Crowdsourcing data to mitigate epidemics

A Perspective on Crowdsourcing and Human-in-the-Loop Workflows in Precision Health

Modern machine learning approaches have led to performant diagnostic models for a variety of health conditions. Several machine learning approaches, such as decision trees and deep neural networks, can, in principle, approximate any function. However, this power can be considered to be both a gift and a curse, as the propensity toward overfitting is magnified when the input data are heterogeneous and high dimensional and the output class is highly nonlinear. This issue can especially plague diagnostic systems that predict behavioral and psychiatric conditions that are diagnosed with subjective criteria. An emerging solution to this issue is crowdsourcing, where crowd workers are paid to annotate complex behavioral features in return for monetary compensation or a gamified experience. These labels can then be used to derive a diagnosis, either directly or by using the labels as inputs to a diagnostic machine learning model. This viewpoint describes existing work in this emerging field and discusses ongoing challenges and opportunities with crowd-powered diagnostic systems, a nascent field of study. With the correct considerations, the addition of crowdsourcing to human-in-the-loop machine learning workflows for the prediction of complex and nuanced health conditions can accelerate screening, diagnostics, and ultimately access to care.

Syndromic surveillance performance in Canada throughout the COVID-19 pandemic, March 1, 2020 to March 4, 2023

The coronavirus disease 2019 (COVID-19) pandemic has highlighted the need for robust surveillance of respiratory viruses. Syndromic surveillance continues to be an important surveillance component recommended by the World Health Organization (WHO). While FluWatchers, Canada's syndromic surveillance system, has been in place since 2015, the COVID-19 pandemic provided a valuable opportunity to expand the program's scope and underlying technology infrastructure. Following some structural changes to FluWatchers syndromic questionnaire, participants are now able to contribute valuable data to the non-specific surveillance of respiratory virus activity across Canada. This article examines the performance of FluWatchers' syndromic surveillance over the three years of the COVID-19 pandemic in Canada. More specifically, this article examines FluWatchers' performance with respect to the correlation between the FluWatchers influenza-like illness (ILI) and acute respiratory infection (ARI) indicators and total respiratory virus detections (RVDs) in Canada, including influenza, respiratory syncytial virus (RSV), severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and other respiratory viruses.

Early childhood tracking application: Correspondence between crowd-based developmental percentiles and clinical tools

Barriers to child developmental screening lead to delayed diagnosis and intervention. babyTRACKS, a mobile application for tracking developmental milestones, presents parents their child's percentiles computed relative to crowd-based data. This study evaluated correspondence between crowd-based percentiles and traditional development measures. Research analyzed babyTRACKS diaries of 1951 children. Parents recorded attainment age for milestones across Gross Motor, Fine Motor, Language, Cognitive, and Social domains. Fifty-seven parents completed the Ages and Stages Questionnaire (ASQ-3), and 13 families participated in the Mullen Scales of Early Learning (MSEL) expert assessment. Crowd-based percentiles were compared with: Centers for Disease Control (CDC) norms for comparable milestones, ASQ-3 and MSEL scores. babyTRACKS percentiles correlated with the percentage of unmet CDC milestones, and with higher ASQ-3 and MSEL scores across several domains. Children who did not meet CDC age thresholds had lower babyTRACKS percentiles by about 20 points and those at ASQ-3 risk had lower babyTRACKS Fine Motor and Language scores. Repeated measures tests showed significantly higher MSEL versus babyTRACKS percentiles in the Language domain. Although ages and milestones in diary varied, the app percentiles corresponded with traditional measures, particularly in fine motor and language domains. Future research is needed for determining referral thresholds while minimizing false alarms.

A Report on Statistics of an Online Self-screening Platform for COVID-19 and Its Effectiveness in Iran

BACKGROUND

The most recent emerging infectious disease, coronavirus disease 2019 (COVID-19), is pandemic now. Iran is a country with community transmission of the disease. Telehealth tools have been proved to be useful in controlling public health disasters. We developed an online self-screening platform to offer a population-wide strategy to control the massive influx to medical centers.

