Maternal Mental Health Symptom Profiles and Infant Sleep: A Cross-Sectional Survey

Abstract

The distinct influence of different, but comorbid, maternal mental health (MMH) difficulties (postpartum depression, anxiety, childbirth-related posttraumatic stress disorder) on infant sleep is unknown, although associations between MMH and infant sleep were reported. This cross-sectional survey aimed: (1) to examine associations between MMH symptoms and infant sleep; (2) to extract data-driven maternal MMH symptom profiles from MMH symptoms; and (3) to investigate the distinct influence of these MMH symptom profiles on infant sleep when including mediators and moderators. Mothers of 3-12-month-old infants (n = 410) completed standardized questionnaires on infant sleep, maternal perception of infant negative emotionality, and MMH symptoms. Data was analyzed using: (1) simple linear regressions; (2) factor analysis; and (3) structural equation modelling. MMH symptoms were all negatively associated with nocturnal sleep duration and only postpartum depression and anxiety symptoms were associated with night waking. Three MMH symptom profiles were extracted: depressive, anxious, and birth trauma profiles. Maternal perception of infant negative emotionality mediated the associations between the depressive or anxious profiles and infant sleep but only for particular infant ages or maternal education levels. The birth trauma profile was not associated with infant sleep. The relationships between MMH and infant sleep may involve distinct mechanisms contingent on maternal symptomatology.

Keywords: City BiTS; EPDS; HADS; PTSD; anxiety; birth trauma; depression; infant sleep; mothers; temperament.

Figure 1

Tested pathways playing a role in the association between a maternal mental health symptom profile and an infant sleep outcome. The coefficients a₁ and a₂ are the coefficients of the two linear relationships between each mediator and the independent variable, while the coefficients b₁, b₂ and c are the coefficients of the linear relationship between the dependent variable, the mediators and the independent variable.

Figure 2

Model (A) displays the path model of the effect of the depressive profile on night waking, including mediators and moderators. Model (B) presents the path model of the effect of nocturnal sleep duration by the depressive profile, taking into account mediators and moderators. Model (C) represents the path model of the effect of the anxious profile on night waking, including mediators and moderators. Maternal perception of infant negative emotionality and the method of falling asleep are tested as mediators for all models, as well as maternal educational level and infant age as moderators. Dashed lines show the direct associations between the maternal mental health symptom profiles and infant sleep indicators, without including the other factors in the model. Non-standardized beta coefficients are reported. * p < 0.05; *** p < 0.001.