Both epidemiological and preclinical studies have shown the benefits of n-3 polyunsaturated fatty acid (n-3 PUFA) on dementia and cognitive impairment, yet the results of clinical randomized controlled trials (RCTs) performed to date are conflicting. The difference in the baseline omega-3 index (O3i) of subjects is a potential cause for this disparity, yet this is usually ignored. The present meta-analysis aimed to evaluate the effect of n-3 polyunsaturated fatty acid (n-3 PUFA) on cognitive function in the elderly and the role of baseline O3i. A systematic literature search was conducted in PubMed, Embase, Cochrane Library, and Web of Science up to June 27th, 2023. The mean changes in the mini-mental state examination (MMSE) score were calculated as weighted mean differences by using a fixed-effects model. Fifteen random controlled trials were included in the meta-analysis. Pooled analysis showed that n-3 PUFA supplementation did not significantly improve the MMSE score (WMD = 0.04, [-0.08, 0.16]; Z = 0.62, P = 0.53; I2 = 0.00%, P(I2) = 0.49). Out of the 15 studies included in the meta-analysis, only 7 reported O3i at baseline and outcome, so only these 7 articles were used for subgroup analysis. Subgroup analysis showed that the MMSE score was significantly improved in the higher baseline O3i subgroup (WMD = 0.553, [0.01, 1.095]; I2 = 0.00%, P(I2) = 0.556) and higher O3i increment subgroup (WMD = 0.525, [0.023, 1.026]; I2 = 0.00%, P(I2) = 0.545). The overall effect demonstrated that n-3 PUFA supplementation exerted no improvement on global cognitive function. However, a higher baseline O3i and higher O3i increment were associated with an improvement in cognitive function in the elderly.