Childhood Poverty and Retrospective Recall of Adverse Childhood Experiences

Background: Despite the consistent finding that adverse childhood experiences (ACEs) and child maltreatment (CM) are associated with poverty, much remains unknown about the potential causal nature and the magnitude of this relationship. The recent work on the nature of CM and its co-occurrence with other forms of adverse experiences makes it important to be able to take this into account in regard to potential prevention and intervention policies or programs. Few longitudinal studies of maltreatment and/or ACEs exist and even fewer have adequate measurements of poverty and other socioeconomic indicators over time. This study aimed to examine the relationship between various family and community economic indicators and later recall of ACEs overall and by type.

Methods: The sample frame for this study included all children from the Child Development Supplement (CDS) I-III originating in 1997 and their caregivers who were subjects of the ongoing Panel Study of Income Dynamic (PSID) household survey (n=3,563). Children were at least 18 in 2014 they were asked to complete the Childhood Retrospective Circumstances Study (CRCS) to report on their ACEs. We linked family-level data from the PSID main survey to children-level data from the CDS (I-III) for repeated measures analyses of poverty and harsh and neglectful parenting approach (a proxy for CM), and individual-level data from CRCS to examine how prospective measures of economic factors were associated with the retrospective recall of ACEs. The final sample consisted of all children participating in the CDS I who also responded to the 2014 self-report survey as adults (n=660). Analyses employed a mixed-model approach with weights to include repeated childhood measures.

Results: Most respondents reported at least one ACE (about 80%), and there was little differentiation possible by type. The results of the mixed regression model showed that an indicator of family poverty at birth was associated with an increase in the number of ACEs reported (b=0.560, p=0.000) while receipt of welfare during pregnancy was associated with a decreased likelihood of later recall of ACEs (b=-1.476, p=0.000). More recent measures of poverty (CDS II or III) appeared to moderate the impact of early poverty (b=-0.024, p=0.000). Childhood community settings were also predictive. Adults who had lived in higher log-median income neighborhoods (b=-7.398, p=0.000), and those living in higher proportion Black neighborhoods were less likely to report ACEs (b=-0.039, p=0.000). The opposite was true for zip codes with higher proportions of Asian (b=0.088, p=0.000) or American Indian communities (0.219, p=0.000).

Conclusions and Implications: Childhood poverty increased the likelihood of later reports of ACEs, although receipt of poverty services prior to birth decreased that likelihood. This study is consistent with the notion that poverty has a causal relationship with ACEs. The fact that it was difficult to differentiate between types of ACEs suggests that anti-poverty efforts may impact multiple forms of ACEs though more research is required to establish the range of benefits.