Preliminary comparison of predictive and empiric lithium dosing: impact on patient outcome.

Objective: To evaluate whether initiating lithium with predictive dosing compared with empiric dosing improves outcome in patients with manic symptoms.

Design: The study was a randomized, single-blind design and used the Modified Slattery predictive method.

Setting And Participants: Eighteen inpatients at an urban psychiatric hospital with a Mania Rating Scale (MRS) score greater than or equal to 24 were enrolled.

Outcome Measures: The study endpoint was defined as an MRS rating less than or equal to 14 or discharge from the hospital. Assessments (MRS, Brief Psychiatric Rating Scale, Clinical Global Impression, Systematic Assessment for Treatment of Emergent Events Scale, quality of life measures) were completed at baseline, on days 3 or 4 and 7 or 8, and weekly thereafter.

Results: The predictive group achieved a therapeutic concentration significantly sooner than did the empiric group (p = 0.004); however, the mean serum lithium concentration at discharge did not differ between the groups. The predictive group was taking significantly higher dosages of antipsychotics in chlorpromazine equivalents on day 3 or 4 (p = 0.05). Significantly fewer gastrointestinal/genitourinary adverse effects on day 3 or 4 were reported by patients in the predictive group (p = 0.04). No difference was found between groups with any rating scale or other pharmacokinetic or medication item. Even though the difference did not meet statistical significance, the predictive group's length of stay in the acute unit was three days shorter than that of the empiric group, which may represent significant cost savings.

Conclusions: The preliminary data do not suggest that patient outcome is improved by using Modified Slattery predictive dosing; however, the suggestion of a shorter length of stay in a restrictive unit merits further evaluation.