Associations between area-level socioeconomic disadvantage and COVID-19 disease consequences in Sydney, Australia: A retrospective cohort analysis

Background: Socioeconomic disparities have shaped COVID-19 outcomes worldwide. Focusing on disease consequences once infected (severity among cases), we examined whether area-level socioeconomic disadvantage was associated with hospitalisation and death among COVID-19 cases in Greater Sydney, Australia. Methods: We conducted a retrospective cohort study of confirmed and probable COVID-19 cases recorded in the New South Wales Notifiable Conditions Information Management System from 2 March 2020–21 February 2022. Area-level disadvantage was measured using the Index of Relative Socio Economic Disadvantage (IRSD). We modelled the odds of (a) hospitalisation and (b) death conditional on infection using logistic regression, adjusting for age, gender and pandemic phase. Results: Among 782,883 included cases, 3.5% were hospitalised and 0.2% died due to COVID-19. After adjustment, a 100-point increase in IRSD score (indicating less socioeconomic disadvantage) was significantly associated with lower odds of hospitalisation (adjusted odds ratio [AOR] 0.80, 95% CI 0.79–0.81) and death (AOR 0.78, 95% CI 0.74–0.83). There was no evidence that these associations differed across pandemic phases. Conclusions: Area‑level socioeconomic disadvantage was associated with higher risks of hospitalisation and death among COVID‑19 cases in Greater Sydney – a setting with public hospital care – with some postal areas having more than twice the odds of hospitalisation and death as others. Given the absence of linked comorbidity and vaccination data, the most plausible explanation is disparities in comorbidity and risk‑factor burden, although contributions from differences in access to testing and quality of care cannot be ruled out. Public health responses should prioritise chronic disease prevention and management in disadvantaged communities to mitigate inequitable outcomes in future pandemics.