Tracking the spread of COVID-19 with digital mobility data

We showed how disease transmission models can be parameterized with usage data of local travel cards (Octopus cards) to track COVID-19 transmissibility in real-time in Hong Kong.
Published in Microbiology
Tracking the spread of COVID-19 with digital mobility data
Like

Share this post

Choose a social network to share with, or copy the URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

In response to the COVID-19 pandemic, countries have sought to control disease transmission by restricting population movement through various non-pharmaceutical interventions (NPIs), thus reducing the number of contacts. However, many NPIs carry high economic and social costs and it is, therefore, important to track their effectiveness to inform countries how to optimize their pandemic control strategies. Although the effectiveness of NPIs could potentially be inferred from temporal changes in the daily number of cases reported to the health system, there is inevitably a delay between infection and reporting which consists of the incubation period (i.e., the time between infection and symptom onset, around 6 days), the time between symptom onset and diagnosis (around 3 days) and the lead time between confirmation and reporting (around 1 day). 

Unlike case data, digital mobility data have no such delay and could serve as an important proxy measure of the transmissibility of COVID-19. In our recent study [1], we showed how disease transmission models can be parameterized with usage data of local travel cards (Octopus cards) to track COVID-19 transmissibility in real-time in Hong Kong (Figure).

Figure. Tracking the spread of COVID-19 with Octopus usage data in Hong Kong

Octopus cards are used by more than 99% of the Hong Kong population for their daily public transport and small retail payments (https://www.octopus.com.hk/). We obtained the daily number of Octopus transport transactions among four types of cards, namely child (for children aged 3 to 11), student (for primary, secondary, and university students aged 12 to 26), adult (for non-student adults aged under 65) and elder (for older adults aged 65 or above). Using the level of Octopus card usage on 1 Jan 2020 as the baseline reference, we estimated the relative changes in mobility of these four age groups over time in the first two waves of the COVID-19 epidemic between January and May 2020. Assuming Octopus card usage is proportional to population mixing, we parameterized the disease transmission model with Octopus data and fit it to the actual COVID-19 case data. By doing so, we were able to nowcast the transmissibility of COVID-19 and provide a short-term forecast of caseload in Hong Kong. Such estimation of NPI effectiveness based on real-time estimates of transmissibility is no longer constrained by the infection-to-reporting delay of 9 to 10 days.

Our findings have helped inform COVID-19 surveillance and response in Hong Kong. Hong Kong experienced its third wave of COVID-19 epidemic in July- September 2020 and fourth wave from November 2020 till now (https://www.chp.gov.hk/files/pdf/local_situation_covid19_en.pdf). Although increasingly stringent NPIs and control measures have been implemented in both the third and fourth wave, real-time Octopus data showed that the drop in population mobility during these waves was slower and smaller than that during the first two waves. Indeed, both the number of cases and the effective reproductive number dropped more slowly in the third and fourth wave, suggesting reduced adherence to NPIs at the population level over time due to response fatigue. As such, the suppression of the current fourth wave in Hong Kong might require more targeted and adaptative NPIs to increase public adherence.

Other types of digital mobility data have been used in the fight against the COVID-19 pandemic since its emergence in late 2019, such as location-based service data from telecom service providers around the world, Alipay and WeChat in mainland China, Apple, Google, and Facebook in the UK and the US, etc. Shortly after Wuhan’s lockdown on 23 Jan 2020, we integrated mobility data provided by WeChat into our disease transmission model and estimated that several hundreds of COVID-19 cases had already spread to major megacities in mainland China before the lockdown. We estimated that nationwide epidemics would be inevitable unless NPIs could be implemented to suppress the spread of COVID-19 infections [2]. These findings alerted the authorities to consider immediate deployment of stringent NPIs across the country, and the first wave of COVID-19 epidemic was suppressed in most parts of China outside Wuhan by March 2020 due to nearly nationwide lockdowns.

A recent study by Nouvellet et al [3] provided a comprehensive evaluation of the relationship between mobility (i.e., mobility data from Apple and Google) and transmission of COVID-19 in 52 countries around the world. They found that although reduced transmission was associated with reduced mobility, not all countries exhibited a clear relationship between mobility and COVID-19 transmission after strict control measures had been relaxed. It will be interesting in the future to further investigate the congruence (or the lack thereof) between mobility and disease transmission, especially after the relaxation of NPIs.

Different types of mobility data will become increasingly available in real-time in many countries that can be utilized by epidemiological studies of COVID-19 and other infectious diseases. Although encouraging results from early mass vaccination programs have now offered a glimmer of light at the end of the COVID-19 tunnel, monitoring of NPI effectiveness via mobility data remains essential until vaccines become widely available to populations across the world.

Reference

1. Leung, K., Wu, J.T. & Leung, G.M. Real-time tracking and prediction of COVID-19 infection using digital proxies of population mobility and mixing. Nat Commun 12, 1501 (2021). https://doi.org/10.1038/s41467-021-21776-2

2. Wu, J. T., Leung, K. & Leung, G. M. Nowcasting and forecasting the potential domestic and international spread of the 2019-nCoV outbreak originating in Wuhan, China: a modelling study. The Lancet, doi:10.1016/S0140-6736(20)30260-9 (2020).

3. Nouvellet, P. et al. Reduction in mobility and COVID-19 transmission. Nature Communications 12, 1090, doi:10.1038/s41467-021-21358-2 (2021).

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Follow the Topic

Microbiology
Life Sciences > Biological Sciences > Microbiology

Related Collections

With collections, you can get published faster and increase your visibility.

Biology of rare genetic disorders

This cross-journal Collection between Nature Communications, Communications Biology, npj Genomic Medicine and Scientific Reports brings together research articles that provide new insights into the biology of rare genetic disorders, also known as Mendelian or monogenic disorders.

Publishing Model: Open Access

Deadline: Jan 31, 2025

Carbon dioxide removal, capture and storage

In this cross-journal Collection, we bring together studies that address novel and existing carbon dioxide removal and carbon capture and storage methods and their potential for up-scaling, including critical questions of timing, location, and cost. We also welcome articles on methodologies that measure and verify the climate and environmental impact and explore public perceptions.

Publishing Model: Open Access

Deadline: Mar 22, 2025