METHODS

We developed a platform operating based on given history by participants, including sex, age, weight, height, location, primary symptoms and signs, and high risk past medical histories. Based on a decision-making algorithm, participants were categorized into four levels of suspected cases, requiring diagnostic tests, supportive care, not suspected cases. We made comparisons with Iran STEPs (STEPwise approach to Surveillance) 2016 study and data from the Statistical Centre of Iran to assess population representativeness of data. Also, we made a comparison with officially confirmed cases to investigate the effectiveness of the platform. A multilevel mixed-effects Poisson regression was used to check the association of visiting platform and deaths caused by COVID-19.

RESULTS

About 310 000 individuals participated in the online self-screening platform in 33 days. The majority of participants were in younger age groups, and males involved more. A significant number of participants were screened not to be suspected or needing supportive care, and only 10.4% of males and 12.0% of females had suspected results of COVID-19. The penetration of the platform was assessed to be acceptable. A correlation coefficient of 0.51 was calculated between suspected results and confirmed cases of the disease, expressing the platform's effectiveness.

CONCLUSION

Implementation of a proper online self-screening tool can mitigate population panic during wide-spread epidemics and relieve massive influx to medical centers. Also, an evidence-based education platform can help fighting infodemic. Noticeable utilization and verified effectiveness of such platform validate the potency of telehealth tools in controlling epidemics and pandemics.

Early-stage spatial disease surveillance of novel SARS-CoV-2 variants of concern in Germany with crowdsourced data

The emergence and rapid spread of novel variants of concern (VOC) of the coronavirus 2 constitute a major challenge for spatial disease surveillance. We explore the possibility to use close to real-time crowdsourced data on reported VOC cases (mainly the Alpha variant) at the local area level in Germany. The aim is to use these data for early-stage estimates of the statistical association between VOC reporting and the overall COVID-19 epidemiological development. For the first weeks in 2021 after international importation of VOC to Germany, our findings point to significant increases of up to 35-40% in the 7-day incidence rate and the hospitalization rate in regions with confirmed VOC cases compared to those without such cases. This is in line with simultaneously produced international evidence. We evaluate the sensitivity of our estimates to sampling errors associated with the collection of crowdsourced data. Overall, we find no statistical evidence for an over- or underestimation of effects once we account for differences in data representativeness at the regional level. This points to the potential use of crowdsourced data for spatial disease surveillance, local outbreak monitoring and public health decisions if no other data on new virus developments are available.

Data Anonymization for Pervasive Health Care: Systematic Literature Mapping Study

BACKGROUND

Data science offers an unparalleled opportunity to identify new insights into many aspects of human life with recent advances in health care. Using data science in digital health raises significant challenges regarding data privacy, transparency, and trustworthiness. Recent regulations enforce the need for a clear legal basis for collecting, processing, and sharing data, for example, the European Union's General Data Protection Regulation (2016) and the United Kingdom's Data Protection Act (2018). For health care providers, legal use of the electronic health record (EHR) is permitted only in clinical care cases. Any other use of the data requires thoughtful considerations of the legal context and direct patient consent. Identifiable personal and sensitive information must be sufficiently anonymized. Raw data are commonly anonymized to be used for research purposes, with risk assessment for reidentification and utility. Although health care organizations have internal policies defined for information governance, there is a significant lack of practical tools and intuitive guidance about the use of data for research and modeling. Off-the-shelf data anonymization tools are developed frequently, but privacy-related functionalities are often incomparable with regard to use in different problem domains. In addition, tools to support measuring the risk of the anonymized data with regard to reidentification against the usefulness of the data exist, but there are question marks over their efficacy.

OBJECTIVE

In this systematic literature mapping study, we aim to alleviate the aforementioned issues by reviewing the landscape of data anonymization for digital health care.

METHODS

We used Google Scholar, Web of Science, Elsevier Scopus, and PubMed to retrieve academic studies published in English up to June 2020. Noteworthy gray literature was also used to initialize the search. We focused on review questions covering 5 bottom-up aspects: basic anonymization operations, privacy models, reidentification risk and usability metrics, off-the-shelf anonymization tools, and the lawful basis for EHR data anonymization.

