. 2022 Jun 14;33(2):393–431. doi: 10.1007/s11065-022-09543-6

Table 3.

Study Characteristics for the 22 Systematic Reviews Included in the Umbrella Review

Citation	Key eligibility criteria for primary studies^a	Primary Studies # Studies^b,c Study designs Sample Size^d	Sample Sex Age % TBI^e	Depression Entry Criteria	Interventions (number of studies)	Comparators^f	Heterogeneity^g
Beedham et al. (2020) Up to Date: January 2019 Protocol: PROSPERO CRD42019122600 Meta-Analysis	Study design: RCT or quasi-experimental; depression as outcome – severity scale or remission rates Age: adults ≥ 18yrs TBI: diagnosis by clinician/ healthcare professional Depression: diagnosis using standardised criteria/ score on validated tools Intervention: TMS, pharmacotherapy or psychological	# Studies: 15 Study Designs: - 6 × RCT - 2 × non-RCT w control group - 7 × non-RCT w/out control group Sample Size: 504	Sex: NR Age: NR % TBI: 100%	DSM MDD III; n = 1 IV; n = 1 IV-SCID (n = 2) IV-MINI (n = 1) HAM-D ≥ 15 & IV-SCID (n = 1) ≥ 17 (n = 1) ≥ 18 & SCID (n = 1) ≥ 18 & DIS Interview (n = 1) PHQ-9 ≥ 10; (n = 1) BDI ≥ 18 (n = 2) Clinical interview (n = 2)	MAOI Moclobemide (n = 1) SNRI Milnacipran (n = 1) SSRI Citalopram (n = 2) Sertraline (n = 6) Escitalopram (n = 1) Stimulant Methylphenidate (n = 2) TCA Amitriptyline (n = 2) Combination Therapy SSRI + Anti-convulsant (n = 1; Citalopram Carbamazepine)	- Comparator condition: TBI group; placebo given (n = 5) - Comparator condition: TBI group; no placebo given (n = 1) - Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 7)	Depression I² = 0—84%
Peppel et al. (2020) Up to Date: September 2018 Protocol: NR Meta-Analysis	- Study design: RCT; primary outcome was depression; baseline and follow-up depression scores for intervention and control - Study aim: having depression was inclusion criterion/reducing depression was aim of study - Age: ≥ 16yrs at injury/ reported separately from other age groups - TBI: mod to sev TBI/ reported separately from other ABI - Intervention: any treatment modality - Comparator: any comparison group - Excluded: studies of mild TBI only; comorbid PTSD; preventative treatment; n < 10 total or n < 5 per group	# Studies: 5 Study Designs: - 5 × RCT Sample Size: 249	Sex: 202/249 male; 81% Age: M 32.4–49.1 % TBI: 100%	NR	SSRI Sertraline (n = 4) Stimulant Methylphenidate (n = 2)	- Comparator condition: TBI group; placebo given (n = 4) - Comparator condition: TBI group; no placebo given (n = 1)	Depression I² = 43.26%
Gao et al. (2019) Up to Date: February 2019 Protocol: NR Meta-Analysis	- Study design: RCT - TBI: any severity - Intervention: Sertraline - Comparator: placebo	# Studies: 5 Study Designs: - 5 × RCT Sample Size: 316	Sex: 213/316 male; 67% Age: M 33.6–54.91 % TBI: 100%	NR (n = 4) N/A (preventative study) (n = 1)	SSRI Sertraline (n = 5)	- Comparator condition: TBI group; placebo given (n = 5)	Depression I² = 4% Harms I² = 0–64%
Kreitzer et al. (2019) Up to Date: September 2017 Protocol: NR Meta-Analysis	-Study design: prospective studies - TBI: any severity - Depression: MDD - Intervention: any Anti-depressant - Excluded: anti-depressant intervention for MDD refractory to other first-line agents; same cohort from prior study	# Studies: 10 Study Designs: - 5 × RCT - 5 × not specified Sample Size: 336	Sex: NR Age: NR % TBI: 100%	NR	MAOI Moclobemide (n = 1) Phenelzine (n = 1) SSRI Escitalopram (n-1) Citalopram (n = 1) Sertraline (n = 5) Stimulant Methylphenidate (n = 1) TCA Amitriptyline (n = 2)	- Comparator condition: TBI group; placebo given (n = 4) - Comparator condition: TBI group; no placebo given (n = 1) - Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 3)	Depression I² = 17%
Liu et al. (2019) Up to Date: NR Protocol: NR Narrative Synthesis	NR	# Studies: 12 Study Designs: - 3 × RCT - 6 X non-RCT w control group - 3 X non-RCT w/out control group Sample Size: 317^h	Sex: NR Age: NR % TBI: Unclearⁱ	NR	SNRI Milnacipran (n = 1) SSRI Citalopram (n = 2) Sertraline (n = 6) Stimulant Methylphenidate (n = 1) TCA Amitriptyline (n = 2) Desipramine (n = 1)	- Comparator condition: TBI group; placebo given (n = 5) - Comparator condition: TBI group; no placebo given (n = 1) - Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 4)	N/A – no meta- analysis
Reyes et al. (2019) Up to Date: NR Protocol: NR Meta-Analysis	- Study design: RCT; depression as a primary or secondary outcome - Age: ≥ 18yrs - TBI: documented LoC/radiological evidence - Depression: diagnosed w MDD using standardized diagnostic criteria - Intervention: Sertraline - Excluded: patients with diagnosis of bipolar disorder, psychotic disorder, substance abuse disorder, bereavement, suicidal ideations or intent; taking anti-depressants; undergoing psychotherapy; serious medical illness; history of allergy/adverse reaction to study drug; pregnant or breastfeeding; prophylactic studies	# Studies: 4 Study Designs: - 4 × RCT Sample Size: 224^j	Sex: 181/224 male; 80.1% Age^k: M 33.6 – 49.1 % TBI: 100%	DSM-IV & HAM-D ≥ 18 (n = 1) HAM-D ≥ 15 (n = 1) BDI > 18 (n = 1) PHQ-9 (n = 1)	SSRI Sertraline (n = 4)	- Comparator condition: TBI group; placebo given (n = 3) - Comparator condition: TBI group; no placebo given (n = 1)	Depression I² = 84 – 98%
