BACKGROUND:Limitations in laboratory diagnostic capacity impact population surveillance of COVID-19. It is currently unknown whether participatory surveillance tools for COVID-19 correspond to government-reported case trends longitudinally and if it can be used as an adjunct to laboratory testing. The primary objective of this study was to determine whether self-reported COVID-19-like illness reflected laboratory-confirmed COVID-19 case trends in Ontario Canada. METHODS:We retrospectively analyzed longitudinal self-reported symptoms data collected using an online tool-Outbreaks Near Me (ONM)-from April 20th, 2020, to March 7th, 2021 in Ontario, Canada. We measured the correlation between COVID-like illness among respondents and the weekly number of PCR-confirmed COVID-19 cases and provincial test positivity. We explored contemporaneous changes in other respiratory viruses, as well as the demographic characteristics of respondents to provide context for our findings. RESULTS:Between 3,849-11,185 individuals responded to the symptom survey each week. No correlations were seen been self-reported CLI and either cases or test positivity. Strong positive correlations were seen between CLI and both cases and test positivity before a previously documented rise in rhinovirus/enterovirus in fall 2020. Compared to participatory surveillance respondents, a higher proportion of COVID-19 cases in Ontario consistently came from low-income, racialized and immigrant areas of the province- these groups were less well represented among survey respondents. INTERPRETATION:Although digital surveillance systems are low-cost tools that have been useful to signal the onset of viral outbreaks, in this longitudinal comparison of self-reported COVID-like illness to Ontario COVID-19 case data we did not find this to be the case. Seasonal respiratory virus transmission and population coverage may explain this discrepancy.

Symptomatic testing programmes are crucial to the COVID-19 pandemic response. We sought to examine United Kingdom (UK) testing rates amongst individuals with test-qualifying symptoms, and factors associated with not testing. We analysed a cohort of untested symptomatic app users (N = 1,237), nested in the Zoe COVID Symptom Study (Zoe, N = 4,394,948); and symptomatic respondents who wanted, but did not have a test (N = 1,956), drawn from a University of Maryland survey administered to Facebook users (The Global COVID-19 Trends and Impact Survey [CTIS], N = 775,746). The proportion tested among individuals with incident test-qualifying symptoms rose from ~20% to ~75% from April to December 2020 in Zoe. Testing was lower with one vs more symptoms (72.9% vs 84.6% p<0.001), or short vs long symptom duration (69.9% vs 85.4% p<0.001). 40.4% of survey respondents did not identify all three test-qualifying symptoms. Symptom identification decreased for every decade older (OR = 0.908 [95% CI 0.883-0.933]). Amongst symptomatic UMD-CTIS respondents who wanted but did not have a test, not knowing where to go was the most cited factor (32.4%); this increased for each decade older (OR = 1.207 [1.129-1.292]) and for every 4-years fewer in education (OR = 0.685 [0.599-0.783]). Despite current UK messaging on COVID-19 testing, there is a knowledge gap about when and where to test, and this may be contributing to the ~25% testing gap. Risk factors, including older age and less education, highlight potential opportunities to tailor public health messages. The testing gap may be ever larger in countries that do not have extensive, free testing, as the UK does.

Background Symptomatic testing programmes are crucial to the COVID-19 pandemic response. We sought to examine United Kingdom (UK) testing rates amongst individuals with test-qualifying symptoms, and factors associated with not testing. Methods We analysed a cohort of untested symptomatic app users (N=1,237), nested in the Zoe COVID Symptom Study (Zoe, N= 4,394,948); and symptomatic survey respondents who wanted, but did not have a test (N=1,956), drawn from the University of Maryland-Facebook Covid-19 Symptom Survey (UMD-Facebook, N=775,746). Findings The proportion tested among individuals with incident test-qualifying symptoms rose from ~20% to ~75% from April to December 2020 in Zoe. Testing was lower with one vs more symptoms (73.0% vs 85.0%), or short vs long symptom duration (72.6% vs 87.8%). 40.4% of survey respondents did not identify all three test-qualifying symptoms. Symptom identification decreased for every decade older (OR=0.908 [95% CI 0.883-0.933]). Amongst symptomatic UMD-Facebook respondents who wanted but did not have a test, not knowing where to go was the most cited factor (32.4%); this increased for each decade older (OR=1.207 [1.129-1.292]) and for every 4-years fewer in education (OR=0.685 [0.599-0.783]). Interpretation Despite current UK messaging on COVID-19 testing, there is a knowledge gap about when and where to test, and this may be contributing to the ~25% testing gap. Risk factors, including older age and less education, highlight potential opportunities to tailor public health messages.

