# Small-area geographical variation in the prevalence of diabetes amongst Australian youth aged <20 years in 2021

## TL;DR

This study maps diabetes prevalence in Australian youth under 20, finding higher rates in rural and regional areas compared to urban centers.

## Contribution

The paper introduces a novel statistical method combining reconstruction and small-area estimation to analyze diabetes prevalence in youth.

## Key findings

- Diabetes prevalence was higher in regional towns of South-East Queensland, New South Wales, and Victoria.
- Urban centers had lower diabetes rates compared to outer suburbs in Australia's five largest cities.
- Rural and peri-urban youth face a systematically higher diabetes burden than urban youth.

## Abstract

To characterise small-area geographical variation in the prevalence of diabetes in Australian youth.

A combined statistical reconstruction and small-area estimation algorithm was applied to privacy-modulated data from the 2021 Australian Census. The census instrument and reconstruction accuracy was examined by comparisons against a hospital-based register and community register. Diabetes prevalence maps were created from the small-area estimates.

The median and interquartile range of estimated diabetes prevalence by small-area unit under our geospatial smoothing model were 1.76 [1.49–1.97] cases per 1000 population for those aged 0–14 years and 5.2 [4.4–5.9] cases per 1000 population for those aged 15–19 years old. Concentrations of elevated prevalence were identified in the vicinities of regional towns across South-East Queensland, regional New South Wales and regional Victoria. Across each of Australia’s five largest cities a gradient of decreasing youth diabetes prevalence from the outer suburbs to the urban centre was identified.

Diabetes burden is systematically higher among rural and peri-urban resident youth in Australia compared with their urban counterparts.

Hotspots of prevalence in regional areas deserve attention from public health authorities.