Slowinski et al. (2019) Up to Date: NR Protocol: NR Meta-Analysis	Study design: depression as a primary or secondary outcome Age: > 18yrs TBI: any severity Depression: any; must have depression diagnosis at baseline Intervention: any Setting: any	# Studies: 12 Study Designs: - 2 × Non-RCT w control group - 3 × Non-RCT w/out control group - 7 × Clinical trial w control group (unclear if RCT) Sample Size: Unclear^l	Sex: NR Age: NR % TBI: 100%	DSM (no further info provided) (n = 11) PHQ-9 (n = 1)	MAOI Phenelzine (n = 1) SNRI Milnacipran (n = 1) SSRI Citalopram (n = 1) Sertraline (n = 5) Stimulant Methylphenidate (n = 2) TCA Amitriptyline (n = 2) Desipramine (n = 1) Combination Therapy SSRI + Anti-convulsant (n = 1; Citalopram Carbamazepine)	- Comparator condition: TBI group; placebo given (n = 6) - Comparator condition: TBI group; no placebo given (n = 1) - Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 3)	Depression I² = 81.45–91.94%
Paraschakis and Katsanos (2017) Up to Date: August 2017 Protocol: NR Meta-Analysis	- Study design: RCT; HAM-D as outcome measure - Age: adult - TBI: any - Depression: major/mod depression, adjustment disorder w depressive symptoms, dysthymic disorder - Intervention: anti-depressants - Comparator: placebo	# Studies: 4 Study Designs: - 4 × RCT Sample Size: 181	Sex: 123/181 male; 68% Age: M 34.5 – 47.7 % TBI: 100%	DSM-IV n = 1 & HAM-D ≥ 18 (n = 1) & HAM-D ≥ 16 (n = 1) N/A (preventative study) (n = 1)	SSRI Citalopram (n = 1) Sertraline (n = 3)	- Comparator condition: TBI group; placebo given (n = 4)	Depression I² = 0%
Yue et al. (2017) Up to Date: September 2016 Protocol: NR Meta-Analysis & Narrative Synthesis^m	NR	# Studies: 12 Study Designs: - 6 × RCT - 6 × Non-RCT w/out control group Sample Size: 650	Sex: 80/650 male; 570/650 NR Age: NR % TBI: 96.6%ⁿ	DSM-IV MDD/ MDE (n = 4) BDI (cut-off ns) (n = 1) MDD (diagnosis method ns) (n = 1) NR (n = 6) N/A (preventative study) (n = 2)	SSRI Sertraline (n = 7) Citalopram (n = 5)	- Comparator condition: TBI group; placebo given (n = 5) - Comparator condition: TBI group; no placebo given (n = 1) - No comparator condition (n = 6)	NR
Maksimowksi and Tampi (2016) Up to Date: December 2015 Protocol: NR Narrative Synthesis	NR	# Studies: 1 Study Designs: - 1 × RCT Sample Size: 30	Sex: NR Age: M 34.8 % TBI: 100%	NR	Stimulant Methylphenidate (n = 1)^o	- Comparator condition: TBI group; placebo given (n = 1)	N/A – no meta- analysis
Plantier et al. (2016) Up to Date: June 2015 Protocol: NR Narrative Synthesis	Study Aim: to treat behavioural disorders post-TBI Exclusion: participants in acute recovery phase in ICU; interventions to improve cognition/ stimulate recovery; absent/insufficient participants w TBI	# Studies: 10 Study Designs: - 4 × RCT - 1 × Non-RCT w control group - 5 × Non-RCT w/out control group Sample Size: 308	Sex: 4/295 male; 291/295 NR Age: n = 1; R 28–74 n = 9; NR % TBI: 100%	NR	MAOI Moclobemide (n = 1) SNRI Milnacipran (n = 1) SSRI Citalopram (n = 2) Sertraline (n = 3) Stimulant Methylphenidate (n = 1) TCA Amitriptyline (n = 1) Desipramine (n = 1) Combination Therapy SSRI + Anti-convulsant (n = 1; Citalopram Carbamazepine)	- Comparator condition: TBI group; placebo given (n = 4) - Comparator condition: non-TBI controls; study drug given (n = 1) - No comparator condition (n = 5)	N/A – no meta- analysis
Salter et al. (2016) Up to Date: October 2014 Protocol: NR Meta-Analysis	Study design: clinical trial Age: adults TBI: any severity; presence of TBI made w comparison to NINDS Common Data Elements definition Depression: diagnosis using standardised criteria/ score on validated tools; must have depression diagnosis at baseline Intervention: any pharmacotherapy	# Studies: 9 Study Designs: - 3 × RCT - 2 × Non-RCT w control group - 4 × Non-RCT w/out control group Sample Size: 245	Sex: NR Age: M 28.5 – 58.3 % TBI: 100%	DSM III-R (n = 1) III-R MDD (n = 1) III MDD & HAM-D > 17 (n = 1) III-R MDD & HAM-D ≥ 18 (n = 1) IV MDE (n = 1) IV MDE & HAM-D > 18 (n = 1) IV MDE/MDD (n = 1) BDI ≥ 18 (n = 1) NR (n = 1)	MAOI Moclobemide (n = 1) Phenelzine (n = 1) SNRI Milnacipran (n = 1) SSRI Citalopram (n = 1) Sertraline (n = 3) Stimulant Methylphenidate (n = 1) TCA Amitriptyline (n = 2) Desipramine (n = 1)	- Comparator condition: TBI group; placebo given (n = 3) - Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 4)	NR
Barker-Collo et al. (2013) Up to Date: NR Protocol: NR Meta-Analysis^p	Study Design: depression/depressive symptoms as an outcome measure; pre- and post-test assessment Age: adult TBI: mTBI (LoC ≤ 30 min/PTA < 24 h) Intervention: pharmacological or non-pharmacological	# Studies: 8 Study Designs: - 2 × RCT - 2 × Non-RCT w control group - 4 × Non-RCT w/out control group Sample Size: 195	Sex: 181/224 male; 80.1% Age^q: M 33.6 – 49.1 % TBI: 100%	MDD & HAM-D ≥ 18 n = 1 DSM III MDD- Feighner (1972) criteria (n = 1) III MDD & HAM-D > 17 (n = 1) III-R MDD & HAM-D > 17 (n = 1) IV MDE (n = 1) IV MDE or ‘minor depression’ (n = 1) IV MDD (n = 2) IV Diagnosis ns (n = 1)	SNRI Milnacipran (n = 1) SSRI Sertraline (n = 3) Citalopram (n = 2) Stimulant Methylphenidate (n = 1) TCA Amitriptyline (n = 2)	- Comparator condition: TBI group; placebo given (n = 2) - Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 4)	Depression I² = 71.1–86.7%^r
Guillamondegui et al. (2011) Up to Date: May 2010 Protocol: NR Narrative Synthesis	- Study design: RCT, cohort w comparison, case control, case series; n ≥ 50 - Age: ≥ 16yrs - TBI: any severity; sustained as adult - Depression: any severity; diagnostic information required - Setting: study conducted in a ‘developed nation’ - Excluded: penetrating TBI; self-report depression	# Studies: 2 Study Designs: - 1 × RCT - 1 × Non-RCT w/out control group Sample Size: 132	Sex: NR Age: M 39.7 – 51.5 % TBI: 100%	DSM IV MDE & HAM-D ≥ 18 (n = 1) IV MDD (5/9 symptoms and depressed mood or anhedonia) (n = 1)	SSRI Citalopram (n = 1) Sertraline (n = 1)	- Comparator condition: TBI group; placebo given (n = 1) - No comparator condition (n = 1)	N/A – no meta- analysis
