Associations of the intake of individual and multiple fatty acids with depressive symptoms among adults in NHANES 2007-2018

Background: Previous studies have mainly focused on the effects of individual fatty acids on
depressive symptoms, while the combined effect of fatty acids on the risk of depressive
symptoms has not yet been extensively reported. This study evaluate the associations between
individual and multiple fatty acids with depressive symptoms in U.S. adults. Methods: Data sets
were obtained from the National Health and Nutrition Examination Survey (NHANES) 2007-2018 cycles. Both males and females aged above 18 years with complete information about
dietary fatty acids intake, depression symptoms, and covariates were included. Weighted linear
regression models were conducted to evaluate the relationships between individual fatty acid
intake and depressive symptoms, and restricted cubic spline (RCS) models were utilized to
explore the corresponding dose-response relationships. Additionally, we implemented the
weighted quantile sum (WQS) regression and quantile g-computation (QGC) models to estimate
the mixed effects of 19 fatty acids and identify the predominant types. Results: After
multivariable adjustments, an increase of one unit in Linoleic acid (LA), Alpha-Linolenic Acid
(ALA), Arachidonic acid (AA), Docosapentaenoic acid(DPA), Docosahexaenoic acid(DHA), was
associated with a decrease in depressive scores by -0.021 (95 % CI: -0.039,-0.003, p = 0.021),-
0.028 (95 % CI: -0.045,-0.011, p = 0.002),-0.026 (95 % CI: -0.044,-0.008, p = 0.005), -0.026 (95
% CI: -0.042,-0.009, p = 0.003), and – 0.022 (95 % CI: -0.041,-0.003, p = 0.022), respectively.
However, a per unit increase in Hexanoic acid and Octanoic acid was associated with an
increase in depressive scores of 0.020 (95 % CI: 0.002,0.038, p = 0.029) and 0.026 (95 % CI:
0.004,0.048, p = 0.020), respectively. Meanwhile, significant dose-response relationships were
supported by the RCS models. As for the mixed effects, both WQS and QGC models
demonstrated that the mixture of polyunsaturated fatty acids (PUFAs) was inversely related to
depressive symptoms, and ALA and DPA were the most critical contributors. DHA was
negatively correlated with depressive symptoms in WQS analysis, but positively correlated with
depressive symptoms in QGC analysis. Limitations: The cross-sectional design limits our ability
to establish causality, and 24-hour dietary recall can lead to potential inaccuracies reflecting
participants' true eating habits. Conclusion: Our study suggests that the single effects of each
PUFA were inversely associated with depressive symptoms, except for octadecatetraenoic acid.
Moreover, higher combined intake of dietary PUFAs is inversely associated with depressive
symptoms in U.S. adults. Among the mixed effects of PUFAs, ALA and DPA may play
predominant roles. However, DHA mixed with other fatty acids may have different effects on
depressive symptoms, and further study is needed.