Addiction is often underdiagnosed in bipolar disorder (BD), although it is frequent and known to complicate its clinical course.

The aim of our study was to study socio-demographic and clinical factors associated with addiction in BD patients.

This is a retrospective, cross-sectional, descriptive and comparative study on 100 patients followed in our department and diagnosed with BD type I according to DSM 5. Demographic and clinical data was compared across the groups: Addiction + (A + ) and Addiction–(A − ).

Nighteen patients had an addiction co-morbidity (A + ), whereas 81 had not (A − ). The mean age of the (A + ) group was 39.47 years whereas it was 42.52 years in the (A − ) group. Males represented 68.4% of the (A + ) group and 48.1% of the (A − ) group. Age of illness onset was lower in the (A + ) group (mean = 23.16, median = 21) compared to the (A − ) group (mean = 26.04, median = 27). Addiction co-morbidity was significantly associated with predominant manic polarity (P = 0.03). All (A + ) patients presented mood episodes with psychotic features, whereas psychotic features were only found in 86.6% of (A − ) patients. Co-morbid addiction was significantly associated with a higher number of mood episodes (P = 0.04), a higher number and duration of hospitalisations (P = 0.02, P = 0.015), and a poorer compliance (P = 0.07). All A+ subjects received antipsychotics, and they were significantly more to receive long-acting antipsychotics (P = 0.06).

Addictions worsen the prognosis of bipolar disorder and require specific therapeutic strategies. They deserve therefore the particular attention of clinicians.

The authors have not supplied their declaration of competing interest.