Background: Temporality of the association of low omega-3 polyunsaturated fatty acid (n-3 PUFA) plasma levels with depression remains questionable. To determine the underlying nature of these associations, this study examined the bidirectional longitudinal associations of n-3 PUFA plasma levels with (presence, onset and course of) depressive disorders and symptoms. Methods: Baseline (n = 2912, 28.6% with current depressive disorder) and 6-year follow-up data (n = 1966, 13.0% with current depressive disorder) of the Netherlands Study of Depression and Anxiety (NESDA) were used. Depression diagnoses and symptoms were based on psychiatric interviews and self-report questionnaires. N-3 PUFA levels (ratio of total fatty acids (mmol%)), were assessed using nuclear magnetic resonance. Results: Using two waves of data, n-3 PUFA levels were lower among depressed persons, as compared to healthy controls (Beta = −0.047, SE = 0.011, p < .001). Nevertheless, baseline n-3 PUFA levels were not consistently associated with subsequent change in depressive symptoms, onset or remission of depressive disorders over 6 years. Furthermore, the difference in n-3 PUFA levels detected at baseline between depressed and non-depressed participants tended to dissipate over 6 years (depression-by-time estimate: p = .011). Finally, subjects depressed both at baseline and at 6-year follow up had consistently lower n-3 PUFA levels over the entire follow-up as compared to those who had never been depressed. Change in depressive disorders across waves was not consistently accompanied by change in n-3 PUFA levels over time. Limitations: No data on intermediate time points and EPA levels were available. Conclusions: Despite significant cross-sectional associations between n-3 PUFA plasma levels and depressive disorders and severity, this 6-year longitudinal study could not confirm an uni- or bidirectional association over time. The association between depression and n-3 PUFA plasma levels is unlikely to be causal.