Intervention Review

You have free access to this content

Exercise for depression

  1. Gary M Cooney1,
  2. Kerry Dwan2,
  3. Carolyn A Greig3,
  4. Debbie A Lawlor4,
  5. Jane Rimer5,
  6. Fiona R Waugh6,
  7. Marion McMurdo7,
  8. Gillian E Mead8,*

Editorial Group: Cochrane Common Mental Disorders Group

Published Online: 12 SEP 2013

DOI: 10.1002/14651858.CD004366.pub6


How to Cite

Cooney GM, Dwan K, Greig CA, Lawlor DA, Rimer J, Waugh FR, McMurdo M, Mead GE. Exercise for depression. Cochrane Database of Systematic Reviews 2013, Issue 9. Art. No.: CD004366. DOI: 10.1002/14651858.CD004366.pub6.

Author Information

  1. 1

    Royal Edinburgh Hospital, NHS Lothian, Division of Psychiatry, Edinburgh, Midlothian, UK

  2. 2

    University of Liverpool, Institute of Child Health, Liverpool, England, UK

  3. 3

    University of Birmingham, Birmingham, UK

  4. 4

    University of Bristol, MRC Centre for Causal Analyses in Translational Epidemiology, School of Social and Community Medicine, Bristol, Avon, UK

  5. 5

    NHS Lothian, University Hospitals Division, Edinburgh, Scotland, UK

  6. 6

    Victoria Hostpital Kirkcaldy, General Surgery, NHS Fife, Kirkcaldy, Fife, UK

  7. 7

    University of Dundee, Centre for Cardiovascular and Lung Biology, Division of Medical Sciences, Dundee, UK

  8. 8

    University of Edinburgh, Centre for Clinical Brain Sciences, Edinburgh, UK

*Gillian E Mead, Centre for Clinical Brain Sciences, University of Edinburgh, Room S1642, Royal Infirmary, Little France Crescent, Edinburgh, EH16 4SA, UK. gillian.e.mead@ed.ac.uk. gmead@staffmail.ed.ac.uk.

Publication History

  1. Publication Status: New search for studies and content updated (no change to conclusions)
  2. Published Online: 12 SEP 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]

MethodsRCT, parallel groups


ParticipantsCommunity volunteers recruited via media. Eligible if had DSM-IV major depressive disorder
Mean age 70 (range 61 to 88)
63% women
N = 156


Interventions1. Group walking or jogging 3 times per week (n = 53 randomised)
2. Sertraline (SSRI) at standard dose (n = 48 randomised)
3. Combined walking or jogging and sertraline (n = 55 randomised)

Duration of interventions: 16 weeks
Exercise intensity was 70% to 85% of target heart rate


Outcomes1. Clinical diagnosis of depression using DSM-IV
2. Hamilton Rating Scale for Depression
3. Beck Depression Inventory


NotesAnalysis intention-to-treat, using last observation carried forward for missing data.
Review authors used group 2 and group 3 in the meta-analysis
Outcome assessor blind


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskThe method of randomisation is not described in the paper

Allocation concealment (selection bias)Low riskThis was categorised as inadequate in the first version of this review, published in the BMJ, after Debbie Lawlor had contacted authors to obtain data on allocation concealment. Further information from the author has enabled us to change this to low risk

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants were aware of the treatment group to which they were allocated

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the interventions were aware of treatment group. It is not clear whether this would have introduced bias.

Blinding (performance bias and detection bias)
outcome assessors
Low riskUsed HAM-D as an outcome measure. This is rated by clinicians. Every effort was made to ensure that clinical raters were unaware of participants' treatments groups after allocation

Incomplete outcome data (attrition bias)
All outcomes
Low riskAlthough not all participants completed the interventions, the authors used last observation carried forward to impute missing values

Selective reporting (reporting bias)Low riskFrom the study report, it appears that all the prespecified outcome measures have been reported,

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsPeople with major depression recruited through television, radio and newspaper. Mean age 52 (SD 8), 76% women, N = 202


Interventions1. Home-based aerobic exercise (same 'exercise prescription' as the supervised aerobic group, but performed it on their own (n = 53)
2. Supervised group aerobic exercise (walking and jogging) (n = 51)
3. Sertraline (n = 49)
4. Placebo (n = 49)
Intervention 16 weeks


OutcomesPrimary endpoint was remission (no MDD) and a HAM score of < 8, and also a continuous severity score on the HAM-D


NotesAnalysis intention-to-treat using last observation carried forward
Blinded outcome assessment


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskConditional randomisation, stratified by age, gender and depression severity

Allocation concealment (selection bias)Low riskCentral allocation by a computer

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants were aware of whether they received exercise or not. It is unclear whether this introduced bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose personnel delivering exercise were aware of group allocation. It is unclear whether this introduced bias

Blinding (performance bias and detection bias)
outcome assessors
Low riskThe outcome assessors were unaware of treatment allocation

Incomplete outcome data (attrition bias)
All outcomes
Low riskImputed missing outcome assessment results by last observation carried forward

Selective reporting (reporting bias)Low riskFrom the study report, it appears that all the prespecified outcome measures have been reported. Authors report protocol on clinicaltrials.gov

Other biasUnclear riskUnclear


MethodsRCT


Participants35 years or older with documented coronary artery disease and depressive symptoms (n=101)


Interventions1. Exercise (group walking, running or jogging on treadmill (n = 37; 65% men)

2. Sertraline (n = 40; 63% men)

3. Placebo (n = 24; 83% men)


OutcomesDepression as diagnosed by DSM-IV and severity of depression as rated on HAM-D; heart disease biomarkers


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-based randomisation

Allocation concealment (selection bias)Low riskAllocation through distribution of sealed envelopes

Blinding (performance bias and detection bias)
participants
Unclear riskNot possible to blind participants receiving exercise; unclear effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskNo detail given

Blinding (performance bias and detection bias)
outcome assessors
Low riskTelephone surveys (HAMD) by psychologists blinded to arm of trial

Incomplete outcome data (attrition bias)
All outcomes
Low risk95 out of 101 completed protocol and outcome assessment

Selective reporting (reporting bias)Low riskPrespecified outcomes reported in results. There is a study protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsUniversity counselling service
Mean age 23.3 years
82% women
N = 11


Interventions1. CBT plus exercise (n = 5)
2. CBT alone (n = 6)

Exercise was walking on a treadmill for 20 minutes, twice a week for 6 weeks

Cognitive therapy: met counsellors once a week for 9 weeks


Outcomes1. DSM-IV MDD, dysthymia or depressive disorder
2. Above cut-off depression on BDI and CES-D


NotesSelf report
Randomisation method not stated
7/11 randomised participants completed the interventions
Data provided for each participant. Mean and SD calculated by us carrying forward baseline data for those who dropped out


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskMethod of randomisation not described

Allocation concealment (selection bias)Unclear riskMethod of randomisation not described

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants were not blind to treatment allocation, but it is unclear whether this introduced bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blinded; the effect of this on bias is unclear

Blinding (performance bias and detection bias)
outcome assessors
High riskThe primary outcome was the BDI, which is a self report measure

Incomplete outcome data (attrition bias)
All outcomes
High risk4/11 dropped out (2/5 in the exercise and CBT; 2/6 in the CBT only group)

Selective reporting (reporting bias)Unclear riskOn the basis of the report, all the prespecified outcomes have been reported. No protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsCommunity-dwelling and nursing home people over 65 with mild depression, recruited through newsletters, newspaper advertisements, distributing flyers at local nursing homes and public presentations

Mean age 73.5 (SD 7.8) in exercise, 76.4 (6.4) in medication and 73.9 (5.8) in control group

62% women

N = 37


Interventions1. Faculty-based group aerobic and resistance training for 60 minutes 3 days a week for 16 weeks (n = 14)

2. Once daily sertraline titrated to response (evaluated at weeks 2, 6, 10 and 14) (n = 11)

3. Control group: contacted at weeks 2, 6, 10, 14 to discuss general health status (n = 12)


OutcomesHAM-D


NotesMethod of randomisation not stated, states intention-to-treat analysis, however data unclear on how many participants dropped out of each group


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-generated random allocation list

Allocation concealment (selection bias)Unclear riskMethod not described in the paper

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants were not blinded to treatment allocation

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind, but it is not clear what influence this had on bias

Blinding (performance bias and detection bias)
outcome assessors
Low riskUsed HDRS as an outcome, assessors were blind to treatment allocation

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskThe authors of the trial state that it is intention-to-treat, but no data are provided on the number who dropped out of the trial

Selective reporting (reporting bias)Unclear riskIt appears that all the prespecified outcomes are reported, but no protocol

Other biasUnclear riskunclear


MethodsRCT


ParticipantsVolunteers aged 18 to 45 recruited via flyers and word of mouth from University and local physician referral. Depression severity mild to moderate, if severe required written permission from physician.

Mean age 26.4 (18 to 43).

100% women.


InterventionsFor 10 weeks:

1. Up to 5 high-intensity aerobic exercise sessions per week (1 supervised) to expend 1000 Kcal per week (n = 15)

2. Up to 5 low-intensity aerobic exercise sessions per week (1 supervised) to expend 1000 Kcal per week (n = 11)

3. Met with investigator once per week for 30 minutes of group stretching exercises (n = 12)


OutcomesBeck Depression Inventory-II


NotesAnalysis not intention-to-treat

BDI-II self-rated depression score


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot described

Allocation concealment (selection bias)Unclear riskUnclear; not described

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind to treatment allocation

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind

Blinding (performance bias and detection bias)
outcome assessors
High riskBDI-II self report was used as the primary outcome

Incomplete outcome data (attrition bias)
All outcomes
High risk16/54 dropped out (3/18 in high dose, 7/18 in low dose and 6/18 in control)

Selective reporting (reporting bias)Unclear riskIt appears that all prespecified outcomes are reported, but no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsCommunity volunteers recruited via media
Mean age 28.5 (SD 4.36)
100% women
N = 40. The number randomised into each group not stated.


Interventions1. Supervised running or walking 4 times a week for 8 weeks
2. Supervised strength training 4 times a week
3. Waiting list control


Outcomes1. Beck Depression Inventory
2. Lubin's Depression Adjective List
3. Hamilton Rating Scale for Depression


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskReport states random assignment after matching participants on baseline BDI scores of intervals of < 19, 20 - 29 and > 30

Allocation concealment (selection bias)High riskInadequate (as assessed by Lawlor and Hopker in BMJ review in 2001)

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind to treatment allocation, but unclear risk of bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind

Blinding (performance bias and detection bias)
outcome assessors
High riskBDI was the primary outcome

Incomplete outcome data (attrition bias)
All outcomes
High riskOf the 57 women who met criteria for inclusion, 40 completed treatment, and 32 completed follow-up. Drop-out rates were 40% in 'track', 29% in 'universal' and 13% in waiting list control

Selective reporting (reporting bias)Unclear riskIt appears that all the prespecified outcome measures are reported, but no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsCommunity volunteers recruited via media. Men or women aged 20 to 45 with mild to moderate depression
Mean age 35.9
75% women
N = 80


Interventions4 different aerobic exercise programmes, that varied in total energy expenditure (7.0 kcal/kg/week or 17.5 kcal/kg/week) and frequency (3 days per week or 5 days per week). The 17.5 kcal/kg/week is consistent with public health recommendations for physical activity and is termed 'public health dose'.
1. Public health dose 3/week (n = 17)
2. Public health dose 5/week (n = 16)
3. Low dose 3/week.(n = 16)
4. Low dose 5/week (n = 18)
5. Control, (flexibility exercise) 3 sessions per week (n = 13)

Exercise was on a treadmill or stationary bike, individually and monitored by laboratory staff.
Duration 12 weeks.


