Hirschfeld 1999.

Methods	Study design: double‐blind, randomised, parallel‐group, study
Participants	Diagnosis: a DSM‐IV diagnosis of bipolar disorder (manic or mixed) and hospitalized for treatment of an acute manic episode Method of diagnosis: participants were required to have manic symptoms of sufficient severity to have a total YMRS score ≥ 14, as assessed by SADS Age: for lithium, median = 36.4 (SD = 8.4) years; for divalproex loading, median = 36.0 (SD = 9.4) years; for divalproex non‐loading, median = 32.4 (SD = 9.1) years; range = 18 ‐ 60 years accepted Sex: lithium 8 women; 11 men, divalproex loading 9 women; 11 men, divalproex non‐loading 8 women; 12 men. Location: not described Co‐morbidities: not described Adjunctive therapy: none Adjunctive medication: lorazepam was allowed to manage agitation, insomnia, restlessness, irritability, and hostility (4 mg/day on days 1‐4 and 2 mg/day on days 5‐7)
Interventions	Participants were randomly assigned to either: Experimental arm ‐ lithium N = 19 Duration: 10 days Treatment protocol: "After a drug washout period of no more that 72 hours and confirmation of the diagnosis of acute mania and of sub‐therapeutic serum concentration of valproate (< 20 ug/mL) and lithium (< 0.2 mEq/L, participants were randomly assigned to 1 of 3 groups. The lithium group (n = 19) received lithium carbonate at the usual initial dose of 300 mg three times daily on days 1 and 2 followed by gradual dose titration on days 3 through 10." Therapist/face‐to‐face contact: not described Comparator arm ‐ divalproex loading N = 20 Duration: 10 days Treatment protocol: "After a drug washout period of no more that 72 hours and confirmation of the diagnosis of acute mania and of sub‐therapeutic serum concentration of valproate (< 20 ug/mL) and lithium (< 0.2 mEq/L, participants were randomly assigned to 1 of 3 groups. The divalproex loading group (N = 20) received oral divalproex administered via a rapid stabilization schedule: 30 mg/kg/day on days 1 and 2 and 20 mg/kg/day on days 3 through 10." Therapist/face‐to‐face contact: not described Comparator arm ‐ divalproex non‐loading N = 20 Duration: 10 days Treatment protocol: "After a drug washout period of no more that 72 hours and confirmation of the diagnosis of acute mania and of sub‐therapeutic serum concentration of valproate (< 20 ug/mL) and lithium (< 0.2 mEq/L, participants were randomly assigned to 1 of 3 groups. The divalproex non‐loading group (n = 20) received oral divalproex at the usual dose of 150 mg three times daily on days 1 and 2 followed by gradual dose titration for the remaining 8 days." Therapist/face‐to‐face contact: not described
Outcomes	Timepoints for assessment: on days 2 through 6 and on days 8 and 10 Primary outcome: SADS‐C Evaluation of YMRS using observed cases GAS Secondary outcome: Discontinuation Adverse events Use of additional medication YMRS
Notes	Date of study: Funding source: sponsored by Abbott Laboratories Declarations of interest among the primary researchers: none given
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"Patients were randomly assigned to 1 of 3 groups"
Allocation concealment (selection bias)	Unclear risk	Not described
Blinding (performance bias and detection bias) All outcomes	Low risk	"Blinded medication was provided. Blinded raters evaluated the participants."
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	"Seven participants (35%) in each of the divalproex‐treated groups and 9 (47%) in the lithium standard‐titration group discontinued the study medication before the end of the study"
Selective reporting (reporting bias)	Unclear risk	Some missing data
Other bias	Low risk	None identified