RESULTS

We identified 239 eligible studies, of which 60 were chosen for general background information; 16 were selected for 7 basic anonymization operations; 104 covered 72 conventional and machine learning-based privacy models; four and 19 papers included seven and 15 metrics, respectively, for measuring the reidentification risk and degree of usability; and 36 explored 20 data anonymization software tools. In addition, we also evaluated the practical feasibility of performing anonymization on EHR data with reference to their usability in medical decision-making. Furthermore, we summarized the lawful basis for delivering guidance on practical EHR data anonymization.

CONCLUSIONS

This systematic literature mapping study indicates that anonymization of EHR data is theoretically achievable; yet, it requires more research efforts in practical implementations to balance privacy preservation and usability to ensure more reliable health care applications.

Fearing fear itself: Crowdsourced longitudinal data on Covid-19-related fear in Sweden

BACKGROUND

The Covid-19 pandemic has had unprecedented effects on individual lives and livelihoods as well as on social, health, economic and political systems and structures across the world. This article derives from a unique collaboration between researchers and museums using rapid response crowdsourcing to document contemporary life among the general public during the pandemic crisis in Sweden.

METHODS AND FINDINGS

We use qualitative analysis to explore the narrative crowdsourced submissions of the same 88 individuals at two timepoints, during the 1st and 2nd pandemic waves, about what they most fear in relation to the Covid-19 pandemic, and how their descriptions changed over time. In this self-selected group, we found that aspects they most feared generally concerned responses to the pandemic on a societal level, rather than to the Covid-19 disease itself or other health-related issues. The most salient fears included a broad array of societal issues, including general societal collapse and fears about effects on social and political interactions among people with resulting impact on political order. Notably strong support for the Swedish pandemic response was expressed, despite both national and international criticism.

CONCLUSIONS

This analysis fills a notable gap in research literature that lacks subjective and detailed investigation of experiences of the general public, despite recognition of the widespread effects of Covid-19 and its' management strategies. Findings address controversy about the role of experts in formulating and communicating strategy, as well as implications of human responses to existential threats. Based on this analysis, we call for broader focus on societal issues related to this existential threat and the responses to it.

Information seeking behavior and COVID-19 pandemic: A snapshot of young, middle aged and senior individuals in Greece

BACKGROUND

The plethora of information in the contemporary digital age is enormous and beyond the capability of the average person to process all the information received. During the COVID-19 pandemic outbreak, huge amount of information is increasingly available in digital information sources and overwhelms the average person. The purpose of this research was to investigate public's information seeking behavior on COVID-19 in Greece.

METHOD

The study was conducted through a web-based survey, facilitated by the use of questionnaire posted on the Google Forms platform. The questionnaire consisted of closed-ended, 7-point Likert scale questions and multiple choice questions and was distributed to all over Greek Regions to almost 3.000 recipients, during the implementation of restrictive measures against the COVID-19 outbreak in Spring 2020. The data collected were subjected to a descriptive statistical analysis. The median was used to present the results. In order to perform analysis between genders, as well as age groups, the non-parametric criteria Mann-Whitney U and Kruskal-Wallis were applied to determine the existence of differences in participants' beliefs.

RESULTS

Responses by 776 individuals were obtained. Individuals dedicated up to 2 h per day to be informed on COVID-19. Television, electronic press and news websites were reported by the participants as more reliable than social media, in obtaining information on COVID-19. Respondents paid attention to official sources of information (Ministry of Health, Civil Protection etc.). Family and friends played an additional role in the participants' information on COVID-19, while the personal doctor, other health workers and pharmacists did not appear to be most preferred sources of information on COVID-19. Participants' most common information seeking strategy in digital environment was keyword searching. Unreliable information, fake news and information overload were the most common difficulties that the participants encountered seeking information on COVID-19. The respondents' views seemed to differ significantly among age groups. The older the participants, the more often they were informed by television (p < 0.001) and the less often by the internet (p < 0.001). Females appear to use more frequently internet (p < 0.001) and social media (p = 0.001) out of habit and visit more often the Ministry of Health (p < 0.001) and the Civil Protection (p=0.005) websites, compared to males. Most of the participants seemed to worry about the fake news phenomenon and agreed that fake news on COVID-19 is being spread in the media and especially social networks.

CONCLUSION

The study revealed that, during the COVID-19 pandemic in Greece, participants obtained information about the disease mainly by television, electronic press and news websites. On the contrary, the limited use of social media demonstrates the participants awareness of the spread of fake news on social media. This observed information seeking behavior might has contributed to individuals' acceptance of the necessary behavioral changes that had led to the Greek success story in preventing spread of the disease.