OutcomesChange in HRSD from baseline to 12 weeks.


NotesIntention-to-treat (though data from the last available exercise session rather than data collected at 12 weeks were used in the analysis)
Outcome assessors blind


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskMethod of sequence generation not stated

Allocation concealment (selection bias)Low riskOpaque sealed envelopes

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind to treatment allocation, but unclear effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose supervising delivery of the intervention were not blind, but it is unclear what effect this had on bias

Blinding (performance bias and detection bias)
outcome assessors
Low riskTrained research assistants applied the HRSD blind to treatment allocation

Incomplete outcome data (attrition bias)
All outcomes
Low riskIntention-to-treat analyses

Selective reporting (reporting bias)Low riskThere is a published protocol (Dunn 2002)

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsCommunity volunteers recruited via media
Mean age 39.4
(range 24 to 60)
92% women


Interventions1. Group walking or jogging for 30 minutes 3 to 5 times a week for 8 weeks (n = 7)
2. Cognitive therapy 1 session of 1.5 hours per week (n = 9)
3. Waiting list control (n = 10)


Outcomes1. Beck Depression Inventory
2. Zung Self Rating Depression Scale


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskInformation not available

Allocation concealment (selection bias)High riskAssessed by Lawlor and Hopker for BMJ review

Blinding (performance bias and detection bias)
participants
Unclear riskInformation not available

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskInformation not available

Blinding (performance bias and detection bias)
outcome assessors
High riskBDI: self report

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInformation not available

Selective reporting (reporting bias)Unclear riskFrom the information available, it appears that all prespecified outcomes were reported

Other biasUnclear riskUnclear


MethodsRCT. n = 21


ParticipantsDepressed people (reactive depression) referred from a University counselling service and recruited via advertisements


Interventions1. Running 4 sessions over 4 weeks (n = 10). Age range 18 - 51
2. Stroking therapy (a type of 'talking' therapy), 4 sessions over 4 weeks (n = 11) (age range 20 - 35)


OutcomesBeck Depression inventory


NotesOutcome assessment not blind (self report)
Analysis not intention-to-treat (only 16/21 randomised participants completed trial and were included in the analysis)


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskMethod not described

Allocation concealment (selection bias)Unclear riskMethod not described

Blinding (performance bias and detection bias)
participants
Unclear riskNot blind to treatment allocation, but unclear effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskNot blind, but unclear risk on bias

Blinding (performance bias and detection bias)
outcome assessors
High riskBDI; self report

Incomplete outcome data (attrition bias)
All outcomes
High riskOnly included the scores for the 16 people who completed 2 or more sessions

Selective reporting (reporting bias)Unclear riskIt appears from the information available that all prespecified outcomes were reported, but no protocol

Other biasUnclear riskUnclear


MethodsRCT (parallel group)


ParticipantsRecruited from media advertisements, pamphlet and poster displays and psychiatric referrals with major depressive episode.

Age range 18 - 55, mean age and gender data not stated
N = 23


Interventions1. Moderate-intensity aerobic exercise. Each session lasted 30 - 40 minutes (n = 10)

2. Mild-intensity stretching (n = 13)

12-week programme of 3 supervised sessions per week


OutcomesBeck Depression Inventory

Montgomery-Asberg Depression Rating scale


NotesIntention-to-treat analysis

Small sample size with insufficient power to detect small differences


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot described

Allocation concealment (selection bias)Unclear riskUnclear

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants were not blind to treatment allocation, but it is unclear what effect this has had on bias.

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention were not blind, but it is unclear what effect this had on bias

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report BDI

Incomplete outcome data (attrition bias)
All outcomes
High risk10/23 dropped out (2/10 in exercise arm and 8/13 in stretching)

Selective reporting (reporting bias)Unclear riskFrom the study report, it appears that all prespecified outcome measures have been reported; no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsCommunity volunteers recruited via media
Data on age and gender not available
N = 61


Interventions1. Group running (3 times a week, for 10 weeks, with a running coach in small groups of 6 - 8 subjects) (n = 15)
2. Cognitive therapy (10 individual 1 hour sessions with a therapist) (n = 16)
3. Combined running and cognitive therapy (n = 18)

10 weeks


OutcomesBeck Depression Inventory


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo information available

Allocation concealment (selection bias)High riskInadequate (as assessed by Lawlor and Hopker in the 2001 BMJ review)

Blinding (performance bias and detection bias)
participants
Unclear riskNot stated

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskNot stated

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report BDI

Incomplete outcome data (attrition bias)
All outcomes
High riskCategorised as not intention-to-treat for the BMJ review (Lawlor and Hopkins). Data on drop-outs not available to lead author

Selective reporting (reporting bias)Unclear riskProtocol not available for scrutiny. From the study report, it appears that all prespecified outcome measures have been reported

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsDepressed people with heart failure (NYHA Class II to III)

42% men

Age 30 - 70


Interventions1. Home exercise programme: 12 weekly face-to-face home visits to monitor walking and to tailor the exercise prescription. Participants were advised to walk for 3 days per week for 12 weeks, and to increase duration to a maximum of 1 hour for 3 days per week at moderate intensity (n = 20).

2. Home exercise programme plus CBT (n = 18)

3. CBT alone, based on Beck's CBT model. Each session lasted 1 hour. Total number of sessions not stated, but we assume this was 12 because in the combined group they were delivered at the same time as the home exercise programme visits (n = 18)

4. Usual care (n = 17)


OutcomesHAM-D, 6-minute walk


NotesSmall sample size


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskReported as randomised, but no details given

Allocation concealment (selection bias)Unclear riskNo details given

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants were not blinded to treatment allocation; unclear effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering treatment were not blinded; unclear effect on bias

Blinding (performance bias and detection bias)
outcome assessors
Low risk"Data collectors were blind to group assessment"

Incomplete outcome data (attrition bias)
All outcomes
High risk68/74 provided outcome data post-intervention. Classified as high risk, as more than 5% did not provide outcome data

Selective reporting (reporting bias)Low riskAll prespecified outcomes were reported.

Other biasLow riskNo other source of bias identified


MethodsRCT


ParticipantsCommunity volunteers
Age range 18 - 30
53.4% women
N = 28


Interventions1. Supervised running (n = 10). running leader met individually with his participants 3 - 4 times per week for 1 hour, then in the 5th week, only 2 sessions were scheduled with the leader, and in the 7th and 8th weeks, only 1 was scheduled.
2. Time-limited psychotherapy (n = 6)
3. Time-unlimited psychotherapy (n = 12)


OutcomesSymptom checklist score


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo information in report

Allocation concealment (selection bias)High riskCategorised in BMJ review (Lawlor and Hopker) as inadequate

Blinding (performance bias and detection bias)
participants
Unclear riskNot blinded to treatment allocation

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering exercise not blinded, effect on bias unclear

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report symptom checklist score

Incomplete outcome data (attrition bias)
All outcomes
High risk6/28 dropped out (2/10 in the running group and 4/18 in the psychological groups)

Selective reporting (reporting bias)High riskOutcome measures were not prespecified. There was no methods section in the paper; after an introduction, the entry criteria were stated and the interventions were described. The first time the outcome measures were described was in the results section.

Other biasHigh riskThe authors conclude the paper by saying that "our bias (and we purposely label it as bias that requires additional evaluation) is that running may prove to have antidepressant properties for many individuals with moderate depression". It is possible that this author bias was present before the trial was completed, and so may have influenced results.


MethodsRCT


ParticipantsUniversity students aged 18 - 25 with depression

100% women


Interventions1. 40 - 60 minutes of running, 3 times a week, supervised. (n = 10)

2. Control group with no active intervention (n = 10)


OutcomesBeck Depression Inventory score


NotesSmall sample size (10 participants in each arm); specific population under study


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskClinician judgement used at recruitment. After reviewing questionnaires psychiatrists "selected" 20 women

Allocation concealment (selection bias)Unclear riskNo information given

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blinded to intervention; unclear effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskNo information given

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report BDI

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo discussion on attrition rate

Selective reporting (reporting bias)Low riskBDI specified at outset and completed in results

Other biasHigh riskThe control group were told not to do so much exercise.


MethodsRCT


ParticipantsCommunity volunteers recruited via media
Data on age and gender distribution not available
N = 17


Interventions1. Running or walking with the instructor for 30 minutes 4 times a week for 8 weeks (n = 5)
2. Cognitive therapy: 1 session of 1 hour and 2 of ½ hour per week (n = 6)
3. Waiting list control (n = 6)


OutcomesBeck Depression Inventory


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskStated that "random assignment" was performed

Allocation concealment (selection bias)High riskCategorised as inadequate in BMJ review (Lawlor and Hopker)

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blinded to treatment allocation

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report BDI

Incomplete outcome data (attrition bias)
All outcomes
High riskPreviously categorised by Lawlor and Hopker as not intention-to-treat, but data on drop-outs are not reported

Selective reporting (reporting bias)Unclear riskThe BDI was listed as the first outcome measure and data on BDI were reported. The authors also mentioned the Zung Self rating Depression Scale, but data were not reported in the abstract. No protocol was available

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsPeople with a history of traumatic brain injury occurring between 6 months and 5 years prior to trial with at least mild depression (n = 80)


Interventions1. Aerobic exercise of participant's choosing. 60 minutes of gym-based supervised exercise per week and 4 x 30-minute home exercise sessions per week (n = 40; 38% men)

2. No intervention in control group (n = 40; 50% men)


OutcomesBeck Depression Inventory Score


NotesThe control group were informed they could participate in exercise programme post-trial intervention period. Authors report that both intervention and control groups showed a substantial increase in exercise.

SDs received from author


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskParticipation open to anyone who met inclusion criteria. Random distribution of sealed envelopes.

