Screening for postpartum depression among low-income mothers using an interactive voice response system

Abstract

This study tested the feasibility of using an interactive voice response (IVR) phone system to screen for postpartum depression among low-income, English- and Spanish-speaking mothers. Newly delivered mothers were interviewed in the hospital. Consenting subjects completed a background questionnaire and were asked to call an automated phone system 7 days postpartum to complete an IVR version of the Edinburgh Postnatal Depression Screen (EPDS). During the phone screen, subjects were branched to different closing narratives based on their depression scores which were later posted to a password protected website. Logistic regression was used to assess relationships between demographic and psychosocial factors, IVR participation, and depression scores. Among 838 ethnically diverse, low income, postpartum mothers, 324 (39%) called into the automated phone screening system. Those who called were more likely to have at least a high school education (OR = 1.63, 95%CI: 1.23, 2.16), be employed (OR = 1.48, 95%CI: 1.08, 2.03) and have food secure households (OR = 1.50, 95%CI: 1.06, 2.13). There was no statistically significant difference between callers and non-callers in terms of marital status, race/ethnicity, parity, or self-reported history of depression. Postpartum depression symptoms were present in 17% (n = 55) and were associated with being single (AOR = 2.41, 95% CI: 1.29, 4.50), first time mother status (AOR = 2.43, 95% CI: 1.34, 4.40), temporary housing (AOR = 2.35, 95% CI: 1.30, 4.26), history of anxiety (AOR = 2.79, 95% CI: 1.69, 6.67), and history of self-harm (AOR = 2.66, 95% C: 1.01, 6.99). Automated phone screening for postpartum depression is feasible among disadvantaged mothers but those with the highest psychosocial risk factors may not choose or be able to access it. IVR could be used to supplement office- and home visit-based screening protocols and to educate patients about mental health resources.