Extracting Medical Information from Paper COVID-19 Assessment Forms

OBJECTIVE

This study examines the validity of optical mark recognition, a novel user interface, and crowdsourced data validation to rapidly digitize and extract data from paper COVID-19 assessment forms at a large medical center.

METHODS

An optical mark recognition/optical character recognition (OMR/OCR) system was developed to identify fields that were selected on 2,814 paper assessment forms, each with 141 fields which were used to assess potential COVID-19 infections. A novel user interface (UI) displayed mirrored forms showing the scanned assessment forms with OMR results superimposed on the left and an editable web form on the right to improve ease of data validation. Crowdsourced participants validated the results of the OMR system. Overall error rate and time taken to validate were calculated. A subset of forms was validated by multiple participants to calculate agreement between participants.

RESULTS

The OMR/OCR tools correctly extracted data from scanned forms fields with an average accuracy of 70% and median accuracy of 78% when the OMR/OCR results were compared with the results from crowd validation. Scanned forms were crowd-validated at a mean rate of 157 seconds per document and a volume of approximately 108 documents per day. A randomly selected subset of documents was reviewed by multiple participants, producing an interobserver agreement of 97% for documents when narrative-text fields were included and 98% when only Boolean and multiple-choice fields were considered.

CONCLUSION

Due to the COVID-19 pandemic, it may be challenging for health care workers wearing personal protective equipment to interact with electronic health records. The combination of OMR/OCR technology, a novel UI, and crowdsourcing data-validation processes allowed for the efficient extraction of a large volume of paper medical documents produced during the COVID-19 pandemic.

Applications of digital health for public health responses to COVID-19: a systematic scoping review of artificial intelligence, telehealth and related technologies

The coronavirus disease 2019 (COVID-19) pandemic has overwhelmed healthcare services, faced with the twin challenges in acutely meeting the medical needs of patients with COVID-19 while continuing essential services for non-COVID-19 illnesses. The need to re-invent, re-organize and transform healthcare and co-ordinate clinical services at a population level is urgent as countries that controlled initial outbreaks start to experience resurgences. A wide range of digital health solutions have been proposed, although the extent of successful real-world applications of these technologies is unclear. This study aims to review applications of artificial intelligence (AI), telehealth, and other relevant digital health solutions for public health responses in the healthcare operating environment amidst the COVID-19 pandemic. A systematic scoping review was performed to identify potentially relevant reports. Key findings include a large body of evidence for various clinical and operational applications of telehealth (40.1%, n = 99/247). Although a large quantity of reports investigated applications of artificial intelligence (AI) (44.9%, n = 111/247) and big data analytics (36.0%, n = 89/247), weaknesses in study design limit generalizability and translation, highlighting the need for more pragmatic real-world investigations. There were also few descriptions of applications for the internet of things (IoT) (2.0%, n = 5/247), digital platforms for communication (DC) (10.9%, 27/247), digital solutions for data management (DM) (1.6%, n = 4/247), and digital structural screening (DS) (8.9%, n = 22/247); representing gaps and opportunities for digital public health. Finally, the performance of digital health technology for operational applications related to population surveillance and points of entry have not been adequately evaluated.

The shortage of hospital beds for COVID-19 and non-COVID-19 patients during the lockdown of Wuhan, China

Background

The 76-day lockdown of Wuhan city has successfully contained the first wave of the coronavirus disease 2019 (COVID-19) outbreak. However, to date few studies have evaluated the hospital bed shortage for COVID-19 during the lockdown and none for non-COVID-19 patients, although such data are important for better preparedness of the future outbreak.

Methods

We built a compartmental model to estimate the daily numbers of hospital bed shortage for patients with mild, severe and critical COVID-19, taking account of underreport and diagnosis delay.

Results

The maximal daily shortage of inpatient beds for mild, severe and critical COVID-19 patients was 43,960 (95% confidence interval: 35,246, 52,929), 2,779 (1,395, 4,163) and 196 (143, 250) beds in early February 2020. An earlier or later lockdown would have greatly increased the shortage of hospital beds in Wuhan. The overwhelmed healthcare system might have delayed the provision of health care to both COVID-19 and non-COVID-19 patients during the lockdown. The second wave in Wuhan could have occurred in June 2020 if social distancing measures had waned in early March 2020. The hospital bed shortage was estimated much smaller in the potential second wave than in the first one.

