Dementia is a disease with no known cure and limited treatment options; thus prevention is a key strategy to reduce dementia burden. Diet is thought to be a modifiable risk factor for dementia, but evidence of the association between overall diet quality in later life and risk of cognitive decline remains limited. This study examined the association between diet quality and neurocognitive outcomes in a large cohort of older adults in Australia (the ASPREE Longitudinal Study of Older Persons, ALSOP). The cohort included 12,357 women and men, with a mean age of 78 years (ranging from 73 to 98), who were community dwelling without dementia. Participants were followed with a cognitive test battery, and dementia was clinically adjudicated by an expert panel according to Diagnostic & Statistical Manual of Mental Disorders-IV criteria. Overall diet quality score was calculated from a 49-item diet questionnaire with scoring based on alignment with the 2013 NHMRC Australian Dietary Guidelines (1). Scores ranged from 1-51, with higher values reflecting better diet quality. Cox proportional hazards regression models were used to examine the association between diet quality and risk of all-cause dementia over a mean 8.9 years of follow up. Cognitive impairment (>1.5 standard deviation decline in cognitive function from baseline, sustained at a subsequent follow up, but not meeting the criteria for dementia, CIND) was a secondary outcome. At baseline, dietary quality was higher among women than men (30.6 (6.6) vs 28.2 (6.9), p<0.001), and older age was associated with lower dietary scores (p=0.003). Diet quality was positively associated with baseline global cognition scores (3MS) (p<0.001). In analyses adjusted for age, gender, and level of education, diet quality was significantly associated with the risk of dementia (Hazard Ratio (HR):0.984; 95% CI: 0.975-0.993, p=0.001) and CIND (HR 0.985; 0.981-0.990, p<0.001). When examined by tertiles, this represented a 14% and 15% lower risk of dementia for the middle and upper tertiles of dietary quality score respectively compared to the lower tertile (p<0.05), and a 14% and 18% lower risk of CIND in the second and third tertiles of diet score respectively (p<0.001). Additional adjustment for other potential confounding factors; physical activity, smoking, alcohol intake, diabetes, area-level socioeconomic status, body mass index, living alone) attenuated the association between diet quality and neurocognitive outcomes, but the association with CIND remained significant (HR 0.86: 0.78-0.94, p=0.001 for upper tertile of diet quality score). These findings suggest that dietary intake aligned with the Australian Dietary Guidelines is associated with a lower risk of cognitive decline. Whilst causation cannot be established from this study, the findings add impetus for further research to determine the extent to which changing dietary patterns in later life can reduce risk of age-related cognitive decline.