Content uploaded by Fiona Barnett
Author content
All content in this area was uploaded by Fiona Barnett on Mar 21, 2014
Content may be subject to copyright.
48 British Journal of Occupational Therapy February 2012 75(2)
Critical review
Introduction
Exercise has long been proven to benefit the general population in terms of
mental health and wellbeing (Stathopoulou et al 2006). Many studies have
examined its effect on anxiety disorders, minor depression and substance
abuse (Stathopoulou et al 2006). In comparison, very little research has
investigated the use of exercise in people who experience more severe and
enduring mental illnesses. This systematic review adopts this focus and
examines the impact that exercise can have on mental health and quality
of life (QOL) for people with a severe mental illness. For the purposes of
this review, the term ‘severe mental illness’ (SMI) encompasses severe and
enduring illnesses, such as severe depression, bipolar disorder, schizophrenia,
schizoaffective disorder and psychosis.
Observational studies of people with SMI have highlighted a link between
reported regular levels of exercise and physical activity and improved mood
and QOL (McCormick et al 2008). Researchers, however, have emphasised
the limited number of intervention studies regarding the effect of exercise
on mental health and QOL in this population group (Fogarty et al 2004,
Richardson et al 2005).
Two main factors support the need for more research on the use of exer-
cise intervention in this population. First, people with SMI are typically pre-
scribed with antipsychotic medications. The risks and side effects of these
medications and polypharmacy warrant a need to explore safer treatment
approaches, such as exercise (Trivedi et al 2006). Secondly, people with SMI
experience greater rates of obesity and risky health behaviours, and poorer
The impact of exercise on the mental health
and quality of life of people with severe
mental illness: a critical review
Kristy Alexandratos,1Fiona Barnett 2and Yvonne Thomas3
Key words:
Occupational therapy,
exercise therapy.
Introduction:
Physical exercise has been proven to benefit the general population
in terms of mental health and wellbeing. However, there is little research
investigating the impact of exercise on mental health and quality of life for
people who experience a severe and enduring mental illness.
Method:
This review aims to describe the effect of physical exercise intervention
on the mental health and quality of life of people with severe mental illness.
Quantitative and qualitative articles published between 1998-2009 were sourced
using electronic databases. Articles were included if the study intervention
involved exercise and the outcome measure included mental health or quality of life.
Sixteen articles were analysed for common themes and appraised critically.
Findings:
The findings show that exercise can contribute to improvements in
symptoms, including mood, alertness, concentration, sleep patterns and psychotic
symptoms. Exercise can also contribute to improved quality of life through social
interaction, meaningful use of time, purposeful activity and empowerment.
Implications:
Future research is warranted to describe the way exercise can
meet the unique needs of this population. Studies with a focus on psychological
outcome measures would provide greater evidence for its use in therapy.
© The College of Occupational Therapists Ltd.
Submitted: 30 January 2011.
Accepted: 2 November 2011.
1Research Worker, Discipline of Occupational
Therapy, School of Public Health, Tropical
Medicine and Rehabilitation Sciences, James
Cook University, Townsville, Queensland,
Australia.
2Senior Lecturer, Discipline of Occupational
Therapy, School of Public Health, Tropical
Medicine and Rehabilitation Sciences, James
Cook University, Townsville, Queensland,
Australia.
3Senior Lecturer, Discipline of Occupational
Therapy, School of Public Health, Tropical
Medicine and Rehabilitation Sciences, James
Cook University, Townsville, Queensland,
Australia.
Corresponding author: Kristy Alexandratos,
Research Worker, Discipline of Occupational
Therapy, School of Public Health, Tropical
Medicine and Rehabilitation Sciences,
James Cook University, Townsville,
Queensland, Australia 4811.
Email: kristy.alexandratos@my.jcu.edu.au
Reference: Alexandratos K, Barnett F,
Thomas Y (2012) The impact of exercise on
the mental health and quality of life of
people with severe mental illness: a critical
review.
British Journal of Occupational
Therapy, 75(2),
48-60.
DOI: 10.4276/030802212X13286281650956
49
British Journal of Occupational Therapy February 2012 75(2)
Kristy Alexandratos, Fiona Barnett and Yvonne Thomas
levels of fitness, compared with the general population (Smith
et al 2007a). There is already an extensive body of research
on the need for exercise interventions to address the physical
health issues in people with SMI (Faulkner et al 2007). Confir-
mation of the benefits of exercise interventions for improv-
ing QOL could potentially add secondary benefits beyond the
recognised improvements in physical health parameters.
The World Health Organization describes quality of life
as a ‘broad ranging concept affected in a complex way by
the person’s physical health, psychological state, level of
independence, social relationships, personal beliefs and
their relationship to salient features of the environment’
(World Health Organization 1997, p1). Occupational therapy
researchers theorise that QOL can be obtained through
engagement and participation in personally meaningful
occupations (Hammell 2004). Occupational therapists use
meaningful activity as a medium through which people can
engage in doing, being, becoming and belonging; these con-
cepts have been described as core aspects in the meaning of
occupation (Wilcock 1998, Hammell 2004). Exercise has been
defined by the American College of Sports Medicine (2006) as
‘… planned, structured, and repetitive bodily movement done
to improve or maintain one or more components of physical
fitness’ (p3). Exercise as a purposeful activity therefore may
be able to provide people with the opportunity to ‘do’ exercise,
to ‘be’ and ‘become’ exercisers, and to ‘belong’ to an exercise
group or culture. The related term ‘physical activity’ extends
not only to exercise, but also to ‘… activities which involve
bodily movement and are done as part of playing, working,
active transportation, house chores and recreational activities’
(World Health Organization 2011).
Method
Aim of review
The aim of this review was to describe the effect of exercise
intervention on the mental health and QOL of people with SMI.
To gain a broad understanding of these issues, both quanti-
tative and qualitative studies were included in the review.
Search strategy
A search was conducted using CINAHL, PsycInfo, the
Cochrane Library, Scopus and Medline. The key words were
‘physical exercise’, ‘exercise’, ‘exercise therapy’, ‘physical
activity’ and ‘exercise movement therapies’. In the mental
health category, literature searches were prefaced with ‘severe’,
‘serious’, ‘enduring’ and ‘persistent’ in relation to the degree
of mental illness. In addition, ‘schizophrenia’ and ‘psychosis’
(and appropriate truncations) were used as key search words.
The word ‘depression’ alone presented articles with mild
and moderate depression. In order to keep with the focus
on SMI, only ‘major’ or ‘severe’ or ‘bipolar’ were included
in the literature search. Alzheimer’s disease and dementia
were listed under SMI in some databases; consequently,
these articles were excluded because they did not meet
the definition of SMI for the purposes of this review. Two
systematic reviews from the Cochrane Library (Campbell
and Foxcroft 2003, Faulkner et al 2007) were examined to
provide an understanding of current knowledge and research.
Reference lists were searched for any additional articles.
In total, 68 articles relevant for this study were sourced,
37 of which were quantitative and 16 qualitative in design.
The remaining articles included six systematic reviews,
five literature reviews, two of mixed-method design and
two opinion pieces.
Inclusion criteria
The following inclusion criteria were placed on all studies:
■Published in a peer-reviewed journal
■Published in English between 1998 and 2009
■Relevant background literature reviewed
■Intervention involved exercise or physical activity (either
independently or as part of a programme involving multiple
intervention approaches).
