Modeling Depressive Symptom Trajectories Among Mothers of Children with ASD from Diagnosis to Age 9

Background: Elevated levels of depressive symptoms among mothers of children with ASD have been well documented (Davis & Carter, 2008; Montes & Halterman, 2007). However, few studies have examined mothers’ depression longitudinally, and most have considered these mothers as a homogeneous group when examining their depressive symptoms. Although several studies suggest some degree of stability in maternal depression over time (e.g., Carter et al., 2009), none have considered the potential heterogeneity of developmental trajectories of maternal depression. Overall sample means and correlations may mask the presence of subgroups of mothers whose depression follows distinct patterns.

Objectives:  The purpose of this study was two-fold: 1) to identify subgroups of mothers with different depression trajectories from the time of child diagnosis (between age 2-4) to age 8.5-9, in a large inception cohort of young children with ASD; and 2) to explore the associations between trajectory membership and key parent/family variables.

Methods: Data were drawn from the Canadian Pathways in ASD study and included the mothers of 284 children with ASD. At the time of initial data collection, which occurred within 4 months of diagnosis, the children’s mean age was 40.33 months (SD=9.3). Mothers completed the Symptom Checklist-90-R (Derogatis, 1994) to assess their depressive symptoms at four time points: within 4 months of diagnosis, 24 months post-diagnosis, when the children were 7.5-8 years old, and when the children were 8.5-9 years old. In addition, at diagnosis, mothers completed a demographic survey, the General Functioning subscale of the McMaster Family Assessment Device (Byles, Byrne, Boyle, & Offord, 1988), and the Social Support survey (NLSCY, 2008-2009). Severity of children’s ASD symptoms  was assessed using the Autism Diagnostic Observation Schedule (Gotham, Pickles, & Lord, 2009). A semi-parametric, group-based analytical strategy in SAS, PROC TRAJ (Jones, Nagin, & Roeder, 2001), was used to examine individual differences in developmental trajectories.

Results: Three distinct trajectory groups provided the best fit to the maternal depression data over time. Group 1 (65.2% of the sample) had the lowest mean depression scores at diagnosis and a stable trajectory. Group 2 (27.3%) had moderate  mean depression scores at diagnosis and a declining trajectory. Group 3 (7.4%) had the highest mean depression score at diagnosis and an inclining trajectory. Both family social support and family functioning distinguished between the trajectories of Group 1 (low, stable), and Groups 2 (medium, declining) and 3 (high, inclining). In addition, older child age at the time of diagnosis was significantly associated with membership in Group 2 compared to Group 1.

Conclusions: There is considerable heterogeneity in the trajectories of maternal depressive symptoms. Lower levels of perceived social support, poorer family functioning, and higher child age at the time of diagnosis all appear to be associated with higher depressive symptomatology. While depression improves over time in some mothers, others experience increased symptoms. Additional research is needed to identify other variables contributing to these trajectories and intervention approaches for assisting these families.