Abstract
BACKGROUND: Reduced well-being and depressive episodes frequently complicate pregnancy and can result in serious adverse outcomes for both mother and infant if left untreated. This study aimed to assess the psychometric validity of the 5-item World Health Organization index (WHO-5), and to evaluate if the WHO-5 index can serve as a proxy for two items of core depressive symptoms from the Major Depression Inventory (MDI), identified as MDI-2. Additionally, the paper aimed to assess well-being and detect risk factors of reduced well-being using the WHO-5 index. METHODS: Using the WHO-5 index, this study analyzed the psychological well-being of a population of 37,129 women in their first trimester of pregnancy. The psychometric validity of the WHO-5 index was evaluated using confirmatory factor analysis and item response theory. Furthermore, cut-off scores of the WHO-5 index previously associated with mental distress, ≤ 50 and ≤ 28, were evaluated in terms of sensitivity, specificity, and positive- and negative predictive values compared to the MDI-2 in a subsample of n = 1001 women. RESULTS: Lower mean score on the WHO-5 index was seen among pregnant women ≤ 25 years old, with lower educational level, who were unemployed, nulliparous, did not understand Danish language, were not cohabitating, were smokers, and women who did not exercise prior to pregnancy. The WHO-5 index was found to be a valid psychometric instrument, however, scores could not be pooled or compared across women who understood Danish and women who did not understand Danish. The sensitivity and specificity of cut-off score of ≤ 50 on the WHO-5 index with a for predicting the presence of one core symptom from the MDI-2 was 0.81 and 0.82, respectively, while the sensitivity and specificity using a cut-off of ≤ 28 was 0.32 and 0.98, respectively. CONCLUSIONS: The WHO-5 index had high degree of acceptability and identified well-known risk factors of mental distress in a pregnant population. The WHO-5 index was found to be a valid psychometric instrument in pregnancy, however, our analysis indicated low predictive value of the investigated cut-off scores on the WHO-5 index in identification of MDI-2 core symptoms.