WAHA 2016.

Methods	The Walnut and Healthy Aging Study (WAHA) 2‐arm, parallel RCT (usual diet plus walnuts vs usual diet), 2 years Summary risk of bias: moderate to high
Participants	Middle‐aged healthy adults N: 362 intervention, 346 control (only preliminary data on 312 participants from one of the two centres is available) Level of risk for CVD: low Men: 32.6% intervention, 31.5% control Mean age in years (SD): 69.4 (3.8) intervention, 68.9 (3.5) control Age range: 63‐79 (inclusion criteria) Smokers: 4.4% intervention, 1.2% control Hypertension: 52.8% intervention, 52.9% control Medications taken by at least 50% of those in the control group: not reported Medications taken by 20%‐49% of those in the control group: not reported Medications taken by some, but less than 20% of the control group: not reported Location: Spain and USA Ethnicity: not reported
Interventions	Type: supplement (food) Comparison: ALA vs unclear Intervention: 15% of daily energy intake as walnuts. The estimated amount of walnuts ranged from about 30–60 g/day (1‐2 ounces). Sachets for daily consumption containing 30 g, 45 g, or 60 g of raw, pieced walnuts were provided as 8‐week allotments to be eaten daily, preferably as the raw product, either as a snack or by incorporating them into shakes, yogurts, cereals, or salads. To improve participants' compliance, 1‐kg extra walnut allowances were provided every 2 months to take into account family needs. Dose: ˜5 g/d ALA Control: usual diet without walnut Compliance: assessed by dietitians through FFQs, recount of empty packages, and changes in FAs concentrations. 95% consumed at least 30 g/d. The proportion of α‐linolenic acid in red blood cells increased in the walnut group by 0.16% (95% CI 0.14 to 0.18) and in the control group by 0.02% (95% CI −0.01 to 0.04; P < 0.001). No data on dietary intake provided. Length of intervention: 2 years (only 1 year results have partly been published)
Outcomes	Main study outcome: change in cognitive decline (results not yet published) Dropouts: 36 intervention, 21 control (after 1 year) Available outcomes: lipids (for TG and HDL only data states "no between diet differences were observed"), weight (waist circumference was provided but without variance, abstract stated that "there were no significant changes in body fat and waist‐to‐hip ratio over time and between the two groups"). Authors provided data on mortality, CVD events, cancer deaths and diagnoses, IBD diagnosis (no CVD deaths). Cognitive, ophthalmological, inflammatory markers, glycaemic status and other outcomes are not yet available. Response to contact: authors provided additional outcome and methodology data.
Notes	Study funding: Calfornia Walnut Commission The 2‐year results as well the full 1‐year results are yet to be published. Outcome data reported are for only for participants from one centre (USA)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "randomized to either the control or walnut group using a computerized random number table with stratification by center, sex, and age range. Couples entering the study were treated as one number and were randomized into the same group".
Allocation concealment (selection bias)	Low risk	Author reply states, "Baseline subject data was collected before randomization. Randomization was done by the clinician, pressing the key on the computer. Since this was a dual center (Barcelona and Loma Linda) trial, a single computer software randomized participants for both the centers."
Blinding of participants and personnel (performance bias) All outcomes	High risk	Single blind. "An unavoidable limitation of the study is not being able to blind participants to the intervention since it consists of a whole food" Rajaram 2017.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Author reply states "Study personnel not in contact with the subjects were blind to the treatment assignment. So (lab technicians, ophthalmology technician, neuro cognitive testers) were not aware of the treatment assignment. Of course clinicians who were visited by subjects every two months, knew the treatment assignment". This suggests that allocation was known by physicians, so high risk for event data
Incomplete outcome data (attrition bias) All outcomes	Low risk	38/362 dropouts in intervention group = 10.5%. 34/346 dropouts in control group = 9.8%. Similar dropout in groups over 2 years.
Selective reporting (reporting bias)	Unclear risk	Although prospectively registered, no full results paper published – results from conference abstracts only report some secondary outcomes
Attention	Unclear risk	Not enough details
Compliance	Low risk	ALA levels were significantly higher in the intervention group
Other bias	Low risk	None noted