Introduction While anxiety, depression and insomnia frequently (co-)occur in Parkinson's disease (PD) patients, little is known about their temporal relationship. In this study, we tested two hypotheses: i) insomnia predicts an increase in symptoms of depression or anxiety and ii) anxiety or depression at baseline predicts insomnia in PD patients six months later. Methods We used longitudinal data from a prospective cohort study of early-stage, medication-naïve PD patients. Primary outcome measures were: anxiety symptoms, measured with the State-Trait Anxiety Inventory (STAI); depressive symptoms, measured with the 15-item Geriatric Depression Scale (GDS-15); and insomnia, defined as a score ≥ 2 on item 1.7 of the Movement Disorder Society – Unified Parkinson's Disease Rating Scale. We performed linear and logistic regression analyses, correcting for baseline value of the respective outcome variable. Results Baseline insomnia was not associated with GDS-15 or STAI total score at follow-up. In a post hoc analysis, we found that insomnia predicted a higher STAI State score (B(SE) = 2.50 (1.07), p < 0.05), while the association with the STAI Trait score was not significant. Baseline STAI scores (B(SE) = 0.02 (0.01), p = 0.001) and GDS-15 score (B(SE) = 0.15 (0.05), p < 0.001) were significantly associated with insomnia at follow-up. Conclusion Symptoms of anxiety and depression may constitute a risk factor for insomnia in PD. The relationship between insomnia and anxiety is bidirectional, which suggests that both anxiety and sleep disorders can start a negative spiral in PD patients, where one enhances the other. Independent clinical attention for these symptoms in PD patients is therefore warranted.