The Geriatric Depression Scale (GDS) is recommended for screening depression in individuals with Parkinson's disease (PD). Empirical evidence, however, is limited regarding its validity and factor structure in PD. Thus, the current study sought to evaluate the convergent and divergent validity of the GDS, as well as the structure and validity of the derived factors. METHOD:Nondemented individuals with PD (n = 158) completed the GDS-30, and items were subjected to a principle component analysis. Geriatric Depression Scale total and factor scores were correlated with depression items from the Movement Disorder Society Unified Parkinson's disease Rating Scale (MDS-UPDRSd) and Hamilton Rating Scale for Depression (HAMDd), as well as with the Apathy Scale (AS), State-Trait Anxiety Inventory (STAI), Modified Fatigue Impact Scale (MFIS), Parkinson's disease Sleep Scale, and a Subjective Cognitive Function composite score. RESULTS:The GDS total score was strongly correlated with divergent neuropsychiatric measures (AS, r = 0.57; STAI, r = 0.66; MFIS, r = 0.60), while only moderately correlated with convergent measures (MDS-UPDRSd, r = 0.36; HAMDd, r = 0.32; Ps < 0.05). Linear regression analyses revealed standardized measures of anxiety, apathy, and fatigue independently predicted the GDS total score, while depression items (MDS-UPDRSd and HAMDd) failed to reach significance. Three independent factors were identified: Anxiety, Apathy, and Fatigue. These factors were significantly predicted by their respective convergent measures. CONCLUSIONS:Taken together, our findings suggest that the GDS and its subscales appear to primarily measure anxiety, apathy, and fatigue in PD, or alternatively, these symptom dimensions may be predominant in PD-depression. Future research with clinically diagnosed samples is needed to confirm these initial findings.