Over the past fifteen years, infectious disease surveillance has evolved to include online citizen science projects that implement active digital data collection. Flu Near You (FNY) is an online participatory surveillance system in the United States that collects weekly health reports from its volunteers. Because high engagement levels and consistent participation of users are essential to accurately track disease and estimate burden, FNY implemented three measures: (1) adding a new feature in that allows website visitors to submit a health report without registering, (2) administering user surveys at the end of each influenza season, and (3) partnering with Science Friday (SciFri) (a weekly public radio program that discusses topics relevant to science, nature, and technology) with the goals of increasing the number of weekly participant reports, improving the frequency of user reporting, and understanding user motivation. In this paper, we evaluate the impact of these three measures. The number of registered users who submitted only one report decreased after allowing users to submit reports without registering. The survey indicated that respondents were primarily motivated by the importance of disease tracking and by a desire to participate in a citizen science project. Finally, users who registered from SciFri were highly engaged and also motivated by a desire to participate in a citizen science project.

IMPORTANCE Tensions around COVID-19 and systemic racism have raised the question: are hospitals advocating for equity for their Black patients? It is imperative for hospitals to be supportive of the Black community and acknowledge themselves as safe spaces, run by clinicians and staff who care about social justice issues that impact the health of the Black community; without the expression of support, Black patients may perceive hospitals as uncaring and unsafe, potentially delaying or avoiding treatment, which can result in serious complications and death for those with COVID-19. OBJECTIVE To explore how hospitals showed public-facing support for the Black community as measured through tweets about social equity or the Black Lives Matter (BLM) movement. DESIGN, SETTING, AND PARTICIPANTS Using a retrospective longitudinal cohort study design, tweets from the top 100 ranked hospitals were collected, starting with the most recent over a 10-year span, from May 3, 2009, to June 26, 2020. The date of the George Floyd killing, May 25, 2020, was investigated as a point of interest. Data were analyzed from June 11 to December 4, 2020. MAIN OUTCOMES AND MEASURES Tweets were manually identified based on 4 categories: BLM, associated with the BLM movement; Black support, expressed support for Black population within the hospital's community; Black health, pertained to health concerns specific to and the creation of health care for the Black community; or social justice, associated with general social justice terms that were too general to label as Black. If a tweet did not contain any hashtags from these categories, it remained unlabeled. RESULTS A total of 281 850 tweets from 90 unique social media accounts were collected. Each handle returned at least 1279 tweets, with 85 handles (94.4%) returning at least 3000 tweets. Tweet publication dates ranged from 2009 to 2020. A total of 274 tweets (0.097%) from 67 handles (74.4%) used a hashtag to support the BLM movement. Among the tweets labeled BLM, the first tweet was published in 2018 and only 4 tweets (1.5%) predated the killing of George Floyd. A similar trend of low signal observed was detected for the other categories (Black support: 244 tweets [0.086%] from 42 handles [46.7%] starting in 2013; Black health: 28 tweets [0.0099%] from 15 handles [16.7%] starting in 2018; social justice: 40 tweets [0.014%] from 21 handles [23.3%] starting in 2015). CONCLUSIONS AND RELEVANCE These findings reflect the low signal of tweets regarding the Black community and social justice in a generalized way across approximately 10 years of tweets for all the hospital handles within the data set. From 2009 to 2020, hospitals rarely engaged in issues pertaining to the Black community and if so, only within the last half of this time period. These later entrances into these discussions indicate that these discussions are relatively recent.

ABSTRACT Background Syndromic surveillance through web or phone-based polling has been used to track the course of infectious diseases worldwide. Our study objective was to describe the characteristics, symptoms, and self-reported testing rates of respondents in three different COVID-19 symptom surveys in Canada. Methods Data sources consisted of two distinct Canada-wide web-based surveys, and phone polling in Ontario. All three sources contained self-reported information on COVID-19 symptoms and testing. In addition to describing respondent characteristics, we examined symptom frequency and the testing rate among the symptomatic, as well as rates of symptoms and testing across respondent groups. Results We found that 1.6% of respondents experienced a symptom on the day of their survey, 15% of Ontario households had a symptom in the previous week, and 44% of Canada-wide respondents had a symptom in the previous month over March-April 2020. Across the three surveys, SARS-CoV-2-testing was reported in 2-9% of symptomatic responses. Women, younger and middle-aged adults (versus older adults) and Indigenous/First nations/Inuit/Métis were more likely to report at least one symptom, and visible minorities were more likely to report the combination of fever with cough or shortness of breath. Interpretation The low rate of testing among those reporting symptoms suggests significant opportunity to expand testing among community-dwelling residents of Canada. Syndromic surveillance data can supplement public health reports and provide much-needed context to gauge the adequacy of current SARS-CoV-2 testing rates.