Price et al. (2011) Up to Date: August 2009 Protocol: NR Meta-Analysis	Study design: RCT; depression primary outcome Age: adult TBI: any neurological disease w biological underpinning Depression: MDD, adjustment disorder, dysthymic disorder; diagnosis using standardised criteria/score on validated tools Intervention: antidepressant Comparator: placebo Excluded: dementia; MCI	# Studies: 1 Study Designs: - 1 × RCT Sample Size: 41	Sex: NR Age: NR % TBI: 100%	NR	SSRI Sertraline (n = 1)	- Comparator condition: TBI group; placebo given (n = 1)	Depression I² = 78%^s
Wheaton et al. (2011) Up to Date: April 2010 Protocol: NR Meta-Analysis	Study Design: treatment and control group; outcomes measures of cognition/behaviour; sufficient data provided to calculate ES Age: ≥ 16yrs/M-1SD ≥ 16yrs TBI: non-penetrating Intervention: any pharmacology; administered ≥ 4wks post TBI Excluded: previous TBI, pre-existing neurological/psychiatric disorder, substance abuse history	# Studies: 6 Study Designs: - 2 × RCT - 2 x non-RCT w control group - 4 x non-RCT w/out control group Sample Size: Unclear^t	Sex: NR Age: M 29.10 – 42 % TBI: 100%	NR	MAOI Phenelzine (n = 1) SNRI Milnacipran (n = 1) SSRI Citalopram (n = 1) Sertraline (n = 1) Stimulant Methylphenidate (n = 1) TCA Amitriptyline (n = 2) Desipramine (n = 1)	- Comparator condition: TBI group; placebo given (n = 2) - Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 2)	NR
Rayner et al. (2010) Up to Date: December 2009 Protocol: Yes Meta-Analysis	Study Design: RCT (cluster and cross-over were eligible); depression as primary outcome Age: > 18yrs Depression: MDD, adjustment disorder, dysthymic disorder using standardised criteria/ score on validated tools; any severity Intervention: anti-depressant prescribed for depression Comparator: placebo	# Studies: 1 Study Designs: - 1 × RCT Sample Size: 52	Sex: 22/52 male; 42% Age: M 46.8 – 51.5 % TBI: 100%	NR	SSRI Sertraline (n = 1)	- Comparator condition: TBI group; placebo given (n = 1)	Depression I² = 54.07–78.27%^u
Fann et al. (2009a) Up to Date: NR Protocol: NR Narrative Synthesis	- Study design: depression as a primary or secondary outcome; quantitative pre/post scores provided - TBI: any severity; reported separately from other ABI - Intervention: any treatment modality	# Studies: 11 Study Designs: - 2 × RCT - 3 × Non-RCT w control group - 6 × Non-RCT w/out control group Sample Size: 243	Sex: 13/243 male; 230/243 NR Age: NR % TBI: 100%	DSM MDD III (n = 2) III-R (n = 3) IV (n = 5) HAM-D > 17 (n = 3) ≥ 17 (n = 1) ≥ 18 (n = 1)	MAOI Moclobemide (n = 1) Phenelzine (n = 1) SNRI Milnacipran (n = 1) SSRI Citalopram (n = 1) Sertraline (n = 4) Stimulant Methylphenidate (n = 1) TCA Amitriptyline (n = 2) Desipramine (n = 1) Combination Therapy SSRI + Anti-convulsant (n = 1; Citalopram Carbamazepine)	- Comparator condition: TBI group; placebo given (n = 3) - Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 6)	N/A – no meta- analysis
Hardy (2009) Up to Date: July 2008 Protocol: NR Narrative Synthesis	Study design: controlled trials, case series, case reports Age: older adults TBI: any Depression: any	# Studies: 1 Study Designs: - 1 × RCT Sample Size: 30	Sex: NR Age: M 34yrs % TBI: 100%	NR	Stimulant Methylphenidate (n = 1)	- Comparator condition: TBI group; placebo given n = 1	N/A – no meta- analysis
Warden et al. (2006) Up to Date: October 2004 Protocol: NR Narrative Synthesis	NR	# Studies: 6 Study Designs: - 1 × RCT - 2 X non-RCT w control group - 3 X non-RCT w/out control group Sample Size: 119	Sex: NR Age: NR % TBI: 100%	DSM III (n = 1) III-R (n = 2) IV (n = 1) NR (n = 2)	SSRI Sertraline (n = 1) TCA Amitriptyline (n = 2) Desipramine (n = 1) MAOI Moclobemide (n = 1) Combination Therapy SSRI + Anti-convulsant (n = 1; Citalopram Carbamazepine)	- Comparator condition: TBI group; placebo given (n = 1) - Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 3)	N/A – no meta- analysis
Comper et al. (2005) Up to Date: 2003 Protocol: NR Narrative Synthesis	Age: 16-65yrs TBI: mTBI; mixed severity with mTBI separately reported; ≤ 5yrs post injury Intervention: any in human TBI populations Excluded: case series and case studies	# Studies: 3 Study Designs: - 2 × Non-RCT w control group - 1 × Non-RCT w/out control group Sample Size: 63	Sex: 47/63 male; 74.6% Age: M 30.7 – 44.2 % TBI: 100%	NR	MAOI Phenelzine (n = 1) SSRI Sertraline (n = 1) TCA Amitriptyline (n = 2)	- Comparator condition: non-TBI controls; study drug given (n = 2) - No comparator condition (n = 1)	N/A – no meta- analysis
Deb and Crownshaw (2004) Up to Date: January 2003 Protocol: NR Narrative Synthesis	Age: majority of sample > 16yrs TBI: any Depression: any Intervention: any drug that may affect behaviour directly or indirectly Excluded: studies using non-psychotropic drugs acute post injury	# Studies: 3 Study Designs: - 1 × non-RCT w control group - 1 × non-RCT w/out control group - 1 × Unclear Sample Size: 61	Sex: NR Age: M 27 – 41.9^v (NR; n = 1) % TBI: 100%	DSM III-R MDD (n = 1) NR (n = 2)	SSRI Sertraline (n = 1) MAOI Moclobemide (n = 1) Combination Therapy SSRI + Anti-convulsant (n = 1; Citalopram Carbamazepine)	- No comparator condition (n = 3)	N/A – no meta- analysis

