Comparative effectiveness of mood stabilizers and antipsychotics in the prevention of hospitalization after lithium discontinuation in bipolar disorder.

The aim of the study was to compare the real-world effectiveness of mood stabilizers and antipsychotics in the prevention of psychiatric hospitalizations and treatment failure after lithium discontinuation in a nationwide bipolar cohort. Using health-care registers, we identified everyone in Finland diagnosed with bipolar disorder during 1987-2018 who discontinued lithium after using it for at least one year (n = 4 052, median period of lithium use before discontinuation 2.7 years). The risk of psychiatric hospitalization and treatment failure (psychiatric hospitalization, death or change in medication) were investigated with within-individual Cox regression. Of mood stabilizer monotherapies, the periods of valproate use (HR = 0.83, 95% CI = 0.71 - 0.97) had lower risk of hospitalization than nonuse of mood stabilizers. Of antipsychotic monotherapies, the use of long-acting injectable (LAI) antipsychotics (HR = 0.48, 95% CI = 0.26 - 0.88) and chlorprothixene (HR = 0.62, 95% CI = 0.44 - 0.88) were associated with lower risk and the use of quetiapine (HR = 1.26, 95% CI = 1.07 - 1.48) and oral olanzapine (HR = 1.23, 95% CI = 1.01 - 1.49) with higher risk of psychiatric hospitalizations than nonuse of antipsychotics. Of mood stabilizer monotherapies, lithium use was associated with lower risk of treatment failure (HR = 0.82, 95% CI = 0.76 - 0.88) than valproate use. The results suggest that antipsychotic LAIs are especially effective in the prevention of psychiatric hospitalizations after lithium discontinuation. The need to alter used medications may be the lowest when lithium is restarted.