February 15, 2022
BEIJING – A team of analysts from Lanzhou University recently predicted that the COVID-19 pandemic is likely to end globally in late 2023, with infections worldwide reaching at least 750 million.
The prediction was made using the Global Prediction System of the COVID-19 Pandemic (GPCP) created at the university.
Following are excerpts of an interview with Huang Jianping, head of the prediction team:Q: Which is the most accurate epidemic forecast made by the system so far?
Huang: Since its launch in May 2020, the GPCP has been making daily, monthly and seasonal predictions for more than 180 countries using epidemic data. Predictions for the epidemic in 2021 in Shijiazhuang, Tonghua, Guangzhou, Nanjing, Zhengzhou and other cities were basically consistent with the actual situation.
Q: What does the system predict?
Huang: In addition to predicting daily, monthly and seasonal waves for more than 180 countries worldwide, the system can also predict important points, such as peak number of daily cases, the ending time and the cumulative number of infections.
Q: What scientific data and related social or economic factors are considered in making predictions?
Huang: This prediction system combines advanced climate prediction techniques with epidemic models. The influence of virus variation, government control measures, people’s self-protection measures, vaccinations and natural factors — such as temperature and humidity — were considered comprehensively.
The epidemic data used in the prediction model are downloaded from Johns Hopkins University. The meteorological data such as temperature and humidity are downloaded from the European Centre for Medium-Range Weather Forecasts and NASA.
Q: Do new virus variants affect accuracy?
Huang: The system also applies to the current outbreak caused by variants of SARS-CoV-2.
Q: The system can be used to predict other emerging outbreaks?
Huang: For other emerging infectious diseases, predictions can be made by adjusting the prediction model and parameter scheme.Q: So far, how has the system helped with outbreak containment?
Huang: The goal of the prediction system is to use scientific methods to predict pandemic development and provide the government with a scientific basis for assessing the epidemic situation and adopting effective control measures.
Q: Compared with other prediction systems worldwide, what’s special about yours? Does any exchange and communications with your peer scientists worldwide improve accuracy?
Huang: Our prediction system is an innovation that combines approaches from fields such as atmospheric sciences and epidemiology to continuously make daily, monthly and seasonal global predictions from the early stage of the global pandemic.
We also share our predictions with the WHO and experts in relevant fields worldwide.
Q: What areas of expertise are on the team?
Huang: The prediction team includes experts and staff specializing in atmospheric sciences, public health, applied statistics and artificial intelligence.
The establishment of the system was fully supported by the School of Public Health at Lanzhou University, First Hospital of Lanzhou University, the Center for Evidence-Based Medicine of Lanzhou University and the School of Politics and International Relations at Lanzhou University.
Q: What are the major challenges for system operation and prediction?
Huang: The biggest challenge for the prediction model is improving accuracy. On the basis of traditional epidemiological models, we use advanced statistical-dynamic climate prediction to improve the traditional epidemiological models.
Using real-time updated epidemic data, we dynamically estimate and revise important parameters in the epidemiological model, such as infection rates, recovery rates and mortality rates, which greatly improves our prediction accuracy.
Q: How does the system predict the pandemic trend worldwide?
Huang: Based on the current transmission of the Omicron variant and control measures around the world, the latest prediction result of GPCP shows that the global pandemic is generally on a downward trend and could end in late 2023, with the total number of infections reaching at least 750 million.Huang Jianping is head of the Collaborative Innovation Center for Western Ecological Safety (CIWES) at Lanzhou University.
Since the outbreak of the global COVID-19 pandemic, CIWES has actively coordinated various scientific research strengths.
Led by Huang, the prediction team worked for three months to develop the Global Prediction System of the COVID-19 Pandemic and launched the model in May 2020.