Identification of risk factors for impulse-control disorder symptoms in patients with Parkinson’s disease

Abstract

Parkinson’s disease (PD) is one of the motoric neurodegenerative disorders that has the fastest-growing prevalence, disability, and death rate compared to other neurological disorders. Globally, from 1990 to 2015, the number of individuals with PD increased by 118%, up to around 6.2 million. There are several motor symptoms and non-motor symptoms that occur in Parkinson’s disease. One of the non-motor symptoms that occur is Impulse-Control Disorder (ICD). In PD, there are 4 main symptoms of ICD that often occur such as pathological gambling, binge-eating, compulsive buying, and compulsive sexual behaviour. ICD is often discovered when the treatment of PD begins, so this research focused on the incidence of ICD as a result of the treatment of PD. The purpose of this study is to identify risk factors for the type and number of ICD symptoms that occur in patients with PD. Both binary and multiclass classification tree methods were implemented to identify risk factors associated with the incidence of ICD for PD patients. Imbalance data problems that arose were handled with Synthetic Minority Over-sampling Technique (SMOTE). The results obtained show that the STAI-Trait total score is a risk factor that always appears for each type of, and the number of type of, ICD symptoms that appear. Moreover, risk factors that only appear in several symptoms are the length of education taken, STAI-State total score, age, duration of PD, SCOPA-AUT total score, MOCA total score, and MDS-UPDRS 3 total score. While, risk factors that only appear for a particular symptom are family history of PD, DAT binding ratio, and dopamine agonist treatment.