McGill University, Department of Epidemiology, Biostatistics and Occupational Health, 2001 McGill College Avenue, Suite 1200, Montreal, H3A 1G1, QC, Canada.
Marianopolis College, 4873 Westmount Avenue, Montreal, H3Y 1X9, QC, Canada.
Spat Spatiotemporal Epidemiol. 2022 Aug;42:100518. doi: 10.1016/j.sste.2022.100518. Epub 2022 May 23.
As of July 2021, Montreal is the epicentre of the COVID-19 pandemic in Canada with highest number of deaths. We aim to investigate the spatial distribution of the number of cases and deaths due to COVID-19 across the boroughs of Montreal. To this end, we propose that the cumulative numbers of cases and deaths in the 33 boroughs of Montreal are modelled through a bivariate hierarchical Bayesian model using Poisson distributions. The Poisson means are decomposed in the log scale as the sums of fixed effects and latent effects. The areal median age, the educational level, and the number of beds in long-term care homes are included in the fixed effects. To explore the correlation between cases and deaths inside and across areas, three different bivariate models are considered for the latent effects, namely an independent one, a conditional autoregressive model, and one that allows for both spatially structured and unstructured sources of variability. As the inclusion of spatial effects change some of the fixed effects, we extend the Spatial+ approach to a Bayesian areal set up to investigate the presence of spatial confounding. We find that the model which includes independent latent effects across boroughs performs the best among the ones considered, there appears to be spatial confounding with the diploma and median age variables, and the correlation between the cases and deaths across and within boroughs is always negative.
截至 2021 年 7 月，蒙特利尔是加拿大 COVID-19 大流行的中心，死亡人数最多。我们旨在研究蒙特利尔各行政区 COVID-19 病例和死亡人数的空间分布。为此，我们建议通过使用泊松分布的双变量分层贝叶斯模型对蒙特利尔 33 个行政区的累计病例和死亡人数进行建模。泊松均值在对数尺度上分解为固定效应和潜在效应的和。固定效应中包括区域的中位年龄、教育水平和长期护理院的床位数量。为了探索区域内和区域间病例和死亡之间的相关性，我们考虑了三种不同的潜在效应的双变量模型，即独立模型、条件自回归模型和允许空间结构和非结构变异性的模型。由于纳入空间效应会改变一些固定效应，我们将 Spatial+方法扩展到贝叶斯区域设置中，以研究是否存在空间混杂。我们发现，在所考虑的模型中，跨行政区独立的潜在效应模型表现最佳，学位和中位年龄变量似乎存在空间混杂，而且区域间和区域内病例和死亡之间的相关性始终为负。
