Background Sleep disorder is one of the most common non-motor symptoms in Parkinson's disease (PD) patients. At present, there are subjective and objective tools to evaluate sleep disorders. Nevertheless, previous studies commonly used single subjective questionnaires or objective examinations. Therefore, we used the combinations of subjective and objective tools to analyze clinical characteristics of sleep disturbances in PD and investigated differences and consistence between subjective and objective tools. Methods One hundred and sixteen PD patients were eligible to participate into this study. All participants were evaluated by Pittsburgh Sleep Quality Index (PSQI), Unified Parkinson's Disease Rating Scale (UPDRS) in on condition, Hoehn-Yahr (H-Y) stage, Hamilton Depression Rating Scale (HAMD 24 items), Montreal Cognitive Assessment (MoCA), Epworth Sleepiness Scale (ESS), and underwent a video-polysomnography (Video-PSG). Results According to PSQI score of 116 PD patients, the proportion of PD patients with sleep disturbances (PSQI ≥ 7) was 50% (N = 58). Compared to PD patients without sleep disturbances, PD patients with sleep disturbances had lower score of MoCA (23.34 ± 3.50 vs 24.89 ± 3.52; t = 2.377, P = 0.019), higher score of UPDRSⅠ[4.00 (2.00, 5.00) vs 3.00 (2.00, 5.00); U = - 2.306, P = 0.021], UPDRSⅡ[12.00 (9.00, 16.00) vs 10.00 (6.00, 13.00); U = - 1.995, P = 0.046], higher levodopa equivalent daily dose [LED, (508.14 ± 335.85) vs (394.06 ± 236.40) mg/d; t = - 2.115, P = 0.037]. Although PD patients with sleep disturbances had more score of UPDSR Ⅲ and higher H-Y stage, the differences were not significant (P > 0.05). On the other hand, decreased total sleep time (TST), reduced sleep efficiency (SE), increased sleep latency (SL), decreased non-rapid eye movement (NREM) sleep stage Ⅱ time were found for PD patients with sleep disturbances (P 0.05, for all). The score of PSQI was positively correlated with the score of ESS (r = 0.200, P = 0.032), HAMD (r = 0.202, P = 0.030), UPDRSⅠ (rs = 0.266, P = 0.004) and Ⅱ (rs = 0.254, P = 0.007), LED (r = 0.213, P = 0.022), SL (rs = 0.211, P = 0.023). Moreover, the score of PSQI was negatively correlated with TST (r = -0.231, P = 0.003), SE (r = -0.192, P = 0.039) and MoCA (r = -0.236, P = 0.011). Conclusion PD patients with sleep disturbances had worse cognition impairment, more mood disorders, decreased activity of daily life. Meanwhile, most of PSG parameters were altered in PD patients with sleep disturbances. Moreover, the severity of sleep disturbances in PD patients was correlated with these factors. Overall sleep quality of PD patients assessed with the objective tool could be predicted by the subjective scale. However, to evaluate sleep architecture and other sleep disorders for PD patients, the objective tools (such as Video-PSG monitoring) are necessary to be used.