Childhood Multiple Risks and Health Trajectories in Adulthood: A Latent Class Analysis

Background/Purpose: The relationship between childhood socioeconomic risks and adult health has been well documented in life course studies. However, these studies focused on few childhood factors or used cumulative index by summing all the risk items and using accumulated mean values, which may miss important information about individuals’ childhood. Guided by a life-course perspective, this study addresses the following gaps in the life course literature. First, some of childhood risks are simultaneous and some forms of risk covariation are sequential such as job loss, poverty, unstable housing, and poor health. Each of these disadvantaged circumstances can be formed like a chain of subsequent life events, which, in turn, might determine subsequent health outcomes in adulthood. Therefore, we examine how childhood multiple risk factors interact with each other and form distinct patterns by identifying subgroups according to socioeconomic risk factors in childhood. Second, many studies still used cross-sectional data and examined the association of multiple risks with health outcomes measured at one point in time, limiting the possibility of variations in the health status of individuals over time. In this study, we investigated the associations between health trajectories in adulthood and childhood multiple risk factors.

Method: Data came from seven waves of the Health and Retirement Study (HRS). We restricted our sample to middle-aged adults between age 51 and 64 who participated in at least three surveys in order to health trajectories (N=6,474; Observations=25,848). A latent class analysis was adopted to identify latent classes according to childhood socioeconomic risk factors. Then, a random coefficient regression model in multi-level growth curve framework was used to analyze the association between the latent classes and health trajectories in adulthood.

Results: We found three groups of latent classes, which were named as the multiple risk group, the low social class group, and the low risk group. The latent classes by childhood socioeconomic risk factors were associated with current socioeconomic factors. Especially, the multiple risk group who experienced all financial and social risks during childhood included significantly more minority individuals with longest poverty experience. Results from the random coefficient growth curve models showed significant association between the latent classes and health trajectories in adulthood (depression, functional limitation, and self-rated health). Furthermore, the differences between the multiple risk group and the other two groups were still strongly significant with current socioeconomic factors controlled.  

Conclusion: Our findings suggest that childhood socioeconomic risk factors can be understood and identified as a form of combinations and each combination has different relations with and has a predictive role in health trajectories in adulthood. The strong association between childhood risk factors and adulthood health trajectories highlight the importance of early life circumstances on subsequent health outcomes. Early stage recognition, prevention, and supportive policy measures against childhood socioeconomic disadvantages need to be promoted in order to prevent poor adulthood health and further diminish socioeconomic inequality in health.