Citation

Key eligibility criteria for primary studies^a

Primary Studies
# Studies^b,c
Study designs
Sample Size^d

Sample
Sex
Age
% TBI^e

Depression Entry Criteria

Interventions
(number of studies)

Comparators^f

Heterogeneity^g

Beedham et al. (2020)

Up to Date: January 2019

Protocol: PROSPERO CRD42019122600

Meta-Analysis

Study design: RCT or quasi-experimental; depression as outcome – severity scale or remission rates

Age: adults ≥ 18yrs

TBI: diagnosis by clinician/ healthcare professional

Depression: diagnosis using standardised criteria/ score on validated tools

Intervention: TMS, pharmacotherapy or psychological

# Studies: 15

Study Designs:

- 6 × RCT

- 2 × non-RCT w control group

- 7 × non-RCT w/out control group

Sample Size: 504

Sex:

Age: NR

% TBI: 100%

DSM MDD

III; n = 1

IV; n = 1

IV-SCID (n = 2)

IV-MINI (n = 1)

HAM-D

≥ 15 & IV-SCID (n = 1)

≥ 17 (n = 1)

≥ 18 & SCID (n = 1)

≥ 18 & DIS Interview (n = 1)

PHQ-9

≥ 10; (n = 1)

BDI

≥ 18 (n = 2)

Clinical interview

(n = 2)

MAOI

Moclobemide (n = 1)

SNRI

Milnacipran (n = 1)

SSRI

Citalopram (n = 2)

Sertraline (n = 6)

Escitalopram (n = 1)

Stimulant

Methylphenidate (n = 2)

TCA

Amitriptyline (n = 2)

Combination Therapy

SSRI + Anti-convulsant (n = 1;

Citalopram

Carbamazepine)

- Comparator condition: TBI group; placebo given

(n = 5)

- Comparator condition: TBI group; no placebo given

(n = 1)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 7)

Depression

I² = 0—84%

Peppel et al. (2020)

Up to Date: September 2018

Protocol: NR

Meta-Analysis

- Study design: RCT; primary outcome was depression; baseline and follow-up depression scores for intervention and control

- Study aim: having depression was inclusion criterion/reducing depression was aim of study

- Age: ≥ 16yrs at injury/ reported separately from other age groups

- TBI: mod to sev TBI/ reported separately from other ABI

- Intervention: any treatment modality

- Comparator: any comparison group

- Excluded: studies of mild TBI only; comorbid PTSD; preventative treatment; n < 10 total or n < 5 per group

# Studies: 5

Study Designs:

- 5 × RCT

Sample Size: 249

Sex:

202/249 male; 81%

Age:

M 32.4–49.1

% TBI: 100%

SSRI

Sertraline (n = 4)

Stimulant

Methylphenidate (n = 2)

- Comparator condition: TBI group; placebo given

(n = 4)