The COVID-19 pandemic is straining public health systems worldwide and major non-pharmaceutical interventions have been implemented to slow its spread. During the initial phase of the outbreak the spread was primarily determined by human mobility. Yet empirical evidence on the effect of key geographic factors on local epidemic spread is lacking. We analyse highly-resolved spatial variables for cities in China together with case count data in order to investigate the role of climate, urbanization, and variation in interventions across China. Here we show that the epidemic intensity of COVID-19 is strongly shaped by crowding, such that epidemics in dense cities are more spread out through time, and denser cities have larger total incidence. Observed differences in epidemic intensity are well captured by a metapopulation model of COVID-19 that explicitly accounts for spatial hierarchies. Densely-populated cities worldwide may experience more prolonged epidemics. Whilst stringent interventions can shorten the time length of these local epidemics, although these may be difficult to implement in many affected settings.

BACKGROUND Racial and ethnic minority groups often face worse patient experiences compared with the general population, which is directly related to poorer health outcomes within these minority populations. Evaluation of patient experience among racial and ethnic minority groups has been difficult due to lack of representation in traditional health care surveys. OBJECTIVE This study aims to assess the feasibility of Twitter for identifying racial and ethnic disparities in patient experience across the United States from 2013 to 2016. METHODS In total, 851,973 patient experience tweets with geographic location information from the United States were collected from 2013 to 2016. Patient experience tweets included discussions related to care received in a hospital, urgent care, or any other health institution. Ordinary least squares multiple regression was used to model patient experience sentiment and racial and ethnic groups over the 2013 to 2016 period and in relation to the implementation of the Patient Protection and Affordable Care Act (ACA) in 2014. RESULTS Racial and ethnic distribution of users on Twitter was highly correlated with population estimates from the United States Census Bureau’s 5-year survey from 2016 (r2=0.99; P<.001). From 2013 to 2016, the average patient experience sentiment was highest for White patients, followed by Asian/Pacific Islander, Hispanic/Latino, and American Indian/Alaska Native patients. A reduction in negative patient experience sentiment on Twitter for all racial and ethnic groups was seen from 2013 to 2016. Twitter users who identified as Hispanic/Latino showed the greatest improvement in patient experience, with a 1.5 times greater increase (P<.001) than Twitter users who identified as White. Twitter users who identified as Black had the highest increase in patient experience postimplementation of the ACA (2014-2016) compared with preimplementation of the ACA (2013), and this change was 2.2 times (P<.001) greater than Twitter users who identified as White. CONCLUSIONS The ACA mandated the implementation of the measurement of patient experience of care delivery. Considering that quality assessment of care is required, Twitter may offer the ability to monitor patient experiences across diverse racial and ethnic groups and inform the evaluation of health policies like the ACA.

