Methods: All available patient data for 14,616 patients with sub-diagnoses type I, type II, or Not Otherwise Specified were extracted from the Swedish national quality assurance register for bipolar disorders (BipoläR). Lithium prescription prevalence for each county (the administrative and financial regional level responsible for the health care system) was calculated. Logistic regression analyses with stepwise adjustments for confounders (age, sex, psychiatric comorbidity, antipsychotic drug treatment, and antidepressant drug treatment) were run, with any type of recurrence as primary and specific episode types as secondary outcomes. The subset of 5,618 patients with bipolar I disorder were also analysed separately. The result of the logistic regression is expressed in odds ratio (OR), which is the quota between odds for populations with 1 percentage of difference in lithium prescription rate. Two-sided testing were used for all analyses and the significance level was set to p<0.05. Results: In the whole group, higher lithium prescription rate was significantly associated with lower odds for any type of recurrence [OR = 0.985 (0.980-0.989), p < 0.0005] with similar results for all adjustment steps. Likewise, higher lithium prevalence was associated with lower odds for depressive and hypomanic episodes. For bipolar I disorder, higher lithium prescription rate was associated with lower risk of overall recurrences [OR=0.979 (0.975-0.987), p < 0.0005], as well as for depressive, manic and hypomanic episodes, specifically. Conclusions: The results suggest that higher regional prescription rates are associated with lower risk recurrence in bipolar disorder. The associations withstood adjusting for demographic variables, medication and comorbid conditions. Notably, the ORs represent the decreased risk with 1% lithium prevalence. Given that the regional differences in Sweden – as well as across countries – are much higher than that, the differences in outcomes might be substantial. With the caveat that this study is based on cross-sectional data, the data suggest that outcomes on a population might improve if more patients receive lithium. The results are in line with meta-analyses of RCTs and lend support for lithium being the first line maintenance treatment in bipolar disorder.