- Comparator condition: TBI group; no placebo given

(n = 1)

Depression

I² = 43.26%

Gao et al. (2019)

Up to Date: February 2019

Protocol: NR

Meta-Analysis

- Study design: RCT

- TBI: any severity

- Intervention: Sertraline

- Comparator: placebo

# Studies: 5

Study Designs:

- 5 × RCT

Sample Size: 316

Sex:

213/316 male; 67%

Age:

M 33.6–54.91

% TBI: 100%

(n = 4)

N/A (preventative study)

(n = 1)

SSRI

Sertraline (n = 5)

- Comparator condition: TBI group; placebo given

(n = 5)

Depression

I² = 4%

Harms

I² = 0–64%

Kreitzer et al. (2019)

Up to Date: September 2017

Protocol: NR

Meta-Analysis

-Study design: prospective studies

- TBI: any severity

- Depression: MDD

- Intervention: any Anti-depressant

- Excluded: anti-depressant intervention for MDD refractory to other first-line agents; same cohort from prior study

# Studies: 10

Study Designs:

- 5 × RCT

- 5 × not specified

Sample Size: 336

Sex: NR

Age: NR

% TBI: 100%

MAOI

Moclobemide (n = 1)

Phenelzine (n = 1)

SSRI

Escitalopram (n-1)

Citalopram (n = 1)

Sertraline (n = 5)

Stimulant

Methylphenidate (n = 1)

TCA

Amitriptyline (n = 2)

- Comparator condition: TBI group; placebo given

(n = 4)

- Comparator condition: TBI group; no placebo given

(n = 1)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 3)

Depression

I² = 17%

Liu et al. (2019)

Up to Date: NR

Protocol: NR

Narrative Synthesis

# Studies: 12

Study Designs:

- 3 × RCT

- 6 X non-RCT w control group

- 3 X non-RCT w/out control group

Sample Size: 317^h

Sex:

Age: NR

% TBI: Unclearⁱ

SNRI

Milnacipran (n = 1)

SSRI

Citalopram (n = 2)

Sertraline (n = 6)

Stimulant

Methylphenidate (n = 1)

TCA

Amitriptyline (n = 2)

Desipramine (n = 1)

- Comparator condition: TBI group; placebo given

(n = 5)

- Comparator condition: TBI group; no placebo given

(n = 1)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 4)

N/A – no meta- analysis

Reyes et al. (2019)

Up to Date: NR

Protocol: NR

Meta-Analysis

- Study design: RCT; depression as a primary or secondary outcome

- Age: ≥ 18yrs

- TBI: documented LoC/radiological evidence

- Depression: diagnosed w MDD using standardized diagnostic criteria

- Intervention: Sertraline

- Excluded: patients with diagnosis of bipolar disorder, psychotic disorder, substance abuse disorder, bereavement, suicidal ideations or intent; taking anti-depressants; undergoing psychotherapy; serious medical illness; history of allergy/adverse reaction to study drug; pregnant or breastfeeding; prophylactic studies

# Studies: 4

Study Designs:

- 4 × RCT

Sample Size: 224^j

Sex: 181/224 male; 80.1%

Age^k:

M 33.6 – 49.1

% TBI: 100%

DSM-IV &

HAM-D ≥ 18 (n = 1)

HAM-D ≥ 15 (n = 1)

BDI > 18 (n = 1)

PHQ-9

(n = 1)

SSRI

Sertraline (n = 4)

- Comparator condition: TBI group; placebo given

(n = 3)

- Comparator condition: TBI group; no placebo given

(n = 1)

Depression

I² = 84 – 98%

Slowinski et al. (2019)

Up to Date: NR

Protocol: NR

Meta-Analysis

Study design: depression as a primary or secondary outcome

Age: > 18yrs

TBI: any severity

Depression: any; must have depression diagnosis at baseline

Intervention: any

Setting: any

# Studies: 12

Study Designs:

- 2 × Non-RCT w control group

- 3 × Non-RCT w/out control group

- 7 × Clinical trial w control group (unclear if RCT)

Sample Size:

Unclear^l

Sex: NR

Age: NR

% TBI: 100%

DSM (no further info provided)

(n = 11)

PHQ-9

(n = 1)

MAOI

Phenelzine (n = 1)

SNRI

Milnacipran (n = 1)

SSRI

Citalopram (n = 1)

Sertraline (n = 5)

Stimulant

Methylphenidate (n = 2)

TCA

Amitriptyline (n = 2)

Desipramine (n = 1)

Combination Therapy

SSRI + Anti-convulsant (n = 1;

Citalopram

Carbamazepine)

- Comparator condition: TBI group; placebo given

(n = 6)

- Comparator condition: TBI group; no placebo given

(n = 1)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 3)

Depression

I² = 81.45–91.94%

Paraschakis and Katsanos (2017)

Up to Date: August 2017

Protocol: NR

Meta-Analysis

- Study design: RCT; HAM-D as outcome measure

- Age: adult

- TBI: any

- Depression: major/mod depression, adjustment disorder w depressive symptoms, dysthymic disorder

- Intervention: anti-depressants

- Comparator: placebo

# Studies: 4

Study Designs:

- 4 × RCT

Sample Size: 181

Sex: 123/181 male; 68%

Age:

M 34.5 – 47.7

% TBI: 100%

DSM-IV

n = 1

& HAM-D ≥ 18 (n = 1)

& HAM-D ≥ 16 (n = 1)

N/A (preventative study)

(n = 1)

SSRI

Citalopram (n = 1)

Sertraline (n = 3)

- Comparator condition: TBI group; placebo given

(n = 4)

Depression

I² = 0%

Yue et al. (2017)

Up to Date:

September 2016

Protocol: NR

Meta-Analysis & Narrative Synthesis^m

# Studies: 12

Study Designs:

- 6 × RCT

- 6 × Non-RCT w/out control group

Sample Size: 650

Sex:

80/650 male; 570/650 NR

Age: NR

% TBI: 96.6%ⁿ

DSM-IV MDD/ MDE

(n = 4)

BDI (cut-off ns)

(n = 1)

MDD (diagnosis method ns)

(n = 1)

(n = 6)

N/A (preventative study)

(n = 2)

SSRI

Sertraline (n = 7)

Citalopram (n = 5)

- Comparator condition: TBI group; placebo given

(n = 5)

- Comparator condition: TBI group; no placebo given

(n = 1)

- No comparator condition

(n = 6)

Maksimowksi and Tampi (2016)

Up to Date: December 2015

Protocol: NR

Narrative Synthesis

# Studies: 1

Study Designs:

- 1 × RCT

Sample Size: 30

Sex:

Age: M 34.8

% TBI: 100%

Stimulant

Methylphenidate (n = 1)^o

- Comparator condition: TBI group; placebo given

(n = 1)

N/A – no meta- analysis

Plantier et al. (2016)

Up to Date: June 2015

Protocol: NR

Narrative Synthesis

Study Aim: to treat behavioural disorders post-TBI

Exclusion: participants in acute recovery phase in ICU; interventions to improve cognition/ stimulate recovery; absent/insufficient participants w TBI

# Studies: 10

Study Designs:

- 4 × RCT

- 1 × Non-RCT w control group

- 5 × Non-RCT w/out control group

Sample Size: 308

Sex: 4/295 male; 291/295 NR

Age:

n = 1; R 28–74

n = 9; NR

% TBI: 100%

MAOI

Moclobemide (n = 1)

SNRI

Milnacipran (n = 1)

SSRI

Citalopram (n = 2)

Sertraline (n = 3)

Stimulant

Methylphenidate (n = 1)

TCA

Amitriptyline (n = 1)

Desipramine (n = 1)

Combination Therapy

SSRI + Anti-convulsant (n = 1;

Citalopram

Carbamazepine)

- Comparator condition: TBI group; placebo given

(n = 4)

- Comparator condition: non-TBI controls; study drug given

(n = 1)

- No comparator condition

(n = 5)

N/A – no meta- analysis

Salter et al. (2016)

Up to Date: October 2014

Protocol: NR

Meta-Analysis

Study design: clinical trial

Age: adults

TBI: any severity; presence of TBI made w comparison to NINDS Common Data Elements definition

Depression: diagnosis using standardised criteria/ score on validated tools; must have depression diagnosis at baseline

Intervention: any pharmacotherapy

# Studies: 9

Study Designs:

- 3 × RCT

- 2 × Non-RCT w control group

- 4 × Non-RCT w/out control group

Sample Size: 245

Sex: NR

Age: M 28.5 – 58.3

% TBI: 100%

DSM

III-R (n = 1)

III-R MDD (n = 1)

III MDD & HAM-D > 17 (n = 1)

III-R MDD & HAM-D ≥ 18 (n = 1)

IV MDE (n = 1)

IV MDE & HAM-D > 18 (n = 1)

IV MDE/MDD (n = 1)

BDI

≥ 18 (n = 1)

(n = 1)

MAOI

Moclobemide (n = 1)

Phenelzine (n = 1)

SNRI

Milnacipran (n = 1)

SSRI

Citalopram (n = 1)

Sertraline (n = 3)

Stimulant

Methylphenidate (n = 1)

TCA

Amitriptyline (n = 2)

Desipramine (n = 1)

- Comparator condition: TBI group; placebo given

(n = 3)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 4)

Barker-Collo et al. (2013)

Up to Date: NR

Protocol: NR

Meta-Analysis^p

Study Design: depression/depressive symptoms as an outcome measure; pre- and post-test assessment

Age: adult

TBI: mTBI (LoC ≤ 30 min/PTA < 24 h)

Intervention: pharmacological or non-pharmacological

# Studies: 8

Study Designs:

- 2 × RCT

- 2 × Non-RCT w control group

- 4 × Non-RCT w/out control group

Sample Size: 195

Sex: 181/224 male; 80.1%

Age^q:

M 33.6 – 49.1

% TBI: 100%

MDD & HAM-D ≥ 18

n = 1

DSM

III MDD- Feighner (1972) criteria (n = 1)

III MDD & HAM-D > 17 (n = 1)

III-R MDD & HAM-D > 17 (n = 1)

IV MDE (n = 1)

IV MDE or ‘minor depression’ (n = 1)

IV MDD (n = 2)

IV Diagnosis ns (n = 1)

SNRI

Milnacipran (n = 1)

SSRI

Sertraline (n = 3)

Citalopram (n = 2)

Stimulant

Methylphenidate (n = 1)

TCA

Amitriptyline (n = 2)

- Comparator condition: TBI group; placebo given

(n = 2)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 4)

Depression

I² = 71.1–86.7%^r

Guillamondegui et al. (2011)

Up to Date: May 2010

Protocol: NR

Narrative Synthesis

- Study design: RCT, cohort w comparison, case control, case series; n ≥ 50

- Age: ≥ 16yrs

- TBI: any severity; sustained as adult

- Depression: any severity; diagnostic information required

- Setting: study conducted in a ‘developed nation’

- Excluded: penetrating TBI; self-report depression

# Studies: 2

Study Designs:

- 1 × RCT

- 1 × Non-RCT w/out control group

Sample Size: 132

Sex: NR

Age:

M 39.7 – 51.5

% TBI: 100%

DSM

IV MDE & HAM-D ≥ 18 (n = 1)

IV MDD (5/9 symptoms and depressed mood or anhedonia) (n = 1)

SSRI

Citalopram (n = 1)

