Bipolar Affective disorder: Lithium/ANti-Convulsant Evaluation

1. Hospital admission: the primary outcome will be the time to hospital admission during the scheduled randomised treatment period. Admission is a useful indicator of a severe relapse of illness because, since bed provision is now minimal, the clinical threshold for admission is high. Admission to hospital is a useful pragmatic outcome: the majority of patients and clinicians view admission as a negative event that an effective maintenance treatment would be expected to prevent. In certain sites, because some mental health services use intensive alternatives to hospital treatment, the primary outcome will need to be a comparable administrative event such as new attendance at a day patient facility or active home treatment. The appropriate primary outcome will be established with each site before the trial starts and, for the purposes of the trial, a proxy must be 7 days per week treatment at day hospital or 7 days per week home treatment or 24 hour admission to flats and hostel accommodation under regular staff supervision.
2. Concurrent use of adjunctive medication: although there is general agreement that admission to hospital is a clinically meaningful and measurable outcome, it is a somewhat insensitive measure of the less severe mood fluctuations that cause considerable disability in bipolar disorder. Furthermore, manic episodes are more likely than depressive episodes to result in hospital admission. The use of adjunctive antidepressant and antipsychotic medication and of mood stabilizers other than lithium and valproate semisodium will provide a measure of the occurrence of mood episodes that are not severe enough to lead to admission.

1. Global Assessment of Functioning Scale (GAF): the GAF is a brief scale of overall functioning of demonstrated reliability and validity that is used in routine clinical practice and in recent trials in bipolar disorder. The GAF will be used to provide an overall estimate of functioning during the previous year.
2. Deliberate self-harm: deliberate self-harm of suicidal intent (including suicide) is a common outcome in bipolar disorder and of obvious clinical importance. There is observational evidence that lithium therapy reduces the incidence of suicide and it is important to measure this outcome, although it is unlikely that BALANCE will have sufficient power to detect a treatment effect reliably.
3. Quality of life: the EuroQol (EQ-5D) will be used to assess quality of life. This questionnaire has been used successfully in two contemporary trials in psychiatry - the ongoing NHS R&D CUtLASS trial of atypical antipsychotics and the recent MRC-funded trial of cognitive behaviour therapy in bipolar disorder.
4. Adverse events: patients have identified adverse events and side-effects as being one of the main negative aspects of long term medication. Valproate semisodium is a newly licensed drug in the UK and it is essential to record the occurrence of all adverse events.
5. Withdrawal from study treatment: withdrawal from allocated treatment is a useful and pragmatic, although non-specific, measure of the overall acceptability and efficacy of a drug
6. Adherence to study medication: adherence to maintenance treatment is frequently overlooked in maintenance trials in bipolar disorder but is essential for interpreting the trial results