Aim To explore the related influencing factors of stroke in middle-aged and elderly population in China, and to construct a nomogram prediction model to provide more personalized reference for the prevention and treatment of stroke. Methods This study included 13 063 participants from the China Health and Retirement Tracking Survey project. This project conducted a cross-sectional survey in 2011 using a multi-stage sampling method, targeting individuals aged 45 and above from 150 counties and 450 communities (villages) in 28 provinces (autonomous regions and municipalities). Detailed data were collected on participants' socio-demographic characteristics, physical measurements, health status, healthcare utilization, household income, and expenditure. The study participants were followed up to assess stroke in 3,5, and 2018. Univariate and multivariate Cox regression analyses were employed to identify the factors associated with stroke incidence and to construct a nomogram predictive model. Results During the follow-up, 774 participants developed to stroke. Multivariate Cox regression results showed that older age (HR＝1.8,5%CI:1.019~1.038), being single (HR＝1.5,5%CI:1.031~1.626), smoking (HR＝1.4,5%CI:1.074~1.489), abnormal body mass index (HR＝1.4,5%CI:1.020~1.420), hypertension (HR＝2.0,5%CI:1.855~2.609) and diabetes (HR＝1.3,5%CI:1.117~1.970) were the risk factors affecting the incidence of stroke, high levels of annual per capita expenditure (HR＝0.3,5%CI:0.642~0.953) are antagonistic factors in the incidence of stroke. The nomogram constructed based on the above factors had good predictive performance, and its area under the curve (AUC) was about 0.700. Conclusion Old age, being single, smoking, abnormal body mass index, history of hypertension and diabetes are independent risk factors for stroke, the nomogram constructed based on these factors can help predict the incidence rate of stroke.