Research report Cross validation with the mood disorder questionnaire (MDQ) of an instrument for the detection of hypomania in Spanish: The 32 item hypomania symptom check list (HCL-32)

Background: The detection and diagnosis of present or past hypomanic episodes is of key importance for the differential diagnosis between depressive disorders and type II bipolar disorder. However, there are few instruments available to satisfactorily screen for the latter condition. The Hypomania Symptom Checklist-32 (HCL-32) is a self-applied questionnaire with 32 hypomania items and 8 severity and functional impact items which is being developed in several European countries for this purpose. Our aim was to develop and validate the psychometric properties of the HCL-32 scale in Spain in patients with bipolar disorder and to compare its properties with other instruments available for the detection of bipolar II disorder. Methods: Patients were selected from 15 psychiatric outpatient departments, diagnosed with type I or type II bipolar disorder (BDI and BDII) and unipolar major depression (MD) according to DSM-IV-TR criteria. A control group of healthy subjects (HS) was likewise assessed. The patient selection criteria included a well-established diagnosis and a stable disorder and pharmacological treatment. The HCL-32 was administered to 237 subjects distributed among the above groups, on two occasions four weeks apart. We analysed the internal consistency, test–retest reliability and discriminative capacity of the HCL-32. Results: The internal consistency of the Spanish version of the HCL-32, evaluated by Cronbach's alpha, was 0.94. Mean of affirmative questions by group were 21.2 (SD 5.8) for BDI, 19.3 (SD 6.2) for BDII, 8.6 (SD 6.6) for MD and 6.6 (SD 6.1) for HS, with statistically significant differences between them (Kruskal–Wallis test, pb0.001). Concurrent validity using the diagnosis variable was 0.72. Test–retest reliability was 0.90. We analysed the best cut-off point by means of a ROC curve analysis; for 14 affirmative responses, a sensitivity of 0.85 95%CI (0.78, 0.91) and specificity of 0.79, 95%CI (0.72, 0.87) were obtained. The