Sertraline (n = 1)

- Comparator condition: TBI group; placebo given

(n = 1)

- No comparator condition

(n = 1)

N/A – no meta- analysis

Price et al. (2011)

Up to Date:

August 2009

Protocol: NR

Meta-Analysis

Study design: RCT; depression primary outcome

Age: adult

TBI: any neurological disease w biological underpinning

Depression: MDD, adjustment disorder, dysthymic disorder; diagnosis using standardised criteria/score on validated tools

Intervention: antidepressant

Comparator: placebo

Excluded: dementia; MCI

# Studies: 1

Study Designs:

- 1 × RCT

Sample Size: 41

Sex: NR

Age: NR

% TBI: 100%

SSRI

Sertraline (n = 1)

- Comparator condition: TBI group; placebo given

(n = 1)

Depression

I² = 78%^s

Wheaton et al. (2011)

Up to Date: April 2010

Protocol: NR

Meta-Analysis

Study Design: treatment and control group; outcomes measures of cognition/behaviour; sufficient data provided to calculate ES

Age: ≥ 16yrs/M-1SD ≥ 16yrs

TBI: non-penetrating

Intervention: any pharmacology; administered ≥ 4wks post TBI

Excluded: previous TBI, pre-existing neurological/psychiatric disorder, substance abuse history

# Studies: 6

Study Designs:

- 2 × RCT

- 2 x non-RCT w control group

- 4 x non-RCT w/out control group

Sample Size: Unclear^t

Sex:

Age: M 29.10 – 42

% TBI: 100%

MAOI

Phenelzine (n = 1)

SNRI

Milnacipran (n = 1)

SSRI

Citalopram (n = 1)

Sertraline (n = 1)

Stimulant

Methylphenidate (n = 1)

TCA

Amitriptyline (n = 2)

Desipramine (n = 1)

- Comparator condition: TBI group; placebo given

(n = 2)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 2)

Rayner et al. (2010)

Up to Date: December 2009

Protocol: Yes

Meta-Analysis

Study Design: RCT (cluster and cross-over were eligible); depression as primary outcome

Age: > 18yrs

Depression: MDD, adjustment disorder, dysthymic disorder using standardised criteria/ score on validated tools; any severity

Intervention: anti-depressant prescribed for depression

Comparator: placebo

# Studies: 1

Study Designs:

- 1 × RCT

Sample Size: 52

Sex: 22/52 male; 42%

Age: M 46.8 – 51.5

% TBI: 100%

SSRI

Sertraline (n = 1)

- Comparator condition: TBI group; placebo given

(n = 1)

Depression

I² = 54.07–78.27%^u

Fann et al. (2009a)

Up to Date: NR

Protocol: NR

Narrative Synthesis

- Study design: depression as a primary or secondary outcome; quantitative pre/post scores provided

- TBI: any severity; reported separately from other ABI

- Intervention: any treatment modality

# Studies: 11

Study Designs:

- 2 × RCT

- 3 × Non-RCT w control group

- 6 × Non-RCT w/out control group

Sample Size: 243

Sex: 13/243 male; 230/243 NR

Age: NR

% TBI: 100%

DSM MDD

III (n = 2)

III-R (n = 3)

IV (n = 5)

HAM-D

> 17 (n = 3)

≥ 17 (n = 1)

≥ 18 (n = 1)

MAOI

Moclobemide (n = 1)

Phenelzine (n = 1)

SNRI

Milnacipran (n = 1)

SSRI

Citalopram (n = 1)

Sertraline (n = 4)

Stimulant

Methylphenidate (n = 1)

TCA

Amitriptyline (n = 2)

Desipramine (n = 1)

Combination Therapy

SSRI + Anti-convulsant (n = 1;

Citalopram

Carbamazepine)

- Comparator condition: TBI group; placebo given

(n = 3)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 6)

N/A – no meta- analysis

Hardy (2009)

Up to Date:

July 2008

Protocol: NR

Narrative Synthesis

Study design: controlled trials, case series, case reports

Age: older adults

TBI: any

Depression: any

# Studies: 1

Study Designs:

- 1 × RCT

Sample Size: 30

Sex:

Age: M 34yrs

% TBI: 100%

Stimulant

Methylphenidate (n = 1)

- Comparator condition: TBI group; placebo given

n = 1

N/A – no meta- analysis

Warden et al. (2006)

Up to Date:

October 2004

Protocol: NR

Narrative Synthesis

# Studies: 6

Study Designs:

- 1 × RCT

- 2 X non-RCT w control group

- 3 X non-RCT w/out control group

Sample Size: 119

Sex:

Age: NR

% TBI: 100%

DSM

III (n = 1)

III-R (n = 2)

IV (n = 1)

(n = 2)

SSRI

Sertraline (n = 1)

TCA

Amitriptyline (n = 2)

Desipramine (n = 1)

MAOI

Moclobemide (n = 1)

Combination Therapy

SSRI + Anti-convulsant (n = 1;

Citalopram

Carbamazepine)

- Comparator condition: TBI group; placebo given

(n = 1)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 3)

N/A – no meta- analysis

Comper et al. (2005)

Up to Date: 2003

Protocol: NR

Narrative Synthesis

Age: 16-65yrs

TBI: mTBI; mixed severity with mTBI separately reported; ≤ 5yrs post injury

Intervention: any in human TBI populations

Excluded: case series and case studies

# Studies: 3

Study Designs:

- 2 × Non-RCT w control group

- 1 × Non-RCT w/out control group

Sample Size: 63

Sex: 47/63 male; 74.6%

Age: M 30.7 – 44.2

% TBI: 100%

MAOI

Phenelzine (n = 1)

SSRI

Sertraline (n = 1)

TCA

Amitriptyline (n = 2)