Conclusions

Our findings suggest that the timing and strength of lockdown is important for the containment of the COVID-19 outbreaks. The healthcare needs of non-COVID-19 patients in the pandemic warrant more investigations.

Examining risk and crisis communications of government agencies and stakeholders during early-stages of COVID-19 on Twitter

During COVID-19, social media has played an important role for public health agencies and government stakeholders (i.e. actors) to disseminate information regarding situations, risks, and personal protective action inhibiting disease spread. However, there have been notable insufficient, incongruent, and inconsistent communications regarding the pandemic and its risks, which was especially salient at the early stages of the outbreak. Sufficiency, congruence and consistency in health risk communication have important implications for effective health safety instruction as well as critical content interpretability and recall. It also impacts individual- and community-level responses to information. This research employs text mining techniques and dynamic network analysis to investigate the actors' risk and crisis communication on Twitter regarding message types, communication sufficiency, timeliness, congruence, consistency and coordination. We studied 13,598 pandemic-relevant tweets posted over January to April from 67 federal and state-level agencies and stakeholders in the U.S. The study annotates 16 categories of message types, analyzes their appearances and evolutions. The research then identifies inconsistencies and incongruencies on four critical topics and examines spatial disparities, timeliness, and sufficiency across actors and message types in communicating COVID-19. The network analysis also reveals increased communication coordination over time. The findings provide unprecedented insight of Twitter COVID-19 information dissemination which may help to inform public health agencies and governmental stakeholders future risk and crisis communication strategies related to global hazards in digital environments.

Assessing the risks of 'infodemics' in response to COVID-19 epidemics

During COVID-19, governments and the public are fighting not only a pandemic but also a co-evolving infodemic-the rapid and far-reaching spread of information of questionable quality. We analysed more than 100 million Twitter messages posted worldwide during the early stages of epidemic spread across countries (from 22 January to 10 March 2020) and classified the reliability of the news being circulated. We developed an Infodemic Risk Index to capture the magnitude of exposure to unreliable news across countries. We found that measurable waves of potentially unreliable information preceded the rise of COVID-19 infections, exposing entire countries to falsehoods that pose a serious threat to public health. As infections started to rise, reliable information quickly became more dominant, and Twitter content shifted towards more credible informational sources. Infodemic early-warning signals provide important cues for misinformation mitigation by means of adequate communication strategies.

Using mHealth to Increase the Reach of Local Guidance to Health Professionals as Part of an Institutional Response Plan to the COVID-19 Outbreak: Usage Analysis Study

BACKGROUND

The ongoing coronavirus disease (COVID-19) pandemic forced health jurisdictions worldwide to significantly restructure and reorganize their medical activities. In response to the rapidly evolving body of evidence, a solid communication strategy is needed to increase the reach of and adherence to locally drafted and validated guidance to aide medical staff with COVID-19-related clinical decisions.

OBJECTIVE

We present a usage analysis of a dedicated mobile health (mHealth) platform as part of an institutional knowledge dissemination strategy of COVID-19-related guidance to all health care workers (HCWs) in a large academic hospital.

METHODS

A multidisciplinary team of experts drafted local guidance related to COVID-19. In total, 60 documents and 17 external links were made available through the platform. Documents were disseminated using a recently deployed mHealth platform for HCWs. Targeted dissemination of COVID-19-related content began on March 22, 2020. Using a third-party statistics tool, data concerning user activity and content use was anonymously collected. A quantitative analysis of user activity was performed over a 4-month period, separated into 3 periods: 2 months before (Period A), 2 weeks after (Period B), and 6 weeks following (Period C) targeted dissemination. Regional epidemiological data (daily new COVID-19 cases and total COVID-19-related hospitalizations) was extracted from an official registry.