Quantitative studies were also required to meet the
following criteria: (1) at least one outcome measure was
used to assess the effect of exercise or physical activity on
mental health, wellbeing or QOL; (2) study design was a
randomised controlled trial (RCT), clinical trial or pre-post
design; and (3) results were reported in terms of statistical
significance. Qualitative studies were required to meet the
following criteria: (1) outcome under study was either the
perceived effect of exercise or physical activity on mental
health, wellbeing or QOL, and (2) research question and
methodology were clearly defined.
Articles dated prior to 1998 were excluded to restrict the
review to literature published over the previous decade. A
total of 16 studies met the inclusion criteria, including nine
quantitative designs, six qualitative designs and one mixed-
method design, although only the quantitative analysis in
this study met the inclusion criteria. The PRISMA flowchart
(Moher et al 2009) (Fig. 1) summarises the assessment and
exclusion of articles.
Methodological quality assessment
Studies were analysed for methodological quality according
to the guidelines developed by the McMaster University
Occupational Therapy Evidence-Based Practice Research
Group (Law et al 1998, Letts et al 2007). In order to illus-
trate the methodological quality of each study, the tool was
modified by awarding a numerical score for criteria met
under this quality assessment tool. The modified version
of the tool was modelled on those used in systematic
reviews by Barras (2005) and Deenadayalan et al (2010).
Appendix 1 details the scoring method and the questions
used to appraise the studies.
Analysis method
The heterogeneity of both interventions and outcome
measures, as well as the consideration of qualitative
studies in this review, rendered a meta-analysis unfeas-
ible. Therefore, a narrative analysis approach was adopted
for this systematic review.
50 British Journal of Occupational Therapy February 2012 75(2)
The impact of exercise on the mental health and quality of life of people with severe mental illness: a critical review
Findings
All 16 articles reviewed are summarised under the headings
of design, intervention, outcome measures, results and
quality score in Table 1.
Methodological design
Quantitative studies
Of the quantitative studies, six were RCTs (Babyak et al 2000,
Beebe et al 2005, Skrinar et al 2005, Brown and Chan 2006,
Duraiswamy et al 2007, Melamed et al 2008) and three were
a pre-post design (McDevitt et al 2005, Trivedi et al 2006,
Smith et al 2007b). The mixed quantitative and qualitative
study also adopted this latter approach (Pelletier et al
2005). Sample size ranged between 10 (Beebe et al 2005)
and 966 participants (Smith et al 2007b). Four articles
were pilot studies (Beebe et al 2005, McDevitt et al 2005,
Pelletier et al 2005, Trivedi et al 2006). Mean age ranges
were between 32 (Duraiswamy et al 2007) and 46 years
(Pelletier et al 2005). Gender distribution in proportion
of female participants ranged from 85% (Brown and Chan
2006) to 20% (Beebe et al 2005).
In terms of validity of the RCTs, one control group had
significantly different baseline characteristics for gender
and subjective scores of health and fitness to the interven-
tion group (Brown and Chan 2006), and two control groups
had issues with contamination (Babyak et al 2000, Melamed
et al 2008). Four trials used single-blind designs whereby
the assessors were blinded (Babyak et al 2000, Beebe et al
2005, Brown and Chan 2006, Duraiswamy et al 2007), while
two trials did not specify if blinding procedures were in place
(Skrinar et al 2005, Melamed et al 2008). Dropout rates
ranged between 13% (McDevitt et al 2005) and 53% (Trivedi
et al 2006). Intention-to-treat analysis was performed on four
studies (Beebe et al 2005, McDevitt et al 2005, Brown and
Chan 2006, Melamed et al 2008). However, two of these
trials also used analysis by compliers (‘per protocol’) only
at the 12-month follow-up period (Melamed et al 2008) and
to calculate measures of change (McDevitt et al 2005).
Qualitative studies
All of the qualitative studies were conducted in the United
Kingdom, with the exception of one Australian study (Fogarty
and Happell 2005). Two studies adopted a case study design
(Carless and Douglas 2008, Carless and Sparkes 2008), while
the remainder consisted of one grounded theory (Crone and
Guy 2008), one ethnography (Faulkner and Sparkes 1999),
one not stated (Fogarty and Happell 2005) and one study that
described itself as descriptive qualitative (Crone 2007).
Sample sizes ranged between 2 (Carless and Douglas
2008) and 12 participants (Fogarty and Happell 2005). One
study had an equal gender ratio (Crone 2007) and another
had one female to two male participants (Faulkner and Sparkes
1999). The remaining qualitative studies had a majority of
male participants. Only two studies reported briefly on par-
ticipant age, with one implying that the participants were in
their forties (Carless and Douglas 2008) and the other stat-
ing an age range of 20-42 years (Fogarty and Happell 2005).
The remaining studies did not report on participant age.
Characteristics of interventions and
associated outcomes
Quantitative studies
Interventions were delivered to participants in a range of
settings. Five interventions were delivered in community or
outpatient facilities (Beebe et al 2005, McDevitt et al 2005,
Brown and Chan 2006, Trivedi et al 2006, Smith et al 2007b),
one was in a supported community living centre (Pelletier
et al 2005), one was in an inpatient setting (Melamed et al
2008), two were in combination inpatient and outpatient
settings (Skrinar et al 2005, Duraiswamy et al 2007) and one
was not stated (Babyak et al 2000).
Interventions varied in the type of exercise offered. Three
studies used solely walking groups as the intervention
(Beebe et al 2005, McDevitt et al 2005, Melamed et al 2008),
two used walking in combination with stationary cycles
and jogging (Babyak et al 2000, Trivedi et al 2006), and a
further two used a mixture of walking, other cardiovascular
exercise forms and weights training (Pelletier et al 2005,
Skrinar et al 2005). One intervention used yoga therapy
(Duraiswamy et al 2007); one used a range of group-based
activities, for example, swimming, aqua aerobics, cycling
and physical activity, such as sightseeing and cinema trips
(Smith et al 2007b); and one used a pre-developed health
promotion programme with an integrated exercise compo-
nent (Brown and Chan 2006). The majority of studies used
Fig. 1. PRISMA (Moher et al 2009) flowchart.
51
British Journal of Occupational Therapy February 2012 75(2)
Kristy Alexandratos, Fiona Barnett and Yvonne Thomas
Table 1. Study design, characteristics and findings
Study and design Intervention Outcome measures* Results Quality score
Quantitative
*Where study focus also included physical health outcome measures, only mental health outcome measures are listed here.
Setting:
Not stated.
Exercise group (study):
Three
supervised sessions per week for
16 weeks. Ten minutes warm-up,
30 minutes walking/cycle
ergometry/jogging, 5 minutes cool
down. Training range 70-85%
maximum heart rate reserve.
Medication group (control):
‘Zoloft’ Sertraline (Selective
serotonin-reuptake inhibitor).
Combined exercise and medication
group:
As above.
Setting:
Outpatients.
Walking group (study):
Supervised
treadmill programme. Three times
per week.Ten minutes warm-up,
30 minutes walking (graded from
5 minutes initially over first
3 weeks of programme),
10 minutes cool down. Walking
at target heart rate.
Control group:
Nil exercise.
Setting:
Community.
Health promotion package (study):
Follows pre-developed programme:
Lilly Meaningful Day Manual.
Programme includes health
education; activity diaries; access
to community facilities. Six by
50-minute one-on-one health
promotion sessions. Subjects
covered include weight control,
healthy eating, exercise, structured
daily activity and substance misuse.
