented;

demographic details collected

Eligible, informed consent provided

Eligible, but does not consent to study;

demographic details collected

Random allocation to groupsPre-intervention assessments completed (AQ, SPIRQ, DASS, neuropsychological tests)

Participant selects a meal to prepare

Intervention: meal preparation sessions completed and videotaped

Intervention: participant completes the SPIRQ and MIRS, therapist completes the MIRS

Group 1: video and verbal feedback provided

Group 2: verbal feedback provided

Group 3: control condition, no feedback provided

Intervention sessions repeated 3 times within 2-3 weeks

Post-intervention assessments completed by (AQ, DASS)

Follow-up (2 months): meal preparation completed and videotapedFollow-up assessments completed (AQ, SPIRQ, DASS, MIRS)

FIGURE 1: Process of participant involvement in the randomised controlled trials. AQ, Awareness Questionnaire; SPIRQ, Self Percep-

tion in Rehabilitation Questionnaire; DASS, Depression Anxiety and Stress Scale; MIRS, Meal Independence Rating Scale.



TABLE1:Secondary

assessments

Assessm

ent

What

itassessed

Details

ofassessmen

tHow

itis

usedin

thestudy

Assessm

entquality

Awaren

essQuestionnaire

(AQ)(Sherer,20

04)

Intellectual

Awaren

essacross

sensory,physical,cognitive

andbeh

aviouraldomains

Standardised

17-item

questionnaire

designed

for

use

afterbrain

injury

which

employsafive-pointscale

Participan

tself-ratingsare

compared

totherap

istratings

toderiveadiscrep

ancy

score

Soundreliab

ilityan

dvalidity

goodinternal

consisten

cy

(Cronbach’s

alphacoefficien

t

of0.88

)(Sherer

etal.,19

98a)

Dep

ressionAnxiety

and

StressScale-21(D

ASS-21)

(Antony,Bieling,Cox,

Enns&

Swinson,19

98)

Self-ratedquestionnaire

measu

ringsymptomsof

emotional

distressonthree

scales,includingdep

ression,

anxiety

andstress

21item

sscoredona

four-pointscalean

dis

a

shorten

edversionofthe

original

DASS-42(Lovibond

&Lovibond,19

95)

Measu

rescu

rren

tem

otional

statusan

dch

angein

emotional

statusover

time

(Hen

ry&

Crawford

,20

05)

Sim

ilar

factorstructure

with

similar

resu

ltsto

the

DASS-42;

highinternal

consisten

cy(A

ntonyet

al.,

1998

)

SelfPerceptionin

Reh

abilitation

Questionnaire

(SPIRQ)

Participan

t’sacceptance

of

theirinjury,theem

ergen

ceof

self-awaren

ess,

theperceived

needforan

dben

efits

of

rehab

ilitation,an

dem

otional

reactions

20-item

selfrated

questionnaire

Completedona

session-by-sessionbasis

to

den

ote

chan

ges

in

self-awaren

essan

d

adjustmen

tforeach

participan

tover

the

interven

tionperiod

Satisfactory

internal

consisten

cyan

dtest-retest

reliab

ility(r

>0.70

)

(Ownsw

orthet

al.,20

09)



TABLE 2: Neuropsychological assessments

Assessment What it assessed How it is used in the study Assessment quality

Behavioural

Assessment of the

Dysexecutive

Syndrome (BADS)

(Wilson, Alderman,

Burgess, Emslie &

Evans, 1996)

Six tests used to assess

difficulties with

high-level tasks such as

planning, organising,

initiating, monitoring,

and adapting

behaviour (Norris &

Tate, 2000)

Two of the six tests have been

selected due to the relevance

of skills: (i) Key search which

investigates participants’

ability to plan a strategy and

solve a problem using an

efficient, effective, and

systematic method, (ii) Zoo

map which investigates the

participants’ ability to plan

and execute the plan

effectively

Superior ecological validity

compared to standard

executive tests in terms of

predicting competency in role

functioning (Norris & Tate,

2000)