RESULTS

During the study period, the platform was downloaded by 1233 new users. Consequently, the total number of users increased from 1766 users before Period A to a total of 2999 users at the end of Period C. We observed 27,046 document views, of which 12,728 (47.1%) were COVID-19-related. The highest increase in activity occurred in Period B, rapidly following targeted dissemination, with 7740 COVID-19-related content views, representing 71.2% of total content views within the abovementioned period and 550 daily views of COVID-19-related documents. Total documents consulted per day increased from 117 (IQR 74-160) to 657 (IQR 481-1051), P<.001. This increase in activity followed the epidemiological curbing of newly diagnosed COVID-19 cases, which peaked during Period B. Total active devices doubled from 684 to 1400, daily user activity increased fourfold, and the number of active devices rose from 53 (IQR 40-70) to 210 (IQR 167-297), P<.001. In addition, the number of sessions per day rose from 166 (IQR 110-246) to 704 (IQR 517-1028), P<.001. A persistent but reduced increase in total documents consulted per day (172 [IQR 131-251] versus 117 [IQR 74-160], P<.001) and active devices (71 [IQR 64-89] versus 53 [IQR 40-70]) was observed in Period C compared to Period A, while only 29.8% of the content accessed was COVID-19-related. After targeted dissemination, an immediate increase in activity was observed after push notifications were sent to users.

CONCLUSIONS

The use of an mHealth solution to disseminate time-sensitive medical knowledge seemed to be an effective solution to increase the reach of validated content to a targeted audience.

A case-based reasoning framework for early detection and diagnosis of novel coronavirus

Coronavirus, also known as COVID-19, has been declared a pandemic by the World Health Organization (WHO). At the time of conducting this study, it had recorded over 11,301,850 confirmed cases while more than 531,806 have died due to it, with these figures rising daily across the globe. The burden of this highly contagious respiratory disease is that it presents itself in both symptomatic and asymptomatic patterns in those already infected, thereby leading to an exponential rise in the number of contractions of the disease and fatalities. It is, therefore, crucial to expedite the process of early detection and diagnosis of the disease across the world. The case-based reasoning (CBR) model is a compelling paradigm that allows for the utilization of case-specific knowledge previously experienced, concrete problem situations or specific patient cases for solving new cases. This study, therefore, aims to leverage the very rich database of cases of COVID-19 to solve new cases. The approach adopted in this study employs the use of an improved CBR model for state-of-the-art reasoning task in the classification of suspected cases of COVID-19. The CBR model leverages on a novel feature selection and the semantic-based mathematical model proposed in this study for case similarity computation. An initial population of the archive was achieved from 71 (67 adults and 4 pediatrics) cases obtained from the Italian Society of Medical and Interventional Radiology (SIRM) repository. Results obtained revealed that the proposed approach in this study successfully classified suspected cases into their categories with an accuracy of 94.54%. The study found that the proposed model can support physicians to easily diagnose suspected cases of COVID-19 based on their medical records without subjecting the specimen to laboratory tests. As a result, there will be a global minimization of contagion rate occasioned by slow testing and in addition, reduced false-positive rates of diagnosed cases as observed in some parts of the globe.

Impact of National Containment Measures on Decelerating the Increase in Daily New Cases of COVID-19 in 54 Countries and 4 Epicenters of the Pandemic: Comparative Observational Study

BACKGROUND

Coronavirus disease (COVID-19) is a worldwide epidemic, and various countries have responded with different containment measures to reduce disease transmission, including stay-at-home orders, curfews, and lockdowns. Comparative studies have not yet been conducted to investigate the impact of these containment measures; these studies are needed to facilitate public health policy-making across countries.

OBJECTIVE

The aim of this study was to describe and evaluate the impact of national containment measures and policies (stay-at-home orders, curfews, and lockdowns) on decelerating the increase in daily new cases of COVID-19 in 54 countries and 4 epicenters of the pandemic in different jurisdictions worldwide.

METHODS

We reviewed the effective dates of the national containment measures (stay-at-home order, curfew, or lockdown) of 54 countries and 4 epicenters of the COVID-19 pandemic (Wuhan, New York State, Lombardy, and Madrid), and we searched cumulative numbers of confirmed COVID-19 cases and daily new cases provided by health authorities. Data were drawn from an open, crowdsourced, daily-updated COVID-19 data set provided by Our World in Data. We examined the trends in the percent increase in daily new cases from 7 days before to 30 days after the dates on which containment measures went into effect by continent, World Bank income classification, type of containment measures, effective date of containment measures, and number of confirmed cases on the effective date of the containment measures.