Standard control group.
Setting:
Inpatient and outpatient.
Yoga therapy (study):
Yoga
positions, breathing practice,
relaxation techniques.
Physical exercise therapy (control):
Brisk walking, jogging, exercises in
standing/sitting postures,
relaxation. Both interventions
involved 3-week training period
followed by 3 months continued
participation (reviewed by
therapist monthly). One hour per
day, 5 days per week.
Diagnostic Interview
Schedule; Hamilton Rating
Scale for Depression
(HRSD); Becks Depression
Inventory (BDI); and
Interview: self-reported
measures of number of
sessions per week, duration
of sessions and type of
physical activity.
Positive and Negative
Syndrome Scale (PANSS).
Initial basic health
questionnaire; Hospital
Anxiety and Depression
Scale (HAD); and Likert
scale for self-reported
measure of current
physical health, physical
fitness and mental health.
Positive and Negative
Syndrome Scale for
Schizophrenia (PANSS);
Social and Occupational
Functioning Scale
(SOFS); and WHO Quality
of Life BREF Version
(WHOQOL-BREF).
Non-significant difference in self-reported
symptoms (BDI scores) between groups.
HRSD and Diagnostic Interview Schedule
scores showed exercise group had
significantly lower rates of depression
than other groups (30% study compared
with 52% control and 55% combined
group). Participants in exercise group
more likely to be partially or fully recovered
at 6-month follow-up when compared
with control or combination group.
Participants who self-reported that they
continued to engage in physical activity
after the study period were less likely to
be classified as ‘depressed’ at a follow-up
6 months later (statistically significant).
Non-significant reduction in PANSS score
for study group. Non-significant increase
in mean PANSS score differences between
groups.
Study group had non-significant
reductions in HAD scores. Study group
had non-significant improvements in
subjective views of mental health.
Both study and control group had
statistically significant drop in total
PANSS score. Study group scored
significantly lower in total PANSS score
than control. Study group scored
significantly lower in PANSS sub-scores
for Negative, Depression Score, Anergia
Score than control (no significant
difference for Positive sub-scores).
Study group scored significantly better
in QOL scores. Study group scored
significantly better in SOFS scores.
10/15
14/15
11/15
10/15
Babyak et al
(2000)
Randomised
controlled trial.
Study duration
10 months
(4 months plus
6 months
post-treatment
follow-up),
n = 156.
Beebe et al
(2005)
Randomised
controlled trial.
Study duration
16 weeks,
n = 10.
Brown and
Chan (2006)
Randomised
controlled trial.
Study duration
6 weeks,
n = 28.
Duraiswamy
et al (2007)
Randomised
controlled trial.
Study duration
4 months,
n = 61.
52 British Journal of Occupational Therapy February 2012 75(2)
The impact of exercise on the mental health and quality of life of people with severe mental illness: a critical review
Table 1 (continued)
Study and design Intervention Outcome measures* Results Quality score
Setting:
Community (outpatient).
Study:
Supervised group walking
programme. Individualised
prescription at orientation.
Four times a week.Warm-up and
cool-down periods. Between 10
and 30 minutes walking. Intensity
between 60 and 79% of predicted
maximal heart rate. Health
workshops on problem solving,
goal setting, overcoming barriers
and planning. Ongoing problem
solving and support.
Setting:
Inpatient (hospital).
Study:
Supervised 30-minute walks
five times per week. Warm-up
period, quick pace, cool-down
period. Aerobic exercise via video
when rain prevented walking
outdoors. Weekly structured
nutrition counselling in small
groups. Group-based behaviour
therapy.
Setting:
Supported community
setting (ICCD Clubhouse).
Study:
Three by 90 minutes
sessions per week. Thirty-minute
warm up (aerobics and flexibility).
Individually customised weight
machines programme. Bottled
water provided. Shirts provided
with team appointed logo printed
on it for group cohesiveness.
Setting:
Inpatient, outpatient and
community.
Health education and exercise
programme (study):
Four sessions
per week.Warm up, cardiovascular
training, cool down and strength
training (two out of every four
sessions only). Graded duration –
starting at 15 minutes and
progressing up to 45 minutes.
Exercise intensity at 70 to 85%
of predicted maximal heart rate.
Weekly health seminar session
on healthy eating, weight
management, exercise levels,
stress relief, spirituality and
wellness, individual planning to
incorporate wellness activities into
one’s life.
Short-Form Health Survey
(SF-12); Profile of Mood
States (POMS); Multnomah
Community Ability Scale;
Outcomes Expectancies
for Exercise Scale; and
Decisional Balance Scale.
Positive and Negative
Syndrome Scale (PANSS);
and abbreviated version of
Quality of Life Enjoyment
and Satisfaction
Questionnaire (Q-LES-Q).
Medical Outcomes Study
(MOS) 36-Item Short Form
(SF-36) Version 2.
Symptom checklist 90R
(SCL-90); Lehman Quality
of Life Questionnaire;
Boston University Making
Decisions Questionnaire;
and Medical Outcomes
Study 36-Item Short-Form
Health Survey (SF-36).
No change in SF-12 scores. Significant
improvement in mood score (POMS) and
psychosocial functioning (Multnomah
Community Ability Scale).
Significantly better quality of life
scores for study group than control
post-intervention (3 months).
Non-significant relationship between
PANSS scores and involvement in the
intervention (3 months). (Twelve-month
review only reported outcome measures
in physical health.)
Significant improvement in mean SF-36
mental health subscale score (from 52.14
to 64.51). Non-significant improvements
in all other aspects of SF-36 scores.
SF-36 revealed significant improvement
in subjective scores of general health
sub-score for study group compared with
control. Significant improvement in
subjective ratings of empowerment for
study group compared with control.
Non-significant improvements for the
study group in most aspects of SCL-90,
SF-36 and quality of life scale.
11/15
9/15
11/15
10/15
McDevitt et al
(2005)
Pre-post design.
Study duration
12 weeks,
n = 15.
Melamed et al
(2008)
Randomised
controlled trial.
Study duration
3 months plus
review 12 months
follow-up post
completion of study,
n = 59.
Pelletier et al
(2005)
Mixed method:
only quantitative
pre-post design
component met
inclusion criteria.
Study duration
16 weeks,
n = 25.
Skrinar et al
(2005)
Randomised
controlled trial.
Study duration
12 weeks,
n = 30.
*Where study focus also included physical health outcome measures, only mental health outcome measures are listed here.
53
British Journal of Occupational Therapy February 2012 75(2)
Kristy Alexandratos, Fiona Barnett and Yvonne Thomas
Table 1 (continued)
Study and design Intervention Outcome measures* Results Quality score
Setting:
Community.
Intervention (study):
Physical
health screening and referral to
either weight management and
physical activity groups and/or
healthy living group. Group structure
varied; physical activity group
included swimming, aqua aerobics,
walking groups, cycle riding, cinema
trips and sightseeing. Healthy
living group included sessions on
eating, exercise, smoking cessation
and alcohol consumption.
Setting:
Community.
Study:
Combination supervised
and home-based sessions. Six
supervised sessions at exercise
laboratory during initial 3 weeks.
Home-based programme
of three to five sessions per week
during weeks 4 to 12. Choice
of treadmills, stationary cycles,
combination of the two or
overground walking. Exercise
intensity at self-selected level.
Setting:
Rehabilitation day centre.
Exercise:
Not described in detail.
Group sport and exercise activities.