Allocation concealment (selection bias)Low riskUse of sealed envelopes

Blinding (performance bias and detection bias)
participants
Unclear riskAll participants informed of nature of the study and both groups increased total amount of exercise over study period

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskNo detail provided

Blinding (performance bias and detection bias)
outcome assessors
High riskBDI self report

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo evidence of missing data

Selective reporting (reporting bias)Low riskAll findings reported

Other biasUnclear riskUnclear


MethodsRCT (parallel group)


ParticipantsCommunity volunteers recruited via media
Mean age 30.1 (SD 6.72)
72% women
N = 74


Interventions1. Supervised running twice a week for 12 weeks (n = 27)
2. Group cognitive therapy for 2 hours once a week (n = 24)
3. Control group: meditation for 1 hour twice weekly (n = 23)


Outcomes1. Symptom checklist
2. Target symptoms
3. Structural Analysis of Social Behaviour
4. Social Adjustment Self reported Questionnaire
5. Cornell Medical Index
6. Role Rating Questionnaire
7. Hamilton Rating Scale
8. Global Assessment Scale


NotesMain outcome assessment not blind. Hamilton Rating Scale administered by interviewer blind to allocation.
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear risk Method not stated

Allocation concealment (selection bias)High riskInadequate. Categorised by Lawlor and Hopker for BMJ review

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind but unclear whether this had an effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report for main outcome

Incomplete outcome data (attrition bias)
All outcomes
High riskOf the 74 randomised, 32 dropped out or never started treatment (12/27 in the running group; 12/23 in the meditation group and 8/24 in the group therapy group)

Selective reporting (reporting bias)Unclear riskIt appears that all outcomes specified in methods are reported. No protocol

Other biasUnclear riskUnclear


MethodsRCT (parallel group)


ParticipantsInpatients with major depression
Mean age 49
55% women
N = 38


Interventions1. Walking training on a treadmill for 10 days (n = 10)
2. Placebo (low-intensity stretching and relaxation) light stretching exercises for the calves, thighs, back, shoulders and pectoral muscles, as well as relaxation exercises, daily for 30 minutes (n = 18)


Outcomes1. Bech-Rafaelsen Scale (BRMS)
2. Center for Epidemiologic Studies Depression Scale (CES-D)


NotesAuthors state intention-to-treat, but of the 39 recruited only 38 were used in the analysis. Outcome assessor for BRMS blinded to treatment allocation


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-generated block list

Allocation concealment (selection bias)Low riskCentral randomisation

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants aware of treatment allocation, unclear effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention aware of allocation, unclear effect on bias

Blinding (performance bias and detection bias)
outcome assessors
Low riskA single psychologist (blinded to treatment allocation) assessed outcome using the BRMS

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskThe report states 39 were randomised, but the outcome data relate to only 38. Three participants dropped out. Missing data were imputed, but authors have not accounted for the 1 participant who seems to have been randomised but was not reported in the tables of results.

Selective reporting (reporting bias)Unclear riskAll prespecified outcomes seem to have been reported, but no protocol

Other biasUnclear riskUnclear


MethodsRCT (parallel group)


ParticipantsReferred from general practitioners, private psychiatrists, psychologists and psychiatric wards institutions. Included if met criteria for major depression

Mean age 38.9

73.9% women

N = 165


Interventions1. Strength circuit training (n = 55)

2. Aerobic (machine-based) training (n = 55)

3. Relaxation control (n = 55

Twice-weekly intervention for 32 sessions delivered over a 4-month period


OutcomesHamilton Rating Scale for Depression


NotesIntention-to-treat analysis

Significant drop-outs in each group

Changed sample size calculation after first 50 participants on basis of observed standard deviation


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputerised restricted randomisation with a block size of 8

Allocation concealment (selection bias)Low riskThe block size and allocation sequence were unknown to the DEMO trial staff

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, but unclear what influence this had on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskPhysiotherapists delivering the intervention were not blind. Unclear how this influenced risk of bias

Blinding (performance bias and detection bias)
outcome assessors
Low riskThe assessor was blind to intervention group. The investigators asked the outcome assessors to guess intervention group. The kappa values for agreement between the right allocation and the guessed allocation were 0.15 and 0.05 for the assessments at 4 and 12 months respectively

Incomplete outcome data (attrition bias)
All outcomes
Low risk137/165 were available for follow-up at the end of the intervention. Eighteen were lost to follow-up and 10 refused to participate (8/55 in strength group, 7/55 in aerobic group and 13/55 in the relaxation group). The authors used a likelihood-based mixed-effect model with an unstructured variance matrix available in SPSS, which is able to handle missing data with higher precision and power than last observation carried forward. The authors reported no significant difference between missing participants and participants included in the analyses at either 4 or 12 months, and concluded that it was reasonable to assume that the missing data were 'missing at random'

Selective reporting (reporting bias)Low riskAll prespecified outcomes seem to have been reported. Protocol was published in advance of the trial

Other biasUnclear riskThe authors repeated power calculations part-way through the trial, and reduced the sample size as the standard deviation was lower than anticipated


MethodsRCT (parallel group)


ParticipantsPsychiatric hospital inpatients
Mean age 40 (range 17 - 60)
Data on sex distribution not available
N = 49


Interventions1. Aerobic exercise with instructor for 1 hour 3 times a week for 9 weeks at 50 - 70% of maximum aerobic capacity (n = 28 randomised)
2. Control group attended occupational therapy whilst intervention group exercised (n = 21 randomised)


OutcomesBeck Depression Inventory


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskBlock randomisation with respect to age

Allocation concealment (selection bias)Low riskCategorised by Lawlor and Hopker as low risk

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report BDI

Incomplete outcome data (attrition bias)
All outcomes
High risk6 participants dropped out, leaving 43 who completed the study

Selective reporting (reporting bias)Unclear riskPrespecified outcome measures were reported, but no protocol

Other biasUnclear riskunclear


MethodsRCT (parallel group)


ParticipantsPrimary care, psychiatric services, advertisement in paper and radio. N = 86 (59 women and 27 men). Mean age 63.7 (range 53 - 78) in exercise and 66.2 (56 - 91) in control group.


Interventions1. Endurance, muscle strengthening and stretching, in a group exercise class, lasting 45 minutes. An instructress ran the class from a podium in the centre of a hall (n = 43). Twice weekly for 10 weeks

2. Health education classes (n = 43), twice weekly for 10 weeks


Outcomes1. Hamilton Rating Scale for Depression
2. Geriatric Depression Scale
3. Clinical Global Impression
4. Patient Global Impression


NotesOutcome assessor blind
Intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-generated random number list

Allocation concealment (selection bias)Low riskSealed envelopes

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, unclear effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention aware of allocation, unclear effect on bias

Blinding (performance bias and detection bias)
outcome assessors
Low riskPrimary outcome was HRSD delivered by one of 2 psychiatrists who were blinded to treatment allocation

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo drop-outs

Selective reporting (reporting bias)Unclear riskAll prespecified outcome measures reported, but no protocol

Other biasUnclear riskUnclear


MethodsRCT (parallel group)


ParticipantsUndergraduate psychology students with a requirement to participate in a research project
No details of age
100% women
N = 47


Interventions1. Aerobic exercise: group running, jogging or dancing for 1 hour twice weekly for 10 weeks (n = 16 randomised)
2. Placebo control group - muscle relaxation for 15 - 20 minutes 4 times a week (n = 15 randomised)
3. Waiting list control (n = 16 randomised)


OutcomesBeck Depression Inventory


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo information

Allocation concealment (selection bias)High riskClassified as inadequate by Lawlor and Hopker

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention not blind

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report depression scores

Incomplete outcome data (attrition bias)
All outcomes
High risk4/47 withdrew (1 from the aerobic exercise, 1 from the placebo condition and 2 from the 'no treatment' condition). 43 remained

Selective reporting (reporting bias)Unclear riskNo prespecified outcomes. Reported depression scores before and after the intervention. No protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsCommunity volunteers from religious and community organisations
Mean age 72.5
Details of gender distribution not provided
N = 30. number randomised into each group not stated, we assume this is 10 in each group)


Interventions1. Walking accompanied by investigator for 20 minutes 3 times a week for 6 weeks
2. Social contact control group (visit by investigator for a "chat" avoiding any discussion of depression or health, twice a week)
3. Waiting list control group


OutcomesBeck Depression Inventory


NotesOutcome assessment not blind
All completed intervention so classified as intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo information

Allocation concealment (selection bias)High riskInadequate (as assessed by Lawlor and Hopker for BMJ review)

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, but this on its own does not necessary imply bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskInvestigator delivering intervention was not blind

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report BDI

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo drop-outs

Selective reporting (reporting bias)Unclear riskAll prespecified outcome measures were reported, but no protocol

Other biasUnclear riskNote Lawlor and Hopker categorised this study as "not intention-to-treat"


MethodsRCT


Participants18 - 60 year-olds with treatment resistant major depressive disorder selected from outpatient setting (n = 33)


Interventions1. Five sessions a week of 30 - 45 minutes moderate intensity walking. Four of these were unsupervised and one was supervised on a hospital gym treadmill (n = 22, 57.9% women)

2. Control group, receiving no exercise (n = 11 : 80% women)


OutcomesHamilton Depression Scale
Global Assessment of Functioning Scale
Clinical Global Impression Scale.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo details provided

Allocation concealment (selection bias)Unclear riskNo details provided

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants were not blinded to intervention, but were not provided with information on how the intervention might benefit them.

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention not blinded to treatment group

Blinding (performance bias and detection bias)
outcome assessors
Low riskInvestigators carrying out rating tests post-intervention were blinded to treatment group

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing data

Selective reporting (reporting bias)Low riskPrespecified outcomes all reported

Other biasLow riskAppears to be free of other bias


MethodsRCT


ParticipantsDepressed people referred to study by general practitioner (primary care physician)
Mean age 42.1
83% women
N = 36


Interventions1. Aerobic exercise - conducted on an individual basis and without group contact, 29 minutes 3 times a week for 4 weeks (n = 9)
2. Strength and stretching exercise completed on an individual basis and without group contact, 20 minutes 3 times a week (n = 8)
3. Waiting list control (n = 7) i.e. delayed treatment


Outcomes1. Beck Depression Inventory
2. Profile of Mood States


NotesOutcome assessment not blind
All completed intervention so analysis intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo data in report

Allocation concealment (selection bias)High riskInadequate (classified by Lawlor and Hopker in BMJ review)

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind

Blinding (performance bias and detection bias)
outcome assessors
High riskBDI self report

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants completed the intervention

Selective reporting (reporting bias)Unclear riskAll prespecified outcomes are reported, but no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsStudent nurses with mild to moderate depressive symptoms
Aged 18 to 20
All women
N = 59


Interventions1. Group jogging 50 minutes a day 5 days a week for 8 weeks (n = 28)
2. Usual care (n = 31)


Outcomes1. CES-D scores (data from means and SD at end of treatment not available so obtained from published graph)


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo information in the paper; wrote to authors but no response

Allocation concealment (selection bias)Unclear riskNo information in the paper

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, unclear effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention not blind, unclear effect on bias

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report using CES-D score, so not blind

Incomplete outcome data (attrition bias)
All outcomes
High risk59 randomised, 7/28 in jogging group not available at follow-up, 3/31 in control group not available for follow-up; leaving data from 49 to be analysed

Selective reporting (reporting bias)Unclear riskAll prespecified outcome measures seem to have been reported, but no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsCollege students with dysphoria or depression
Mean age 22
27% women
N = 11


Interventions1. Jogging 5 times a week for 30 minutes over 4 weeks (n = 3)
2. Meditation (n = 3)
3. Self-chosen activity (n = 3)
4. Self monitoring (control) (n = 2)


Outcomes1. Depression Adjective Checklist
2. Minnesota Multiphasic Personality Inventory


NotesNot stated whether intention-to-treat though all participants allocated control and running provided data at baseline and post-intervention


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskUnclear, not specified. Too old a study to contact authors

Allocation concealment (selection bias)Unclear riskUnclear

Blinding (performance bias and detection bias)
participants
Unclear riskUnclear

Blinding (performance bias and detection bias)
those delivering intervention
Low riskNot relevant as activities done on own

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf-report outcomes

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants completed the study, data from all reported

Selective reporting (reporting bias)Unclear riskAll prespecified outcomes reported, but no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsRecruited from clinical activity registries of the University psychiatric unit if diagnosed with major depression.