Wisconsin Card

Sorting Test (WCST)

(Lezak, Howieson &

Loring, 2004)

Assesses the

participants’ ability to

display flexible thought

process despite

changing schedules of

reinforcement

Assesses abstract reasoning,

ability to change problem

solving strategies, and

attention

Good internal consistency and

reliability (Greve, Stickle,

Love, Bianchini & Stanford,

2005)

Wechsler Memory

Scale, 3rd Edition

(WMS�III)

(Wechsler, 2002)

Assess the participants’

immediate and delayed

verbal memory skills

One of the primary subtest of

the WMS�III, the Logical

Memory Test have been

selected for the study

Internal consistency score of

0.87, a test–retest value of

0.82, and internal consistency

of 0.81, which indicates good

reliable and validity

(Wechsler, 2002)

Treatment group one:videotape plus verbal feedback

The participant and the therapist will watch the video of the participant's

functional task performance

The participant and the therapist will each complete the MIRS

The therapist and the participant will discuss each item on the MIRS and the differences between the therapist's and

the participant's ratings

The discussion will follow the feedback guidelines developed for the study

The participant will then complete the SPIRQ

Treatment group two:verbal feedback

The participant and the therapist will each complete the MIRS after the

participant completes the functional task

The therapist and the participant will discuss each item on the MIRS and the differences between the therapist's and

the participant's ratings

The discussion will follow the feedback guidelines developed for the study

The participant will then complete the SPIRQ

Treatment group three: experiential feedback

The therapist will complete the MIRS after the participant completes the

functional task

The participant will then complete the MIRS and the SPIRQ

FIGURE 2: Treatment groups. SPIRQ, Self Perception in Rehabilitation Questionnaire; DASS, Depression Anxiety and Stress Scale.



Randomisation and techniques to minimisebias

The study incorporates several aspects designed to

minimise bias or confounds in the results. These

include:

1. Use of a random allocation list, prepared according

to a 1:1:1 randomisation sequence from a computer

generated random number table; this method of ran-

domisation is designed to minimise bias as all partic-

ipants have an equal chance to be allocated into any

one of the groups (Altman & Bland, 1999).

2. Concealed allocation, which means the allocation to

group is not known by anyone involved in the

research and therefore there is no bias in the tim-

ing of involving participants in the study (Altman

& Schulz, 2001). To obtain group allocation, the

therapist will contact an independent and off-site

investigator who is not involved in assessment or

treatment of participants. This investigator will

open sealed opaque envelopes that are numbered

consecutively and contain group allocation details.

3. Blinding of participants, which will be employed to

minimise bias and improve the quality of the study’s

results (Day & Altman, 2000). In the consent infor-

mation, participants will be told that an intervention

will occur after the meal preparation. However, they

will not be told the type of intervention that could

occur. As all participants will be video-taped during

the meal preparation task, it is unlikely that they will

perceive differences between their intervention and

that of other participants in the study. To assess if

the participants have remained blind to group alloca-

tion, a follow up questionnaire will be given which

asks the participant about their perception of the

intervention and if they have spoken about the inter-

vention to any other participant that is involved in

the study.

4. Blinding of assessors, in which the assessor who

views the videotape of the participant’s meal prepa-

ration and rates the number of errors is blind to the

participant’s random allocation (see Data Collection

next (Day & Altman, 2000)).

Data collection

Following the consent and random allocation processes,

the pre-intervention assessment will be completed, con-

sisting of the secondary outcome measures as described

in Table 1.

A blinded assessor will watch all five videotapes of

the participant’s functional task performance (four

intervention sessions and one follow-up session) and

complete an Error Rating Form, which has been devel-

oped to collate the types of errors and record number

of checks. A portion of the videos will be rated by a

second researcher to examine inter-rater reliability of

ratings.