AbstractImportanceAccess to testing is key to a successful response to the COVID-19 pandemic.ObjectiveTo determine the geographic accessibility to SARS-CoV-2 testing sites in the United States, as quantified by travel time.DesignCross-sectional analysis of SARS-CoV-2 testing sites as of April 7, 2020 in relation to travel time.SettingUnited States COVID-19 pandemic.ParticipantsThe United States, including the 48 contiguous states and the District of Columbia.ExposuresPopulation density, percent minority, percent uninsured, and median income by county from the 2018 American Community Survey demographic data.Main OutcomeSARS-CoV-2 testing sites identified in two national databases (Carbon Health and CodersAgainstCovid), geocoded by address. Median county 1 km2 gridded friction surface of travel times, as a measure of geographic accessibility to SARS-CoV-2 testing sites.Results6,236 unique SARS-CoV-2 testing sites in 3,108 United States counties were identified. Thirty percent of the U.S. population live in a county (N = 1,920) with a median travel time over 20 minutes. This was geographically heterogeneous; 86% of the Mountain division population versus 5% of the Middle Atlantic population lived in counties with median travel times over 20 min. Generalized Linear Models showed population density, percent minority, percent uninsured and median income were predictors of median travel time to testing sites. For example, higher percent uninsured was associated with longer travel time (β = 0.41 min/percent, 95% confidence interval 0.3-0.53, p = 1.2×10−12), adjusting for population density.Conclusions and RelevanceGeographic accessibility to SARS-Cov-2 testing sites is reduced in counties with lower population density and higher percent of minority and uninsured, which are also risk factors for worse healthcare access and outcomes. Geographic barriers to SARS-Cov-2 testing may exacerbate health inequalities and bias county-specific transmission estimates. Geographic accessibility should be considered when planning the location of future testing sites and interpreting epidemiological data.Key PointsSARS-CoV-2 testing sites are distributed unevenly in the US geography and population.Median county-level travel time to SARS-CoV-2 testing sites is longer in less densely populated areas, and in areas with a higher percentage of minority or uninsured populations.Improved geographic accessibility to testing sites is imperative to manage the COVID-19 pandemic in the United States.

On March 16, 2020, the United States government issued new guidelines promoting public health social social distancing interventions to reduce the spread of the COVID-19 epidemic in the country [1]. In addition, many state and local governments in the United States have enacted stay-at-home policies banning mass gatherings, enforcing school closures, and promoting smart working. So far, however, the extent to which these policies have resulted in reduced people’s mobility has not been quantified. By analyzing data from millions of (anonymized, aggregated, privacy-enhanced) devices, we estimate that by March 23 the the policies have generally reduced by half the overall mobility in several major US cities. In order to gauge the observed results we know events, we note that the commuting volume on Monday, March 16, approached those of a typical snow day or analogous day when public schools are partially closed (ie January 2). By Friday, March 20, we observe commuting numbers that resemble those measured on federal holidays (ie Martin Luther King Jr. Day in January or Presidents’ Day in February). Currently, we are unable to quantify the extent∗ a. vespignani@ northeastern. edu

In March 2020, many state and local governments in the United States enacted stay-at-home policies banning mass gatherings, closing schools, and promoting remote working. By analyzing anonymized location data from millions of mobile devices, we quantify how much people have reduced their daily mobility and physical contacts in accordance with these guidelines. At the regional level, we measure declines in daily commute volume as well as transit between major urban areas. At the individual level, we measure changes in the average user’s daily range of mobility, number of unique contacts, and number of co-location events. According to these five measures, we estimate that the average person in the United States had reduced their daily mobility by between 45-55% as of late April, 2020, and had reduced their daily contacts between 65-75%. The United States’ physical distancing guidelines expired on April 30, 2020 and are not set to be renewed; as of early May, 2020, we report increases in mobility and contact patterns across most states (up to 10-14%, compared to the last week of April), though we do not observe a commensurate increase in commute volume. The response to the COVID-19 pandemic has amounted to one of the largest disruptions of economic, social, and mobility behavior in history, and quantifying these disruptions is vital for forecasting the further spread of this pandemic and crafting our collective response. ∗Equal contribution. §Correspondence: m.chinazzi@northeastern.edu

BACKGROUND:Media reporting on communicable diseases has been demonstrated to affect the perception of the public. Communicable disease reporting related to foreign-born persons has not yet been evaluated.OBJECTIVE:Examine how political leaning in the media affects reporting on tuberculosis (TB) in foreign-born persons.METHODS:HealthMap, a digital surveillance platform that aggregates news sources on global infectious diseases, was used. Data was queried for media reports from the U.S. between 2011-2019, containing the term "TB" or "tuberculosis" and "foreign born", "refugee (s)," or "im (migrants)." Reports were reviewed to exclude duplicates and non-human cases. Each media source was rated using two independent media bias indicators to assess political leaning. Forty-six non-tuberculosis reports were randomly sampled and evaluated as a control. Two independent reviewers performed sentiment analysis on each report.RESULTS:Of 891 TB-associated reports in the US, 46 referenced foreign-born individuals, and were included in this analysis. 60.9% (28) of reports were published in right-leaning news media and 6.5% (3) of reports in left-leaning media, while 39.1% (18) of the control group reports were published in left- leaning media and 10.9% (5) in right-leaning media (p < .001). 43% (20) of all study reports were posted in 2016. Sentiment analysis revealed that right-leaning reports often portrayed foreign-born persons negatively.CONCLUSION:Preliminary data from this pilot suggest that political leaning may affect reporting on TB in US foreign-born populations. Right-leaning news organizations produced the most reports on TB, and the majority of these reports portrayed foreign-born persons negatively. In addition, the control group comprised of non-TB, non-foreign born reports on communicable diseases featured a higher percentage of left-leaning news outlets, suggesting that reporting on TB in foreign-born individuals may be of greater interest to right-leaning outlets. Further investigation both in the U.S. and globally is needed.

