Recent studies have shown that the abnormal accumulation of endogenous formaldehyde could be a critical factor in age-related cognitive decline. The aim of this study was to estimate the correlation between uric formaldehyde and general cognitive abilities in a community-based elderly population, and to measure the extent and direction in which the correlation varied with demographic characteristics. Using a double-blind design, formaldehyde in human urine was analyzed by high-performance liquid chromatography (n = 604), and general cognitive abilities were measured using the Montreal Cognitive Assessment (MoCA). Demographic characteristics, in terms of age, gender, residential region, and education were taken into consideration. We found that uric formaldehyde levels were inversely correlated with the MoCA score, and the concentration varied with demographic features: higher odds of a high formaldehyde level occurred among the less educated and those living in old urban or rural areas. In cytological experiments, the level of cellular formaldehyde released into the medium increased as SH-SY5Y and BV2 cells were incubated for three days. Formaldehyde in excess impaired the processes of N2a cells and neurites of primary cultured rat hippocampal cells. However, removal of formaldehyde markedly rescued and regenerated the processes of N2a cells. These results demonstrated a negative correlation between the endogenous formaldehyde and general cognitive abilities. High formaldehyde levels could be a risk factor for cognitive impairment in older adults, and could be developed as a non-invasive marker for detection and monitoring of age-related cognitive impairment.