Data analysis

All measures will be scored according to established pro-

tocols. An intention-to-treat approach will be used in data

analyses, which means that data from participants will be

analysed according to group allocation, regardless of

whether or not they completed the intervention (Hollis &

Campbell, 1999). Data will be screened for missing values

and distribution. Multiple imputation approach will be

used to deal with missing data (Sterne et al., 2009).

Appropriate transformations will be applied to primary

and secondary outcome variables that do not meet

assumptions of normality. In preliminary analyses,

descriptive statistics will be used to summarise the demo-

graphic and diagnostic data of participants, including

neuropsychological test scores for the three groups.

Groups will be compared on demographic and clinical

variables. If groups differ significantly on any of the

demographic, diagnostic or neuropsychological vari-

ables, these will be accounted for as covariates in subse-

quent analyses examining intervention effects. Between-

and within-group comparisons will be made for the pri-

mary outcome measures using linear mixed effects mod-

els (Brown & Prescott, 2006) with time as the within-

subjects variable (pre- and post-intervention) and group

as the between-subjects variable (video and verbal, ver-

bal, control), controlling for any potential confound. Sim-

ple effects analyses will be conducted to determine where

significant differences exist.

Discussion

This article describes the protocol of a RCT designed to

evaluate the effectiveness of different types of feedback

interventions for improving on-line awareness during

occupational performance. The RCT will also explore

the impact of feedback intervention on emotional status

and adjustment to determine whether this type of inter-

vention has any negative impact on people with TBI.

Incorporating a measure of emotional status (the DASS-

21) will assess for any negative effects of feedback on

emotional status. Case by case examination of the par-

ticipant responses to videotaped and verbal feedback

might provide valuable information as to the types of

people who may not be suited to receiving direct feed-

back in these formats because of heightened emotional

reactions.

Although on-line self-awareness (i.e. error correction)

is the primary outcome measures; the study will also

measure intellectual self-awareness to determine if there

is any generalisation of treatment effects to broader self-

knowledge (Ownsworth, Fleming, Stewart & Griffin,

2009; Sherer, Bergloff, Boake, High & Levin, 1998a).

Toglia and Kirk (2000) proposed that the experience of

task performance interacts with the self-knowledge of a

person with a TBI, so that on-line monitoring shapes

the development of intellectual awareness. This study



will specifically test the extent to which this interaction

occurs under different feedback conditions.

There is limited high quality evidence supporting the

use of feedback as an awareness intervention, despite

the widespread use of various forms of feedback in

neuro-rehabilitation (Fleming & Ownsworth, 2006).

However, previous single-case studies and clinical

observations suggest that video feedback is a powerful

self-awareness intervention (Ownsworth et al., 2010).

The methodological design of a RCT is selected to pro-

vide a scientifically rigorous evaluation for this area. It

is anticipated that the results will be of considerable use

clinically for guiding therapists’ choice of methods of

delivering feedback on functional task performance in

brain injury rehabilitation. The results of the study have

the potential to contribute to improved rehabilitation

outcomes for this significant group.

There are a number of potential limitations to this

study. The length of participant recruitment (i.e. over

two years of recruitment) will be a hindrance to the time

that the results are available to clinicians. The length of

time post-injury may be quite different between the

community participants and those from the inpatient

setting, which could be a potential limitation as the par-

ticipants may not be similar in terms of level of recov-

ery. However, we anticipate the level of impaired self-

awareness to be similar at baseline, as this is screened as

part of the eligibility criteria for participation in the

study. Another limitation is that we are comparing three

different types of feedback interventions. Therefore, we

require more participants in the study to achieve a suffi-

cient sample size to adequately compare the effective-

ness. However, comparing three types of feedback

interventions is a benefit of the design, as it will deter-

mine not only if feedback is effective, but also what type

is most effective.

It is important to note that this study specifically

focuses on a meal preparation task and does not assess

generalisation to other skills. Future research will be

needed to assess translation of gains in on-line aware-

ness to other functional tasks. It is estimated that the

trial will be completed in 2012. The results will be com-

piled and published in a peer-reviewed journal.