- Comparator condition: non-TBI controls; study drug given

(n = 2)

- No comparator condition

(n = 1)

N/A – no meta- analysis

Deb and Crownshaw (2004)

Up to Date: January 2003

Protocol: NR

Narrative Synthesis

Age: majority of sample > 16yrs

TBI: any

Depression: any

Intervention: any drug that may affect behaviour directly or indirectly

Excluded: studies using non-psychotropic drugs acute post injury

# Studies: 3

Study Designs:

- 1 × non-RCT w control group

- 1 × non-RCT w/out control group

- 1 × Unclear

Sample Size: 61

Sex:

Age: M 27 – 41.9^v (NR; n = 1)

% TBI: 100%

DSM

III-R MDD (n = 1)

(n = 2)

SSRI

Sertraline (n = 1)

MAOI

Moclobemide (n = 1)

Combination Therapy

SSRI + Anti-convulsant (n = 1;

Citalopram

Carbamazepine)

- No comparator condition

(n = 3)

N/A – no meta- analysis

Non-RCT w/out control group includes open trials, cohort studies, case series

ABI Acquired Brain Injury, BDI Beck Depression Inventory, DSM Diagnostic and Statistical Manual of Mental Disorders (III – 3^rd Edition; III-R – 3^rd Edition, Revised; IV – 4^th Edition; IV-SCID – Structured Clinical Interview for DSM-IV; IV-MINI – Mini-International Neuropsychiatric Interview), ES Effect Size, HAM-D Hamilton Rating Scale for Depression, ICU Intensive Care Unit, LoC Loss of Consciousness, M Mean, MAOI Monoamine Oxidase Inhibitors, MCI Mild Cognitive Impairment, MDD Major Depressive Disorder, MDE Major Depressive Episode, mTBI Mild Traumatic Brain Injury, n Number, NINDS National Institute of Neurological Disorders and Stroke, PHQ-9 Patient Health Questionnaire (9 Questions), SNRI Serotonin and Norepinephrine Reuptake Inhibitors, NR Not Reported, N/A Not Applicable, PTA Post-Traumatic Amnesia, PTSD Post-Traumatic Stress Disorder, R Range, RCT Randomized Controlled Trial, SD Standard Deviation, SSRI Selective Serotonin Reuptake Inhibitor, TBI Traumatic Brain Injury, TCA Tricyclic Antidepressant, TMS Transcranial Magnetic Stimulation

^aEligibility criteria had to be clearly provided in the methods section of the review

^bThe number of studies does not always equate to the number of interventions as some studies used multiple intervention groups

^cOnly primary studies that met the umbrella review’s inclusion criteria were extracted from each systematic review

^dThis refers to the overall sample size; including both the treatment group and control group

^e% of participants included in the extracted data that have TBI (as opposed to other non-TBI ABI)

^fThe category of ‘no comparator condition’ is used for single arm study such as cohort study

^gWe have only extracted the I² value as this was consistently reported across reviews. Some reviews reported other heterogeneity statistics that are not included in the table

^hLiu et al. (2019) only included the number of cases in the primary studies (i.e. did not report the number of healthy control participants)

ⁱLiu et al. (2019) included the entire mixed ABI sample from Turner-Stokes et al. (2002). Data from the TBI group was not reported separately

^jWe have extracted the sample size as stated in the abstract of this paper and Table 1. However, it is noted that the overall sample size should be 223. There were three studies with loss to follow-up. For two studies (Ashman et al., 2009; Fann et al., 2017), all participants enrolled in the study have been included in Table 1 and counted in the overall sample size. In contrast, for the remaining study (Ansari et al., 2014) with loss to follow-up, only those who completed the study have been included. As such, if the 9 people lost to follow-up are included in the overall sample size, this brings the sample size to 233. However, the Lee et al. (2005) study’s sample size should only be 20 participants as only 20 participants received sertraline or placebo interventions, with the remaining 10 participants receiving methylphenidate and therefore not being included in this review. With removal of these 10 participants, this bring the sample size back down to 223

^kAnsari et al. (2014) was not included in calculating participant average age, as review extracted frequency of age brackets with no M or SD provided

^lSlowinski et al. (2019) only included the number of cases in the primary studies (i.e. did not report the number of healthy control participants)

^mYue et al. (2017) included 12 studies. Two studies were included in a meta-analysis and the remaining 10 studies were only included in a narrative synthesis

ⁿYue et al. (2017) included the entire mixed ABI sample from Turner-Stokes et al. (2002). As Turner-Stokes et al. (2002) was not included in their meta-analysis, this did not impact the meta-analysis findings

^oMaksimowski and Tampi (2016) also made reference to the SSRI sertraline as the one included study (Lee et al., 2005) had two treatment arms; methylphenidate and sertraline. However, given the focus of their review was on stimulants, and the sertraline intervention group was considered another comparator group by Maksimowksi and Tampi (2016), we have only included methylphenidate in our extraction

^pBarker-Collo et al. (2013) pooled pharmacological and non-pharmacological interventions in their meta-analyses, with no separate meta-analyses conducted for pharmacological interventions only. As such, no pooled estimates could be extracted from this review

^qAnsari et al. (2014) not included in calculating participant average age, as review extracted frequency of age brackets with no M or SD provided

^sThe meta-analysis in Price et al. (2011) included other neurological disorders. This heterogeneity figure is therefore not specific to the TBI study only

^tWheaton et al. (2011) only included the sample size for the treatment group in Supplemental Table B. Given there were two RCTs with a control group it is not clear what the overall sample size is

^uThe meta-analyses in Rayner et al. (2010) included other physical illnesses. These heterogeneity figures are therefore not specific to the TBI study only

^vThe average age reported here from the Deb and Crownshaw (2004) review is based on only two of the three studies. No details about age were provided for the third study