Data on age not stated (inclusion criteria range 40 - 60)

100% women
N = 30


Interventions1. Physiological strengthening exercises plus pharmacological treatment (n=10). The physical activity programme included 2 60-minute lessons per week, held by skilled an instructor, with ISEF (Physical Education) diploma, Psychology degree and post-degree diploma in sport Psychopathology (MS). Each session was set in three steps:
Step I: welcome and warming up (about 5 minutes)
Step II: physiological strengthening (about 50 minutes)
Step III: stretching, cooling down, goodbye (about 5 minutes).

2. Pharmacological treatment only (n = 20)

Intervention delivered twice per week for 8 months


OutcomesHamilton Rating Scale for Depression


NotesNot stated if intention-to-treat analysis

No information given regarding drop-out rates


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Unclear riskUnclear

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, unclear risk of bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention were not blind, unclear risk of bias

Blinding (performance bias and detection bias)
outcome assessors
Unclear riskUnclear. Two different trained psychiatrists rated outcomes, including the HAM-D, but authors did not report whether these psychiatrists were blind to treatment allocation

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll those starting the intervention had outcome data reported

Selective reporting (reporting bias)High riskThe authors stated that a structured diagnostic interview was performed to make a diagnosis of depression at 8 months (follow-up), but these data were not reported

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsSeveral groups including one with depression in the absence of seasonal affective disorder. Also 1 group of depressed people fulfilling criteria for seasonal affective disorder mean age 35.2.
100% women
N = 63


Interventions1. 54 minutes per day of cycling on stationary bicycle for 1 week
2. Bright light therapy

7 groups, 9 participants in each

We are using data from the cycling and bright light therapy


Outcomes1. HDRS score
2. Body weight
3. Oxygen consumption


NotesRandomisation method unclear
Unclear if outcome assessment was blind


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo information given

Allocation concealment (selection bias)Unclear riskUnclear

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThe authors did not state how the intervention was delivered

Blinding (performance bias and detection bias)
outcome assessors
Unclear riskHDRS, but not stated whether outcome assessor blind

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk63 participants were recruited, and were included in 7 groups, 9 per group. Data not provided on number still in the trial at the end of the interventions

Selective reporting (reporting bias)Unclear riskInsufficient information provided to make this judgement

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsUniversity students presenting to mental health clinic with depression
Details of age and gender distribution not provided
N = 18


Interventions1. Supervised running for at least 20 minutes 3 times a week for 10 weeks plus counselling
2. Counselling only


OutcomesBeck Depression Inventory


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo information to make this judgement

Allocation concealment (selection bias)High riskNot used

Blinding (performance bias and detection bias)
participants
Unclear riskNo information

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind, but the effect on bias is uncertain

Blinding (performance bias and detection bias)
outcome assessors
High riskJudged by Lawlor and Hopker in BMJ review as not blind

Incomplete outcome data (attrition bias)
All outcomes
High riskJudged by Lawlor and Hopker as not intention-to-treat

Selective reporting (reporting bias)Unclear riskInsufficient information; no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsHospital inpatients being treated for severe depression with conventional therapy (n = 26)

Gender not specified


Interventions1. 3 sessions per week of participant-selected aerobic exercise (n = 15)

2. Control group receiving conventional therapy (i.e. pharmacotherapy/ECT only) (n = 11)


OutcomesDepressive symptom rating by psychiatrist using HAM-D scoring


NotesIntervention continued 'until discharge' but no further information on length of intervention. No detail on exercise compliance rates. Of 40 originally invited to take part, 14 declined at outset.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskRandomisation stated but details not given

Allocation concealment (selection bias)Unclear riskNo details on how participants were allocated

Blinding (performance bias and detection bias)
participants
Unclear riskNo details given on to what extent participants were aware of the theoretical effects of exercise on depression. Participants not blinded to intervention but unclear of the effect of this on bias.

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering exercise not blinded to group; effect of this on bias is unclear

Blinding (performance bias and detection bias)
outcome assessors
Unclear riskNo detail given on whether those assessing outcome measures were blinded to group

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants remained in trial throughout intervention

Selective reporting (reporting bias)Unclear riskReported on all measures outlined at start of trial

Other biasUnclear riskUnclear


MethodsRCT (parallel group)


ParticipantsCommunity volunteers recruited via the media
Age range 18 - 35 (mean age not stated)
26% women
N = 150


Interventions1. Cognitive therapy and aerobic dance classes (n = 30)
2. Aerobic dance classes only (n = 30)
3. Cognitive therapy only using the principles of A Beck (n = 30)
4. Cognitive therapy and non-aerobic exercise classes (n = 30)
5. Non-aerobic exercise only (arts and crafts) (n = 30)
6. No intervention (n =30)

Duration of interventions was 10 weeks


OutcomesMinnesota Multiphasic Personality Inventory


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskParticipants blocked by gender; male-female ratios kept constant

Allocation concealment (selection bias)Unclear riskUnclear

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, unclear risk on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskGroup intervention, unclear if those delivering the intervention were blind

Blinding (performance bias and detection bias)
outcome assessors
Unclear riskNot stated

Incomplete outcome data (attrition bias)
All outcomes
High risk180 randomised, 150 completed trial

Selective reporting (reporting bias)Unclear riskReported prespecified outcomes; no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsOlder members of a local cultural community

Age 60 - 80

100% women all with geriatric depression score greater than 10

(n = 70)


Interventions1. 10 sessions of aerobic group exercise, each 30 minutes in duration, including jogging and stretching (n = 23)

2 "Laughter yoga" - 10 sessions of structured group activity which includes laughing, clapping, chanting and positive discussion (n = 23)

3. Control (n = 24)


OutcomesDepression scoring on Geriatric Depression Scale, Life Satisfaction scale


NotesOverall drop-out of 10 participants


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNo detail given

Allocation concealment (selection bias)Unclear riskNo detail given

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blinded to intervention; unclear what effect this may have on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskNot blinded; unclear effect on bias

Blinding (performance bias and detection bias)
outcome assessors
High riskGeriatric depression scale; self-reported

Incomplete outcome data (attrition bias)
All outcomes
High risk14% attrition rate

Selective reporting (reporting bias)Low riskAll measures listed at outset reported in results

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsRecruited from hospital databases of stroke patients discharged in last year, general practitioners and newspaper articles. Had to be over 6 months post-stroke and have depression confirmed by a psychiatrist.

Mean age 67.13 (range 21 to 93)
40% women
N = 45


Interventions1. Group-based moderate-intensity strengthening exercises twice a week for 10 weeks.The PRT programme included 2 high-intensity sessions/week for 10 weeks at a community-based gymnasium. (n = 23).

2. Usual care (n = 22)


OutcomesCentre for Epidemiologic Studies for Depression scale


NotesIntention-to-treat analysis

Outcome was self-rated symptoms of depression by CES-D scale


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskBlock randomised list

Allocation concealment (selection bias)Low riskRandomisation was conducted centrally by an independent person

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, unclear risk of bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention were not blind, unclear risk of bias

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf report outcome (depressive symptoms by CES-D scale)

Incomplete outcome data (attrition bias)
All outcomes
High riskBaseline assessment was performed in 45 people; complete data were available for 43 people at 6 months (23/23 in intervention group and 20/22 in the control)

Selective reporting (reporting bias)Unclear riskReported all prespecified outcome (though we do not have access to the protocol)

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsCommunity volunteers from 2 registers of individuals interested in participation in research
Mean age 70 (range 61 - 88)
63% women
N = 32


Interventions1. Supervised non-aerobic progressive resistance training 3 times a week for 10 weeks (n = 17)
2. Control group received health seminars twice a week in which depression and mental health were not discussed (n = 15)


Outcomes1. Beck Depression Inventory
2. Hamilton Rating Scale of Depression


NotesOutcome assessment not blind
Intention-to-treat analysis


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-generated list of blocks of 5

Allocation concealment (selection bias)Low riskAssessed by Lawlor and Hopker as adequate

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention were not blind

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf-rated BDI

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll completed the study

Selective reporting (reporting bias)Unclear riskAll prespecified outcomes reported, though no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsPeople responding to a postal questionnaire who had DSM-IV depression or dysthymia
Mean age 69
55% women
N = 60


Interventions1. Progressive resistance training at 80% of 1 repetition max (n = 20)
2. Resistance training at 20% of 1 repetition max (n = 20)
3. Usual care (n = 20)

Each intervention group held 3 times a week for 8 weeks


Outcomes1. Hamilton Rating Scale for depression
2. Geriatric Depression score


NotesNot intention-to-treat (50/60 completed the study and were available for assessment)
Outcome assessment blind


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-generated random numbers

Allocation concealment (selection bias)Low riskAdequate. Sealed opaque envelopes open after baseline assessment

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, unclear effect on bias

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind to treatment allocation, unclear effect on bias

Blinding (performance bias and detection bias)
outcome assessors
Low riskHRSD performed by blinded outcome assessors

Incomplete outcome data (attrition bias)
All outcomes
High risk6/60 dropped out (2 from the high-dose, 3 from the low-dose and 1 from the usual care group)

Selective reporting (reporting bias)Unclear riskPrespecified outcomes in paper were reported, but no protocol

Other biasUnclear riskUnclear


MethodsRCT


ParticipantsPsychiatric hospital outpatients and hospital day-patients
Mean age 35.5 (range 19 - 58)
64% women
N = 83


Interventions1. Group running 3 times a week for 12 weeks, plus routine care (n = 48)
2. Control group: routine care only (n = 35)


Outcomes1. Beck Depression Inventory
2. State-Trait Anxiety Inventory


NotesOutcome assessment not blind
Analysis not intention-to-treat


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot stated

Allocation concealment (selection bias)Low riskAdequate (categorised by Lawlor and Hopker)

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, unclear risk

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering the intervention were not blind, but the influence on bias is unclear

Blinding (performance bias and detection bias)
outcome assessors
High riskSelf-reported outcomes

Incomplete outcome data (attrition bias)
All outcomes
High riskA total of 18 dropped out, leaving 65 for analyses

Selective reporting (reporting bias)Unclear riskAll the prespecified outcomes stated in the paper were reported, but no protocol

Other biasUnclear riskunclear


MethodsRCT


ParticipantsResidents recruited from 8 long-term care facilities if clinical evidence of Alzheimer's Disease dementia and depression

