Back

Estimating changes in life expectancy in Hong Kong during the COVID-19 pandemic

Law, A. H. T.; Presanis, A. M.; Cheung, J. K.; Wu, P.; Schooling, M.; Cowling, B. J.; Wong, J. Y.

2024-12-13 public and global health
10.1101/2024.12.12.24318910 medRxiv
Show abstract

BackgroundHong Kong has one of the longest life expectancies in the world but was heavily impacted by COVID-19 in 2022. MethodsWe constructed sex-specific life tables from 1998-2023 using parametric bootstrapping to account for statistical uncertainty in mortality rates. We used Arriagas decomposition method to estimate age-specific contributions to overall changes in life expectancy over this period. We also estimated cause-specific mortality rates. ResultsHong Kong reported 50666 deaths in 2020, 51354 in 2021, 63692 in 2022, and 54731 in 2023. Estimates of life expectancy in males and females in 2020 and 2021 were similar to the pre-pandemic trend from 1998-2019 but declined significantly in 2022. Life expectancy for males was 82.3 years in 2021 and 80.4 years in 2022, and for females it was 87.9 in 2021 and 86.4 years in 2022. Compared to the pre-pandemic trend, the 2022 values corresponded to reductions by 2.22 (95% CI: 2.08, 2.36) years in males and 2.30 (95% CI: 2.17, 2.43) years in females. The loss in life expectancy in 2022 was mainly attributed to increased respiratory mortality rates in adults aged 65 or above. In 2023 life expectancy increased by 0.60 (95% CI: 0.46, 0.75) years in males and by 1.10 (95% CI: 0.95, 1.26) years in females. ConclusionsIn 2022 a very high respiratory mortality rate in older adults in Hong Kong during the COVID-19 pandemic was associated with a reduction in life expectancy by more than 2 years. In 2023 life expectancy increased towards the pre-pandemic trend.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.