RESULTS

We included 122,366 patients with confirmed COVID-19 infection from 54 countries and 24,071 patients from 4 epicenters on the effective dates on which stay-at-home orders, curfews, or lockdowns were implemented between January 23 and April 11, 2020. Stay-at-home, curfew, and lockdown measures commonly commenced in countries with approximately 30%, 20%, or 10% increases in daily new cases. All three measures were found to lower the percent increase in daily new cases to <5 within one month. Among the countries studied, 20% had an average percent increase in daily new cases of 30-49 over the seven days prior to the commencement of containment measures; the percent increase in daily new cases in these countries was curbed to 10 and 5 a maximum of 15 days and 23 days after the implementation of containment measures, respectively.

CONCLUSIONS

Different national containment measures were associated with a decrease in daily new cases of confirmed COVID-19 infection. Stay-at-home orders, curfews, and lockdowns curbed the percent increase in daily new cases to <5 within a month. Resurgence in cases within one month was observed in some South American countries.

Using Reports of Symptoms and Diagnoses on Social Media to Predict COVID-19 Case Counts in Mainland China: Observational Infoveillance Study

BACKGROUND

Coronavirus disease (COVID-19) has affected more than 200 countries and territories worldwide. This disease poses an extraordinary challenge for public health systems because screening and surveillance capacity is often severely limited, especially during the beginning of the outbreak; this can fuel the outbreak, as many patients can unknowingly infect other people.

OBJECTIVE

The aim of this study was to collect and analyze posts related to COVID-19 on Weibo, a popular Twitter-like social media site in China. To our knowledge, this infoveillance study employs the largest, most comprehensive, and most fine-grained social media data to date to predict COVID-19 case counts in mainland China.

METHODS

We built a Weibo user pool of 250 million people, approximately half the entire monthly active Weibo user population. Using a comprehensive list of 167 keywords, we retrieved and analyzed around 15 million COVID-19-related posts from our user pool from November 1, 2019 to March 31, 2020. We developed a machine learning classifier to identify "sick posts," in which users report their own or other people's symptoms and diagnoses related to COVID-19. Using officially reported case counts as the outcome, we then estimated the Granger causality of sick posts and other COVID-19 posts on daily case counts. For a subset of geotagged posts (3.10% of all retrieved posts), we also ran separate predictive models for Hubei province, the epicenter of the initial outbreak, and the rest of mainland China.

RESULTS

We found that reports of symptoms and diagnosis of COVID-19 significantly predicted daily case counts up to 14 days ahead of official statistics, whereas other COVID-19 posts did not have similar predictive power. For the subset of geotagged posts, we found that the predictive pattern held true for both Hubei province and the rest of mainland China regardless of the unequal distribution of health care resources and the outbreak timeline.

CONCLUSIONS

Public social media data can be usefully harnessed to predict infection cases and inform timely responses. Researchers and disease control agencies should pay close attention to the social media infosphere regarding COVID-19. In addition to monitoring overall search and posting activities, leveraging machine learning approaches and theoretical understanding of information sharing behaviors is a promising approach to identify true disease signals and improve the effectiveness of infoveillance.

Evaluation of the Secondary Transmission Pattern and Epidemic Prediction of COVID-19 in the Four Metropolitan Areas of China

