ADCS 2010

Methods	Alzheimer's Disease Cooperative Study (ADCS) RCT, parallel, (n‐3 DHA vs n‐6 LA), 18 months Summary risk of bias: low
Participants	Individuals with mild to moderate Alzheimer's disease N: 238 intervention, 164 control Level of risk for CVD: low Men: 52.9% intervention, 40.2% control Mean age in years (SD): 76 (9.3) intervention, 76 (7.8) control Age range: unclear Smokers: 24.4% intervention, 21.9% control Hypertension: not reported Medications taken by at least 50% of those in the control group: cholinesterase inhibitor, memantine Medications taken by 20%‐49% of those in the control group: none Medications taken by some, but less than 20% of the control group: none Location: USA Ethnicity: not reported
Interventions	Type: supplement (capsule) Comparison: DHA vs omega 6 Intervention: 2 × 1 g algal‐derived DHA capsules (Neuromins) per day for a total daily dose of 2 g, each capsule contain 45% to 55% of DHA and does not contain EPA (950 mg soft‐gel capsules which contain approximately 510 mg DHA). Dose: +DHA 1.02 g/d. Control: 2 × 1 g placebo capsules per day (made up of corn or soy oil) Compliance: measured by pill counts at every visit Length of intervention: 18 months
Outcomes	Main study outcome: change in the cognitive subscale of the Alzheimer's Disease Assessment Scale (ADAS‐cog) and change in the Clinical Dementia Rating (CDR) Dropouts: 67 intervention, 40 control (discontinued treatment but included in main analyses) Available outcomes: mortality, measures of cognition, baseline & change in plasma DHA, adverse events Response to contact: no data provided
Notes	Study funding; quote: "grant UO1‐AG10483 from the National Institute on Aging. The National Institute on Aging was not otherwise involved in the design and conduct of the study, or in the analysis of data or preparation of the manuscript". "The placebo and DHA study drugs were provided by Martek Biosciences. Martek also provided plasma and cerebrospinal fluid measurements of fatty acids, as well as partial financial support for the magnetic resonance imaging sub study. (Martek Biosciences produces nutritional supplements from cultivated fungi and microalgae). Martek employees participated in design of the study and in revision of the manuscript, but were not involved in data management or data analysis." (Quinn 2010, p. 1910).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was achieved with a centralised interactive voice response system, using a block design with a block size of 5 (3 in the DHA group and 2 in the placebo group.
Allocation concealment (selection bias)	Low risk	Randomisation was achieved with a centralised interactive voice response system, using a block design with a block size of 5 (3 in the DHA group and 2 in the placebo group.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Placebo capsules (made up of corn or soy oil) were identical in appearance. The adequacy of blinding was assessed by questionnaires completed by caregivers, study coordinators, and site physicians.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	The adequacy of blinding was assessed by questionnaires completed by caregivers, study coordinators, and site physicians with results showing no difference between groups and the majority did not know.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Intention‐to‐treat analysis. At 12 months data were available for > 80% (ITT analysis)
Selective reporting (reporting bias)	Low risk	Prospectively registered February 2007, study started February 2007, completed May 2009. Primary outcomes were rate of change in ADAS‐Cog11 and CDR‐SOB, which are both reported in main report. NPI and ADL were secondary outcomes also reported.
Attention	Low risk	Both study arms had the same follow‐up and care.
Compliance	Unclear risk	Measured by pill count at every visit. 28% intervention and 24% control discontinued supplement with a minority discontinuing due to adverse events. A further 8% were excluded for < 80% compliance in both intervention and control arms.
Other bias	Low risk	None noted