Lloret 2009

Methods	Design: randomised, placebo‐controlled, double‐blind study Duration: 6 months
Participants	Country: Spain Number of centres: 1 Diagnosis: probable AD according to NINCDS‐ADRDA Number of participants: 57 participants total randomised (completers: 19 in vitamin E group, 14 in placebo group), 18 healthy controls Age: not described Gender: not described Ethnicity: not described Cognitive status: severity of participants based on the Geriatric Dementia Scale ‐ mild (n = 25), moderate (n = 26) and severe (n = 6) Inclusion criteria: not described Exclusion criteria: people taking antioxidant supplements or any medication other than cholinesterase inhibitors
Interventions	Treatment 1: vitamin E 800 IU/day Treatment 2: placebo daily
Outcomes	Clock Drawing Test (months 0 and 6) BDS (months 0 and 6) MMSE (months 0 and 6) Blood total glutathione levels and oxidised glutathione (months 0 and 6) Plasma malondialdehyde (months 0 and 6)
Notes	‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomization was performed using a randomized list of numbers."
Allocation concealment (selection bias)	Unclear risk	No reference made to how allocation concealment was ensured.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Study described as 'double blind' but authors did not describe how blinding was maintained and whether it was successful. It is unclear whether the placebo was identical in appearance to the vitamin E treatment.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Study described as 'double blind' but authors did not describe how blinding was maintained and whether it was successful.
Incomplete outcome data (attrition bias) All outcomes	High risk	24/57 AD participants did not complete the research. "Of the patients who finished the study, 14 had been treated with placebo and 19 with vitamin E." The reasons for participant dropout were not described.
Selective reporting (reporting bias)	Unclear risk	Outcomes reported, though the level of detail lacking. For example, means and SDs not described.
Other bias	Low risk	No evidence of other bias.