This study shows that a CBT-based intervention is clinically effective in the treatment of ICB in PD although it draws on techniques developed for ICB in the general population. The use of a mix of self-report and clinician-rated measures enabled estimation of measures important to patient, carer, and clinician, the majority of which improved significantly. Additionally, the intervention appeared to decrease psychiatric morbidity in the carers of patients with PD ICB. Our study also shows that spontaneous recovery from these harmful behaviors in PD which persist after optimization of medication is rare. To our knowledge, the only other relevant study is a recent trial of amantadine which led to a reduction in pathologic gambling,32
which awaits confirmation. Moreover, in a large cross-sectional study of ICB, amantadine use was associated with a higher incidence of at least one active ICB when compared to no amantadine use.33
In addition to ICB-specific measures, the CBT-based intervention also seemed to benefit depression and anxiety, which are commonly reported comorbidites,6
as well as measures relating to work and social function. These findings are in keeping with a conceptual model which makes dysphoria a central component of PD-ICB.8
Low mood, anxiety, and loss or avoidance of previously rewarding activities may be important factors in the maintenance of ICB, making them valid targets within treatment. For example, increasing purposeful day-to-day activity could relieve dysphoria and increase enjoyment in neglected pastimes, while reducing the time spent on and need for ICB-related behaviors.
While the intervention proved effective in reduction of psychological symptoms in patients with PD, we did not demonstrate an improvement in our primary outcome measures of carer burden or distress, although changes were in the desired direction. The secondary outcome of carer psychiatric morbidity (GHQ) did show significant improvement. A recent study demonstrated patient depression and levodopa end-of-dose dysphoria to be most predictive of carer burden in a PD ICB sample.5
Improvement in patient depression scores occurred without any changes in dopamine replacement therapies in our study, but this was not associated with a reduction in carer-rated burden. It is possible that a longer follow-up period would be required to demonstrate change in perceptions of burden by carers, especially in couples where the ICB had been going for many months or years.
Although the results are encouraging, several limitations apply. The study was small and designed to test the feasibility of a large-scale multicenter trial. The sample size and 6-month duration of follow-up limit the conclusions that can be drawn. However, the observation of impact on a range of patient indices relating to differing aspects of outcome suggests that the findings are reliable. A further limitation of sample size was that it precluded detailed examination of factors that may predict individual response to treatment. The intervention appeared to be acceptable and well-tolerated and dropouts were few.
Regarding trial design, referrals were made from a variety of sources but information was not systematically collected on the total number of potentially eligible cases or on those who declined the offer of referral to the trial. Patients with greater severity or those with less insight may have been excluded. Furthermore, it was not possible to maintain blindness to group allocation from the assessor or patient, which may have led to reporting bias in favor of the intervention group. This is offset to some extent by the use of self-report measures and a high level of inter-rater reliability in the evaluation of outcome. Additionally, the study did not include an active control condition, relying on a comparison between the intervention and waiting list (plus SMC). This control condition was, however, reflective of current clinical practice in PD ICB as it stands, given the absence of an evidence base for management options for this complex range of conditions.
The follow-up duration of 6 months was on average 2 months after end of treatment. Future studies would provide a more clinically useful picture of treatment efficacy with follow-up for at least a year with note of factors such as relapse or new onset ICB in the intervention group. In addition, further work could investigate what proportion of clinical improvement was due to the CBT component and how much was due to other aspects of the psychosocial intervention.
Finally, the present study was able to demonstrate treatment efficacy in patients with PD-ICB, a condition associated with considerable morbidity and distress. Future work will also need to consider cost effectiveness and questions of treatment delivery. That is, who is best placed to deliver the intervention—a psychiatric nurse with training in CBT, a PD nurse specialist, another health care professional, or even lay group, and the extent of training and supervision required.