Brief Report Hallucinations, Sleep Fragmentation, and Altered Dream Phenomena in Parkinson’s Disease Eric J. Pappert, MD, Christopher G. Goetz, MD, Francie G. Niederman, BSN, *Rema Raman, MS, and *Sue Leurgans, PhD Departments of Neurological Sciences and *Preventive Medicine, Rush-Presbyterian-St. Luke’s Medical Center, Chicago, Illinois, U.S.A. Summary: In a series of consecutively randomized outpa- tients who had Parkinson’s disease (PD), we examined the association of three behaviors: sleep fragmentation, altered dream phenomena, and hallucinations/illusions. Using a log- linear model methodology, we tested the independence of each behavior. Sixty-two percent of the subjects had sleep fragmen- tation, 48% had altered dream phenomena, and 26% had hallucinations/illusions. Eighty-two percent of the patients with hallucinations/illusions experienced some form of sleep disor- der. The three phenomena were not independent. The interac- tion between sleep fragmentation and altered dream phenom- ena was strongly statistically significant. Likewise, a signifi- cant interaction existed between altered dream phenomena and hallucinations/illusions. No interaction occurred between sleep fragmentation and hallucinations/illusions. Sleep fragmenta- tion, altered dream phenomena, and hallucinations/illusions in PD should be considered distinct but often overlapping behav- iors. The close association between altered dream phenomena and hallucinations suggests that therapeutic interventions aimed at diminishing dream-related activities may have a spe- cific positive impact on hallucinatory behavior. Key Words: Hallucinations—Illusions—Sleep fragmentation—Dream phenomenon—Parkinson’s disease. Hallucinations and illusions are a common complica- tion of the chronic pharmacologic treatment of idiopathic Parkinson’s disease (PD) occurring in approximately one third of patients. 1,2 Whereas the pathogenesis, patho- physiology, and natural history of hallucinations and il- lusions are not fully understood, 2 this serious complica- tion limits drug therapy of motor disability and is a sig- nificant risk factor for permanent nursing home placement 3 and its related increased mortality. 4 Promi- nent abnormalities of sleep hygiene 5 and altered dream phenomena 6 are also commonly encountered in PD. Some investigators have suggested that sleep disruptions and altered dream phenomena are linked to hallucina- tions either as precursors or comorbid features. 7 Inde- pendence of these three phenomena has not been for- mally tested. Because sleep disorders are potentially treatable, a full delineation of their relationship to hallu- cinations has implications for understanding pathogen- esis and therapy. METHODS Patient Selection Outpatients with the diagnosis of idiopathic PD (based on the presence of at least three of the following features: rest tremor, bradykinesia, rigidity, and postural reflex impairment) and a Hoehn and Yahr stage of two or three while ‘‘on’’ qualified for study participation. Patients with clinical evidence of non-idiopathic PD, including diffuse Lewy body disease and Alzheimer’s disease as well as clinically significant stroke, were excluded. Ad- ditional patient eligibility criteria required the use of levodopa/carbidopa, with or without other medications, levodopa responsiveness, and the availability of a spouse Received April 9, 1998; revision received September 15, 1998. Ac- cepted September 17, 1998. Address correspondence and reprint requests to Eric J. Pappert, MD, at the Movement Disorders Section, Department of Neurological Sci- ences, Rush-Presbyterian-St. Luke’s Medical Center, 1725 W. Harrison St., Suite 1106, Chicago, IL 60612, U.S.A. Movement Disorders Vol. 14, No. 1, 1999, pp. 117–121 © 1999 Movement Disorder Society 117