The aim of this study was to identify sociodemographic and clinical predictors of excellent response, that is, 'cure' of future affective episodes, to lithium in monotherapy. We used nationwide registers to identify all patients with a diagnosis of bipolar disorder in psychiatric hospital settings who were prescribed lithium from 1995 to 2006 in Denmark (N=3762). Excellent lithium responders were defined as patients who after a stabilization lithium start-up period of 6 months, continued lithium in monotherapy without getting hospitalized. The rate of excellent response to lithium in monotherapy was 8.9% [95% confidence interval (CI): 7.9-9.9] at 5-year follow-up and 5.4% (95% CI: 4.4-6.3) at 10-year follow-up. The rate of nonresponse to lithium monotherapy was significantly increased for female patients [hazards ratio (HR)=1.12, 95% CI: 1.04-1.21) and for patients with a depressive index episode compared with patients in remission or with a diagnosis of other or unspecified bipolar disorder before first lithium purchase (HR=1.13, 95% CI: 1.03-1.25). The rate of nonresponse increased by 3% (95% CI: 2-5%) for every psychiatric hospitalization before first purchase of lithium. Patients with somatic comorbidity had increased rates of non-response to lithium compared with patients without somatic comorbidity (HR=1.23, 95% CI: 1.00-1.52).It is concluded that the prevalence of excellent response to lithium monotherapy is low and such patients are characterized by few earlier psychiatric hospitalizations, a manic index episode before lithium and reduced somatic comorbidity.