Background: In Parkinson's disease (PD), there is heterogeneity in the clinical presentation and underlying biology. Research on PD subtypes aims to understand this heterogeneity with potential contribution for the knowledge of disease pathophysiology, natural history and therapeutic development. There have been many studies of PD subtypes but their impact remains unclear with limited application in research or clinical practice. Objective: To critically evaluate PD subtyping systems. Methods: We conducted a systematic review of PD subtypes, assessing the characteristics of the studies reporting a subtyping system for the first time. We completed a critical appraisal of their methodologic quality and clinical applicability using standardized checklists. Results: We included 38 studies. The majority were cross-sectional (n=26, 68.4%), used a data-driven approach (n=25, 65.8%), and non-clinical biomarkers were rarely used (n=5, 13.1%). Motor characteristics were the domain most commonly reported to differentiate PD subtypes. Most of the studies did not achieve the top rating across items of a Methodologic Quality checklist. In a Clinical Applicability Checklist, the clinical importance of differences between subtypes, potential treatment implications and applicability to the general population were rated poorly, and subtype stability over time and prognostic value were largely unknown. Conclusion: Subtyping studies undertaken to date have significant methodologic shortcomings and most have questionable clinical applicability and unknown biological relevance. The clinical and biological signature of PD may be unique to the individual, rendering PD resistant to meaningful cluster solutions. New approaches that acknowledge the individual-level heterogeneity and that are more aligned with personalized medicine are needed.