ObjectiveTo analyze the prevalence of cognitive impairment in patients with type 2 diabetes mellitus (T2DM) aged 60 years and above, and to evaluate the effect of length of sleep at night on cognitive impairment.MethodsA multi-stage cluster random sampling method was used to select 1 149 diabetic patients aged 60 years and above from 6 counties and urban areas from May to October 2023. The basic information, lifestyle, disease history and length of sleep at night of the interviewees were obtained through a questionnaire survey, and the existence of cognitive impairment was screened using AD8+CSI-D. Multivariate Logistic regression analysis and restricted cubic spline plot were used to study the effect of length of sleep at night on cognitive impairment.ResultsThe detection rate of cognitive impairment was 43.52%. Multi-factor Logistics regression analysis showed that the risk of cognitive impairment was 1.550 times [95% CI (1.031-2.330)] and 1.555 times [95% CI (1.104-2.189)] for≤5 h and 7.1-8.0 h compared with 6.1-7.0 h for length of sleep at night. Stratified analysis showed that the incidence of cognitive impairment in women with length of sleep at night≤5 h was higher than that in men, and the risk of cognitive impairment in women ≥80 years old with length of sleep at night≤5 h and 7.1-8.0 h was higher than that in 70-79 years old. The results of restricted cubic spline plot showed that the relationship between length of sleep at night and cognitive impairment in elderly diabetic patients was approximately U-shaped.ConclusionLong or short night sleep length is positively associated with cognitive impairment, and there are gender and age differences.