Brignole‐Baudouin 2011.

Methods	Study design: randomized, parallel‐group, controlled trial Study site(s): multi‐center (9 sites) Number randomized (total and per group): 138 participants in total Unit of randomization (individual or eye): individual Exclusions after randomization: 1 participant (unknown group) discontinued; 5 participants withdrew (2 due to adverse events; 1 lack of improvement; 2 withdrew consent) in the treatment group; 4 participants withdrew (2 due to adverse events; 1 because of their worsening condition; 1 withdrew consent) in the control group Losses to follow‐up: 1 participant in the treatment group Unit of analysis (individual or eye): individual (worse eye) Number analyzed: 58 participants in the treatment group; 63 participants in the control group (15 participants did not have 2 evaluable conjunctival impression cytologies, for reasons of early participant withdrawal, or lost or impaired samples) Reported power calculation? (Y/N): Y (80% power) Reported subgroup analysis? (Y/N): N
Participants	Baseline characteristics Countries: France and Italy Age (mean ± SD, range): 60 ± 11.75 years in the omega‐3 treatment group; 59.7± 11.95 years in the control group Gender: 1 man and 57 women in the omega‐3 treatment group; 3 men and 60 women in the control group Inclusion criteria: 1. Scores for at least 2 of the following 4 objective tests: Schirmer test values < 10 mm/5 min; TBUT values < 10 s; corneal fluorescein staining score ≥ 1 and < 4; lissamine green by van Bijsterveld score > 3 and < 6 2. Score of at least 1 for at least 2 of the 5 following subjective tests (scored from 0 to 3): foreign body sensation; dryness; burning; stinging; photophobia 3. For those having systemic treatment, no change in treatment for at least a month before inclusion 4. Written informed consent Exclusion criteria: 1. Severe dry eye defined as lissamine green > 6 or corneal staining ≥ 4 2. Uncontrolled inflammatory disease 3. Drastic change in food and/or food supplements within the last month 4. Other food supplement with eicosapentaenoic acid and docosahexaenoic acid 5. Evidence of acute ocular infection and/or intraocular inflammation within 1 month before the start of this study 6. Ocular surgery within the last 6 months 7. Change in ocular treatment within the last month 8. Patients currently using any ophthalmic medication including any ocular ointment except artificial tear preparation and eye cleaning solution for treatment of dry eye syndrome 9. Patients treated with topical ocular, steroidal or non‐steroidal, anti‐inflammatory treatment within the last month 10. Patients treated with ocular topical cyclosporine within the last month 11. Occlusion therapy with lacrimal or punctum plugs within the last 3 months 12. Patients currently wearing contact lenses 13. Pregnant or lactating women Equivalence of baseline characteristics? (Y/N): Y
Interventions	Intervention #1 (treatment group): oral capsule containing fish oil (omega‐3 fatty acids, average of 285 mg including eicosapentaenoic acid 142.5 mg and docosahexaenoic acid 95 mg) and omega‐6 average of 5 mg; Medilar, Bausch and Lomb, R. P. Scherer GmbH & Co., 3 capsules daily Intervention #2 (control group): placebo capsule containing medium‐chain triglycerides, 3 capsules daily (daily dose: 575 mg/d) Length of follow‐up: 3 months Notes: participants were instructed to take 3 capsules daily during meals with a glass of water; daily dosage of 855 mg of omega‐3 fatty acids and 15 mg of omega‐6 fatty acids
Outcomes	Primary outcome(s): "reduction in" HLA‐DR expression (%) Secondary outcome(s): change from baseline in HLA‐DR arbitrary units of fluorescence (AUF); global subjective dry eye symptoms score (foreign body sensation, dry eye sensation, burning, stinging, and photophobia); subjective dry eye score for each symptom; Schirmer test; TBUT; corneal fluorescein staining; conjunctival lissamine green staining; QOL questionnaire; frequency of artificial tear usage (ie, "number of daily instillations of tear substitutes") Safety endpoints: occurrence of adverse events; changes in external eye exam and biomicroscopy Adverse events reported? (Y/N): Y Measurement time points (specify intervals at which outcomes were assessed): baseline, 6 weeks, 3 months Other issues with outcome assessment (eg, quality control for outcomes, if any): none
Notes	Study dates: not reported Funding source(s): "this study was sponsored and funded by Bausch and Lomb Inc Montpellier, France" Conflicts of interest: 1 author is affiliated with a pharmaceutical firm Publication language: English Registered on clinical trials registry? (Y/N): N
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"The treatments were consecutively numbered and allocated to study participants in ascending order using the next available consecutive number from a randomization list (1:1) established prior to study enrolment (sas® 8.2, Cary, NC, USA) on Solaris 2.8 (Redwood Shores, CA, USA) and Framemaker 5.5.6 (San Jose, CA, USA)"
Allocation concealment (selection bias)	Unclear risk	Allocation concealment was not reported
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"Double‐masked" study "The treatments were identical in appearance, packaging and treatment regimen for the placebo to preserve masking" "The Investigator, patients, and Bausch and Lomb personnel involved in the monitoring or conduct of the study were blinded to the study drug codes. Study drug codes were not available to the above personnel until after the study was completed and the database was finalized"
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"Double‐masked" study "The treatments were identical in appearance, packaging and treatment regimen for the placebo to preserve masking" "The Investigator, patients, and Bausch and Lomb personnel involved in the monitoring or conduct of the study were blinded to the study drug codes. Study drug codes were not available to the above personnel until after the study was completed and the database was finalized"
Incomplete outcome data (attrition bias) All outcomes	High risk	"Analyses were performed on the full analysis set (FAS) and per‐protocol set (PPS). The FAS included all randomized patients who received the study treatment at least once, had at least one follow‐up visit available and had two assessable analyses of conjunctival IC (at day 0 and month 3)" "Fifteen patients did not have two evaluable conjunctival IC for reasons of early patient withdrawal or lost or impaired samples. Thus, only 121 patients were included in the FAS (n = 58 fatty acids; n = 63 placebo)" 15/138 (10.9%) patients who were randomized were not included in the analysis
Selective reporting (reporting bias)	Unclear risk	No access to study protocol or clinical trials registry
Other bias	High risk	One author is affiliated with a pharmaceutical firm