Table 1 Summary of included studies

Asia

Studies linked to primary publication reported by Lee et al. [14]

Lee et al. [14], multinational (China, Hong Kong, Korea, Malaysia, Singapore, and Taiwan)

Patients aged ≥ 18 years who presented with new or first episode of MDD (DSM-IV-TR or ICD-10 definition) and were prepared to take ADT. CGI-S score ≥ 4 and at least 2 months free of depressive symptoms before onset of current episode. Age, mean (SD): 45.1 (14.1) years. Females, n (%): 625 (68.8)

Prospective, noninterventional, epidemiological observational study; 909 inpatients and outpatients enrolled from 40 practice sites in 6 East Asian countries from June 14, 2006, to February 15, 2007, and followed for 3 months: China (n = 299), Hong Kong (n = 90), Korea (n = 198), Malaysia (n = 98), Singapore (n = 30), and Taiwan (n = 194)

EQ-5D-3L^a utility scores were reported at baseline and at 6 months stratified by the presence of PPS

Ang et al. [11], multinational (China, Hong Kong, Korea, Malaysia, Singapore, and Taiwan)

Baseline EQ-VAS scores stratified by the prescribed intervention (SSRI or SNRI) and change from baseline in EQ-VAS stratified by prescribed intervention and presence of PPS were reported

Lee et al. [13], Korea subanalysis

Subgroup analysis of 198 Korean patients from study conducted in Lee et al. [14]

Baseline EQ-5D-3L^a utility scores and change from baseline at 6 months were reported, stratified by the presence of PPS

Chen et al. [12], Taiwan subanalysis

Subgroup analysis of 194 Taiwanese patients from study conducted in Lee et al. [14]

Baseline EQ-5D-3L^a utility scores and change from baseline at 3 months were reported, stratified by the presence of PPS and prescribed intervention (SNRI or SSRI)

Li et al. [15], China subanalysis

Subgroup analysis of 299 Chinese patients from study conducted in Lee et al. [14]

EQ-VAS scores were reported at baseline and at 3 months for the overall population, stratified by the presence of PPS and prescribed intervention (SSRI or SNRI)

Novick et al. [17], multinational (China, Hong Kong, Malaysia, Singapore, South Korea, and Taiwan)

Subgroup analysis of 426 patients who started ADT at the baseline visit and had information on adherence during the follow-up period in the study conducted in Lee et al. [14]

EQ-5D-3L^a utility scores were reported at baseline and 3 months for patients with clinically reported adherence or nonadherence to ADT

Novick et al. [16], China subanalysis

Subgroup analysis of 300 Chinese patients from study conducted in Lee et al. [14]

EQ-5D-3L^a utility and EQ-VAS scores were reported at baseline and at 3 months for the overall population; scores stratified by the presence of PPS were reported at 3 months

Studies linked to primary publication reported by Kuga et al. [19]

Kuga et al. [19], Japan

Patients ≥ 20 years with at least moderate depression (QIDS ≥ 16) and at least moderate PPS (BPI-SF average pain ≥ 3) presenting with an episode of MDD without psychotic traits, defined by the DSM-IV. Age, mean (SD): 42.9 (14.6) years. Female: 51.2%

Prospective, observational, 12-week study conducted at 39 sites, including psychiatry and psychosomatic outpatient/inpatient clinics/hospitals. 523 patients diagnosed by the investigator were recruited, with patient visits occurring between February 13, 2014, and February 26, 2016

Baseline EQ-5D^b utility scores and change from baseline at 2, 4, 6, and 12 weeks were reported, stratified by the prescribed intervention (duloxetine or SSRIs)

Kuga et al. [18], Japan

Subgroup analyses of study conducted in Kuga et al. [19]

EQ-5D^b utility scores stratified by the prescribed intervention (duloxetine or SSRIs) were reported at week 12 for the following subgroups (at baseline): number of MDEs, BPI-SF average pain score, and HAMD-17 total score

Unique studies

Kim et al. [20], South Korea

Patients aged 19–65 years with a diagnosis of MDD according to the DSM-IV, confirmed by the Mini-International Neuropsychiatric Interview and started on ADT monotherapy either as first-line therapy or as first treatment switch from previous ADT. Age, mean (SD): 45.2 (13.1) years. Female: 74.0%

PERFORM-K was an observational, cross-sectional, multisite study. 343 patients were recruited from 29 psychiatric departments in university or general hospitals throughout South Korea between October 2013 and January 2014

Baseline EQ-5D^b utility scores were reported, stratified by severity of depression (MADRS score 0–25, 26–29, 30–34, and 35–60) and severity of perceived cognitive dysfunction (PDQ-D score 0–12, 13–27, 28–43, and 44–80)

