ABSTRACT
BACKGROUND: Depending on the classification system used, 5-40% of manic subjects present with concomitant depressive symptoms. This post-hoc analysis evaluates the hypothesis that (hypo)manic subjects have a higher burden of depression than non-(hypo)manic subjects. METHODS: Data from 806 Bipolar I or II participants of the Stanley Foundation Bipolar Network (SFBN) were analyzed, comprising 17,937 visits. A split data approach was used to separate evaluation and verification in independent samples. For verification of our hypotheses, we compared mean IDS-C scores ratings of non-manic, hypomanic and manic patients. Data were stored on an SQL-server and extracted using standard SQL functions. Linear correlation coefficients and pivotal tables were used to characterize patient groups. RESULTS: Mean age of participants was 40 ± 12 years (range 18-81). 460 patients (57.1%) were female and 624 were diagnosed as having bipolar I disorder (77.4%) and 182 with bipolar II (22.6%). Data of 17,937 visits were available for analyses, split into odd and even patient numbers and stratified into three groups by YMRS-scores: not manic < 12, hypomanic < 21, manic < 30. Average IDS-C sum scores in manic or hypomanic states were significantly higher (p < .001) than for non-manic states. (Hypo)manic female patients were likely to show more depressive symptoms than males (p < .001). Similar results were obtained when only the core items of the YMRS or only the number of depressive symptoms were considered. Analyzing the frequency of (hypo)manic mixed states applying a proxy of the DSM-5 mixed features specifier extracted from the IDS-C, we found that almost 50% of the (hypo)manic group visits fulfilled DSM-5 mixed features specifier criteria. CONCLUSION: Subjects with a higher manic symptom load are also significantly more likely to experience a higher number of depressive symptoms. Mania and depression are not opposing poles of bipolarity but complement each other.
ABSTRACT
BACKGROUND: Bipolar disorders are often not recognized. Several screening tools have been developed, e.g., the Hypomania Checklist-32 (HCL-32) and the Mood Disorder Questionnaire (MDQ) to improve this situation. Whereas the German HCL-32 has been used in non-clinical samples, neither the HCL-32 nor the MDQ has been validated in German samples of mood-disordered patients. Additionally, hardly any prior study has included patients with non-mood disorders or has considered potential effects of comorbid conditions. Therefore the goal of this study was to test the validity of both scales in a diverse patient sample while also taking into account psychiatric comorbidity. METHOD: A multi-site study was conducted involving seven centers. Patients (n=488) completed the HCL-32 and MDQ and were independently interviewed with the Structured Clinical Interview for DSM (SCID). RESULTS: Sensitivity for bipolar I was similar for HCL-32 and MDQ (.88 and .84) but slightly different for bipolar II (.90 and .83), specificity, however, was higher for MDQ. In general, a comorbid condition led to increased scores in both tools regardless of whether the primary diagnosis was bipolar or not. LIMITATIONS AND DISCUSSION: Although we included not just mood-disordered patients, detailed subgroup analyses for all diagnostic categories were not possible due to sample sizes. In summary, HCL-32 and MDQ seem fairly comparable in detecting bipolar disorders although their effectiveness depends on the goal of the screening, psychiatric comorbidity, and potentially the setting.
