Cumulative pramipexole exposure raises the odds of gambling, especially in bipolar disorder
STUDY: Lindström S et al, Journal of Affective Disorders 2026 (in press)
STUDY TYPE: Nationwide register-based cohort study
FUNDING: University Regional Hospital Unit (USVE); Fredrik and Ingrid Thuring Foundation; AB Svenska Spel Research Council (one co-author holds a position sponsored by this state-owned gambling operator, which had no role in study conduct).
Background
Pramipexole is a dopamine D3 agonist approved for Parkinson’s disease and restless legs syndrome. It has a large, sustained effect size in treatment-resistant depression, and smaller trials find a large effect in bipolar depression as well.
Like other D3 agonists (eg, aripiprazole), pramipexole can cause hedonic dysregulation, including gambling disorder. The rate is much higher in Parkinson’s disease (20-30%) than mood disorders (1-3%). This study looks at how the risk changes with dose and time.
The Study
- 18,732 adults in Sweden with at least one psychiatric diagnosis who filled a pramipexole prescription between 2006 and 2021; patients with Parkinson’s disease were excluded.
- No comparator group: the analysis compared low, intermediate, and high cumulative pramipexole exposure within the cohort, using time-varying Cox regression models.
- Median follow-up was 5.6 years (105,650 person-years total).
- Primary outcome: incident gambling disorder diagnosed in a specialized healthcare setting.
Results
Only 28 patients (0.15%) received a gambling disorder diagnosis during follow-up, a number that likely understates the true rate, given how rarely people with gambling problems seek formal care.
Even so, a clear dose-response pattern emerged. Compared to the lowest-exposure group, intermediate cumulative exposure (one to five years of standard daily dosing) was associated with a 2.8-fold higher risk of gambling disorder (adjusted hazard ratio 2.75). High cumulative exposure (more than five years equivalent) was associated with a 6.6-fold higher risk (adjusted HR 6.60). Each doubling of cumulative dose raised the hazard by 75%.
Bipolar disorder was the strongest psychiatric risk factor, tripling the risk of gambling disorder independent of dose (adjusted HR 3.02). Male sex and younger age also predicted higher risk.
Limitations
- Only 28 outcome events, limiting statistical power and precision of the estimates.
- Treatment indication was unknown: many patients likely received pramipexole for restless legs syndrome, not depression, which could mix distinct risk profiles.
- Gambling disorder is heavily underdiagnosed in Swedish healthcare registers, so the absolute estimates are conservative.
- Detection bias: patients on higher doses may have more clinical contact, increasing the chance a gambling problem gets noticed and recorded.
Practice Implications
- The risk is low, and rises with dose and duration.
- I agree with the dose-relationships, but the increase with duration may reflect that fact that this side effect takes time (and, often, interviews with family) to diagnose.
- Although bipolar disorder is a risk, most cases of hedonic dysregulation do not involve other manic symptoms.
- Check for hedonic dysregulation before prescribing, and follow up with a simple question: “Have you noticed any changes in gambling, spending, or sexual behavior on the medication?”
— Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
