ABSTRACT
BACKGROUND: The anxiety and their related disorders (AD) are the most prevalent of all mental health conditions, disproportionately affecting women. The value of perinatal AD screening is well established but there is very limited evidence to support the applicability of existing anxiety screening instruments. To our knowledge, no previous studies have evaluated an AD screening instrument in a perinatal population using full gold standard methodology. OBJECTIVE: To assess the accuracy of the most commonly used and/or recommended screening tools for perinatal AD (i.e., the Edinburgh Postnatal Depression Scale (EPDS) and its anxiety subscale (EPDS-3A), and the Generalized Anxiety Disorder 7 and 2-item Scales (GAD-7 and GAD-2) alongside a clinically derived alternative; the Anxiety Disorder - 13 (AD-13). METHODS: 310 Canadian women completed mood and anxiety questionnaires at approximately 3-months postpartum. Those scoring at/above cut-off on one or more questionnaire completed a diagnostic interview for depression and all AD (nâ¯=â¯115). The accuracy of each scale was assessed via ROC analyses. RESULTS: Only the AD-13 met the standard of a clinically useful screening measure, with an area under the curve (AUC) above 0.8. This was achieved with and without the inclusion of the related disorders. No other measure demonstrated an AUC above 0.8, either including or excluding the related disorders. CONCLUSIONS: Neither the EPDS/EPDS 3-A, nor the GAD-7/GAD-2 can be recommended for widespread use as a perinatal AD screening tool. The high performance of the AD-13 is a good indication that an effective alternative is well within reach.
