Yichun Fan

Introduction: Physical activity (PA) is a critical health behaviour often disrupted by public health crises. Prior research has primarily relied on self-reported surveys, limiting timely and scalable insights. This study examines how large-scale digital exercise data can enhance real-time PA prediction and support interventions to strengthen PA resilience during global public health crises. Methods: We analysed PA data from 5.17 million individuals across 65 countries, collected via a smartphone PA tracking app between January 2019 and June 2021. The dataset includes longitudinal exercise histories, sociodemographic characteristics, geographic locations and social network connections. We developed statistical and machine learning models—including linear regressions, neural networks (NNs) and graph neural networks (GNNs)—to predict weekly PA adherence (defined as at least 150 min of exercise per week in line with WHO recommendations) before and during the COVID-19 pandemic. Models predicted each week’s adherence using the individual’s exercise history from the preceding 1–3 weeks, together with sociodemographic, weather and pandemic-related variables. We also applied surrogate models and econometric analyses to assess the associations of individual and urban-level attributes with PA resilience globally. Results: In a US county subsample where all model classes could be compared (Middlesex County, MA), the GNN achieved the highest predictive performance for weekly PA adherence, with 89.9% accuracy pre-pandemic and 82.1% during the pandemic. At the global level, the NN model demonstrated similar predictive performance (accuracy 87.6% and 82.8%). Personal exercise history was the strongest predictor, improving model accuracy by 20 percentage points, while increased model complexity or the addition of social network features yielded modest gains. Marked global disparities were observed: cities in lower- and upper-middle-income countries experienced greater reductions in PA during the pandemic compared with cities in high-income countries. Females, socially connected individuals and residents in cities with better access to outdoor sports infrastructure were more likely to maintain regular PA during the pandemic. Conclusion: Digital PA data can enable accurate, real-time prediction of PA adherence and inform individual and environmental factors influencing global PA resilience during crises. This approach holds promise for informing targeted interventions and urban policy strategies to enhance health resilience and reduce global health disparities.

The COVID-19 pandemic has created unprecedented burdens on people’s physical health and subjective well-being. While countries worldwide have developed platforms to track the evolution of COVID-19 infections and deaths, frequent global measurements of affective states to gauge the emotional impacts of pandemic and related policy interventions remain scarce. Using 654 million geotagged social media posts in over 100 countries, covering 74% of world population, coupled with state-of-the-art natural language processing techniques, we develop a global dataset of expressed sentiment indices to track national- and subnational-level affective states on a daily basis. We present two motivating applications using data from the first wave of COVID-19 (from 1 January to 31 May 2020). First, using regression discontinuity design, we provide consistent evidence that COVID-19 outbreaks caused steep declines in expressed sentiment globally, followed by asymmetric, slower recoveries. Second, applying synthetic control methods, we find moderate to no effects of lockdown policies on expressed sentiment, with large heterogeneity across countries. This study shows how social media data, when coupled with machine learning techniques, can provide real-time measurements of affective states.

Significance: When the COVID-19 pandemic ends and it becomes safe to resume economic behavior, we will need to find effective ways of communicating that it is truly safe to do so. In this study, we tested a simple “nudge” that informed people of the proportion of their neighbors who were planning to visit a restaurant over the weekend, so that those who felt that restaurants were very unsafe would realize that many others felt comfortable visiting them. Our nudge successfully motivated such hesitant individuals to increase restaurant visits by 37%, and additional analyses indicate they indeed felt that visiting restaurants was safer.

BACKGROUND COVID-19, as a global health crisis, has triggered the fear emotion with unprecedented intensity. Besides the fear of getting infected, the outbreak of COVID-19 also created significant disruptions in people’s daily life and thus evoked intensive psychological responses indirect to COVID-19 infections. OBJECTIVE This study aims to develop novel digital trackers of public fear emotion during the COVID-19 pandemic, and to uncover meaningful topics that the citizens are concerned about to inform policy decision-making. METHODS We construct an expressed fear database using 16 million social media posts generated by 536 thousand users in China between January 1st, 2019 and August 31st, 2020. We employ Bidirectional Encoder Representations from Transformers (BERT) to detect the fear emotion within each post and apply BERTopic to extract the central fear topics. RESULTS We find that on average, 2.45% of posts per day having fear as the dominant emotion in 2019. This share spiked after the COVID-19 outbreak and peaked at 9.1% on the date that China’s epi-center Wuhan city announced lockdown. Among the fear posts, topics related to health takes the largest share (39%). Specifically, we find that posts regarding sleep disorders (Nightmare and Insomnia) have the most significant increase during the pandemic. We also observe gender heterogeneity in fear topics, with females being more concerned with health while males being more concerned with job. CONCLUSIONS Our work leverages the social media data coupled with computational methods to track the emotional response on a large scale and with high temporal granularity. While we conduct this research in a tracing back mode, it is possible to use such a method to achieve real-time emotion monitoring, thus serving as a helpful tool to discern societal concerns and aid for policy decision-making.

Previous literature suggests that active commuting has substantial health benefits. Yet, in polluted regions, it can also cause additional health risks by increasing riders’ pollution exposure and raising their inhalation rate. We examine the effect of perceived air pollution on stated commuting choices using an on-site survey experiment for 2285 non-automobile commuters in Zhengzhou, a heavily polluted city in central China. We integrate a sequential randomized controlled trial in a survey where individuals in the treatment group received tailored information on their commuting-related pollution exposure, based on our 2 week peak-hour pollution monitoring campaign across transportation modes in the city. We find that travelers in Zhengzhou have already adopted pollution prevention actions by favoring indoor commuting modes on polluted days. Individuals receiving personalized pollution exposure information by mode further decrease active commuting by 8.4 percentage points (95% CI: 5.1, 11.6), accompanied by a 14.7 percentage points (95% CI: 10.7, 18.3) increase in automobile commuting. Travellers make sub-optimal, overly risk averse choices by reducing active commuting even for trips where epidemiological research suggests the exercise benefits outweigh pollution exposure risks. This pollution avoidance tendency significantly attenuates the effect of policies encouraging active commuting. Our findings show the intricately intertwined relationships between the public health targets of promoting active lifestyles and reducing pollution exposure, and between individual pollution avoidance and societal pollution mitigation.