Epidemiology and public health impact of coronavirus disease 2019

Abstract

With the emergence of severe acute respiratory syndrome coronavirus 2 (SARS‑CoV‑2) and the rapid evolution of the virus, it is important to monitor the epidemiology of COVID-19 over time by examining key epidemiological parameters of individual epidemic waves in order to understand the transmission dynamics and to inform control strategies. In addition to the disease burden measured by confirmed infections, we shouldn’t negelect the indirect health impact of the COVID-19 pandemic on the population potentially resulting from the stringent implementation of various social and public health control measures and population behavioral changes during the pandemic. During my PhD study, I continuously monitored the key epidemiological parameters of COVID-19 epidemics in the context of the evolution of SARS-CoV-2, and estimated the population health impact of the pandemic on hospitalizations and deaths in Hong Kong. I first conducted a systematic review and meta-analysis on the incubation period of the ancestral strain of SARS-CoV-2. By fitting meta-regression models, I found that a shorter mean and 95th percentile of the incubation period was estimated with the data collected before the peak of an epidemic than after the peak. Then I used the information on the time of exposure to the virus and the onset time of viral shedding and symptoms from individual COVID-19 cases retrospectively collected and estimated that the mean and 95th percentile of the latent periods was shorter than the estimates of the incubation period for infections with the ancestral strain. I further analyzed the data of individual COVID-19 cases and their close contacts collected from outbreaks caused by Delta and Omicron variants, and found that following the evolution of SARS-CoV-2, infections with more recently emerged virus variants appeared to show a shorter latent and incubation period and the serial interval, the infectious period was shorter but the per-contact transmission rate increased, and the estimated basic and instantaneous reproduction number increased accordingly. Inactivated COVID-19 vaccination with incomplete, primary, and booster doses reduced viral loads among individuals with a breakthrough infection, implying a possible reduction in transmitting the virus to their close contacts. Lastly, weekly data on hospitalizations and deaths in Hong Kong population from 2010 to 2020 were analyzed using interrupted time-series regression models. Reductions in hospital admission were identified during the first year of the pandemic, accompanied by a substantial increase in excess deaths, largely occurred in patients hospitalized or died due to cardiovascular diseases. In conclusion, my PhD work demonstrated that the key epidemiological parameters of infections with SARS-CoV-2 might change following the evolution of the virus, which provided critical information for adjustment of control measures and inference of transmission dynamics of new virus variants. The estimated excess hospitalization and mortality burden highlighted the importance of assessing the indirect health impact of the COVID-19 pandemic, in addition to the direct morbidity and mortality related to the infection, and the need for pandemic preparedness in response to population changes in healthcare behaviors and implementation of stringent control measures during public health emergencies in the future.