# The role of natural outdoor environment on COVID-19 mortality and hospitalisations among older community-dwellers in the pre-vaccination period: the Register RELOC-AGE cohort study in Sweden

## TL;DR

This study explores how living near natural outdoor environments might affect the risk of dying or being hospitalized from COVID-19 among older adults in Sweden before vaccines were available.

## Contribution

The study provides new evidence on the potential protective role of natural outdoor environments on severe COVID-19 outcomes in older adults.

## Key findings

- Higher PSD-scores were associated with slightly lower risks of death and hospitalisation from COVID-19, though results were not statistically significant.
- No evidence was found that income or education levels modified the relationship between natural environments and COVID-19 outcomes.
- The study highlights the need for further research on how living environments may influence pandemic outcomes.

## Abstract

Most COVID-19-related deaths and hospitalisations occurred among older people. Natural outdoor environments influence health outcomes, but the evidence of their effect on COVID-19 outcomes among older adults is limited. This study aimed to investigate associations between residential natural outdoor environment and COVID-19-related deaths or hospitalisations among community dwellers aged 59 years or older, and whether income and education moderated these associations.

The current study is part of the Register RELCOC-AGE cohort and included ≥ 59-year-old community-dwellers (N = 299,219) in Scania, Sweden. The main exposure was the Perceived Sensory Dimension Score (PSD-score), an aggregated area-level score indicating the presence of residential natural outdoor environment, measured between 2008 and 2019, grouped into low, intermediate, and high PSD-score. COVID-19 deaths and hospitalisations were assessed from 1 January to 31 December 2020. Hazard ratios (HRs) were estimated using Cox proportional hazard regression, and moderation by income or education was assessed using interaction terms. Models were adjusted for demographic and socioeconomic characteristics, population density and comorbidities.

The adjusted HRs for COVID-19 deaths and hospitalisations among intermediate– as compared to low PSD-score residents– were 0.90 (95% CI: 0.71–1.15) and 0.88 (95% CI: 0.77–1.01), respectively. Among high PSD-score residents, adjusted HRs were 0.88 (95% CI: 0.62–1.24) and 0.92 (95% CI: 0.76–1.12) for COVID-19 deaths and hospitalisations, respectively. No evidence of moderation of associations by income and education was found.

A protective effect from residential natural outdoor environment on COVID-19 hospitalisation and death was suggested, but statistical uncertainty was substantial and the evidence thus inconclusive. Future investigations across diverse populations could shed more light on the role of residential living environments in mitigating the consequences of epidemics and pandemics.

The online version contains supplementary material available at 10.1186/s12889-025-25515-w.