Structural equation modelling of holistic health variables, healthy diet, physical activity, and quality of life: comparison between Malaysian and Nigerian undergraduate students

• Main Author: Abdulrahman, Abdulwali Sabo
• Format: Thesis
• Language: English
• Published: 2025
• Subjects: R Medicine; RA Public aspects of medicine
• Online Access: http://eprints.usm.my/63133/

Holistic health is increasingly recognized as a comprehensive approach that considers the whole person and the interplay of multiple life dimensions. Well-being arises from the interaction between individual potentials (IP), demands of life (DL), social determinants (SDH), and environmental determinants (EDH). Healthy diet (HD) and physical activity (PA) further contribute to overall quality of life. This study therefore aims to evaluate how SDH, EDH, IP, DL, HD, and PA interact within a holistic health model to influence quality of life. The study applied a mixed-methods approach, conducted in three phases, to develop and validate a holistic health model. In Phase I, four new instruments were developed to measure key determinants of holistic health: the Social Determinants of Health Questionnaire (SDHQ), Environmental Determinants of Health Questionnaire (EDHQ), Demands of Life Questionnaire (DLQ), and Individual Potentials Questionnaire (IPQ). These tools were developed through extensive literature review, in-depth interviews, and expert evaluations to ensure content and face validity. Phase II covers a validation study using a cross-sectional survey of 1,460 undergraduate students (730 each from Nigeria and Malaysia). This phase employed exploratory and confirmatory factor analyses (EFA and CFA), alongside assessments of reliability and validity, including composite reliability (CR), average variance extracted (AVE), Cronbach’s alpha, and test-retest reliability. Phase III investigated the structural relationships between SDH, EDH, DL, IP, HD, PA, and quality of life, using independent samples of 1,140 students (570 from each country), as well as the measurement and structural invariance to confirm the cross-cultural applicability of the instruments. Additionally, multigroup SEM was conducted to compare structural relationships across Nigerian and Malaysian samples. In Phase I, four holistic health questionnaires were developed: SDHQ (20 items, 2 factors), EDHQ (18 items, 2 factors), DLQ (18 items, 3 factors), and IPQ (14 items, 2 factors), all showing acceptable content and face validity (indices = 0.83–1.00). In Phase II, demonstrated adequate validity and reliability across all instruments, with adequate fit indices (CFI = 0.928–0.967; TLI = 0.910–0.957; SRMR = 0.039–0.080; RMSEA = 0.041–0.068) and high reliability (Cronbach’s α = 0.815–0.947; CR = 0.760–0.950; ICC = 0.765–0.987). In Phase III, structural equation modeling supported 8 of 10 hypothesized pathways in both Malaysian and Nigerian samples, with good model fit (CFI = 0.972-0.989, TLI = 0.954-0.982, SRMR = 0.021-0.026, RMSEA = 0.039-0.060, RMSEA p-value = 0.110-0.879). The instruments demonstrated measurement and structural invariance across both groups (ΔCFI and ΔTLI < 0.01, ΔRMSEA < 0.015), and multigroup SEM confirmed six similar path relationships (CFI = 0.982, TLI = 0.969, SRMR = 0.020, RMSEA = 0.052, RMSEA p-value = 0.360). The study confirmed that the SDHQ, EDHQ, DLQ, and IPQ are valid, reliable, and invariant across Nigerian and Malaysian students, making them suitable for assessing holistic health factors. It also introduced one of the first quantitative models relating these variables to HD, PA, and quality of life, offering a foundation for future interdisciplinary research on the interrelationship of social, environmental, and individual determinants of health.