Our objective was to quantify the cross-sectional associations between dietary fatty acid (DFA) patterns and cognitive function among Hispanic/Latino adults. This study included data from 8942 participants of the Hispanic Community Health Study/Study of Latinos, a population-based cohort study (weighted age 56·2 years and proportion female 55·2 %). The National Cancer Institute method was used to estimate dietary intake from two 24-h recalls. We derived DFA patterns using principal component analysis with twenty-six fatty acid and total plant and animal MUFA input variables. Global cognitive function was calculated as the average z-score of four neurocognitive tests. Survey linear regression models included multiple potential confounders such as age, sex, education, depressive symptoms, physical activity, energy intake and CVD. DFA patterns were characterised by the consumption of long-chain SFA, animal-based MUFA and trans-fatty acids (factor 1); short to medium-chain SFA (factor 2); very-long-chain n-3 PUFA (factor 3); very-long-chain SFA and plant-based MUFA and PUFA (factor 4). Factor 2 was associated with greater scores for global cognitive function (β = 0·037 (sd 0·012)) and the Digit Symbol Substitution (DSS) (β = 0·56 (sd 0·17)), Brief Spanish English Verbal Learning-Sum (B-SEVLT) (β = 0·23 (sd 0·11)) and B-SEVLT-Recall (β = 0·11 (sd 0·05)) tests (P < 0·05 for all). Factors 1 (β = 0·04 (sd 0·01)) and 4 (β = 0·70 (sd 0·18)) were associated with the DSS test (P < 0·05 for all). The consumption of short to medium-chain SFA may be associated with higher cognitive function among US-residing Hispanic/Latino adults. Prospective studies are necessary to confirm these findings.

Dietary assessment of individual fatty acid intake is difficult due to a number of limitations. Information regarding the type, quantity and brand-name of fat used in cooking and at the table is required. In addition, margarine manufacturers may change the component oils used for reasons of cost, which changes the fatty acid composition of their products from season-to-season. Independent markers of fatty acid intake are required, therefore, to compensate for these limitations. Adipose tissue concentrations have been used as a measure of habitual intake of fatty acid groups and individual fatty acids in numerous studies. Saturated (SFA) and monounsaturated fatty acids (MUFA) are generally poorly correlated with adipose tissue concentrations, which can be explained partly by endogenous synthesis. In general, adipose tissue concentrations of exogenously-produced fatty acids (n-3 and n-6 polyunsaturated fatty acids (PUFA)) are well correlated with estimates of habitual intake. Correlations between dietary trans unsaturated fatty acids (TUFA) and adipose tissue concentrations vary between countries, which may be due to differences in dietary sources. Correlations may be affected by differences in bioavailability or selective retention of fatty acids in certain tissue lipids.