Background <p>Previous double-blind randomized controlled trials (RCTs) have shown that sodium benzoate, a D-amino acid oxidase inhibitor enhancing D-serine availability and NMDA receptor-mediated function, is able to improve cognition in patients with cognitive decline. Accordingly, this study aimed to compare cognitive outcomes between patients receiving sodium benzoate and those receiving placebo.</p> Methods <p>Following the PRISMA guidelines, this meta-analytic study utilized appropriate keyword strings to systematically search the PubMed, Embase, and Cochrane databases for RCTs published in all languages from inception to June 2024. Included criteria were: (1) patients aged 50 or older; (2) those diagnosed with dementia (probable Alzheimer’s disease or vascular dementia) or those with mild cognitive impairment (MCI) having a clinical dementia rating score of 0.5; (3) RCTs comparing the effect of benzoate treatment with that of placebo controls; and (4) the use of cognitive tests as therapeutic effect outcomes.</p> Results <p>Of 351 articles screened, five RCTs were included (246 sodium benzoate, 178 placebo; mean age 72.6 years; mean education 6.01 years; 62% female). The overall effect size for the therapeutic effect of sodium benzoate on cognitive improvement was significant (<i>p</i> = 0.02), with females outperforming males (<i>p</i> = 0.02). Additionally, regression analysis found the cognitive outcome generated from sodium benzoate was not influenced by the use of anti-dementia medication.</p> Conclusion <p>Our findings highlighted a better understanding of the effect of sodium benzoate on cognitive improvement, particularly in female patients with MCI and dementia.</p>