• antidepressants;
  • audit;
  • bipolar;
  • lithium;
  • prescribing practice;
  • quality improvement;
  • serum levels


The study was designed to test an audit-based quality improvement programme (QIP) addressing lithium prescribing and monitoring in UK mental health services.


A baseline clinical audit was conducted against the following standards: (i) measurement of renal and thyroid function before initiating treatment with lithium and (ii) recommended monitoring of serum lithium and renal and thyroid function during maintenance treatment. A re-audit was conducted at 18 months and a supplementary audit at three years.


Data were submitted for patients at baseline (n = 3,373), re-audit (n = 3,647), and supplementary audit (n = 5,683), 57% of whom had bipolar disorder. The baseline findings prompted a patient safety alert issued by the National Patient Safety Agency. By supplementary audit, the proportion of patients having four serum lithium tests over the previous year had increased from 30% at baseline to 48%, and the respective proportions that had two tests of renal function from 55% to 70% and thyroid function from 49% to 66%. Elderly patients and those prescribed a drug known to interact with lithium were not more likely to be monitored in line with the audit standards. Between baseline and supplementary audit, the proportion of patients with a diagnosis of bipolar disorder prescribed an antidepressant increased from 36% to 41%.


Improvements in biochemical monitoring of lithium treatment were achieved over time with participation in a QIP that included benchmarking of performance against clinical standards and customized change interventions. Nevertheless, gaps remain between the standard and current practice. Antidepressants are frequently prescribed in patients with bipolar disorder despite a paucity of evidence supporting their efficacy.