Adverse childhood experiences and premature mortality through mid-adulthood: A five-decade prospective study

Abstract

Background: Adverse childhood experiences (ACEs) can have lasting effects on adult health and survival. In this study, we aimed to examine how the cumulative number and clustering patterns of ACEs were related to premature mortality.

Methods: Participants (N=46 129; 45% White, 48% Black; 49·5% females) were offspring (born in 1959-1966) of participants enrolled in the Collaborative Perinatal Project (CPP). We conducted latent class analysis to examine the clustering patterns of ACEs assessed between children's birth and age seven. We also calculated the cumulative ACE scores of 13 individual ACEs. Cox regression models were used to examine the associations of ACE clusters and scores with risk of premature mortality from adolescence to mid-adulthood.

Findings: At the start of the follow-up for mortality in 1979, participants were 12-20 years old (Mean=15·99 years), and within the 38-year follow-up through 2016, 3 344 deaths were observed among the 46 129 CPP offspring. Five latent classes of ACEs were identified. Compared to children with Low Adversity (48% of the sample), children in Family Instability (9%, HR=1·28, 95%CI 1·07-1·53), Poverty & Crowded Housing (21%, HR=1·41, 95%CI 1·24-1·62), and Poverty & Parental Separation (19%, HR=1·50, 95%CI 1·33-1·68) classes had higher hazards of premature mortality. In addition, children with 2 (HR=1·27, 95%CI 1·14-1·41), 3 (HR=1·29, 95%CI 1·15-1·45), and 4+ (HR=1·45, 95%CI 1·30-1·61) ACEs had higher hazards of mortality than those with no ACE. The clusters of Poverty & Crowded Housing (HR=1·28, 95%CI 1·10-1·49) and Poverty & Parental Separation (HR=1·23, 95%CI 1·02-1·48) remained associated with higher risk of premature mortality, beyond the cumulative risk of higher number of ACEs (HR=1·05, 95%CI 1·01-1·08).

Interpretation: About half of the CPP cohort experienced early life adversities that clustered into four distinct patterns, which were associated with different risk of premature mortality. It is important to deepen our understanding of how specific clusters of childhood adversities affect health and premature mortality to better inform approaches to prevention and interventions.

Figure 1

Five latent classes of early childhood adversities identified in the CPP cohort. Note: Children in Class 1 had low risk of experiencing any adversity. Children in Class 2 were highly likely to experience physical or emotional harshness and physical neglect. Children in Class 3 had relatively high probabilities of experiencing family instability (e.g., marital changes, residential changes, parent/sibling death, foster care, or divorce/separation). Children in Class 4 were mainly exposed to poverty and crowded housing. Children in Class 5 had particularly high probabilities of experiencing poverty, welfare use, and parental divorce/separation and some risk of marital and residential changes as well.

Figure 2

Risk of premature mortality for individuals in each latent class of ACEs. Note: All the other ACE classes had higher mortality risk than the low adversity class. The mortality rate (i.e., number of deaths per 100 000 individuals per year) was 183 for the Poverty & Crowded Housing class, 177 for the Poverty & Parental Separation class, 135 for the Family Instability class, 139 for the Parental Harshness & Neglect class, and 114 for the Low Adversity class.