Football, swimming, cycling,
badminton, walking group.
Setting:
Rehabilitation day centre.
Exercise:
Not described in detail.
Group sport and exercise activities.
Football, swimming, cycling,
badminton, walking group.
Liverpool University Side
Effect Rating Scale
(LUNSERS); and Likert scale
for subjective report of
self-esteem ranging from
poor to good.
Structured Clinical Interview
for DSM-IV Axis I Disorders
– Clinician Version (SCID-CV)
(for initial diagnosis);
Hamilton Rating Scale for
Depression (HRSD); 30 Item
Inventory of Depressive
Symptomatology – Self
Report (IDS-SR); and
General Activities Form of
the Quality of Life
Enjoyment and Satisfaction
Scale (Q-LES-Q general).
Researcher immersed in
field; participated in
exercise sessions; participant
observation and informal
interviews (conversation);
document review of medical
records;single semi-structured
interview with participant;
and single semi-structured
interview with mental
health professional in close
contact with participant
(care coordinator, physio-
therapist, exercise leader).
In-depth semi-structured
interview (45-90 minutes);
one participant had two
follow-up open-ended
interviews of 45 minutes
each.
Significant improvement in self-reported
measures of self-esteem. Significantly
higher prevalence of moderate to high
self-esteem (from 9.5% of participants at
baseline to 23% of participants at final
follow-up).
Significant changes in both clinician
scored and self-report measures.
Intention-to-treat analysis:
Significant
mean HRSD score decrease of 5.8 points.
Significant mean IDS-SR score decrease
of 13.9 points. Twenty-nine per cent of
participants met HRSD criteria for
remission post-treatment.
Analysis of completing participants:
Significant mean HRSD score decrease of
10.4 points. Significant mean IDS-SR
score decrease of 18.8 points. Significant
improvement in mean Q-LES-Q general
score. Sixty-three per cent of completing
participants met HRSD criteria for
remission post-treatment.
Emerging themes include:
Sense of
achievement and satisfaction; exercise
provided social support networks;
increased level of control over own life;
improvements in interpersonal skills;
constructive use of time; satisfaction from
contributing to a ‘team’; mental health
professionals believe exercise contributed
to recovery.
In their stories, all three men described:
Positive experiences of physical activity
during childhood; ‘valued physical activity’
(Carless and Sparkes 2008, p206);
enjoyed the social aspect of exercising; no
longer experienced psychotic symptoms;
and increased level of alertness and
concentration.
10/15
9/15
15/23
15/23
Smith et al
(2007b)
Pre-post design.
Study duration
2 years,
n = 966.
Trivedi et al
(2006)
Pre-post design
(pilot study).
Study duration
12 weeks,
n = 17.
Qualitative
Carless and
Douglas (2008)
Case studies.
Study duration
18 months,
n = 2.
Carless and
Sparkes (2008)
Case studies.
Study duration at
least 6 months in
the programme;
18 months full
study duration,
n = 3.
*Where study focus also included physical health outcome measures, only mental health outcome measures are listed here.
54 British Journal of Occupational Therapy February 2012 75(2)
The impact of exercise on the mental health and quality of life of people with severe mental illness: a critical review
Table 1 (continued)
Study and design Intervention Outcome measures* Results Quality score
Setting:
Not stated; participants
recruited from supported living
residence or day centres; referred
from primary or secondary care
agencies.
Exercise:
Walking programme in bird
parks, gardens, historic places, trails
in woods, lakes and coastlines.
Walks included educational talks
on wildlife, herbal plants, fauna
and flora. Transport provided.
Setting:
Inpatient and outpatient.
Exercise:
Group sessions two times
a week. Badminton, gym, water
aerobics, 10-pin bowling.
Participants paid for sessions at
community facilities.
Setting:
Shared supported housing.
Exercise:
Thirty minutes continuous
moderate activity twice weekly.
Walking in parks. Swimming.
Activities chosen by participants.
Setting:
Residential community
care unit.
Exercise:
Individual exercise
programmes prescribed and
conducted by qualified exercise
physiologists. No details given on
form or intensity of exercise.
Individual interviews
(20-45 minutes);
open-ended questions.
Two focus groups (50-60
minutes); open-ended
questions; focus groups
facilitated by previous
service user who used
experience to empathise
and gain rich data.
Participant observations;
interview with participants;
interview with workers; and
field diary.
One semi-structured,
informal focus group with
participants, exercise
physiologists and nursing
staff involved in facilitating
programme.
Participants liked the opportunity to be
involved in something.
Perceived benefits and outcomes
included:
Enjoyment – in the intervention
itself; opportunity to meet people and for
social interaction; purposeful activity and
sense of achievement; ‘help with sleeping’
(p177); researcher suggests walking
programme may provide mental health
benefits and positive emotional
experiences.
Conceptual framework developed to
explain experiences of sports therapy.
Core theme of framework was ‘taking
part’ – being involved in the exercise
programme.
Perceived benefits include:
Social
interaction; accomplishment – doing
something purposeful; wellbeing;
self-esteem; positivity; improved mood;
reduced feelings of anger; mental
alertness; increased energy; and
distraction from illness.
Effects of exercise include:
Identified as
coping strategy for symptom control (for
example, hearing voices); improved sleep
patterns; and increased social behaviours.
Possible explanations for effects include:
Increased individual control; offered
distraction from hallucinations; improved
self-esteem (body image and autonomy);
and benefits of social interactions.
Themes surrounding psychological
impacts include:
Participants enjoyed the
group dynamics, offering a team
environment and partners for support and
encouragement; and participants viewed
their involvement in the programme
positively.
21/23
21/23
16/23
14/23
Crone (2007)
Descriptive.
Study duration
not stated,
n = 4.
Crone and
Guy (2008)
Grounded theory.
Study duration:
participants had
taken part in
sports therapy
between 2 months
and 4 years,
n = 11.
Faulkner and
Sparkes (1999)
Ethnography.
Study duration
10 weeks,
n = 3.
Fogarty and
Happell (2005)
Qualitative
(not specified).
Study duration
3 months,
n = 12 (six
participants with
SMI; six mental
health staff).
*Where study focus also included physical health outcome measures, only mental health outcome measures are listed here; SMI = severe mental illness.
supervised exercise programmes, except for one that used
six supervised sessions prior to a home exercise programme
(Trivedi et al 2006) and another that included advice and
counselling on structured exercise and facilitated access to
local facilities (Brown and Chan 2006).
The majority of studies measured the effect on mental
health and QOL in combination with measures of physical
health, with only three focusing solely on psychological out-
comes (Babyak et al 2000, Trivedi et al 2006, Duraiswamy
et al 2007). Tools used to measure outcomes varied across
the range of studies. Some trends were noted in the number
of studies that used the measures of Positive and Negative
Syndrome Scale for Schizophrenia (PANSS) (n = 3) (Beebe
et al 2005, Duraiswamy et al 2007, Melamed et al 2008);
Hamilton Rating Scale for Depression (n = 2) (Babyak et al
2000, Trivedi et al 2006); and Medical Outcomes Study 36
Item Short Form questionnaire version 2 (n = 2) (Pelletier
et al 2005, Skrinar et al 2005).