Mean age 87.9 (range 71 to 101)

89% women

N = 45


Interventions1. Comprehensive exercise - strength, balance, flexibility and walking (n = 16)

2. Supervised walking at pace of individual (n = 17)

3. Control group of casual conversation (n = 12)

Intervention delivered individually 5 days per week for 16 weeks


OutcomesCornell Scale for Depression in Dementia


NotesAnalysis intention-to-treat

Substantial drop-out rate


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskMethod of random sequence generation not stated

Allocation concealment (selection bias)Low riskAdequate. Participants were assigned a code number, which was drawn by a research assistant who had no access to pretest results

Blinding (performance bias and detection bias)
participants
Unclear riskParticipants not blind, though effect on bias unclear

Blinding (performance bias and detection bias)
those delivering intervention
Unclear riskThose delivering intervention were not blind, unclear risk on bias

Blinding (performance bias and detection bias)
outcome assessors
Low riskObservational assessment by raters blind to treatment group allocation

Incomplete outcome data (attrition bias)
All outcomes
High risk9/45 dropped out (1/16 in comprehensive exercise group, 6/17 in walking group, 2/12 in conversation)

Selective reporting (reporting bias)Unclear riskAll prespecified outcome seem to be reported (although we do not have the trial protocol)

Other biasUnclear riskUnclear

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Abascal 2008Did not have to have depression to enter trial

Akandere 2011Did not have to have depression to enter trial

Annesi 2010subgroup analysis from a trial in people with obesity

Arcos-Carmona 2011Did not have to have depression to enter trial

Armstrong 2003Postnatal depression

Armstrong 2004Postnatal depression

Asbury 2009Did not have to have depression to enter trial

Attia 2012Not RCT for exercise in depression

Aylin 2009Did not have to have depression to enter trial

Badger 2007Did not have to have depression to enter trial

Baker 2006Did not have to have depression to enter trial

Bartholomew 2005Single bout of exercise

Beffert 1993Trial involving adolescents

Berke 2007Did not have to have depression to enter trial

Berlin 2003Duration of exercise was only 4 days

Biddle 1989Non-systematic review

Blumenthal 2012bDid not have to have depression to enter trial

Bodin 2004Single bout of either martial arts or stationary bike

Bosch 2009Did not have to have depression to enter trial

Bosscher 1993Comparing different types of exercise with no non-exercising control group

Bowden 2012Not exercise according to ACSM; all arms received an intervention

Boyll 1986College students, did not have to have depression

Brittle 2009Did not have to have depression to enter trial

Bromby 2010Non-randomised study

Broocks 1997Non-systematic review

Brown 1992Trial involving adolescents with diagnoses of dysthymia and conduct disorder

Burbach 1997Non-systematic review

Burton 2009Did not have to have depression to enter trial and multimodal intervention

Carney 1987Participants were those undergoing haemodialysis and did not have to have depression to be included

Chalder 2012Intervention not exercise; multimodal intervention including motivational interviewing, life coaching, support

Chan 2011Intervention is Dejian mind-body intervention, not exercise.

Chen 2009Did not have to have depression to enter trial and intervention was yoga

Chou 2004Exercise intervention was Tai Chi

Chow 2012Exercise intervention was Qigong

Christensen 2012Did not have to have a depression to enter trial

Ciocon 2003Published in abstract form only, intervention appeared not to be exercise according to ACSM definition, and no further information available from the authors

Clegg 2011Did not have to have depression to enter trial

Courneya 2007Did not have to have depression to enter trial

Dalton 1980Trial in a "wheelchair bound population" with diverse aetiologies

Demiralp 2011Did not have to have depression to enter trial

Deslandes 2010Not randomized; participants chose their intervention

DeVaney 1991A trial of reducing exercise in those exercising more than 6 hours per week

DiLorenzo 1999People with depression were excluded

Eby 1985Trial of exercise in students who did not have to have depression to enter trial

Elavsky 2007Did not have to have depression to enter trial

Emery 1990aParticipants did not have to have depression

Emery 1990bParticipants did not have to have depression

Ersek 2008Did not have to have depression to enter trial

Fitzsimmons 2001Not exercise (the participant was placed in wheelchair adapted for connection to the front of a bicycle, the carer pedaled and steered the bicycle)

Fox 2007Did not have to have depression to enter trial

Gary 2007Did not have to have depression to enter trial

Ghroubi 2009Did not have to have depression to enter trial

Gottlieb 2009Did not have to have depression to enter trial

Gusi 2008Did not all have to have depression to enter trial

Gustafsson 2009Single bout of exercise with no non-exercising control group

Gutierrez 2012Did not have to have depression to enter trial

Haffmans 2006Did not have to have depression to enter trial (mixed population of people with affective disorders)

Hannaford 1988General mental health patients with no separation of those with depression

Haugen 2007Did not have to have depression to enter trial

Hedayati 2012Not a trial

Hembree 2000Participants were ageing female population residing in a retirement home environment who did not have diagnosis of depression to enter the trial

Herrera 1994Participants did not have to have depression to enter the trial

Hughes 1986Effect of exercise on mood in people free from psychopathology

Hughes 2009Outline trial involving adolescents

Immink 2011Yoga as intervention

Jacobsen 2012Participants did not have to have depression to enter the trial

Johansson 2011Did not have to have depression to enter trial; intervention Qigong

Karlsson 2007Did not have to have depression to enter trial

Kerr 2008Did not have to have depression to enter trial

Kerse 2010Did not all have diagnosis of depression to enter trial

Kim 2004Effect of exercise on mental distress in healthy participants

Knapen 2003Non-psychotic psychiatric patients with no separation of those with depression

Knapen 2006Did not have to have depression to enter trial (mixed depression and/or anxiety and/or personality disorders)

Kubesh 2003Outcome was executive function; Mood was measured using a subjective mental state scale. People with depression and some controls underwent neuropyschological testing prior to, or after exercise. Participants were randomly allocated to 2 doses of exercise. Thus, people with depression were not randomly allocated to exercise or control

Kulcu 2007Did not have to have depression to enter trial

Kupecz 2001Participants were veterans and did not have diagnosis of depression to enter the trial

Labbe 1988Comparison of exercise with exercise and instructions about how to improve compliance to exercise

Lacombe 1988Three types of exercise, no non-exercising control

Lai 2006Trial in stroke patients. Did not have to have depression to be eligible

Latimer 2004Did not have to have depression to enter trial

Lautenschlager 2008Did not have to have depression to enter trial

Lavretsky 2011Control is health education, an active intervention

Legrand 2009Comparing 2 exercise regimens (of walking intensity) with no non-exercising control group

Leibold 2010Qualitative analysis

Leppämäki 2002Effects of exercise on symptoms of mental distress in healthy participants

Levendoglu 2004Did not have to have depression to enter trial

Lever-van Milligen 2012Did not have to have depression to be included in trial

Levinger 2011Did not have to have depression to enter trial

Lin 2007Did not have to have depression to enter trial

Littbrand 2011Did not have to have depression; control group undertake activity as intervention

Lolak 2008Did not have to have depression to enter the trial

Machado 2007Did not have to have depression to enter trial

MacMahon 1988Trial involving adolescents

Mailey 2010Participants were having mental health counselling, but there is no statement that they had to have depression to enter the study

Martin 2009Did not have to have depression to enter trial

Martinsen 1988aNon-systematic review

Martinsen 1988bRCT (block randomisation with respect to sex) but compared different
types of exercise without including a non-exercising control group.

Martinsen 1989cNon-systematic review

Martinsen 1993Non-systematic review

Matthews 2011Educational intervention and stretching exercises in control; did not have to be depressed to enter trial

McClure 2008intervention included a combination of interventions including a pedometer, Step Up program workbook,
and a series of counselling calls from a study counsellor

Midtgaard 2011Did not have to have depression to enter trial

Milani 2007Retrospective evaluation of patients with depressive symptoms who participated in cardiac rehabilitation programme post major cardiac event

Morey 2003Older sedentary adults who did not have a diagnosis of depression to enter the trial

Motl 2004Older adults who did not have to be depressed to be included in the trial

Mudge 2008Did not have to have depression to enter trial

Munro 1997Cost-effectiveness analysis of the likely public health benefits of purchasing exercise for over 65s

Mutrie 2007Did not have to have depression to enter trial

NCT00416221Study not randomised

NCT00546221Comparing 2 different exercise interventions with no non-exercising control arm

NCT00964054Trial in adolescents

NCT01152086Comparing 2 exercise regimens (of hiking programme) with no non-exercising control group, in participants with chronic suicidality, not depression

Neidig 1998Participants had HIV infection and did not have diagnosis of depression to enter trial

Netz 1994General mental health patients with no separation of those with depression

Neuberger 2007Did not have to have depression to enter trial

Nguyen 2001Trial in people with chronic obstructive pulmonary disease who did not have to have depression to enter trial

O'Neil 2011Multimodal intervention - telephone based lifestyle advice as well as exercise

Oeland 2010Combination of people with depression and/or anxiety disorders

Oretzky 2006Exercise intervention was yoga

Ouzouni 2009Did not have to have depression to enter trial

Pakkala 2008Did not have to have depression to enter trial

Palmer 2005Participants were recovering from substance abuse

Passmore 2006Aerobic exercise versus aerobic and resistance exercise; no non-exercising control

Peacock 2006The methodology fulfilled criteria but the study was not completed due to staff sickness

Pelham 1993General mental health patients with no separation of those with depression

Penninx 2002Retrospective subgroup analysis of patients who participated in a randomised trial of exercise for knee osteoarthritis who also had depression

Penttinen 2011A study in survivors of breast cancer, not depression

Perna 2010A study in breast cancer, not depression

Perri 1984No outcome measure of depression. This must have been excluded by Debbie Lawlor.

