Berson 2004.

Methods	RCT, parallel, (n‐3 DHA vs n‐6 LA), 48 months Summary risk of bias: low
Participants	People with retinitis pigmentosa aged 18‐55 years N: 221 randomised overall, analysed 105 intervention, 103 control Level of risk for CVD: low Men: 48% intervention, 54% control Mean age in years (SD): 37.8 (6.5) intervention, 36.0 (7.2) control Age range: unclear (18‐55 inclusion criterion) Smokers: not reported Hypertension: not reported Medications taken by at least 50% of those in the control group: vitamin A Medications taken by 20%‐49% of those in the control group: multivitamins Medications taken by some, but less than 20% of the control group: not reported Location: USA Ethnicity: unclear (6% of the study population were minorities)
Interventions	Type: supplement (DHA capsules) Comparison: DHA vs omega 6 Intervention: 6 × 500 mg capsules/d of DHA (1.2 g/d DHA plus 1.8 g vegetable oil) plus < 0.0006 mg/d tocopherols plus 15,000 IU retinyl palmitate (vitamin A). Dose: +1.2 g/d DHA Control: 6 × 500 mg capsules/d of soy and corn oils (half each) with 120 mg/d ALA, plus < 0.0006 mg/d tocopherols plus 15000 IU retinyl palmitate (vitamin A) Compliance: 92% of capsules taken by both intervention and control groups (assessed by monthly calendars), Plasma DHA much higher in intervention than control Length of intervention: 48 months
Outcomes	Main study outcome: retinal degeneration Dropouts: 5 or 6 intervention, 7 or 8 control Available outcomes: mortality, cancer diagnoses, lipids, eyesight Response to contact: yes (no data provided)
Notes	Study funding: National Eye Institute and Foundation Fighting Blindness
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random numbers
Allocation concealment (selection bias)	Low risk	Random numbers available only to programmer who provided assignments to data manager, all staff in contact with patients were masked to group assignment
Blinding of participants and personnel (performance bias) All outcomes	Low risk	States that all staff in contact with participants were masked to group assignment, as were participants. However no information was provided regarding the taste, smell and appearance of the active and placebo capsules
Blinding of outcome assessment (detection bias) All outcomes	Low risk	All assessments were performed blind to study allocation. Each ocular examination was performed without review of previous records. All serum samples were analysed without knowledge of treatment group assignment.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Numbers of dropouts and reasons for dropouts not stated. 221 participants randomised, data presented on 208 participants
Selective reporting (reporting bias)	Unclear risk	No trials registry entry or protocol found.
Attention	Low risk	Staff in contact with patients were masked, so unable to bias time, etc.
Compliance	Low risk	92% of capsules taken by both intervention and control groups (assessed by monthly calendars), Plasma DHA much higher in intervention than control
Other bias	Low risk	None noted