# Determinants of COVID-19 prevalence in Central Java, Indonesia: An ecological study of socio-demographic, environmental, and healthcare factors

## TL;DR

This study explores how social, environmental, and healthcare factors influenced the high rate of COVID-19 in Central Java, Indonesia, and suggests policy actions to reduce disparities.

## Contribution

The study introduces a spatially aware regression approach to identify regional disparities in factors affecting COVID-19 prevalence in Central Java.

## Key findings

- Tourist arrival changes and healthcare workforce distribution are linked to increased COVID-19 prevalence.
- Environmental health workers are associated with a protective effect against higher prevalence.
- Geographically Weighted Regression (GWR) better captures regional variations compared to OLS regression.

## Abstract

Central Java, Indonesia, experienced a 40.9 % COVID-19 positivity rate in 2022, exceeding the WHO benchmark. This study examines the association between changes in sociodemographic, environmental, and healthcare factors and the rise in COVID-19 prevalence, focusing on regional disparities across Central Java. Variables from public datasets were chosen based on the Social Determinants of Health (SDOH) framework. Data analysis begins with variable identification via Pearson correlation, followed by an Ordinary Least Squares (OLS) regression employing Stepwise Backward Elimination, and subsequent assumption tests including Jarque-Bera, Breusch-Pagan, Moran's I, and multicollinearity checks. Upon identifying spatial autocorrelation and heteroscedasticity, Geographically Weighted Regression (GWR) was applied to address spatial heterogeneity. Ordinary Least Squares (OLS) analysis identified Change in tourist arrival ratio per population, environmental health workforce ratio per land area, and community healthcare workforce ratio per land area as associated factors with change in COVID-19 prevalence. The Geographically Weighted Regression (GWR) model, with a higher R2 value of 0.66, better accounted for regional variations, especially in central and eastern regions. The findings indicate that traveler mobility and the spatial distribution of community health workers are linked to increased COVID-19 prevalence, whereas environmental health workers are associated with a protective result, but these are associations at the aggregate (district/city) level and may be influenced by confounding or reverse causation. Structural factors such as unequal access to resources, healthcare, and sanitation, driven by tourism-induced social inequality, contribute to the disproportionate impact of COVID-19 on vulnerable communities, making it essential for policymakers to address these disparities to protect both local populations and visitors. The study recommends regulating risk-based tourist activities, expanding the environmental health workforce, and enhancing spatial monitoring systems to inform evidence-based health policy.