More than 40 years after its efficacy in the management of acute mania was first described, lithium remains first-line drug treatment for bipolar mood disorders. Studies conducted under natural conditions have shown that one of the main causes of failure is associated with poor treatment compliance. The poor compliance rate observed with treatment of this nature currently stands between 18 and 47%. The differences highlighted in the various studies can be explained, at least in part, by the definitions and assessment methods employed in each study. It is generally acknowledged that the longer the evaluation period, the worse the compliance. Certain factors accounting for poor compliance point to the patients themselves whilst others are directly treatment-related, i.e, therapeutic efficacy and tolerance. Compliance with two pharmaceutical dosage forms of lithium carbonate, namely the standard (250 mg) and sustained-release (400 mg) formulations, was assessed under natural conditions in patients satisfying DSM IV diagnostic criteria for bipolar disorder type I, II or non-specified or for major recurrent depression. Evaluation via a <<heteroquestionnaire>> was conducted retrospectively over 3 months and prospectively over 6 months. Assessments were carried out at baseline and after 3 and 6 months. Patients who responded at a minimum of two evaluations were included in the statistical analysis. Forty-four (44) of the 50 patients selected were included. A global compliance rate of 74% was recorded at 6 months and this was deemed to be excellent. These results corroborate published data relating to the overall good compliance observed with lithium patients. Tolerance and compliance were better in the group receiving the sustained-release formulation (400 mg per tablet). A certain number of factors likely to influence the quality of compliance were measured. No correlation was established between the quality of compliance assessed by the doctor and data relating to the disorder or to the patients' understanding of and attitude towards treatment. Two variables were significantly correlated with the extent of patient-assessed compliance, namely those patients displaying better compliance generally started treatment later in life (35.7 years +/- 12.0 compared with 25 years + 15.6; p = 0.03) and took fewer tablets (1.7 tablets/day +/- 1.1 versus 2.7 tablets/day 1.6; p = 0.03). These factors are discussed.
