Graduation Date

Fall 12-16-2022

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Programs

Epidemiology

First Advisor

Shinobu Watanabe-Galloway

Second Advisor

Melissa Tibbits

Third Advisor

Kendra Schmid

Fourth Advisor

Kendra Ratnapradipa

Abstract

To comprehensively capture the risk and protective factors of ACE exposure and ACE-related outcomes, we propose to consider a social-ecological model framework. The concept of the Social-Ecological Model was originally introduced in the 1970s by psychologist Urie Bronfenbrenner and designed to theoretically capture the multifaceted influence different personal, social, and environmental factors have on health behaviors and outcomes. Additionally, this would help to determine what prevention and intervention methods would be more appropriate to implement. This conceptual framework had five system levels: microsystem (individual), mesosystem (interpersonal), exosystem (community context), macrosystem (societal and cultural influences), and chronosystem (internal and external elements of time and historical content; policy).34 CDC adapted and recently modified this model into four main levels: individual (genetics and personal factors), relationships (family and social networks), community (neighborhood and school/workplace), and societal (policy).32

Further, we considered the pathways between ACEs risk and protective factors, ACE prevalence and depression. We hypothesized that the relationship between ACEs and depression can be more accurately explained by conducting a comprehensive examination of social-ecological factors. To assess these relationships, we incorporated the SEM approach. SEM is a multivariate modeling approach that allows for the examination of complex pathways of observed (measured) and unobserved (latent) effects. This approach incorporates theory, path diagrams, and linked regression-style equations to understand the relationships between variables in the model.35,36 Unlike a traditional multivariate regression analysis, where the independent and dependent variables are evident, with SEM a dependent variable in one model can become independent in another which enables it to infer causal relationships.35 SEM provides us with the opportunity to address the knowledge gap of understanding the multivariate influence different social- ecological variables have on depression.

Furthermore, SEM can account for measurement error when estimating effects, test the model fit of the data, and specify statistical models derived from theory.36-38 However, it is important to understand that determining causal relationships in non-experimental studies should be cautioned. Because our study incorporates the social-ecological theory to comprehensively examine the pathways between ACEs and depression, we believed this statistical approach is the most appropriate in addressing our study aim. By utilizing the SEM approach, we anticipated our results providing us with a more well-rounded picture of ACEs and their impact on children and adolescents.

The specific aims for this dissertation were as follows:

Aim 1: Examine what methodological approaches have been used to determine the ACE prevalence among persons 17 years or younger in the U.S. and investigate the prevalence of different ACEs among persons 17 years or younger in the U.S. For this aim, we conducted a scoping literature review to understand how ACE prevalence is determined.

Aim 2: Examine the trend of ACE prevalence among children and adolescents aged 6-17 years in the U.S. from 2016-2019. For this aim, we conducted a cross-sectional study of ACE prevalence using the 2016-2019 NSCH surveys.

Aim 3: Examine the influence social-ecological factors have on ACEs and mental health outcomes, among children and adolescents aged 6-17 years in the U.S. For this aim, we used the 2019 NSCH to conduct a structural equation model analysis to examine the pathways from ACEs and mental health outcomes.

Comments

2022 Copyright, the authors

Available for download on Wednesday, June 07, 2023

Share

COinS