Husain et al. [21], Pakistan

Patients aged 18–65 years with DSM-5 MDE that had failed to respond to ≥ 2 ADTs. Age, median (IQR): 40 (30–46) years. Male, n (%): 11 (55)

Multisite, 12-week, double-blind, placebo-controlled pilot trial. 41 patients were recruited from outpatient psychiatric clinics at Abbasi Shaheed Hospital, Karwan-e-Hayat Hospital, Civil Hospital, and the Institute of Behavioural Sciences between October 2014 and March 2016

EQ-VAS scores stratified by the prescribed intervention (minocycline + TAU or placebo + TAU) were reported at baseline and at 12 weeks

Europe

Studies linked to primary publication reported by Montgomery et al. [22]

Montgomery et al. [22], multinational (Austria, Belgium, Bulgaria, Czech Republic, Estonia, Germany, Italy, Lithuania, Poland, Romania, Russia, Spain, Sweden, and UK)

Patients aged ≥ 18 years and ≤ 75 years with single-episode MDD or recurrent MDD (and a current MDE < 12 months) according to DSM-IV-TR and a MADRS total score ≥ 22 and item 1 score ≥ 3. Eligible patients had an inadequate response to a SSRI/SNRI monotherapy at approved doses for ≥ 6 weeks prior to the screening visit. Age, mean (SD): 47 (12) years. Female, n (%): 195 (77.1)

Double-blind, randomized, flexible-dose, active comparator study including 495 patients recruited from 71 psychiatric inpatient and outpatient settings in 14 countries from January 2012 to December 2012. Patients were recruited via advertisements (in Austria, Germany, Estonia, Russia, Sweden, and UK) or referrals from primary care physicians

EQ-VAS scores at baseline and 8-week change from baseline were reported, stratified by prescribed intervention (vortioxetine or agomelatine)

Papakostas et al. [23], multinational (Austria, Belgium, Bulgaria, Czech Republic, Estonia, Germany, Italy, Lithuania, Poland, Romania, Russia, Spain, Sweden, and UK)

Subgroup analyses of study conducted in Montgomery et al. [22]

EQ-VAS scores at baseline and 8-week change from baseline were reported, stratified by prescribed intervention (vortioxetine or agomelatine) and previous treatment (SSRI or SNRI)

Studies linked to primary publication reported in Garcia-Cebrian et al. [24]

Garcia-Cebrian et al. [24], multinational (Austria, Belgium, France, Germany, Ireland, Italy, Netherlands, Norway, Portugal, Sweden, Switzerland, and UK)

Patients aged ≥ 18 years with clinical diagnosis of depression by their physician and about to start ADT for a first or subsequent episode of depression. Age, mean (SD): 46.8 (14.7) years Female, 68.2%. Duration of depressive illness, mean (SD), 8.5 (10.4) years, Duration of current episode, mean (SD): 13.6 (16.5) weeks

European, prospective, observational study (FINDER). 3515 patients presenting during the normal course of care were enrolled by 437 primary care physicians or specialists (mostly psychiatrists) from 12 countries between May 2004 and September 2005. Data were collected at baseline, 3 months, and 6 months during routine visits

EQ-5D-3L^a utility and EQ-VAS scores were reported at baseline

Reed et al. [25], multinational (Austria, Belgium, France, Germany, Ireland, Italy, Netherlands, Norway, Portugal, Sweden, Switzerland, and UK)

Subsequent analysis of study reported in Garcia-Cebrian et al. [24]

EQ-5D-3L^a utility and EQ-VAS scores were reported at baseline, 3 months, and 6 months for patients receiving ADT

Unique studies

Kuyken et al. [27], UK

Patients aged ≥ 18 years with recurrent MDD in full or partial remission according to DSM-IV currently on a therapeutic dose of ADT. Patients had ≥ 3 previous MDEs in which depression was the primary disorder and was not secondary to substance abuse, bereavement, or a general medical condition

MBCT-TS:

 Female, n (%): 151 (71)

 Age, mean (SD): 50 (12) years

ADM:

 Female, n (%): 174 (82)

 Age, mean (SD): 49 (13) years

Economic evaluation alongside a clinical trial. Two-arm, multicenter, single-blind superiority trial. Patients identified through physician and telephone assessment. 95 patients recruited from primary care in urban and rural settings in 4 UK centers: Bristol, Exeter and East Devon, North and Mid Devon, and South Devon

