Cvijanovic Dane, Grubor Nikola, Rajovic Nina, Vucevic Mira, Miltenovic Svetlana, Laban Marija, Mostic Tatjana, Tasic Radica, Matejic Bojana, Milic Natasa
Department of Cardiovascular Diseases, University Clinical Center Zvezdara, Belgrade, Serbia.
Institute of Medical Statistics and Informatics, Faculty of Medicine, University of Belgrade, Dr Subotica 15, Belgrade, Serbia, 381 63367700.
JMIR Public Health Surveill. 2025 Apr 17;11:e56877. doi: 10.2196/56877.
Concerns have been raised about discrepancies in COVID-19 mortality data, particularly between preliminary and final datasets of vital statistics in Serbia. In the original preliminary dataset, released daily during the ongoing pandemic, there was an underestimation of deaths in contrast to those reported in the subsequently released yearly dataset of vital statistics.
This study aimed to assess the accuracy of the final mortality dataset and justify its use in further analyses. In addition, we quantified the relative impact of COVID-19 on the death rate in the Serbian capital's population. In the process, we aimed to explore whether any evidence of cause-of-death misattribution existed in the final published datasets.
Data were sourced from the electronic databases of the Statistical Office of the Republic of Serbia. The dataset included yearly recorded deaths and the causes of death of all citizens currently living in the territory of Belgrade, the capital of the Republic of Serbia, from 2015 to 2021. Standardization and modeling techniques were utilized to quantify the direct impact of COVID-19 and to estimate excess deaths. To account for year-to-year trends, we used a mixed-effects hierarchical Poisson generalized linear regression model to predict mortality for 2020 and 2021. The model was fitted to the mortality data observed from 2015 to 2019 and used to generate mortality predictions for 2020 and 2021. Actual death rates were then compared to the obtained predictions and used to generate excess mortality estimates.
The total number of excess deaths, calculated from model estimates, was 3175 deaths (99% CI 1715-4094) for 2020 and 8321 deaths (99% CI 6975-9197) for 2021. The ratio of estimated excess deaths to reported COVID-19 deaths was 1.07. The estimated increase in mortality during 2020 and 2021 was 12.93% (99% CI 15.74%-17.33%) and 39.32% (99% CI 35.91%-39.32%) from the expected values, respectively. Those aged 0-19 years experienced an average decrease in mortality of 22.43% and 23.71% during 2020 and 2021, respectively. For those aged up to 39 years, there was a slight increase in mortality (4.72%) during 2020. However, in 2021, even those aged 20-39 years had an estimated increase in mortality of 32.95%. For people aged 60-79 years, there was an estimated increase in mortality of 16.95% and 38.50% in 2020 and 2021, respectively. For those aged >80 years, the increase was estimated at 11.50% and 34.14% in 2020 and 2021, respectively. The model-predicted deaths matched the non-COVID-19 deaths recorded in the territory of Belgrade. This concordance between the predicted and recorded non-COVID-19 deaths provides evidence that the cause-of-death misattribution did not occur in the territory of Belgrade.
The finalized mortality dataset for Belgrade can be safely used in COVID-19 impact analysis. Belgrade experienced a significant increase in mortality during 2020 and 2021, with most of the excess mortality attributable to SARS-CoV-2. Concerns about increased mortality from causes other than COVID-19 in Belgrade seem misplaced as their impact appears negligible.
人们对新冠疫情死亡率数据的差异表示担忧，尤其是塞尔维亚生命统计初步数据集和最终数据集之间的差异。在疫情期间每日发布的原始初步数据集中，与随后发布的年度生命统计数据集相比，死亡人数被低估。
本研究旨在评估最终死亡率数据集的准确性，并证明其在进一步分析中的适用性。此外，我们量化了新冠疫情对塞尔维亚首都人口死亡率的相对影响。在此过程中，我们旨在探究最终发布的数据集中是否存在死因误判的证据。
数据来源于塞尔维亚共和国统计局的电子数据库。该数据集包括2015年至2021年居住在塞尔维亚共和国首都贝尔格莱德境内的所有公民的年度死亡记录和死因。利用标准化和建模技术量化新冠疫情的直接影响，并估计超额死亡人数。为了考虑逐年趋势，我们使用混合效应分层泊松广义线性回归模型预测2020年和2021年的死亡率。该模型根据2015年至2019年观察到的死亡率数据进行拟合，并用于生成2020年和2021年的死亡率预测值。然后将实际死亡率与获得的预测值进行比较，并用于生成超额死亡率估计值。
根据模型估计，2020年超额死亡总数为3175人（99%置信区间1715 - 4094），2021年为8321人（99%置信区间6975 - 9197）。估计的超额死亡人数与报告的新冠死亡人数之比为1.07。2020年和2021年死亡率较预期值的估计增幅分别为12.93%（99%置信区间15.74% - 17.33%）和39.32%（99%置信区间35.91% - 39.32%）。0至19岁年龄组在2020年和2021年的死亡率分别平均下降了22.43%和23.71%。对于39岁及以下人群，2020年死亡率略有上升（4.72%）。然而，在2021年，即使是20至39岁的人群，估计死亡率也上升了32.95%。对于60至79岁的人群，2020年和2021年的估计死亡率增幅分别为16.95%和38.50%。对于80岁以上人群，2020年和2021年的增幅分别估计为11.50%和34.14%。模型预测的死亡人数与贝尔格莱德境内记录的非新冠死亡人数相符。预测的和记录的非新冠死亡人数之间的这种一致性证明在贝尔格莱德境内没有发生死因误判。
贝尔格莱德最终确定的死亡率数据集可安全用于新冠疫情影响分析。贝尔格莱德在2020年和2021年死亡率显著上升，大部分超额死亡归因于新冠病毒。对贝尔格莱德除新冠疫情以外其他原因导致死亡率上升的担忧似乎没有依据，因为其影响似乎微不足道。
