[Quality of reporting in studies on bipolar disorders: implications for the development of guidelines]

Selective publishing as well as inadequate reporting of clinical trials entail a risk of bias in clinical decision making. Therefore the CONSORT statement was introduced to improve the quality of reporting of randomized controlled trials (RCT). This study aimed to assess the quality of reporting of RCTs on pharmacological treatment of bipolar disorder in relation to publication period and endorsement of publication guidelines.

In the context of the development of the German evidence and consensus-based S3 guidelines for diagnosis and therapy of bipolar disorders a systematic literature search was carried out to identify all RCTs published between 2000 and 2010 relevant to the pharmacological treatment of bipolar disorders. An adapted checklist based on the CONSORT statement was used to assess the quality of reporting.

A total of 134 RCTs were included in this analysis. Of the 72 checklist items, 43% were generally reported adequately (reported in  ≥ 75% of all trials) and 25% inadequately (reported in  < 25% of all trials). Reporting was generally poor for randomization, effect size (reported in 22%) and number needed to treat (NNT 16%). No consistent trend could be shown for improvement in quality of reporting over time or for journals that do or do not endorse the URM (uniform requirements for manuscripts submitted to biomedical journals).

Clinical investigators as well as editors and reviewers should be further encouraged to follow publication guidelines otherwise trials have to be downgraded or excluded from systematic evaluations.