Piette 2011Does not fulfill ACSM criteria for exercise; intervention largely psychological

Raglin 1990Non-systematic review

Rhodes 1980Not randomised, participants not depressed

Robledo Colonia 2012Did not have to have depression to enter trial

Rofey 2008Trial involving adolescents

Roshan 2011Trial in adolescents

Roth 1987No outcome measure of depression. This must have been excluded by Debbie Lawlor

Ruunsunen 2012Did not have to have depression to enter trial; intervention multimodal

Salminen 2005Coronary heart disease patients with no separation of those with depressive symptoms
Intervention described by authors as health advocacy, counselling and activation programme

Salmon 2001Non-systematic review

Sarsan 2006Did not have to have depression to enter trial

Schwarz 2012Unable to get access to full text of study; attempts made to contact authors were unsuccessful

Sexton 1989Comparing different types of exercise with no non-exercising group

Silveira 2010Unable to assess if randomized; attempts made to contact authors unsuccessful, assumed to be non-random

Sims 2006Did not all have diagnosis of depression to enter trial

Skrinar 2005DSM-IV or psychotic disorders; no separation of those with depression

Smith 2008Did not have to have depression to enter trial

Sneider 2008intervention was combined diet and exercise

Songoygard 2012Did not have to have depression to be included in trial

Stein 1992Not described as randomised. Did not have to be depressed to participate

Stern 1983Trial in patients with myocardial infarction who did not have to be depressed to enter trial

Strömbeck 2007Did not have to have depression to enter trial

Sung 2009Did not have to have depression to enter trial

Tapps 2009Did not have to have depression to enter trial

Taylor 1986Trial in patients with myocardial infarction (no diagnosis of depression to enter trial)

Tenorio 1986Trial in subclinical depression

Thomson 2010Did not have to have depression to enter the trial

Tomas-Carus 2008Did not have to have depression to enter trial

TREAD 2003Ongoing trial comparing 2 intensities of exercise

Trivedi 2011No true control group; both arms of trial receive exercise

Tsang 2003Participants had chronic physical disease not depression

Tsang 2006Exercise intervention was Qigong

Underwood 2013Did not have to have depression to enter trial

Van de Vliet 2003Single study design

Van der Merwe 2004Intervention was a manual-based therapy programme not exercise

Vickers 2009Intervention was not exercise (exercise counselling)

Weinstein 2007Did not have to have depression to enter trial

Weiss 1989Not randomised controlled trial

White 2007Did not have to have depression to enter trial

Whitham 2011Trial in bipolar affective disorder

Wieman 1980Jogging versus racket ball so no non-exercising control

Wilbur 2009Did not have to have depression to enter trial

Williams 1992Aerobic versus low-intensity exercise. No control group

Wipfli 2008Did not have to have depression to enter trial

Wipfli 2011Participants did not have to be depressed to enter trial; control arm had intervention of yoga and stretching

 
Characteristics of studies awaiting assessment [ordered by study ID]

MethodsRandomised trial

ParticipantsPeople with MI in selected hospitals in Iran

InterventionsEducation programme, including 'exercises'

OutcomesHospital anxiety and depression score

NotesUnlikely to fulfil inclusion criteria, as participants did not have to have depression and the intervention was multimodal


MethodsRandomised Controlled Trial

ParticipantsOutpatients with major depressive disorder (DSM-IV)

InterventionsSupervised stretching or supervised aerobic fitness programme

OutcomesHamilton depression rating Scale

NotesUnlikely to fulfil inclusion criteria, as the control arm also received exercise


MethodsClinical trial

ParticipantsPeople > 65 years with a recent decline in memory or thinking

InterventionsAerobic or stretching exercise group, 60 minutes 3 times a week for 12 weeks

OutcomesDepression included as an outcome measure

Notesunlikely to be eligible for inclusion as it appears that people did not have to have depression to enter the trial


MethodsRandomised trial, 75 adults

ParticipantsMajor depression

Interventions9-week 'chronotherapeutic intervention' or 9 weeks of daily exercise. Both groups received duloxetine

Outcomes17-item Hamilton depression rating scale

NotesLikely to be eligible for inclusion

N=75

Study unlikely to contribute to current review comparisons. Further scrutiny required.


MethodsRandomised clinical trial

ParticipantsPeople with 1 of several conditions (e.g. mental health problems, coronary heart disease)

InterventionsWelsh exercise referral scheme or usual care, Euroquol 5D, Hospital anxiety and depression score, Client Service Receipt Inventory questionnaire,

OutcomesTotal minutes of physical activity per week

NotesUnlikely to be eligible for inclusion as the 'mental health' subgroup included people with depression, anxiety and stress, and because we have previously excluded trials that reported subgroups with depression


MethodsRandomised trial; three groups. (n = 97)

Participantspeople with "self-declared" depression

Interventions6-week programme of Argentine tango dance, mindfulness meditation, waiting list control

OutcomesDepression, anxiety and stress scale, self esteem scale, satisfaction with life scale, and mindful attention awareness scale

NotesLikely to be eligible for inclusion

N=66

Study could contribute data to two comparisons, although three-arm trial with small sample size. Results positive for intervention involving exercise. Unlikely to affect review conclusions.


MethodsRandomised trial (n = 20)

ParticipantsPeople with previous suicidal attempts and clinically diagnosed with "hopelessness"

Interventions9-week hiking, 9-week control

OutcomesHopelessness, depression, physical endurance, suicidal ideation

NotesLikely to be eligible for inclusion

N=20

Study could contribute data for one comparison, although very small sample size. Results positive for intervention involving exercise. Unlikely to affect review conclusions.

 
Characteristics of ongoing studies [ordered by study ID]

Trial name or titleDoes a home-based physical activity programme improve function and depressive symptomatology in older primary care patients: a randomised controlled trial

MethodsRandomised controlled trial

ParticipantsThose aged 75 or older with depression

InterventionsHome-based physical activity programme

OutcomesChange in geriatric depression score

Starting date2006

Contact informationKaren Hayman

k.hayman@auckland.ac.nz

Notes


Trial name or titlePromoting physical activity to improve the outcome of depression in later life (ACTIVEDEP)

MethodsRandomised controlled trial - parallel

ParticipantsAge 50 or over with DSM-IV diagnosis of depression

InterventionsMixed aerobic and strength training programme

OutcomesMontgomery-Asperg Depression Rating Scale; remission of symptoms

Starting date2009

Contact informationOsvaldo Ameida

osvaldo.almeida@uwa.edu.au

Notes


Trial name or titleA randomised controlled trial to improve depression in family carers through a physical activity intervention

IMPACCT Study

MethodsRandomised controlled trial

ParticipantsDepressed carers over 60 and their care recipients

Interventions6-month physical activity programme

OutcomesRating on Geriatric Depression Scale

Starting date2012

Contact informationMs Kirsten Moore

k.moore@nari.unimelb.edu.au

NotesCompleted, unreported study


Trial name or titleEffect of sprint interval training on depression: a randomised controlled trial

MethodsRandomisation, blinding

ParticipantsAge 20 35, male, diagnosis of depression

InterventionsSprint training exercise or aerobic exercise

OutcomesDepression scale

Starting dateSeptember 2012

Contact informationDr Khaled Badaam

khalid_badaam@yahoo.com

NotesCompleted, unreported study


Trial name or titleEffect of running therapy on depression (EFFORT-D). Design of a randomised controlled trial in adult patients [ISCRCTN 1894]

MethodsRandomised controlled trial

ParticipantsPeople with a depressive disorder

InterventionsGroup-based 1-hour exercise sessions of increased intensity over 6 months

OutcomesReduction in depressive symptoms as measured by HAM-D

Starting date2012

Contact informationFrank Kruisdijk

f.kruisdijk@ggzcentraal.nl

Notes


Trial name or titleThe effect of regular exercise on the depression of haemodialysis patients

Methods

ParticipantsAge 15 - 65, chronic kidney disease receiving haemodialysis

InterventionsExercise

OutcomesBeck Depression Inventory Score

Starting dateJune 2012

Contact informationAlireza Abdi

alireza_abdi61@yahoo.com

NotesDepression not specified in inclusion criteria

Completed, unreported study


Trial name or titleA comparative study of the efficiency of group cognitive-behavioural therapy with aerobic exercise in treating major depression

MethodsRandomised

Participants18 - 25 with depressive symptoms

InterventionsEither exercise or CBT or no intervention as control

OutcomesBeck Depression Inventory score

Starting dateJuly 2012

Contact informationKheirollah Sadeghi

khsadeghi@kums.ac.ir

NotesCompleted, unreported study


Trial name or titlePsycho-education physical exercise effects: does treating subsyndromal depression improve depression- and diabetes-related outcomes? PEPEE

MethodsRandomised controlled three-arm study

Participants18 - 60 years with depression and Type II diabetes

InterventionsPsycho-education or exercise intervention or control

OutcomesDepressive symptoms

Starting date2010

Contact informationMirjana Pibernik-Okanovic

mirjana.pibernik@idb.hr

Notes


Trial name or titleRandomized clinical trial Investigating the effect of different exercise forms on depression

MethodsRandomised interventional model

Participants18 - 55 with depression

InterventionsStrength endurance training

OutcomesHamilton Depression Scoring Scale

Starting date2006

Contact informationMerete Nordentoft

Bispebjerg Hospital

Notes


Trial name or titleEfficacy of an exercise intervention to decrease depressive symptoms in veterans with hepatitis C

MethodsRandomised interventional model

ParticipantsAdults positive for hepatitis C with depression

InterventionsHome-based walking programme

OutcomesReduction in Beck Depression Scale

Starting date2008

Contact informationPatricia Taylor-Young

Portland VA Medical Centre

Notes


Trial name or titleEffects of exercise on depression symptoms, physical function, and quality of life in community-dwelling elderly

MethodsRandomised interventional model

ParticipantsCommunity-dwelling participants age 65 or older

InterventionsGroup exercise 3 times per week

OutcomesTaiwanese Geriatric Depression Scale

Starting date2009

Contact informationYing-Tai Wu

National Taiwan University

Notes


Trial name or titleExercise study to help patients who have type 2 diabetes and depression

MethodsRandomised interventional model

ParticipantsEomen 21 - 65, with depression and diabetes

InterventionsExercise group

OutcomesDepression symptoms

Starting date2009

Contact informationKristin Schneider

University of Massachusetts, Worcester

Notes


Trial name or titleClinical and neuroendocrine/metabolic benefits of exercise in treatment resistant depression (TRD): a feasibility study

MethodsRandomised controlled trial

ParticipantsSedentary adults with depression

InterventionsModerate intensity aerobic exercise

OutcomesChange from baseline depression score on Hamilton Scale

Starting date2011

Contact informationRavi Singareddy

Penn State University College of Medicine

Notes


Trial name or titleEfficacy of exercise and counselling Intervention on relapse in smokers with depressive disorders STOB-ACTIV

MethodsRandomised, intervention model

Participants18-65 with depression

InterventionsExercise

OutcomesEvaluation of depression

Starting date2010

Contact informationXavier Quantin

x-quantin@chu-montpellier.fr

NotesMay be multimodal with counselling for smoking cessation


Trial name or titleThe impact of psychological interventions (with and without exercise) on psychometric and immunological measures in patients with major depression

MethodsRandomised controlled trial

Participants18 - 65 year-olds with depression diagnosed according to DSM-IV

InterventionsExercise and psychological input

OutcomesChange in depression symptomology

Starting date2011

Contact informationFrank Euteneuer

frank.euteneuer@staff.uni-marburg.de

NotesIntervention may be multimodal


Trial name or titleAn internet-administered, therapist-supported physical exercise program for the treatment of depression

MethodsRandomised intervention model

ParticipantsAdults with DSM-IV diagnosis of depression

InterventionsPhysical exercise programme

OutcomesChange in MADRS rating scale for depression

Starting dateMarch 2012

Contact informationProf Per Carlbing

per.carlbring@psy.umu.se

Notes


Trial name or titleRole of exercise in depression in middle aged and older adults

MethodsRandomised, double-blind

ParticipantsAge 46 or older; diagnosis of depression

InterventionsLow-dose exercise

OutcomesPHQ 9 depression score

Starting dateMay 2012

Contact informationDaniel O Clark

Indiana University School of Medicine

Notes


Trial name or titleThe effects of behavioral activation and physical exercise on depression