BACKGROUND:News media coverage is a powerful influence on public attitude and government action. The digitization of news media covering the current opioid epidemic has changed the landscape of coverage and may have implications for how to effectively respond to the opioid crisis.OBJECTIVE:This study aims to characterize the relationship between volume of online opioid news reporting and opioid-related deaths in the United States and how these measures differ across geographic and socioeconomic county-level factors.METHODS:Online news reports from February 2018 to April 2019 on opioid-related events in the United States were extracted from Google News. News data were aggregated at the county level and compared against opioid-related death counts. Ordinary least squares regression was used to model opioid-related death rate and opioid news coverage with the inclusion of socioeconomic and geographic explanatory variables.RESULTS:A total of 35,758 relevant news reports were collected representing 1789 counties. Regression analysis revealed that opioid-related death rate was positively associated with news reporting. However, opioid-related death rate and news reporting volume showed opposite correlations with educational attainment and rurality. When controlling for variation in death rate, counties in the Northeast were overrepresented by news coverage.CONCLUSIONS:Our results suggest that regional variation in the volume of opioid-related news reporting does not reflect regional variation in opioid-related death rate. Differences in the amount of media attention may influence perceptions of the severity of opioid epidemic. Future studies should investigate the influence of media reporting on public support and action on opioid issues.

Americans spend trillions of minutes in cars annually, tens of billions of minutes traveling to health care providers, and hundreds of millions of minutes ridesharing (e.g., with Uber or Lyft). From July to October 2017, we recruited rideshare users (e.g., Uber or Lyft) to participate in a survey about health messaging during rides. Responses were collected anonymously on tablet devices. We interviewed 170 ridesharers and assessed their interest in health messaging delivered during rides. Participants ranged from 19 to 79 years of age, and most (87%) reported using their smartphones to search for health information. More than 70% expressed interest in health messaging during rides, and 55% of current smokers expressed interest in quit-smoking messaging. The most popular of suggested health topics included healthy eating (61.8%), exercise (60.6%), and weight loss (40.0%), and the preferred message formats were video (33.5%), smartphone apps (33.5%), and online social network (25.3%). Free time spent riding in cars represents an enormous untapped resource, and our findings suggest that riders are receptive to the idea of health messaging delivered during rideshare trips.

Purpose: Participatory disease surveillance actively engages the public in reporting on symptoms of their health to provide community-level data that complements traditional healthcare-based surveillance. Participatory surveillance is timely, low-cost, can account for non-medically attended populations, and allows for direct engagement with local populations. Three examples of participatory surveillance systems – Flu Near You in North America, Influenzanet in Europe, and Flu Tracking in Australia and New Zealand – have recently collaborated to develop Global Flu View, a shared platform for aggregation and dissemination of crowdsourced data on influenza-like illness. Methods & Materials: The collaborating surveillance systems have developed a shared application programming interface (API) for data exchange among systems. The API specifications outline specific variables that are called into a shared, cloud-based database. Data is processed via an ETL layer resulting in aggregate data at the postcode level. Key variables include reporting week, postal and country codes, user birth month and birth year, gender, vaccination status, and a series of symptom variables that include fever, cough, sore throat, headache, fatigue, and other indications of influenza-like illness. Processed data will be available to view in both graph and map formats on the website www.globalfluview.org. Results: The Global Flu View platform connects community-level influenza surveillance efforts across 15 countries on 3 continents. Tens of thousands of weekly reports are made available for users to explore through graphs, maps, and various filtering tools. The Global Flu View database also serves as a repository of influenza surveillance data that can be shared with disease modelers, researchers, and epidemiologists around the world. Conclusion: Global Flu View can highlight the value of the participatory surveillance approach, serving as a model for other countries. The platform is poised to incorporate data from additional countries with similar tools and serves as a case study for collaborative, multi-country data sharing for global health surveillance. As of July 2018 the platform is in beta-testing mode and will be made publicly available by late August 2018.