Objective
While treatment with antipsychotics and antiepileptics have been associated with an increased risk of diabetes mellitus (DM), lithium may have the opposite effect via inhibition of glycogen synthase kinase‐3. The aim of this study was to investigate whether treatment of bipolar disorder with lithium, antipsychotics, or antiepileptics is associated with the risk of DM in a real‐world clinical setting.
Methods
Using nationwide registers, we identified all patients diagnosed with bipolar disorder in Danish Psychiatric Services from January 1, 1996, to January 1, 2019 (N = 30,451). The risk of developing DM was operationalized via hospital diagnoses and redeemed prescriptions for glucose‐lowering drugs. For lithium, antipsychotics, valproate, and lamotrigine, we calculated hazard rate ratios (HRR) for developing DM via adjusted Cox proportional hazards models. Potential cumulative dose–response‐like associations were examined using the log‐rank test.
Results
During follow‐up (245,181 person‐years), 2107 (6.9%) patients developed DM. Compared with non‐users of the respective drugs, we found no clinically or statistically significant difference in the risk of developing DM among patients receiving lithium (n = 11,690; incidence rate of DM/1000 person‐years (IR) = 8.87, 95% CI: 8.02–9.90; HRR = 0.94, 95% CI: 0.84–1.06) or lamotrigine (n = 11,785; IR = 7.58, 95% CI: 6.69–8.59; HRR = 0.89, 95% CI: 0.77–1.02), respectively. Conversely, for patients receiving valproate (n = 5171; IR = 12.68, 95% CI: 10.87–14.80; HRR = 1.34, 95% CI: 1.14–1.58) and antipsychotics (n = 22,719; IR = 12.00, 95% CI: 11.14–12.94; HRR = 1.65, 95% CI: 1.45–1.88), respectively, there was increased risk of developing DM. For antipsychotics, we observed a clear cumulative dose–response‐like association with the risk of DM.
Conclusions
Treatment with valproate and antipsychotics—but not with lithium and lamotrigine—was associated with increased risk of DM in a real‐world cohort of patients with bipolar disorder.