55
British Journal of Occupational Therapy February 2012 75(2)
Kristy Alexandratos, Fiona Barnett and Yvonne Thomas
Qualitative studies
Findings of the qualitative studies fell under one of two main
concepts. The first concept was the effect of exercise on mental
health symptoms. In terms of the effect on symptoms, two
studies described improvements in illness management and
symptom levels, including mood and feelings of anger, as well
as offering a distraction from psychotic symptoms, such as
hallucinations (Faulkner and Sparkes 1999, Crone and Guy
2008). In one study the participants reported no longer expe-
riencing psychotic symptoms, and in another study mental
health professionals believed that the exercise intervention
had contributed to the recovery of their patients (Carless and
Douglas 2008, Carless and Sparkes 2008). Additionally, two
studies revealed an improvement in sleep patterns as a result
(Faulkner and Sparkes 1999, Crone 2007), and another two
noted improvements in concentration levels and alertness
(Carless and Sparkes 2008, Crone and Guy 2008).
The second concept explored the perceived effect of
exercise on QOL. Most significantly, in four of the studies,
the participants described the positive social interactions
that the exercise intervention offered (Faulkner and Sparkes
1999, Fogarty and Happell 2005, Crone 2007, Carless and
Douglas 2008). Secondly, three studies revealed that a per-
ceived benefit of exercise was doing something purposeful
and using their time constructively (Crone 2007, Carless and
Douglas 2008, Crone and Guy 2008). Thirdly, participants
from two separate studies described the way in which the
exercise intervention had helped them to increase the level
of control they had over their own life (Faulkner and Sparkes
1999, Carless and Douglas 2008). Lastly, improvements in
self-esteem were described in two studies (Faulkner and
Sparkes 1999, Crone and Guy 2008).
Discussion
Methodological design
Quantitative studies
First, the studies revealed a trend in relatively small sample
sizes. A number of these smaller studies were pilot studies.
Although these present important data of their own, they serve
to remind us that it is a relatively underexplored topic for
which researchers are still trialling the most effective means
of testing (Beebe 2007). It can also be said that the difficult
nature of recruiting participants with a SMI may have con-
tributed to the small sample sizes in the studies (Beebe 2007).
Secondly, with regard to methodological quality, more than
half of the RCTs employed single blinding techniques. The
use of blinded assessors in these studies improved the quality
of their evidence by avoiding issues of measurement bias
(Glasziou et al 2001). Thirdly, half of the RCTs had method-
ological concerns with their control groups. Control groups
were either not matched for characteristics; had received
aspects of the intervention; or had participants self-enrol in
similar interventions outside the study. More methodolog-
ically sound control groups would greatly support the asso-
ciated reports of effectiveness (Law et al 1998) of the exercise
interventions. Furthermore, the pre-post trials may have pro-
vided significant results, but the research design limits the
power to claim that positive outcomes were a result of the
treatment alone and not due to other factors (Law et al 1998).
Fourthly, less than half of the quantitative studies employed
intention-to-treat analysis, and half of these did not do so
for all their outcomes. Intention-to-treat analysis would
have provided a more realistic representation of the inter-
vention effect as it would occur in real life (Riegelman
2005) and, given the high dropout rate in the studies, it
could be more confidently assumed that the same phe-
nomena would occur in clinical practice. Those studies
that accounted for dropouts in their analysis may provide
practitioners with a more accurate picture of the clinical
significance of exercise interventions. Lastly, all studies
used a volunteer method of sampling, except for one that
used a sampling method of convenience by using partici-
pants already enrolled in a wellbeing support programme
(Smith et al 2007b). This method of sampling could have
potentially caused an overestimation of the effect (Law et al
1998) of exercise, because participants may have been more
motivated and interested in exercise. However, volunteer
sampling is common practice in RCTs due to the require-
ment for informed consent (Riegelman 2005). Given the
high dropout rates and small sample sizes, it would appear
at this stage that volunteer sampling would be a required
means for this population group. Beebe (2007) indicated
that the difficulties in obtaining people with SMI for research
may, in fact, call for more creative recruitment methods.
It is important to note that only three studies explored
the effect specifically on mental health and QOL (Babyak
et al 2000, Trivedi et al 2006, Duraiswamy et al 2007). The
remainder of the quantitative evidence comes from studies
that explored both the physical and mental health benefits
of exercise. Studies with a sole focus on mental health and
QOL would provide greater evidence for the use of exercise
in therapy, because they would offer a deeper insight and
more detailed analysis of these benefits. Although it may
prove difficult, there is certainly a need for larger RCTs
using solely psychological outcome measures and sound
research methodology.
Qualitative studies
The purpose and nature of qualitative research is different
to that relying on quantitative methods and, as such, cannot
be either compared or evaluated by the same means. When
it comes to issues of bias in qualitative research, confirma-
bility aims to ensure that the findings presented are neutral
(Letts et al 2007). In one study, the experience of one
researcher as a previous SMI service user certainly added
a level of insight and trustworthiness that could not be
ignored (Crone and Guy 2008). However, half of the studies
did not address strategies used to combat bias, making it
difficult to rely on the evidence these produce when evalu-
ating the effects of exercise intervention.
When approaching qualitative research, one important
factor is to ensure that results are interpreted in context (Taylor
56 British Journal of Occupational Therapy February 2012 75(2)
The impact of exercise on the mental health and quality of life of people with severe mental illness: a critical review
2007). It is mainly the responsibility of the reader to make
certain that he or she does not misinterpret the intentions of
the study; however, researchers must provide sufficient infor-
mation for the reader to be able to do so (Taylor 2007). The
majority of the studies described in this report had difficulty
achieving transferability, that is, the researchers did not
provide a detailed description of the participants, setting
and /or intervention (Letts et al 2007). In all of these cases,
it was due to a poor or absent description of the participants
or intervention settings. Many studies did not describe the
exercise intervention used adequately and, as demonstrated
by the quantitative studies, the form of exercise can poten-
tially affect the outcome. The absence of participant diagnosis
in half the studies creates difficulties when attempting to
apply findings to the target population. Furthermore, the
over-representation of male participants in the studies restricts
the degree to which results can be applied to the wide pop-
ulation of people with SMI, particularly given that research
suggests that women with SMI have different experiences
of exercise intervention to men (McDevitt et al 2006).
Research findings
The results of the studies suggest that exercise can lead to
improvements in QOL by offering an avenue for social
interaction and goal-directed activity. It can also contri-
bute towards a sense of empowerment and improved self-
confidence for people who experience SMI. In addition,
the study results indicate that exercise can improve some
symptoms of SMI.
Quantitative findings
Cardiovascular: walking exercise
When extrapolating the evidence for exercise to improve the
mental health and QOL of people with SMI, a variety of out-
comes emerged. The results of the walking interventions
suggested that it is not an effective form of exercise to improve
the symptoms of mental illness. Results did suggest that
walking could have a positive effect on QOL; however, as
only one study used this outcome measure (Melamed et al
2008), the evidence base to support its use is limited. Mood
and psychosocial functioning were also found to improve
after the walking intervention but, again, only one study
examined this area (McDevitt et al 2005). All the walking
studies explored the effect on physical health as well, and
it could be seen that this style of intervention was chosen
specifically for its physical health benefits without much
consideration for its psychological effect. Stronger evidence
would be required to support the use of walking as an effec-
tive means of improving mental health and wellbeing.
Weights training and mixed cardiovascular:
walking, cycling and jogging
The studies that used a variety of cardiovascular exercises,
including walking, cycling and jogging, were more successful
than the interventions that used walking alone. Results
demonstrated the effectiveness in reducing the severity of
major depression when measured using clinical scales.