EQ-5D-3L utility scores were reported at baseline and at 1, 9, 12, 18, and 24 months by prescribed intervention (MBCT-TS or maintenance ADM)

Serfaty et al. [29], UK

Patients ≥ 65 years with a primary diagnosis of depressive disorder made by the researcher who administered the Geriatric Mental State and History and Etiology Schedule and a BDI-II score of ≥ 14. Age, mean (SD): 75 (7.1) years. Female, n (%): 50 (74.6)

Single-blind, randomized controlled trial that took place between April 2004 and September 2007. Patients were recruited by self-referral, primary care referral, and by database searches. Patients who scored ≥ 5 on GDS-15 were offered a further interview to see whether they satisfied entry criteria. 204 patients were randomized

EQ-5D-3L utility scores were reported at baseline, 4, and 10 months by prescribed intervention (CBT + TAU, TC + TAU, or TAU)

Morriss et al. [28], UK

Patients ≥ 18 years with persistent moderate or severe primary unipolar depression with a current MDE (DSM-IV), met 5 of 9 NICE criteria for symptoms of moderate depression, had HAMD-17 ≥ 16, and had GAF ≤ 60. Age, mean (SD): 46 (11.3) years. Female n (%): 60 (64%)

Economic evaluation alongside a clinical trial. Patients currently under the care of a secondary care mental health team were recruited from 3 sites (Cambridge, Derby, and Nottingham) for a multicenter, single-blind, patient-level, parallel, randomized controlled trial. 187 patients were randomized

EQ-5D-3L utility scores were reported at baseline and at 6, 12, and 18 months by prescribed intervention (TAU or SDS)

Sapin et al. [30], France

Patients aged 18–92 years who consulted a primary care physician for a new episode of MDD according to DSM-IV and had not previously received ADT. Age, mean (SD): 44.2 (14.1) years. Sex ratio (males/females): 0.4. MADRS score, mean (SD): 32.7 (7.7)

National, multicenter, prospective, noncomparative cohort study with a scheduled follow-up period of 2 months. 250 outpatients were enrolled by 95 physicians between May and November 2002

EQ-5D-3L utility scores were reported for the total population at baseline, and at baseline, 4 weeks and 8 weeks, stratified by clinical response, (responder remitters, responder nonremitters, and nonresponders) according to changes in MADRS scores

Fernandez et al. [26], multinational (Europe)

Patients aged 18–85 years with moderate to severe MDD (DSM-IV), without suicidal tendencies, and a MADRS total score ≥ 18 at screening 1 week before start of ADT and at start of ADT

Venlafaxine:

 Female n (%): 89 (71.2)

 Age, mean (SD): 46.5 (13.5) years

Escitalopram:

 Female n (%): 95 (75.4)

 Age, mean (SD): 48.4 (14.7) years

Economic evaluation alongside a clinical trial reported by Montgomery et al. [22]. 293 outpatients recruited for a randomized, double-blind, flexible-dose, clinical trial across 8 European countries

EQ-5D-3L^a utility scores were reported by prescribed intervention (escitalopram or venlafaxine) at baseline and 8 weeks

Saragoussi et al. [31], multinational (France, Germany, Spain, Sweden, and UK)

Patients aged 18–65 years with MDD according to DSM-IV initiating or switching to an ADT. Age, mean (SD): 44.3 (12.0) years. Female: (73.2%)

2-years, multicenter, prospective, noninterventional cohort study (PERFORM) that enrolled 1159 outpatients by either a primary care physician or a psychiatrist at 194 sites in France, Germany, Spain, Sweden, and the UK

EQ-5D-3L^a scores were reported at baseline, and at 2, 6, 12, 18, and 24 months for patients receiving antidepressant monotherapy or undergoing first switch of ADT

The Americas

Soares et al. [32], multinational (Argentina, Chile, Colombia, Mexico, and US)

Postmenopausal women aged 40–70 years with a primary diagnosis of MDD based on MINI and DSM-IV criteria, and depressive symptoms for ≥ 30 d before screening visit and a MADRS total score of ≥ 22 at screening and baseline, with a ≤ 5-point improvement from screening to baseline. Age, mean (SD): 55 (6) years in escitalopram double blind/desvenlafaxine OL cohort and 54 (6) years in desvenlafaxine DB/desvenlafaxine OL

Randomized, phase 3b, parallel group, comparator-controlled, multicenter study with an 8-week, double-blind acute phase followed by a 2-arm, 6-month extension phase conducted from December 2006 to September 2008. 607 patients were enrolled from 72 sites across Argentina, Chile, Colombia, Mexico, and the US and randomized in the acute phase. 129 patients entered the open-label extension phase

