Su 2008.

Methods	RCT: parallel: NCT00618865
Participants	36 women randomised Inclusion criteria: pregnant women aged 18‐40 years, with major depressive disorder (DSM‐IV) with onset between 16 weeks and 32 weeks GA; and to not have taken psychotropic agents for at least 1 month, to have a score of at least 18 on the 21‐item HAM‐D at screening; and to have good physical health as determined by medical history, physical examination, blood laboratory results, electrocardiogram, chest radiography and urinalysis. No psychotropic agents were given during the study period. Exclusion criteria: DSM‐IV diagnosis of bipolar disorder, psychotic disorder, or substance abuse/dependence or any Axis II diagnosis of borderline or antisocial personality disorder. Setting: China Medical University Hospital, Taiwan (June 2004 to June 2006).
Interventions	SUPPLEMENTATION: DHA + EPA versus placebo Group 1: omega‐3 LCPUFA (3.4 g/day; 2.2 g EPA and 1.2 g DHA) (produced from Menhaden fish body oil concentrate). Total number randomised: n = 18 (13 completed) Group 2: placebo: 5 identical gelatin capsules per day (olive‐oil ethyl‐esters). Total number randomised: n = 18 (11 completed) Timing of supplementation: 8 weeks to ˜30‐32 weeks GA Before random assignment, all consenting participants took part in a placebo trial for 1 week – those showing a decrease of 20% or more in HAM‐D scores (placebo responders) were not permitted to proceed to the randomisation stage. All capsules (fish oil and placebo) were vacuum deodorised, amended by blending with orange flavour and supplemented with tertiary butylhydroquinone (0.2 mg/g) and tocopherols (2 mg/g) as antioxidants DHA + EPA dose/day: high: 1.2 g DHA + 2.2 g EPA
Outcomes	Women/birth: HAM‐D (every other week); EPDS (Taiwanese version); Beck Depression Inventory and brief adverse effect checklist at week ‐1 (lead‐in period), week 0 (baseline) and weeks 2, 4, 6 and 8; blood samples taken at week 1 and week 8 for omega‐3 PUFA analysis (EPA; DHA) “No obstetric complication was noted in any participant” Babies/infants/children: “all the newborns were normal in general physical and neurobehavioral examination at birth”
Notes	Funding: National Science Council, Department of Health, and China Medical University and Hospital, Taiwan Declarations of interest: none reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Placebo used
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Not reported, but probably done.
Incomplete outcome data (attrition bias) All outcomes	High risk	The intention‐to‐treat population included all women who had a baseline and at least 1 post baseline observation. The per protocol population included all women who completed 8 weeks of treatment. 12/36 (33%) lost to follow‐up: Omega‐3: 5/18 (28%) lost to follow‐up: 2 did not return 3 ‘unsatisfactory response’ Placebo: 7/18 (39%) lost to follow‐up: 2 did not return 3 ‘unsatisfactory response' 1 severe nausea
Selective reporting (reporting bias)	High risk	Only depressive symptoms, adverse events and omega‐3 status assessed, e.g. no birth outcomes reported numerically.
Other bias	Unclear risk	Baseline characteristics similar between study groups. Not clear what constituted “unsatisfactory response” (3 women in each group).