References

Altman, D. & Bland, M. (1999). Treatment allocation in con-

trolled trials: Why randomise? British Medical Journal,

318, 1209.

Altman, D. & Schulz, K. (2001). Statistics notes: Concealing

treatment allocation in randomised trials. British Medical

Journal, 323, 446–447.Antony, M. M., Bieling, P. J., Cox, B. J., Enns, M. W. &

Swinson, R. P. (1998). Psychometric properties of the 42-

item and 21-item versions of the Depression Anxiety

Stress Scales (DASS) in clinical groups and a community

sample. Psychological Assessment, 10, 176–181.

Barco, P. P., Crosson, B., Bolesta, M. M., Werts, D. & Stout,

R. (1991). Training awareness and compensation in post-

acute head injury rehabilitation. In: J. S. Kreutzer & P. H.

Wehman (Eds.), Cognitive rehabilitation for persons with

traumatic brain injury: A functional approach (pp. 129–146).Baltimore, MD: Paul H.

Baum, C. M., Connor, L. T., Morrison, T., Hahn, M., Drom-

erick, A. W. & Edwards, D. F. (2008). Reliability, validity,

and clinical utility of the executive function performance

test: A measure of executive function in a sample of peo-

ple with stroke. American Journal of Occupational Therapy,

62, 446–455.Brown, H. & Prescott, R. (2006). Applied mixed models in

medicine. New York: Wiley.

Day, S. J. & Altman, D. G. (2000). Statistics notes: Blinding

in clinical trials and other studies. British Journal of Clini-

cal Psychology, 321, 504.

Fleming, J. (2009). Metacognitive occupation-based training

in traumatic brain injury. In: I. Soderback (Ed.), Interna-

tional handbook of occupational therapy interventions (pp.

225–231). New York, NY: Springer.

Fleming, J. & Ownsworth, T. (2006). A review of awareness

interventions in brain injury rehabilitation. Neuropsycho-

logical Rehabilitation, 16 (4), 474–500.Fleming, J., Strong, J. & Ashton, R. (1998). Cluster analysis

of self-awareness levels in adults with traumatic brain

injury and relationship to outcome. The Journal of Head

Trauma Rehabilitation, 13 (5), 39–51.Fleming, J., Lucas, S. & Lightbody, S. (2006). Using occupation

to facilitate self-awareness in people who have acquired

brain injury: a pilot study. Canadian Journal of Occupational

Therapy - Revue Canadienne d Ergotherapie, 73 (1), 44–55.Greve, K. W., Stickle, T. R., Love, J., Bianchini, K. J. & Stan-

ford, M. S. (2005). Latent structure of the Wisconsin Card

Sorting Test: A confirmatory factor analytic study.

Archives of Clinical Neuropsychology, 20, 355–364.Henry, J. D. & Crawford, J. R. (2005). The 21-item version of

the Depression Anxiety Stress Scales (DASS–21): Norma-

tive data and psychometric evaluation in a large non-clini-

cal sample. British Journal of Clinical Psychology, 44, 227–239.Hollis, S. & Campbell, F. (1999). What is meant by intention

to treat analysis? Survey of published randomized con-

trolled trials. British Medical Journal, 319, 670–674.Langer, K. G. & Padrone, F. J. (1992). Psychotherapeutic

treatment of awareness in acute rehabilitation of traumatic

brain injury. Neuropsychological Rehabilitation, 2, 59–70.Lezak, M., Howieson, D. B. & Loring, D. W. (2004). Neuro-

psychological assessment (4th ed.). New York: Oxford Uni-

versity Press.

Lovibond, S. H. & Lovibond, P. F. (1995). Manual for the

Depression Anxiety Stress Scales (2nd ed.). Sydney: Psy-

chology Foundation.

Mateer, C. A. (1999). The rehabilitation of executive disor-

ders. In: D. T. Stuss, G. Winocur & I. H. Robertson

(Eds.), Cognitive neurorehabilitation (pp. 314–332). New

York, NY: Cambridge University Press.

