Design limitations to bipolar II treatment efficacy studies: A challenge and a revisionist strategy

Trials examining medication efficacy for bipolar II disorder commonly employ a set of standardized interval measures to assess outcomes. The key issue is whether such interval measures pick up changes in the severity, duration and frequency of depressive, hypomanic and euthymic episodes.

We examine the application of measures most commonly used to monitor progress in nine studies involving participants with a bipolar II disorder and published in journals with a moderate to high impact factor.

Studies rarely provided interval details for assessing depressive and hypomanic symptoms. None specified whether ratings of depressive and hypomanic symptoms were based on severity, duration or number of symptoms, and none recorded any data on euthymic periods.

Our sample of reviewed studies was small and our analyses focused only on the three most commonly used outcome measures. We advocate for complementary subjective daily mood monitoring strategies but recognize that such strategies need to be validated in future studies.

We argue that interval ratings undertaken weekly or over longer periods may compromise efficacy data. We recommend that userguides be developed to ensure standard outcome measures are employed consistently across trials, and that specific details be published in trial papers about how measures were employed and what mood episode characteristics were measured at each assessment. We also argue for daily ratings to be used as an outcome measure to provide data on severity, frequency and duration of depressive, hypomanic and euthymic periods in intervention studies of those with a bipolar II disorder.