MethodsRandomised, Intervention Model

ParticipantsParticipants with depression

InterventionsBehavioural activation, motivational interviewing, physical activity

OutcomesChange from baseline in Patient Health Questionnaire

Starting dateAugust 2012

Contact informationProfessor Per Carlbring

per.carlbring@psy.umu.se

Notes


Trial name or titleEffect of a supervised exercise program during whole pregnancy on outcomes and level of depression. A randomized controlled trial

MethodsRandomised, intervention

ParticipantsPregnant women

InterventionsExercise group

OutcomesChange in level of depression

Starting date2009

Contact informationMaria Perales

m.perales.santaella@gmail.com

Notes


Trial name or titleEffects of Cognitive Behavioural Therapy and exercise on depression and cognitive deficits in Multiple Sclerosis

MethodsRandomised, single-blind, intervention model

Participants18 - 50 years-old with multiple sclerosis and depression

InterventionsExercise, CBT

OutcomesChange in Hamilton Depression rating, cognitive scoring

Starting dateJanuary 2013

Contact informationBethany Lerman

bethany.lerman@sunnybrook.ca

Notes


Trial name or titleThe effects of exercise in depression symptoms using levels of neurotransmitters and EEG as markers

MethodsRandomised, intervention model

Participants18 - 65 years old with depression

InterventionsExercise

OutcomesBeck Depression Inventory score, serum levels serotonin and catecholamines

Starting dateMarch 2013

Contact informationDr Timothy Barclay

thbarclay@liberty.edu

Notes


Trial name or titleExercise training in depressed traumatic brain injury survivors

MethodsRandomised, single-blind intervention model

ParticipantsDepressed, sedentary survivors of traumatic brain injury

InterventionsAerobic exercise

OutcomesMood assessment, MRI, neuropsychology testing, biochemical assays, suicide severity rating

Starting dateFebruary 2013

Contact informationJustin Alicea

joalicea@vcu.edu

Notes


Trial name or titleA randomised controlled trial of exercise class for older persons with mild depression

MethodsParallel randomised trial

Participants60 - 86 year-olds with mild depression

InterventionsLow-intensity exercise programme

OutcomesHamilton Depression Rating Scale

Starting date2009

Contact informationKazushige Ihara

ihara1@med.toho-u.ac.jp

Notes

 
Comparison 1. Exercise versus 'control'

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Reduction in depression symptoms post-treatment351353Std. Mean Difference (IV, Random, 95% CI)-0.62 [-0.81, -0.42]

 2 Reduction in depression symptoms follow-up8377Std. Mean Difference (IV, Random, 95% CI)-0.33 [-0.63, -0.03]

 3 Completed intervention or control291363Risk Ratio (M-H, Random, 95% CI)1.00 [0.97, 1.04]

 4 Quality of life4Std. Mean Difference (IV, Fixed, 95% CI)Subtotals only

    4.1 Mental
259Std. Mean Difference (IV, Fixed, 95% CI)-0.24 [-0.76, 0.29]

    4.2 Psychological
256Std. Mean Difference (IV, Fixed, 95% CI)0.28 [-0.29, 0.86]

    4.3 Social
256Std. Mean Difference (IV, Fixed, 95% CI)0.19 [-0.35, 0.74]

    4.4 Environment
256Std. Mean Difference (IV, Fixed, 95% CI)0.62 [0.06, 1.18]

    4.5 Physical
4115Std. Mean Difference (IV, Fixed, 95% CI)0.45 [0.06, 0.83]

 
Comparison 2. Exercise versus psychological therapies

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Reduction in depression symptoms post-treatment7189Std. Mean Difference (IV, Random, 95% CI)-0.03 [-0.32, 0.26]

 2 Completed exercise or pyschological therapies4172Risk Ratio (M-H, Random, 95% CI)1.08 [0.95, 1.24]

 3 Quality of life1Mean Difference (IV, Fixed, 95% CI)Totals not selected

    3.1 Physical
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

    3.2 Mental
1Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]

 
Comparison 3. Exercise versus bright light therapy

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Reduction in depression symptoms post-treatment118Mean Difference (IV, Fixed, 95% CI)-6.4 [-10.20, -2.60]

 
Comparison 4. Exercise versus pharmacological treatments

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Reduction in depression symptoms post-treatment4300Std. Mean Difference (IV, Random, 95% CI)-0.11 [-0.34, 0.12]

 2 Completed exercise or antidepressants3278Risk Ratio (M-H, Random, 95% CI)0.98 [0.86, 1.12]

 3 Quality of Life1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    3.1 Mental
125Mean Difference (IV, Fixed, 95% CI)-11.90 [-24.04, 0.24]

    3.2 Physical
125Mean Difference (IV, Fixed, 95% CI)1.30 [-0.67, 3.27]

 
Comparison 5. Reduction in depression symptoms post-treatment: Subgroup analyses

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Exercise vs control subgroup analysis: type of exercise35Std. Mean Difference (IV, Random, 95% CI)Subtotals only

    1.1 Aerobic exercise
281080Std. Mean Difference (IV, Random, 95% CI)-0.55 [-0.77, -0.34]

    1.2 Mixed exercise
3128Std. Mean Difference (IV, Random, 95% CI)-0.85 [-1.85, 0.15]

    1.3 Resistance exercise
4144Std. Mean Difference (IV, Random, 95% CI)-1.03 [-1.52, -0.53]

 2 Exercise vs control subroup analysis: intensity35Std. Mean Difference (IV, Random, 95% CI)Subtotals only

    2.1 light/moderate
376Std. Mean Difference (IV, Random, 95% CI)-0.83 [-1.32, -0.34]

    2.2 moderate
12343Std. Mean Difference (IV, Random, 95% CI)-0.64 [-1.01, -0.28]

    2.3 hard
11595Std. Mean Difference (IV, Random, 95% CI)-0.56 [-0.93, -0.20]

    2.4 vigorous
5230Std. Mean Difference (IV, Random, 95% CI)-0.77 [-1.30, -0.24]

    2.5 Moderate/hard
266Std. Mean Difference (IV, Random, 95% CI)-0.63 [-1.13, -0.13]

    2.6 Moderate/vigorous
242Std. Mean Difference (IV, Random, 95% CI)-0.38 [-1.61, 0.85]

 3 Exercise vs control subroup analysis: number of sessions35Std. Mean Difference (IV, Random, 95% CI)Subtotals only

    3.1 0 - 12 sessions
5195Std. Mean Difference (IV, Random, 95% CI)-0.42 [-1.26, 0.43]

    3.2 13 - 24 sessions
9296Std. Mean Difference (IV, Random, 95% CI)-0.70 [-1.09, -0.31]

    3.3 25 - 36 sessions
8264Std. Mean Difference (IV, Random, 95% CI)-0.80 [-1.30, -0.29]

    3.4 37+ sessions
10524Std. Mean Difference (IV, Random, 95% CI)-0.46 [-0.69, -0.23]

    3.5 unclear
373Std. Mean Difference (IV, Random, 95% CI)-0.89 [-1.39, -0.40]

 4 Exercise vs control subroup analysis: diagnosis of depression35Std. Mean Difference (IV, Random, 95% CI)Subtotals only

    4.1 clinical diagnosis of depression
23967Std. Mean Difference (IV, Random, 95% CI)-0.57 [-0.81, -0.32]

    4.2 depression categorised according to cut points on a scale
11367Std. Mean Difference (IV, Random, 95% CI)-0.67 [-0.95, -0.39]

    4.3 unclear
118Std. Mean Difference (IV, Random, 95% CI)-2.00 [-3.19, -0.82]

 5 Exercise vs control subgroup analysis: type of control351353Mean Difference (IV, Fixed, 95% CI)-1.57 [-1.97, -1.16]

    5.1 placebo
2156Mean Difference (IV, Fixed, 95% CI)-2.66 [-4.58, -0.75]

    5.2 No treatment, waiting list, usual care, self monitoring
17563Mean Difference (IV, Fixed, 95% CI)-4.75 [-5.72, -3.78]

    5.3 exercise plus treatment vs treatment
6225Mean Difference (IV, Fixed, 95% CI)-1.22 [-2.21, -0.23]

    5.4 stretching, meditation or relaxation
6219Mean Difference (IV, Fixed, 95% CI)-0.09 [-0.65, 0.48]

    5.5 occupational intervention, health education, casual conversation
4190Mean Difference (IV, Fixed, 95% CI)-3.67 [-4.94, -2.41]

 
Comparison 6. Exercise versus control: sensitivity analyses

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Reduction in depression symptoms post-treatment: peer-reviewed journal publications and doctoral theses only341335Std. Mean Difference (IV, Random, 95% CI)-0.59 [-0.78, -0.40]

 2 Reduction in depression symptoms post-treatment: studies published as abstracts or conference proceedings only118Std. Mean Difference (IV, Random, 95% CI)-2.00 [-3.19, -0.82]

 3 Reduction in depression symptoms post-treatment: studies with adequate allocation concealment14829Std. Mean Difference (IV, Random, 95% CI)-0.49 [-0.75, -0.24]

 4 Reduction in depression symptoms post-treatment: studies using intention-to-treat analysis11567Std. Mean Difference (IV, Random, 95% CI)-0.61 [1.00, -0.22]

 5 Reduction in depression symptoms post-treatment: studies with blinded outcome assessment12658Std. Mean Difference (IV, Random, 95% CI)-0.36 [-0.60, -0.12]

 6 Reduction in depression symptoms post-treatment: allocation concealment, intention-to-treat, blinded outcome6464Std. Mean Difference (IV, Random, 95% CI)-0.18 [-0.47, 0.11]

 7 Reduction in depression symptoms post-treatment: Lowest dose of exercise351347Std. Mean Difference (IV, Fixed, 95% CI)-0.44 [-0.55, -0.33]

 
Summary of findings for the main comparison. Exercise compared to control for adults with depression

Exercise compared to no intervention or placebo for adults with depression

Patient or population: adults with depression
Settings: any setting
Intervention: Exercise
Comparison: no intervention or placebo

OutcomesIllustrative comparative risks* (95% CI)No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

No intervention or placeboExercise

Symptoms of depression
Different scales
Follow-up: post-treatment
The mean symptoms of depression in the control groups was
0
The mean symptoms of depression in the intervention groups was
0.62 standard deviations lower
(0.81 to 0.42 lower)1
1353
(35 studies)
⊕⊕⊕⊝
moderate2,3,4
SMD -0.62 (95% CI: -0.81 to -0.42).

The effect size was interpreted as 'moderate' (using Cohen's rule of thumb)

Symptoms of depression (long-term)
different scales
The mean symptoms of depression (long-term) in the control groups was
0
The mean symptoms of depression (long-term) in the intervention groups was
0.33 standard deviations lower
(0.63 to 0.03 lower)
377
(8 studies)
⊕⊕⊝⊝
low4,5
SMD -0.33 (95% CI: -0.63 to -0.03).

