Adaptation and validation of the Spanish version of the Mood Disorder Questionnaire for the detection of bipolar disorder

Objective: The Mood Disorder Questionnaire (MDQ) is an instrument for the detection of patients with bipolar disorder (BD). The original English version is validated in both the psychiatric and the general population, but a validated Spanish version is not yet available. Psychometric properties of the Spanish adaptation of the MDQ in psychiatry are described. Methods: The MDQ is a self‐administered questionnaire comprising a list of 13 hypomanic symptoms and two questions about concurrence of symptoms and functional impairment caused by the symptoms. We selected patients from 15 psychiatric outpatient departments, diagnosed with BD type I and II (BDI and BDII) and major depression (MD) according to DSM‐IV‐TR criteria (concurrent validity instrument). A control group of healthy subjects (HS) was selected. The patient‐selection criteria included stability of the disorder and pharmacological treatment. The MDQ was administered to 236 subjects, distributed among the four groups, on two occasions, four weeks apart. We analysed the internal consistency, test–retest reliability, and discriminative capacity of the MDQ for the detection of patients with BD. Results: Concurrent validity based on diagnosis according to DSM‐IV‐TR was 0.83. The internal consistency, evaluated by Cronbach’s α, was 0.90. The mean (SD) number of affirmative responses by group was: 9.8 (2.4) for BDI, 8.5 (2.8) for BDII, 2.7 (2.2) for MD, and 1.02 (1.9) for HS. Statistically significant differences between all the groups were found (Kruskal–Wallis test, p < 0.001). Concurrent validity using the diagnostic variable was 0.83. Test–retest reliability was 0.92. We analysed the scale’s discriminative capacity, revealing a sensitivity value of 0.60 95% confidence interval (CI) = 0.51–0.69] and a specificity value of 0.98 (95% CI = 0.94–0.99) in the detection of BD. The positive and negative probability ratios were 35.5 and 2.4, respectively. If we consider only seven positive responses as the discriminative criterion, sensitivity increases to 0.81 (95% CI = 0.73–0.88), the specificity value is 0.95 (95% CI = 0.89–0.98) and the positive and negative probability quotients are 16 and 5.3. Conclusions: The psychometric characteristics of the Spanish version are similar to those of the original version. In the Spanish adaptation of the MDQ, seven positive responses to hypomanic symptoms show a good discriminative capacity for BD in patients attending psychiatric outpatient facilities; therefore, this cut‐off score is proposed for the detection of BD in psychiatric outpatients.