Structural Equation Modelling of Quality of Life in Adults with Down Syndrome in Nigeria.

Abstract

Current research on the quality of life (QoL) of individuals with Down syndrome (DS) primarily focuses on high-income countries, leaving a significant gap in understanding the unique challenges faced by this demographic in developing countries. This thesis provides empirical evidence on the measurement of QoL in adults (aged ≥ 18 years) with DS across Nigeria’s six geopolitical zones. It develops and tests a conceptual framework based on Schalock and Verdugo’s QoL model. A total of 294 participants, including 224 adults with DS and 70 caregivers, were surveyed using the Personal Outcomes Scale to evaluate eight QoL domains. The cross -sectional study employed advanced statistical methods including Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), Structural Equation Modelling (SEM), and Latent Dirichlet Allocation (LDA) for Topic Modelling. EFA revealed 10 factors in self-reported data and 11 in proxy-reported data, suggesting a more complex QoL structure than the hypothesised eight domains. CFA demonstrated a moderate fit across various indices (such as CFI, GFI, and RMSEA) for both datasets after adjustments, highlighting the model’s adaptability and interconnectedness of QoL domains. Key findings indicate that interpersonal relations (IR) and social inclusion (SI) were supported across the domains in self-reported data, while self-determination (SD) and IR were supported in proxy-reported data. Disparities include non-significant SD↔PW and RI↔PW in self-reported data, but significant in proxy reported data, showing the need for multiple perspectives in measuring QoL for adults with DS. This highlights the complexity of QoL measurement and the importance of iterative model development. The SEM analysis revealed that 22 of 28 domain pairs in self-reported data and 23 of 28 pairs in proxy reported data were statistically significant. IR consistently showed significant relationships across domains in both data types. Positive relationships between IR and domains like SI, rights (RI), emotional well-being (EW), physical well-being (PW), and material well-being (MW) show the pivotal role of IR in causally influencing various aspects of the QoL for adults with DS. The thesis developed a novel normalised Quality of Life Index (nQoLI) which revealed that 70% of adults with DS reported high QoL based on self-reports, and 71% based on proxy reports, with no significant differences between groups. LDA Topic Modelling consistently highlighted the word “want” across both datasets, indicating a desire for autonomy in various aspects of life among adults with DS. The analysis revealed similarities in both groups, with adults with DS expressing a desire to marry (IR domain). Differences between the groups were noted, particularly regarding challenges in controlling meal portions (PW domain). These analyses have not been previously conducted in studies in African countries. Overall, this thesis systematically examines the complexity of QoL in adults with DS. Recommendations include increased funding for DS research; inclusive education policies, vocational training, and employment; mobile technology implementation; region-specific resource allocation; dedicated healthcare regimes; and stigma reduction to improve QoL for this population globally.