Novi Reandy Sasmita, Rahmatil Adha Phonna, Zurnila Marli Kesuma, Saiful Kamal, Nudzran Yusya

Spatial-Temporal Epidemiology of COVID-19 in Aceh, Indonesia: A Statistical Perspective

Introduction

Spatial-temporal epidemiology of covid-19 in aceh, indonesia: a statistical perspective. Explore spatial-temporal COVID-19 epidemiology in Aceh, Indonesia. This study uses statistical models to identify key factors influencing case spread, including healthcare access and population data.

Abstract

The development of COVID-19 cases in Aceh for each region based on spatio-temporal is vital information to know. Spatio-temporal mapping is carried out to knowthe distribution of cases in diversity based on regional and time conditions. The timeseries design study was used as the research design in this study. This study aims toobtain factors that influence the incidence of COVID-19 cases in Aceh using paneldata regression analysis and the GTWR model for more accurate results. There arenine variables from 23 districts/cities in Aceh Province in 2020 and 2021. Based onpartial panel data regression analysis, of the eight independent variables that arefactors for analysis, it shows that only the variable number of doctors (p < 0.000),number of Tuberculosis Cases (p < 0.000), Number of Villages with Puskesmas (p< 0.026), and Number of Poor population (p < 0.035) have a significant effect onthe increase in COVID-19 cases in Aceh. The number of Tuberculosis Cases is avery dominant variable. Then, the results of the GTWR analysis using the AdaptiveKernel Exponential weighting function show that regional and time diversity affectthe factors that cause an increase in COVID-19 cases in Aceh. These factors need tobe a concern in controlling COVID-19 cases in Aceh in the future.

Review

This study, "Spatial-Temporal Epidemiology of COVID-19 in Aceh, Indonesia: A Statistical Perspective," addresses a highly relevant and important topic by investigating the spatio-temporal dynamics and influencing factors of COVID-19 cases in Aceh, Indonesia. The application of both panel data regression and the Geographically and Temporally Weighted Regression (GTWR) model is a commendable approach, aiming to capture both fixed effects and local variations for more accurate results. The abstract successfully identifies several statistically significant variables—number of doctors, Tuberculosis cases, villages with Puskesmas, and the poor population—as influential factors, with Tuberculosis cases highlighted as a particularly dominant variable. This initial identification of key socio-economic and healthcare infrastructure variables affecting COVID-19 incidence provides valuable preliminary insights for local public health strategies. While the study's ambition is commendable, the abstract presents several ambiguities that require clarification. For instance, it initially mentions "nine variables from 23 districts/cities" but then refers to "eight independent variables that are factors for analysis," creating an inconsistency regarding the total number and nature of variables. It is unclear what these "eight independent variables" are, beyond the four significant ones mentioned, and a comprehensive list or clear categorization would be beneficial. Furthermore, while the abstract states that the GTWR analysis shows "regional and time diversity affect the factors," it does not provide specific examples or elaborate on *how* these factors' influence varies across space and time. This lack of concrete findings from the GTWR, beyond a general statement of diversity, diminishes the perceived value of applying this advanced model. Quantifying the "dominance" of the Tuberculosis Cases variable (e.g., through effect size or relative contribution) would also add significant empirical weight to this claim. Overall, this study presents a valuable contribution to understanding the complex epidemiology of COVID-19 in a specific Indonesian context, utilizing advanced statistical methods to identify critical influencing factors. The findings on socio-economic and healthcare infrastructure variables offer actionable insights for policymakers. However, to enhance the clarity and impact of the work, the full manuscript should provide more detailed methodological descriptions, a clearer rationale for variable selection, and a more comprehensive discussion of the GTWR results, specifically highlighting the spatial and temporal heterogeneity of effects rather than just stating its existence. With these major revisions, particularly in elaborating on the methodology and providing more specific results from the GTWR analysis, this paper has the potential to be a significant addition to the literature.

Full Text

Comments

You need to be logged in to post a comment.