However, subjective reports of symptom control were not as
positive, with only one study recording significant improve-
ments (Trivedi et al 2006). Given the subjectivity of QOL,
it is just as important to consider these self-reported mea-
sures when weighing up the evidence. Cardiovascular
exercise combined with mixed weights training provided
some degree of evidence for the positive effect on QOL,
but with significant changes in only some components of
self-reported measures.
Low intensity exercise and physical activity
The broad lower intensity interventions appeared more
successful in improving QOL and reducing symptoms in
comparison to other forms of exercise. Notably, the yoga
therapy intervention was the most successful intervention
of all studies, with positive improvements in measures of
symptoms, wellbeing and QOL (Duraiswamy et al 2007).
However, given that the intervention also included aspects
of relaxation and breathing exercises, it could be possible
that these contributed to the positive effects as well and that
it was not solely the exercise itself. The community-based
exercise groups that used swimming, aqua aerobics and
recreation-based physical activity, such as sightseeing and
educational walks, gained significant improvements in
subjective ratings of self-esteem (Smith et al 2007b). However,
as this was the only outcome measure used, it is impossible
to comment on its effect on other psychological parameters.
Intervention settings
The majority of interventions were delivered in community
or outpatient settings, and this dictates the degree to which
these results can be generalised to the population with SMI.
The research highlighted the effective use of exercise with
consumers in community settings. It can be assumed that
the participants in these studies would have been manag-
ing their illness at a high functional level, which could
have aided positive outcomes. Very little evidence exists to
support the use of exercise or physical activity to improve
QOL for people with SMI in acute inpatient settings, and
there is room for investigation into its feasibility. The main
difference between interventions delivered in a supervised
setting compared with those offering advice and home-based
exercise programmes was evident in the dropout rates. The
two studies that used unsupervised and self-directed exercise
programmes had the highest dropout rates of all studies
(Brown and Chan 2006, Trivedi et al 2006). This phenom-
enon suggests that adherence to exercise programmes in
people with SMI is greater with the external motivation
and reinforcement provided by supervised programmes.
Long-term feasibility
Most of the interventions were short-term programmes, with
limited follow-up. Although some of these studies proved
effective in demonstrating immediate short-term gains in
mental health and QOL, many were not able to report on
the sustainability of such improvements. While this provides
57
British Journal of Occupational Therapy February 2012 75(2)
Kristy Alexandratos, Fiona Barnett and Yvonne Thomas
some degree of evidence for the use of exercise as a thera-
peutic medium, it does not offer practitioners the confi-
dence that it will provide long-term benefits for their clients.
Only one study measured the maintenance of mental health
improvements, because it reviewed participants 6 months
after the completion of the exercise intervention (Babyak
et al 2000). Further research is necessary into the long-term
effects of exercise on the wellbeing of people with SMI.
The short duration of most of the studies may also have
contributed to the statistical non-significance of results.
When considering long-term benefits, one cannot ignore
the high dropout rate in the studies and, as such, the feasi-
bility of long-term programmes. While the dropout rate
could, in part, have been attributed to the nature of SMI,
researchers have also identified a number of barriers to par-
ticipation, such as fatigue, illness, symptoms of mental
illness, side effects of medications and weight gain (McDevitt
et al 2006, Usher et al 2007). Drawing on meaningful and
personally motivated mediums of exercise and physical
activity may help to address the high dropout rates.
Qualitative findings
The qualitative studies focused directly on mental health and
QOL outcomes, with only one study that assessed subjective
reports of physical health as well (Fogarty and Happell 2005).
The restricted range of diagnostic groups makes it difficult
to determine the impact of exercise intervention in all people
with SMI. Of the studies that specified participant diagnosis,
all investigated the effect of exercise on people with a diag-
nosis of schizophrenia. Although these studies provide a good
deal of evidence for the use of exercise in this group, they
do not provide any evidence for its use in people with other
illnesses, such as major depression and bipolar disorder.
Future qualitative studies are required to explore the effects
of exercise intervention on SMIs collectively and separately.
Mental illness symptoms
In terms of symptoms, the research suggested that exercise
can contribute to subjective improvements in mood, alert-
ness, concentration and sleep patterns. Research suggests
that exercise can have a positive effect on sleep patterns
for the general population (Youngstedt 2005). It could be
possible that improvements in the sleep patterns of people
with SMI are due more to this general exercise benefit than
to actual reductions in symptoms themselves. Additionally, the
exercise interventions appeared to have improved psychotic
symptoms, such as hallucinations, in some cases simply by
providing a distraction. In one study, health professionals
suggested that exercise had contributed to the recovery of
patients with SMI (Carless and Douglas 2008); however,
stronger evidence would be required to support this concept.
Quality of life
Quality of life improvements were similar across the studies.
The research suggested that, most notably, exercise can improve
QOL by providing people with SMI a medium through which
to foster social interaction. However, without sufficient details
on the interventions, it cannot be determined whether this
occurred through group interventions, connection with
community facilities or involvement in the study process
itself, and warrants further research. Indeed, researchers
have recognised the need to investigate the role of physical
activity in reducing social isolation for this population further
(Richardson et al 2005). The studies also suggested that
exercise intervention can contribute to improved QOL by pro-
viding people with SMI an avenue through which to spend
their time meaningfully. It could be seen that it was the process
of being involved in something purposeful, rather than the
exercise itself, that contributed to these improvements.
It appears to be the perception of some mental health
staff that improvements are attributable solely to the dis-
traction from daily life and structure that exercise provides
(Faulkner and Biddle 2001, 2002). However, it is possible
that the occupation of exercise itself, being a purposeful and
goal-directed activity, has a much more direct effect than
these studies suggest. Further research into the mechanism
responsible for QOL improvements would help to shed
light on this issue.
Empowerment
Participants described the way in which the exercise inter-
vention empowered them and increased their level of per-
sonal control and self-confidence. The mechanism through
which this confidence improvement occurs is unclear. Some
researchers believe that it is secondary to the effect of exercise,
linked to improvements in, and control over, their own body
image (Faulkner and Sparkes 1999). Other research has
described the way in which improved self-confidence from
the exercise intervention transferred into other domains of life
(Shiner et al 2008). Certainly, there is room for researchers
to explore the role of exercise to provide a medium through
which people can develop the skills, attitude and identity
to improve their QOL in all areas.
Implications
The exercise experiences of people with SMI link closely
with the concepts of the meaning of occupation: doing, being,
becoming and belonging (Wilcock 1998, Hammell 2004). The
concept of ‘doing’ is illustrated by the participants’ engage-
ment in purposeful activity, and meaningful and construc-
tive use of their time. ‘Being’ and ‘becoming’ are seen where
people with SMI reshape their identity and form perceptions
of the self through the exercise experience. The strong
emphasis on the value of group exercise and partners for
motivation shows that exercise provides a medium through
which these people can experience ‘belonging’.
There is an opportunity for occupational therapy to look
towards embedding exercise and physical activity interven-
tion into meaningful everyday occupations. Sorensen (2006)
argued that the motivation to participate in exercise in people
with SMI is similar to the general population, and that exercise
needs to be self-determined and an enjoyable act in itself. If
people with SMI were offered a range of ways to exercise, and
were able to select those that are intrinsically motivating, it may
58 British Journal of Occupational Therapy February 2012 75(2)
The impact of exercise on the mental health and quality of life of people with severe mental illness: a critical review
improve the acceptance of such interventions and help to
address the high dropout rate. A deeper insight into the expe-
riences of exercise for people with SMI would help researchers
to design more acceptable intervention strategies, and aid
health professionals’ understanding of their clients’ needs.
