Objective: To explore the longitudinal patterns and influencing factors of depression and cognitive impairment in patients with Parkinson's disease (PD), verify their bidirectional predictive relationship, and preliminarily analyze the modifying effects of medication on depression and cognitive impairment. Methods: A total of 86 de novo drug-naïve patients with idiopathic PD were enrolled and assessed for motor and non-motor symptoms at initial consultation (baseline) and follow-up, including the Unified Parkinson's Disease Rating Scale part Ⅲ for motor symptoms, Hamilton Depression Rating Scale for depressive symptoms, and Montreal Cognitive Assessment for cognitive function. Spearman correlation analysis was used to preliminarily screen clinical characteristics associated with depressive and cognitive impairment in PD, and generalized linear models were used to analyze the associations between longitudinal changes in cognitive impairment and depressive symptoms. Results: Patients' motor symptoms significantly deteriorated, while depressive symptoms exhibited dynamic conversion characteristics. Motor symptoms and other non-motor symptoms (anxiety, sleep disturbances) demonstrated statistically significant differences across groups stratified by varying degrees of depression severity. Patients with mild cognitive impairment primarily exhibited impairments in visuospatial and executive functions and delayed memory. Dopaminergic medication showed no significant effects on depression or cognitive function. Cognitive impairment and depressive symptoms progressed synchronously and had bidirectional predictive effects. Conclusion: Depressive symptoms in PD patients exhibit dynamic progression patterns. Anxiety and sleep disturbances may serve as critical indicators for depression screening in PD. There exists a bidirectional predictive relationship between cognitive impairment and depressive symptoms. Dopaminergic drug therapy fails to improve cognitive function and depressive symptoms.