Temporal trends, geographic patterns and predictors of chronic health conditions in Australian children: a mixed multilevel analysis.

OBJECTIVE

This study examined temporal trends, spatial distribution and multilevel predictors of chronic conditions using nationally representative data from the Longitudinal Study of Australian Children.

STUDY

DESIGN

Longitudinal population-based cohort study using multilevel and spatial analyses.

METHODS

Multilevel mixed-effects logistic regression models were used to identify child-level, parent-level and context-level predictors across waves 2-8. Model fit was assessed using the intraclass correlation coefficient and Akaike Information Criterion. Spatial mapping was performed using Statistical Area Level 2. Significance was set at p < 0.05.

RESULTS

The prevalence of chronic conditions rose from 32.1% to 58.7%. Significant child-level predictors included severe sleep problems (adjusted odd ratio (AOR) = 2.98), sedentary behaviour (AOR = 1.27), poor diet (low fruit/vegetable intake: AOR = 1.14; high-fat foods: AOR = 1.88), sugary drink intake (AOR = 1.23) and obesity (AOR = 1.43). Parental and contextual predictors included Indigenous status (AOR = 1.68), financial hardship (AOR = 1.16), maternal medical conditions (AOR = 1.48), paternal depression (AOR = 1.61), low household income ($A <500: AOR = 1.26), paternal smoking (AOR = 1.41), heavy alcohol consumption by mothers (AOR = 1.77) and fathers (AOR = 1.72), residence in non-metropolitan areas (AOR = 1.53) and living in moderately to very remote areas (AOR = 2.01).

CONCLUSION

Chronic conditions among Australian children are increasing, driven by interconnecting child, family and geographic risk factors. Multilevel interventions are needed to address these rising inequalities.