Unpacking analyses relying on area-based data: are the assumptions supportable?

Area of Science:

• Public Health
• Health Services Research
• Socioeconomic Determinants of Health

Background:

• Australian health records lack direct individual socioeconomic status (SES) measures.
• Area-based SES is commonly used as a proxy in health research.
• This study assesses the reliability of area-based SES as a proxy for socioeconomic disadvantage.

Purpose of the Study:

• To evaluate the reliability of residential address as a proxy for socioeconomic disadvantage in health analyses.
• To examine how hospital inpatient separation rates vary by area-based SES at different geographic aggregation levels.
• To assess the temporal stability of address as a measure of SES.

Main Methods:

• Utilized the Western Australian Data Linkage System for hospital inpatient data from Perth residents.
• Calculated hospital separation rates across varying geographic aggregation levels: Census Collection Districts, postcode areas, and Statistical Local Areas (SLAs).
• Analyzed the association between area of residence SES and hospitalisation patterns, including repeat admissions and residential mobility over five years.

Main Results:

• A strong association exists between individual SES and usual address SES at the smallest geographic level (Census Collection Districts).
• Aggregating data to larger areas (postcodes, SLAs) weakens this association but still provides a reliable indication.
• Individuals from most disadvantaged areas showed 13% higher admission rates and 47% more overall separations due to multiple admissions.

Conclusions:

• Postcode and SLA level data offer a reliable, albeit understated, measure of area socioeconomic disadvantage.
• Most hospitalised Perth residents maintained the same address for repeat admissions within five years.
• Residential moves were predominantly within or between areas of similar SES, with less mobility from affluent areas.