There were several ICD studies in 2010.

I believe the numbers are 7% on non-agonists, 17% agonists. The REAL figures are no doubt much higher

Impulse control disorders and link to treatment in Parkinson disease: cross-sectional study Reference: Arch Neurol 2010; 67: 589-595.

Source: Arch Neurol

Date published: 11/05/2010 16:40
Summary
by: Yuet Wan Impulse control disorders (ICDs), such as problem or pathological gambling, compulsive buying, compulsive sexual behaviour, and binge or compulsive eating, have been reported in Parkinson disease (PD). A link has also been reported between ICDs in PD and use of dopamine agonist treatment, levodopa treatment and post deep-brain stimulation surgery.



The DOMINION Study assessed the point frequency of four ICDs in PD and their association with dopamine-replacement therapies and other clinical measures. It involved 3090 patients with treated idiopathic PD from 46 centres in the US and Canada. The main outcome measures were:



• The Massachusetts Gambling Screen score for current problem/pathological gambling

• The Minnesota Impulsive Disorders Interview score for compulsive sexual behaviour and buying

• Diagnostic and Statistical Manual of Mental Disorders research criteria for binge-eating disorder



The following findings were reported:



• An ICD was identified in 13.6% of patients (gambling 5.0%, compulsive sexual behaviour 3.5%, compulsive buying 5.7%, and binge-eating disorder4.3%), and 3.9% had 2 or more ICDs.



• ICDs were more common in patients treated with a dopamine agonist than not (17.1% vs. 6.9%; odds ratio, 2.72; 95% CI, 2.08 to 3.54; p <0 .001).



• ICD frequency was similar for pramipexole and ropinirole (17.7% vs. 15.5%; 1.22; 0.94 to 1.57; p = 0.14).



• Other variables independently associated with ICDs were levodopa use, living in the US, younger age, being unmarried, current cigarette smoking, and family history of gambling problems.



The researchers conclude from these findings that “dopamine agonist treatment in PD is associated with 2- to 3.5-fold increased odds of having an ICD. This association represents a drug class relationship across ICDs. The association of other demographic and clinical variables with ICDs suggests a complex relationship that requires additional investigation to optimise prevention and treatment strategies.”