Norris, G. & Tate, R. L. (2000). The behavioural assessment

of the dysexecutive syndrome (BADS): Ecological, con-



current and construct validity. Neuropsychological Rehabili-

tation, 10 (1), 33–45.Ownsworth, T., Fleming, J., Desbois, J., Strong, J. & Kui-

pers, P. (2006). A metacognitive contextual intervention

to enhance error awareness and functional outcome

following traumatic brain injury: a single-case experi-

mental design. Journal of the International Neuropsychologi-

cal Society, 12 (1), 54–63.Ownsworth, T., Fleming, J., Shum, D., Kuipers, P. & Strong,

J. (2008). Comparison of individual, group and combined

intervention formats in a randomized controlled trial for

facilitating goal attainment and improving psychosocial

function following acquired brain injury. Journal of Reha-

bilitation Medicine, 40, 81–88.Ownsworth, T., Fleming, J., Stewart, E. & Griffin, J. (2009).

The Self-Perceptions in Rehabilitation Questionnaire: A new

measure of therapy progress in brain injury rehabilitation.

Paper presented at the Sixth Symposium on Neuropsy-

chological Rehabilitation; Tallinn, Estonia. 3 August.

Ownsworth, T., Quinn, H., Fleming, J., Kendall, M. &

Shum, D. (2010). Error self-regulation following traumatic

brain injury: A single case study evaluation of metacog-

nitive skills training and behavioural practice interven-

tions. Neuropsychological Rehabilitation, 20 (1), 59–80.Roland, M. & Torgerson, D. J. (1998). Understanding con-

trolled trials: What are pragmatic trials? British Medical

Journal, 316 (7127), 285.

Schlund, M. W. (1999). Self awareness: Effects of feedback

and review on verbal self reports and remembering fol-

lowing brain injury. Brain Injury, 13 (5), 375–380.Schmidt, J., Lannin, T., Fleming, J. & Ownsworth, T. (2011).

A systematic review of feedback interventions for

impaired self-awareness following brain injury. Journal of

Rehabilitation Medicine, 43, 673–680.Sherer, M. (2004). The Awareness Questionnaire. The Center

for Outcome Measurement in Brain Injury. Retrieved 6

June, 2008, from http://www.tbims.org/combi/aq

Sherer, M., Bergloff, P., Boake, C., High, W. & Levin, E.

(1998a). The Awareness Questionnaire: Factor structure

and internal consistency. Brain Injury, 12 (1), 63–68.Sherer, M., Bergloff, P., Levin, E., High, W., Oden, K. E. &

Nick, T. G. (1998b). Impaired awareness and employment

outcome after traumatic brain injury. Journal of Head

Trauma Rehabilitation, 13 (5), 52–61.Sterne, J. A. C., White, I. R., Carlin, J. B., Spratt, M., Roy-

ston, P., Kenward, M. G. et al. (2009). Multiple imputa-

tion for missing data in epidemiological and clinical

research: Potential and pitfalls. British Medical Journal,

339, 157–160.Tango, T. (2009). Sample size formula for randomized con-

trolled trials with counts of recurrent events. Statistics

and Probability Letters, 79, 466–472.Tham, K. & Tegner, R. (1997). Video feedback in the reha-

bilitation of patients with unilateral neglect. Archives of

Physical Medicine & Rehabilitation, 78 (4), 410–413.Toglia, J. & Kirk, U. (2000). Understanding awareness

deficits following brain injury. NeuroRehabilitation, 15 (1),

57–70.Wechsler, D. (2002). WAIS-III/WMS-III technical manual:

Updated. San Antonio, TX: Psychological Corporation.

Wilson, B. A., Alderman, N., Burgess, P. W., Emslie, H. &

Evans, J. (1996). Behavioural assessment of the dysexecutive

syndrome. Edmunds, UK: Thames Valley Test.



Documents

Feedback interventions for improving self-awareness after brain injury: A protocol for a pragmatic randomised controlled trial