Aggregation Bias and Socioeconomic Gradients in Waiting Time for Hospital Admissions

Area of Science:

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

Background:

• Waiting times in publicly funded healthcare systems aim for equitable access based on need.
• Existing research indicates socioeconomic status (SES) influences healthcare waiting times, often using aggregated residential data.
• Aggregate SES measures may introduce aggregation bias, potentially skewing findings on healthcare access disparities.

Purpose of the Study:

• To quantify the aggregation bias in measuring socioeconomic gradients of healthcare waiting times.
• To compare socioeconomic gradients in waiting times using individual-level versus aggregate-level SES data.
• To assess the impact of different data aggregation levels (individual, population cell, municipal) on estimating SES-based waiting time disparities.

Main Methods:

• Analysis of socioeconomic gradients in healthcare waiting times across three data aggregation levels: individual, population cell, and municipal.
• Measurement of socioeconomic status using education and income data at each respective level.
• Comparison of the magnitude and accuracy of socioeconomic gradients observed at different aggregation levels.

Main Results:

• The socioeconomic gradient in waiting times is modest when SES is measured at the individual level.
• Aggregate-level measurements of SES (population cell and municipal) show stronger associations with waiting times.
• Aggregate data leads to less accurate estimates and risks overstating the magnitude of socioeconomic gradients in healthcare access.

Conclusions:

• Measuring socioeconomic status at an aggregate level can significantly inflate perceived disparities in healthcare waiting times.
• Individual-level data provides a more accurate representation of the relationship between socioeconomic status and healthcare access.
• Researchers must be cautious of aggregation bias when analyzing socioeconomic determinants of health using publicly available data.