The effect size was interpreted as 'small' (using Cohen's rule of thumb)

Adverse eventsSee commentSee comment0
(6 studies)
⊕⊕⊕⊝
moderate
Seven trials reported no difference in adverse events between exercise and usual care groups. Dunn 2005 reported increased severity of depressive symptoms (n = 1), chest pain (n = 1) and joint pain/swelling (n = 1); all these participants discontinued exercise. Singh 1997 reported that 1 exerciser was referred to her psychologist at 6 weeks due to increasing suicidality; and musculoskeletal symptoms in 2 participants required adjustment of training regime. Singh 2005

reported adverse events in detail (visits to a health professional, minor illness, muscular pain, chest pain, injuries requiring training adjustment, falls, deaths and hospital days) and found no difference between the groups. Knubben 2007 reported "no negative effects of exercise (muscle pain, tightness or fatigue)"; after the training had finished, 1 person in the placebo group required gastric lavage and 1 person in the exercise group inflicted a superficial cut on her arm. Sims 2009

reported no adverse events or falls in either the exercise or control group. Blumenthal 2007 reported more side effects in the sertraline group (see comparison below) but there was no difference between the exercise and control group. Blumenthal 2012a reported more fatigue and sexual dysfunction in the sertraline group than the exercise group.

Acceptability of treatmentStudy population1363
(29 studies)
⊕⊕⊕⊝
moderate2
RR 1
(95% CI: 0.97 to 1.04)

865 per 1000865 per 1000
(839 to 900)

Quality of lifeThe mean quality of life in the intervention groups was
0 higher
(0 to 0 higher)
0
(4 studies)
See commentThere was no statistically significant differences for the mental (SMD -0.24; 95% CI -0.76 to 0.29). psychological (SMD 0.28; 95% CI -0.29 to 0.86) and social domains (SMD 0.19; 95% CI -0.35 to 0.74). Two studies reported a statistically significant difference for the environment domain favouring exercise (SMD 0.62; 95% CI 0.06 to 1.18) and 4 studies reported a statistically significant difference for the physical domain favouring exercise (SMD 0.45; 95% CI 0.06 to 0.83).

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 Effect estimate calculated by re-expressing the SMD on the Hamilton Depression Rating Scale using the control group SD (7) from Blumenthal 2007 (study chosen for being most representative). The SD was multiplied by the pooled SMD to provide the effect estimate on the HDRS.
2 Lack of blinding of outcome assessors probably increased effect sizes and drop-out rates were high. Also sequence generation was considered unclear in 23 studies.
3 I² = 63% and P < 0.00001, indicated moderate levels of heterogeneity
4 Population size is large, effect size is above 0.2 SD, and the 95% CI does not cross the line of no effect.
5 Lack of blinding of outcome assessors probably increased effect sizes and drop-out rates were high. Also sequence generation was considered unclear in 4 studies.
 
Summary of findings 2. Exercise compared to psychological treatments for adults with depression

Exercise compared to cognitive therapy for adults with depression

Patient or population: adults with depression
Settings:
Intervention: Exercise
Comparison: cognitive therapy

OutcomesIllustrative comparative risks* (95% CI)No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

Cognitive therapyExercise

Symptoms of depressionThe mean symptoms of depression in the intervention groups was
0.03 standard deviations lower
(0.32 lower to 0.26 higher)
189
(7 studies)
⊕⊕⊕⊝
moderate1,2,3
SMD -0.03 (95% CI: -0.32 to 0.26)

Acceptability of treatmentStudy population172
(4 studies)
⊕⊕⊕⊝
moderate1
RR 1.08
(95% CI: 0.95 to 1.24)

766 per 1000827 per 1000
(728 to 950)

Quality of LifeThe mean quality of life in the intervention groups was
0 higher
(0 to 0 higher)
0
(1 study)
⊕⊕⊕⊝
moderate1
One trial reported changes in the Minnesota Living with Heart Failure Questionnaire, a quality of life measure (Gary 2010). There was no statistically significant difference for the physical domain (MD 0.15; 95% CI: -7.40 to 7.70) or the mental domain (MD -0.09; 95% CI: -9.51 to 9.33).

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 Lack of blinding of outcome assessors probably increased effect sizes and drop-out rates were high. Also sequence generation was considered unclear in 7 studies.
2 I² = 0% and P = 0.62, indicated no heterogeneity
3 The studies included were all relevant to the review question, particularly given that all studies had to meet the criteria of the ACSM definition of exercise.
 
Summary of findings 3. Exercise compared to bright light therapy for adults with depression

Exercise compared to bright light therapy for adults with depression

Patient or population: adults with depression
Settings:
Intervention: Exercise
Comparison: bright light therapy

OutcomesIllustrative comparative risks* (95% CI)No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

Bright light therapyExercise

Symptoms of depressionThe mean symptoms of depression in the intervention groups was
6.4 lower
(10.2 to 2.6 lower)
18
(1 study)
⊕⊝⊝⊝
very low1,2,3
MD -6.40 (95% CI: -10.20 to -2.60).

Although this trial suggests a benefit of exercise, it is too small to draw firm conclusions

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 Lack of blinding of outcome assessors probably increased effect sizes and drop-out rates were not reported. Also sequence generation and concealment was considered unclear.
2 The study included was relevant to the review question, particularly given that all studies had to meet the criteria of the ACSM definition of exercise.
3 Based on 18 people
 
Summary of findings 4. Exercise compared to pharmacological treatments for adults with depression

Exercise compared to antidepressants for adults with depression

Patient or population: adults with depression
Settings:
Intervention: Exercise
Comparison: antidepressants

OutcomesIllustrative comparative risks* (95% CI)No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

AntidepressantsExercise

Symptoms of depressionThe mean symptoms of depression in the intervention groups was
0.11 standard deviations lower
(0.34 lower to 0.12 higher)
300
(4 studies)
⊕⊕⊕⊝
moderate1,2,3
SMD -0.11 (95% CI: -0.34 to 0.12)

Acceptability of treatmentStudy population278
(3 studies)
⊕⊕⊕⊝
moderate1
RR 0.98
(95% CI: 0.86 to 1.12)

891 per 1000873 per 1000
(766 to 997)

Quality of lifeThe mean quality of life in the intervention groups was
0 higher
(0 to 0 higher)
0
(1 study)
⊕⊕⊕⊝
moderate1
One trial, Brenes 2007, reported no difference in change in SF-36 mental health and physical health components between medication and exercise groups.

Adverse eventsSee commentSee comment0
(3 studies)
⊕⊕⊕⊝
moderate1
Blumenthal 1999 reported that 3/53 in exercise group suffered musculoskeletal injuries; injuries in the medication group were not reported.

Blumenthal 2007 collected data on side effects by asking participants to rate a 36-item somatic symptom checklist and reported that "a few patients reported worsening of symptoms"; of the 36 side effects assessed, only 1 showed a statistically significant group difference (P = 0.03), i.e. that the sertraline group reported worse post-treatment diarrhoea and loose stools.

Blumenthal 2012a assessed 36 side effects; only 2 showed a significant group difference: 20% of participants receiving sertraline reported worse post-treatment fatigue compared with 2.4% in the exercise group and 26% reported increased sexual problems compared with 2.4% in the exercise group.

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 Lack of blinding of outcome assessors probably increased effect sizes and drop-out rates were high. Also sequence generation was considered unclear in 1 study.
2 I² = 0% and P = 0.52, indicated no heterogeneity
3 The studies included were all relevant to the review question, particularly given that all studies had to meet the criteria of the ACSM definition of exercise.
 
Table 1. Number screened; number still in trial and exercise intervention at end of trial

Trial IDScreenedRandomisedAllocated exerciseCompleted trialCompleted comparator group, e.g. control, other treatment (as a proportion of those allocated)Completed exercise (as a proportion of those allocated)

Blumenthal 1999604 underwent telephone screening1565513341/48 (medication)44/55 (exercise plus medication)

39/53 (exercise alone)

Blumenthal 200745720251 (supervised), 53 home-based18342/49 (placebo)

45/49 (sertraline)
45/51 (supervised), 51/53 home-based

Blumenthal 2012b1680 enquired about the study101379523/24 completed 'placebo' and 36/40 completed the medication36/37 completed the exercise

Brenes 2007Not reported3714Not reportedNot reportedNot reported

Bonnet 2005Not reported11574/63/5

Chu 2008104 responded to adverts54363812/1826/36 (both exercise arms combined)

15/18 in the high-intensity arm

Dunn 20051664 assessed for eligibility8017459/1311/17 (public health dose 3 times per week)

Doyne 1987285 responded to adverts57Not reported40 completed treatment or control27 (denominator not known)13 (denominator not known)

Epstein 1986250 telephone inquiries received337Not reportedNot reported7

Fetsch 1979Not reported2110168/118/10

Foley 2008215 responded to adverts2310135/138/10

Fremont 198772 initially expressed an interest61214931/4018/21

Gary 2010982 referred, 242 had heart failure, 137 had a BDI > 10 and 74 eligible and consented742068/74 completed post-intervention assessments and 62 completed follow-up assessmentsusual care 15/17exercise only: 20/20

Greist 1979Not reported28102215/188/10

Hemat-Far 2012350 screened201020not statednot stated

Hess-Homeier 1981Not reported175Not reportedNot reportedNot reported

Hoffman 2010253 screened, 58 ineligible84427639/42 (2 were excluded by the trialists and 1 did not attend follow-up)37/42 of exercise group provided data for analysis

Klein 1985209 responded to an advertisement74274211/23 (meditation)

16/24 (group therapy)
15/27

Knubben 2007Not reported39 (note data on only 38 reported)203516/1819/20

Krogh 2009390 referred16511013742/5595/110 (both exercise arms combined)

47/55 (strength)

48/55 (aerobic)

Martinsen 1985Not reported43243717/1920/24

Mather 20021185 referred or screened86438642/4343/43

McCann 1984250 completed BDI, 60 contacted47164314/15 completed placebo

14/16 completed 'no treatment'
15/16

McNeil 19918230103010/10 (waiting list)

10/10 (social contact)
10/10

Mota-Pereira 2011150332229/3310/1119/22

Mutrie 198836249247/79/9

Nabkasorn 2005266 volunteers screened59284928/3121/28

Orth 19791711372/23/3

Pilu 2007Not reported30103020/2010/10

Pinchasov 2000Not reported189Not reportedNot reportedNot reported

Reuter 1984Not reportedNot reported9Not reportedNot reported9

Schuch 201114/40 invited patients were not interested in participating2615"no patient withdrew from intervention""no patient withdrew from intervention""no patient withdrew from intervention"

Setaro 1985211 responses to advertisement18030150Not reported25/30

Shahidi 201170 older depressed women chosen from 500 members of a district using the geriatric depression scale702360/7020/2420/23

Sims 20091550 invitations, 233 responded45234322/2221/23

Singh 1997Letters sent to 2953 people, 884 replied32173215/1517/17

Singh 200545160205419/20 (GP standard care)18/20 (high-intensity training)

Veale 1992Not reported83485729/3536/48

Williams 200896 in parent study4333348/1026/33 (both exercise groups combined)

15/16 exercise

11/17 walking

 BDI: Beck Depression Inventory