This review raises several questions, warranting further
research to explore the use of exercise as an augmentative
treatment strategy. Further study is needed on the mecha-
nism through which holistic improvements in psychosocial
domains occur, as well as the most effective styles of exercise
intervention. From an occupational therapy perspective, more
research is needed to explore the role that purposeful activity,
meaningful use of time, social interaction and empowerment
play towards improving QOL and psychological wellbeing.
There is a clear path for researchers to explore the unique
way in which exercise is experienced by people with SMI
and, as such, how this occupation affects QOL for this client
group in comparison to the general population.
Furthermore, given the mixed results from the different
interventions, further research is warranted into comparing the
benefits of different mediums of exercise, as well as research
into other types of exercise not previously explored. The high
proportion of male participants in exercise studies also high-
lights a gap in the evidence from the perspective of women
with SMI. Research into the exercise experiences of these
women is needed in order to provide a more comprehensive
understanding, because the experiences would most likely
differ from those of their male counterparts.
Methodological limitations
The following limitations of this review should be acknowl-
edged. First, restricting the initial search to English language
articles only may have meant the exclusion of relevant
research printed in other languages. Secondly, as the litera-
ture search was completed in December 2009, any relevant
research published since then was not included. Lastly,
because some exercise interventions were included as part
of wider healthy living interventions, interventions such as
behavioural therapy and nutrition education could have
acted as potential confounders.
Conclusion
This systematic review highlights that existing research
warrants the use of exercise intervention to relieve symptoms
of SMI and to improve QOL. First, the research shows that
exercise can contribute to improvements in mental illness
symptoms, including mood, alertness, concentration and
sleep patterns. It also suggests that it may reduce or provide
a distraction from psychotic symptoms, such as hallucinations.
Furthermore, findings suggest that exercise may lead to a
reduction in the severity of depression, and improvements in
psychosocial functioning. Secondly, the research shows that
exercise can contribute to improved QOL for people who expe-
rience SMI. These improvements appear to occur through
exercise as it provides opportunity and avenues for social
interaction, meaningful use of time and purposeful and goal-
directed activity. The studies also describe how exercise can
lead to empowerment and increased personal control.
The literature suggests that mixed styles of exercise, such
as cardiovascular and weights training, yoga and swimming,
possibly provide greater benefits than walking alone. In
spite of this, the small number of studies limits the strength
of this argument, and calls for research comparing the
outcomes of different mediums of exercise. In addition, it
is evident that further research is needed to determine the
long-term effectiveness of any improvements and the mech-
anism through which these improvements occur.
Given the subjective nature of measuring QOL, there is
an opportunity for qualitative researchers to explore the
experiences of exercise for individuals with SMI. This would
help to give depth and an interpretation of existing research
from the experiences of people with SMI themselves. However,
if qualitative approaches are to be used to add strength to
the argument for exercise intervention in treating SMI, there
is a need for a clearer picture of the context of the study and
greater confirmability.
The positive effect of exercise on the mental health and
wellbeing of the general population is already widely
researched. This review suggests that exercise may be an
effective complementary intervention strategy to reduce
symptoms and to improve QOL for people experiencing SMI.
The limited number of studies on people with SMI high-
lights that future studies are needed to describe the way in
which exercise can meet the unique needs of this population.
Acknowledgement
This article is based on work carried out as part of a BOT (Hons) degree at
James Cook University, Queensland, Australia.
Conflict of interest:
None declared.
Key findings
■Exercise can improve quality of life through increasing social inter-
action and purposeful activity, and providing a sense of empowerment
and improved self-confidence.
■Exercise can improve some symptoms of severe mental illness.
What the study has added
The study contributes to the understanding of how exercise can improve
quality of life and reduce symptoms for people who experience a severe
mental illness.
References
American College of Sports Medicine (2006)
ACSM’s guidelines for exercise
testing and prescription.
Baltimore: Lippincott Williams and Wilkins.
Babyak M, Blumenthal J, Herman S, Khatri P, Doraiswamy M, Moore K,
Craighead E, Baldewicz T, Ranga Krishnan K (2000) Exercise treatment
for major depression: maintenance of therapeutic benefit at 10 months.
Psychosomatic Medicine, 62(5),
633-38.
Barras S (2005)A systematic and critical review of the literature: the effectiveness
of occupational therapy home assesment on a range of outcome measures.
Australian Occupational Therapy Journal, 52(4),
326-36.
59
British Journal of Occupational Therapy February 2012 75(2)
Kristy Alexandratos, Fiona Barnett and Yvonne Thomas
Beebe L (2007) What can we learn from pilot studies?
Perspectives in Psychiatric
Care, 43(4),
213-18.
Beebe L,Tian L, Morris N, Goodwin A, Allen S, Kuldau J (2005) Effects of exercise
on mental and physical health parameters of persons with schizophrenia.
Issues in Mental Health Nursing, 26(6),
661-76.
Brown S, Chan K (2006) A randomized controlled trial of a brief health promotion
intervention in a population with serious mental illness.
Journal of Mental
Health, 15(5),
543-49.
Campbell P, Foxcrof t D (200 3) E xercise therapy for schizophrenia.
Cochrane Database of Systematic Reviews (4).
Art. No.: CD004412.
DOI: 004410.001002/14651858.CD14004412.
Carless D, Douglas K (2008) The role of sport and exercise in recovery from serious
mental illness:twocase studies.
InternationalJournal of Men’sHealth,7(2),
137-56.
Carless D,Sparkes A (2008) The physical activity experiences of men with serious
mental illness: three short stories.
Psychology of Sport and Exercise, 9(2),
191-210.
Crone D (2007) Walking back to health: a qualitative investigation into service
users’ experiences of a walking project.
Issues on Mental Health Nursing,
28(2),
167-83.
Crone D, Guy H (2008) ‘I know it is only exercise, but to me it is something
that keeps me going’: a qualitative approach to understanding mental
health service users’ experiences of sports therapy.
International Journal
of Mental Health Nursing, 17(3),
197-207.
Deenadayalan Y, Perraton L, Machotka Z, Kumar S (2010) Day therapy
programs for adolescents with mental health problems: a systematic review.
Internet Journal of Allied Health Sciences and Practice, 8(1),
1-14. Available at:
http://ijahsp.nova.edu/articles/Vol8Num1/kumar.htm
Accessed 17.01.11.
Duraiswamy G, Thirthalli J, Nagendra H, Gangadhar B (2007) Yoga therapy as an
add-on treatment in the management of patients with schizophrenia –a
randomized controlled trial.
Acta Psychiatrica Scandinavica, 116(3),
226-32.
Faulkner G, Biddle S (2001) Exercise and mental health:it’s just not psychology!
Journal of Sports Sciences, 19(6),
433-44.
Faulkner G, Biddle S (2002) Mental health nursing and the promotion of physical
activity.
Journal of Psychiatric and Mental Health Nursing, 9(6),
659-65.
Faulkner G, Sparkes A (1999) Exercise as therapy for schizophrenia: an
ethnographic study.
Journal of Sport and Exercise Psychology, 21(1),
52-69.
Faulkner G, Cohn T, Remington G (2007) Interventions to reduce weight gain
in schizophrenia.
Cochrane Database of Systematic Reviews (1).
Art. No.:
CD005148. DOI: 005110.001002/14651858.CD14005148.pub14651852.