Change in EQ-5D-3L^a utility scores from acute-phase baseline and extension-phase baseline were reported

Revicki and Wood [33], multinational (Canada and US)

Patients aged 18–65 years with DSM-III-R diagnosis of MDD based on clinician interview using the SCID and completed at least 8 weeks of ADT or had completed a regimen within the last 2 months. Patients with dysthymia were allowed. N = 70. Age, mean (SD): 42 (11) years; male: 23%; married: 48%; HAMD-17 score, mean (SD): 11.65 (8.2)

Prospective observational study, 70 outpatients enrolled from a university family practice clinic in Toronto or a community-based primary care practice in San Diego; Canada (n = 40) and US (n = 30)

Standard gamble utility scores were reported for the total population and for a series of hypothetical health states based on HAMD-17 depression severity: patients with severe depression, untreated; patients with moderate depression, mild depression, and in remission, all stratified by intervention received (nefazodone, fluoxetine, or imipramine); patients in remission receiving no treatment

Multiregional

Studies linked to primary publication reported by Duenas et al. [34]

Duenas et al. [34], multinational (China, Hong Kong, Malaysia, Philippines, Taiwan, Thailand, Singapore, Saudi Arabia, United Arab Emirates, Mexico, Israel, and Austria)

Outpatients aged ≥ 18 years with a primary diagnosis of MDD (excluding treatment-resistant depression) according to ICD-10 or DSM-IV-TR criteria who were at least moderately depressed (CGI-S score ≥ 4), presenting with an episode of MDD and initiating, or switching to, any available SSRI or SNRI. Patients were sexually active without sexual dysfunction (sexual dysfunction defined as ASEX total score ≥ 19, ASEX score ≥ 5 on any item, or ASEX score ≥ 4 on any 3 items). Age, mean (SD): 38.2 (10.3), years. Female, n (%): 319 (51.8)

6-month, prospective, observational study. 1659 patients were enrolled from 89 sites in 12 countries between November 15, 2007, and November 28, 2011; China (n = 199), Hong Kong (n = 17), Malaysia (n = 33), Philippines (n = 110), Taiwan (n = 180), Thailand (n = 8), Singapore (n = 2), Saudi Arabia (n = 168), United Arab Emirates (n = 116), Mexico (n = 581), Israel (n = 9), and Austria (n = 42)

Change from baseline in EQ-5D-3L^a utility scores and EQ-VAS at 8 weeks were reported by prescribed intervention (duloxetine or SSRI) for all patients and stratified for patients with and without TESD

Duenas et al. [35], multinational (China, Hong Kong, Malaysia, Philippines, Taiwan, Thailand, Singapore, Saudi Arabia, United Arab Emirates, Mexico, Israel, and Austria)

Change from baseline in EQ-5D-3L^a utility scores and EQ-VAS at 6 months were reported by prescribed intervention (duloxetine or SSRI) for all patients and stratified for patients with and without TESD

Hong et al. [37], multinational (China, Hong Kong, Malaysia, Philippines, Taiwan, Thailand, Singapore, Saudi Arabia, United Arab Emirates, Mexico, Israel, and Austria)

Post-hoc analysis of the study reported in Duenas et al. [34]

EQ-5D-3L^a utility scores and EQ-VAS scores were reported at baseline, 8 weeks, 16 weeks, and 24 weeks for the total population and PPS + and PPS − populations, stratified by the prescribed intervention (duloxetine or SSRI)

Hong et al. [36], East Asia subanalysis (China, Hong Kong, Malaysia, the Philippines, Taiwan, Thailand, and Singapore)

Post-hoc subgroup analysis of 587 East Asia patients from study conducted in Hong et al. [37]; China (n = 205), Hong Kong (n = 18), Malaysia (n = 33), the Philippines (n = 113), Taiwan (n = 199), Thailand (n = 17), and Singapore (n = 2)

EQ-5D-3L^a utility scores and EQ-VAS scores were reported at baseline, stratified by the prescribed intervention (duloxetine or SSRI), and at 24 weeks, stratified by the prescribed intervention and presence of PPS

Unique studies

Florea et al. [38], multinational

MDD patients aged ≥ 18 years who received the approved doses of vortioxetine 5, 10, 15, and 20 mg/d

Across-study comparison of HRQoL included in 5 short-term (6–8 week) studies (NCT00672958, NCT00735709, NCT00635219, NCT00839423, NCT01140906) and 1 dedicated study in elderly patients (NCT00811252)

EQ-5D-3L^a health state score change from baseline at 6 weeks for different doses of vortioxetine versus placebo