Objective To explore the correlation between vascular cognitive impairment and tea-drinking. Methods A total of 1865 elder individuals aged over 60 years were sampled in 2004 in Beijing. We conducted a population based case-control study on 87 patients with vascular cognitive impairment (case group) and 87 sex- and age-matched controls（control group）. Univariate analysis was conducted firstly, and then multivariate analysis for financial situation，hypertension，tea drinking，education and living area was conducted using logistic model. Results In cross-link analysis, the ORs for tea drinkers who had better financial situation or a higher education level were the least (OR=0.190 and 0.204 respectively). In univariate analysis, tea drinkers in case group were less than those in controls (＝10.508, P＝0.015), and the ORs decreased as the decrease of frequency of tea-drinking. But there was no difference in the sorts of the tea. There were less individuals who had hypertension or worse financial situation in case group than in control group. In multivariate analysis, tea intake was associated with a lower prevalence of cognitive impairment. The OR was 0.513 for individuals who drink tea almost everyday, compared with non-tea-drinkers. Conclusion Tea-drinking may be the protective factor of vascular cognitive impairment.