Abstract
BACKGROUND: Sleep disorders are a common complication in elderly patients with Parkinson's disease and cognitive impairment. This retrospective cohort study investigates the factors associated with sleep disorders in elderly patients with Parkinson's disease and cognitive impairment and proposes a framework for a future comprehensive relaxation training intervention based on the identified factors, to inform subsequent clinical studies. METHODS: A retrospective study was conducted on 108 elderly patients with Parkinson's disease and cognitive impairment who visited the outpatient department of our hospital from January 2021 to December 2024. All patient data were obtained from the electronic medical record system. Patients were divided into a sleep disorder group (n = 40) and a non-sleep disorder group (n = 68) based on the presence or absence of sleep disorders. General information differences between the two groups were collected and compared. Collinearity analysis was performed on indicators with significant differences between the two groups. Logistic regression analysis was used to identify the primary factors associated with sleep disorders in elderly patients with Parkinson's disease and cognitive impairment. A line chart was established using R software for validation. Finally, a framework for a comprehensive relaxation training intervention was proposed as a potential future clinical application based on the model's findings. RESULTS: There were statistically significant differences between the sleep disorder group and the non-sleep disorder group in terms of Hoehn-Yahr staging, equivalent dose of levodopa, Hamilton Anxiety Scale (HAMA), Hamilton Depression Scale (HAMD), and chronic pain (p < 0.05). No collinearity was observed among the indicators. Multivariate logistic regression analysis revealed that Hoehn-Yahr staging, equivalent dose of levodopa, HAMA, HAMD, and chronic pain were all risk factors for sleep disorders in elderly Parkinson's disease patients with cognitive impairment (OR = 6.327, 2.698, 3.203, 1.041, 1.217, p < 0.05). Based on the results of the logistic regression analysis, a risk prediction nomogram model for sleep disorders in elderly patients with Parkinson's disease and cognitive impairment was constructed. The receiver operating characteristic (ROC) curve showed an area under the curve (AUC) value of 0.963 (95% CI, 0.931-0.955). The calibration curve indicated that the model's predictive results were well aligned with the actual occurrence of sleep disorders in elderly patients with Parkinson's disease and cognitive impairment, with a Brier Score of 0.051 and a model fit p-value of 0.925. The statistic was 2.688. The clinical decision curve was generally higher than the two extreme curves, indicating that the factors included in the plot diagram have a high net benefit in predicting sleep disorders in elderly patients with Parkinson's disease cognitive impairment. CONCLUSION: There are numerous factors associated with sleep disorders in elderly patients with Parkinson's disease and cognitive impairment, primarily including Hoehn-Yahr staging, equivalent dose of levodopa, HAMA, HAMD, and chronic pain. The risk prediction nomogram model constructed based on these factors has certain predictive value for the occurrence of sleep disorders, can assist in the early screening of high-risk populations in clinical practice, and provides a basis for developing corresponding relaxation training interventions to reduce the occurrence of sleep disorders.