Fogarty M, Happell B (2005) Exploring the benefits of an exercise program
for people with schizophrenia: a qualitative study.
Issues in Mental Health
Nursing, 26(3),
341-51.
Fogarty M, Happell B, Pinikahana J (2004)The benefits of an exercise program
for people with schizophrenia: a pilot study.
Psychiatric Rehabilitation
Journal, 28(2),
173-76.
Glasziou P, Irwig L, Bain C, Colditz G (2001)
Systematic reviews in health care:
a practical guide.
Cambridge: Cambridge University Press.
Hammell K (2004) Dimensions of meaning in the occupations of daily life.
Canadian Journal of Occupational Therapy, 71(5),
296-305.
Law M, Stewart D, Pollock N,Letts L, Bosch J, Westmorland M (1998)
Guidelines
for critical review form: quantitative studies.
Available at:
http://www.srs-
mcmaster.ca/Default.aspx?tabid=630
Accessed 25.11.09.
Letts L,Wilkins S, Law M,Stewart D,Bosch J, Westmorland M (2007)
Guidelines
for critical review form: qualitative studies (Version 2.0).
Available at:
http://www.srs-mcmaster.ca/Default.aspx?tabid=630
Accessed 25.11.09.
McCormick B, Frey G, Lee C, Chun S, Sibthorp J, Gajic T, Stamatovic-Gajic B,
Maksimovich M (2008) Predicting transitory mood from physical activity
level among people with severe mental illness in two cultures.
International
Journal of Social Psychiatry, 54(6),
527-38.
McDevitt J,Wilbur J, Kogan J, Briller J (2005) A walking program for outpatients
in psychiatric rehabilitation: pilot study.
Biological Research for Nursing,
7(2),
87-97.
McDevitt J, Snyder M, Miller A, Wilbur J (2006) Perceptions of barriers and
benefits to physical activity among outpatients in psychiatric rehabilitation.
Journal of Nursing Scholarship, 38(1),
50-55.
Melamed Y, Stein-Reisner O, Gelkopf M, Levi G, Sivan T, Ilievici G, Rosenberg R,
Weizman A, Bleich A (2008) Multi-modal weight control intervention for
people with persistent mental disorders.
Psychiatric Rehabilitation Journal,
31(3),
194-200.
Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009) Preferred
reportingitems for systematic reviews and meta-analyses:the PRISMAstatement.
PLoS Medicine, 6(7):
e1000097. doi:10.1371/journal.pmed.1000097
Pelletier J, Nguyen M, Bradley K, Johnsen M, McKay C (2005) A study of a
structured exercise program with members of an ICCD Certified Clubhouse:
program design, benefits, and implications for feasibility.
Psychiatric
Rehabilitation Journal, 29(2),
89-96.
Richardson C, Faulkner G, McDevitt J, Skrinar G, Hutchinson D, Piette J (2005)
Integrating physical activity into mental health services for persons with
serious mental illness.
Psychiatric Services, 56(3),
324-31.
Riegelman R (2005)
Studying a study and testing a test: how to read the
medical evidence.
Philadelphia: Lippincott Williams and Wilkins.
Shiner B,Whitley R,VanCittersA, Pratt S, Bartels S (2008) Learning what matters
for patients: qualitative evaluation of a health promotion program for those
with serious mental illness.
Health Promotion International, 23(3),
275-82.
Skrinar G, Huxley N, Hutchinson D, Menninger E, Glew P (2005)The role of
a fitness intervention on people with serious psychiatric disabilities.
Psychiatric Rehabilitation Journal, 29(2),
122-27.
Smith S,Yeomans D, Bushe C, Eriksson C, Harrison T, Holmes R, Mynors-Wallis L,
Oatway H,Sullivan G (2007a)A well-being programme in severe mental
illness. Baseline findings in a UK cohort.
International Journal of Clinical
Practice, 61(12),
1971-78.
Smith S,Yeomans D, Bushe C, Eriksson C, Harrison T, Holmes R, Mynors-Wallis L,
Oatway H,Sullivan G (2007b)A well-being programme in severe mental
illness. Reducing risk for physical ill-health: a post-programme service
evaluation at 2 years.
European Psychiatry, 22(7),
413-18.
Sorensen M (2006) Motivation for physical activity of psychiatric patients
when physical activity was offered as part of treatment.
Scandinavian
Journal of Medicine and Science in Sports, 16(6),
391-98.
Stathopoulou G, Powers M, BerryA, Smits J, Otto M (2006) Exercise interventions
for mental health: a quantitative and qualitative review.
Clinical Psychology:
Science and Practice, 13(2),
179-93.
Taylor M (2007)
Evidence-based practice for occupational therapists.
Oxford: Blackwell Publishing.
Trivedi M, Greer T, Grannemann B, Chambliss H, Jordan A (2006) Exercise
as an augmentation strategy for treatment of major depression.
Journal
of Psychiatric Practice, 12(4),
205-13.
Usher M, Stanbury L, Cheeseman V, Faulkner G (2007) Physical activity
preferences and perceived barriers to activity among persons with severe
mental illness in the United Kingdom.
Psychiatric Services, 58(3),
405-08.
Wilcock AA (1998) Reflections on doing, being and becoming.
Canadian
Journal of Occupational Therapy, 65(5),
248-56.
World Health Organization (1997)
WHOQOL measuring quality of life.
Available at:
http://www.who.int/mental_health/media/68.pdf
Accessed
04.12.09.
60 British Journal of Occupational Therapy February 2012 75(2)
The impact of exercise on the mental health and quality of life of people with severe mental illness: a critical review
Appendix 1. Criterion scoring system (yes = 1; no = 0; not applicable/
not stated = 0)
Quantitative criteria
Total quality score out of 15
1. Study purpose was stated clearly
2. Relevant background literature was reviewed
3. Research design was appropriate
4. Sample was described in adequate detail
5. Sample size was justified
6. Outcome measures were valid
7. Outcome measures were reliable
8. Intervention was described in adequate detail
9. Contamination was avoided
10. Co-intervention was avoided
11. Results were reported in statistical significance
12. Analysis method used was appropriate
13. Clinical significance of findings was reported
14. Number of dropouts was reported
15. Conclusions were appropriate to study findings
World Health Organization (2011)
Global strategy on diet, physical activity
and health. Available at: http://www.who.int/dietphysicalactivity/pa/en/
index.html
Accessed 22.06.11.
Youngstedt S (2005) Effects of exercise on sleep.
Clinics in Sports Medicine,
24(2),
355-65.
Qualitative criteria
Total quality score out of 23
1. Study purpose was stated clearly
2. Relevant background literature was reviewed
3. Study design was appropriate to study question
4. Theoretical perspective for study was identified
5. Study methods were congruent with study purpose
6. Selection process was described in detail
7. Selection occurred until data redundancy was reached
8. A clear site description was provided
9. A clear participant description was provided
10. Role of researcher was clearly described
11. Any assumptions and biases of researcher described
12. Procedural rigour used in data collection methods
13. Analysis of data was inductive
14. Findings were reflective of data
15. Decision trail was developed during analysis
16. Data analysis process was clearly described
17. A meaningful picture of phenomenon under study emerged
18. Methods used to ensure credibility were described
19. Methods used to ensure transferability were described
20. Methods used to ensure dependability were described
21. Methods used to ensure confirmability were described
22. Conclusions were appropriate to study findings
23. Findings contributed to future research and practice
