Background: Depression and anxiety are prevalent and can cause suffering in patients with Parkinson’s disease (PD). The Korean version of the Hospital Anxiety and Depression Scale (K-HADS) has been widely used to assess depression and anxiety symptoms in Korean patient with PD. The present study aimed to assess the reliability and validity of the K-HADS using Rasch measurement analysis. Methods: A total of 106 PD patients (54 males, 52 females) who met the diagnostic criteria of the United Kingdom Brain Bank were recruited. Unidimensionality, the Rasch model fit, response category functioning, patient-item distribution, and the separation reliability of the K-HADS depression (K-HADS-D) and anxiety (K-HADS-A) subscales were statistically evaluated. Results: The mean K-HADS-D and K-HADS-A scores were 8.08±4.69 (mean±standard deviation) and 5.44±4.18, respectively. Cronbach’s α coefficients of the K-HADS-D and K-HADS-A were 0.82 and 0.83. The Rasch analysis revealed that the K-HADS-D and K-HADS-A showed unidimensionality and no disordered functioning was observed in the 4-point polytomous scale. However, both K-HADS-D and K-HADS-A exhibited suboptimal separation reliability, while the K-HADS-A showed inadequate scale targeting with floor effect. Conclusions The present study comprises the first validation of the K-HADS using the Rasch measurement model, suggesting that the K-HADS-D and K-HADS-A are clinimetrically acceptable and reliable scales for use in Korean patients with PD. However, the moderate person separation indices implicate the relatively low discriminatory ability of the K-HADS in our study patients.