Patients with mild cognitive impairment (MCI) are at high risk of dementia, but early identification and active intervention can reduce its morbidity and the incidence of dementia. There is currently no suitable neuropsychological assessment scale to effectively identify MCI in neurological outpatient departments in China. The Mini-Mental State Examination (MMSE) is often used to screen for MCI in outpatient departments in China.To compare the value of Mini-Cog and MMSE in screening patients for MCI in a neurological outpatient department, and determine differences in the value of Mini-Cog for different ages and educational levels.This was a retrospective study of 229 patients with suspected MCI who visited the Cangzhou Central Hospital between March 2012 and April 2016. The MCI group included 119 patients diagnosed with MCI and 110 cases without MCI (non-MCI group). The MCI patients were subgrouped as 40 to 60 years of age, 61 to 80 years, and >80 years; and as without education, ≤6 years education, and >6 years education. All subjects were assessed using the Mini-Cog and MMSE.There were significant differences in Mini-Cog (P < .05) and MMSE (P < .05) between the MCI and non-MCI groups. The sensitivity, specificity, positive predictive value, negative predictive value, and Youden index (85.71%, 79.41%, 0.8108, 0.8438, and 0.6550) of Mini-Cog were all higher than those of MMSE (64.76%, 71.57%, 0.7010, 0.6364, and 0.3370) in identifying MCI, but there was no significant difference in specificity (P > .05). Mini-Cog was better than MMSE (P < .05) for MCI patients with different ages and education levels.These results showed that the Mini-Cog was superior to MMSE in identifying MCI patients. Mini-Cog was less affected by age and education level than MMSE. The Mini-Cog assessment was short (3-4 minutes) and easily accepted by the patients. Mini-Cog could be more suitable for application in outpatient department in primary hospitals.