Understanding the transmission dynamics of COVID-19 is crucial for evaluating its spread pattern, especially in metropolitan areas of China, as its spread could lead to secondary outbreaks. In addition, the experiences gained and lessons learned from China have the potential to provide evidence to support other metropolitan areas and large cities outside China with their emerging cases. We used data reported from January 24, 2020, to February 23, 2020, to fit a model of infection, estimate the likely number of infections in four high-risk metropolitan areas based on the number of cases reported, and increase the understanding of the COVID-19 spread pattern. Considering the effect of the official quarantine regulations and travel restrictions for China, which began January 23~24, 2020, we used the daily travel intensity index from the Baidu Maps app to roughly simulate the level of restrictions and estimate the proportion of the quarantined population. A group of SEIR model statistical parameters were estimated using Markov chain Monte Carlo (MCMC) methods and fitting on the basis of reported data. As a result, we estimated that the basic reproductive number, R₀, was 2.91 in Beijing, 2.78 in Shanghai, 2.02 in Guangzhou, and 1.75 in Shenzhen based on the data from January 24, 2020, to February 23, 2020. In addition, we inferred the prediction results and compared the results of different levels of parameters. For example, in Beijing, the predicted peak number of cases was 467 with a peak time of March 01, 2020; however, if the city were to implement different levels (strict, moderate, or weak) of travel restrictions or regulation measures, the estimation results showed that the transmission dynamics would change and that the peak number of cases would differ by between 54% and 209%. We concluded that public health interventions would reduce the risk of the spread of COVID-19 and that more rigorous control and prevention measures would effectively contain its further spread, and awareness of prevention should be enhanced when businesses and social activities return to normal before the end of the epidemic. Further, the experiences gained and lessons learned from China offer the potential to provide evidence supporting other metropolitan areas and big cities with their emerging cases outside China.

Modeling analysis of COVID-19 based on morbidity data in Anhui, China

Since the first case of coronavirus disease (COVID-19) in Wuhan Hubei, China, was reported in December 2019, COVID-19 has spread rapidly across the country and overseas. The first case in Anhui, a province of China, was reported on January 10, 2020. In the field of infectious diseases, modeling, evaluating and predicting the rate of disease transmission is very important for epidemic prevention and control. Different intervention measures have been implemented starting from different time nodes in the country and Anhui, the epidemic may be divided into three stages for January 10 to February 11, 2020, namely. We adopted interrupted time series method and develop an SEI/QR model to analyse the data. Our results displayed that the lockdown of Wuhan implemented on January 23, 2020 reduced the contact rate of epidemic transmission in Anhui province by 48.37%, and centralized quarantine management policy for close contacts in Anhui reduced the contact rate by an additional 36.97%. At the same time, the estimated basic reproduction number gradually decreased from the initial 2.9764 to 0.8667 and then to 0.5725. We conclude that the Wuhan lockdown and the centralized quarantine management policy in Anhui played a crucial role in the timely and effective mitigation of the epidemic in Anhui. One merit of this work is the adoption of morbidity data which may reflect the epidemic more accurately and promptly. Our estimated parameters are largely in line with the World Health Organization estimates and previous studies.

Evaluation of the Secondary Transmission Pattern and Epidemic Prediction of COVID-19 in the Four Metropolitan Areas of China

Understanding the transmission dynamics of COVID-19 is crucial for evaluating its spread pattern, especially in metropolitan areas of China, as its spread could lead to secondary outbreaks. In addition, the experiences gained and lessons learned from China have the potential to provide evidence to support other metropolitan areas and large cities outside China with their emerging cases. We used data reported from January 24, 2020, to February 23, 2020, to fit a model of infection, estimate the likely number of infections in four high-risk metropolitan areas based on the number of cases reported, and increase the understanding of the COVID-19 spread pattern. Considering the effect of the official quarantine regulations and travel restrictions for China, which began January 23~24, 2020, we used the daily travel intensity index from the Baidu Maps app to roughly simulate the level of restrictions and estimate the proportion of the quarantined population. A group of SEIR model statistical parameters were estimated using Markov chain Monte Carlo (MCMC) methods and fitting on the basis of reported data. As a result, we estimated that the basic reproductive number, R ₀, was 2.91 in Beijing, 2.78 in Shanghai, 2.02 in Guangzhou, and 1.75 in Shenzhen based on the data from January 24, 2020, to February 23, 2020. In addition, we inferred the prediction results and compared the results of different levels of parameters. For example, in Beijing, the predicted peak number of cases was 467 with a peak time of March 01, 2020; however, if the city were to implement different levels (strict, moderate, or weak) of travel restrictions or regulation measures, the estimation results showed that the transmission dynamics would change and that the peak number of cases would differ by between 54% and 209%. We concluded that public health interventions would reduce the risk of the spread of COVID-19 and that more rigorous control and prevention measures would effectively contain its further spread, and awareness of prevention should be enhanced when businesses and social activities return to normal before the end of the epidemic. Further, the experiences gained and lessons learned from China offer the potential to provide evidence supporting other metropolitan areas and big cities with their emerging cases outside China.