Population-scale Longitudinal Mapping of COVID-19 Symptoms, Behavior, and Testing Identifies Contributors to Continued Disease Spread in the United States

Allen, William E.; Altae-Tran, Han; Briggs, James; Jin, Xin; McGee, Glen; Raghavan, Rumya; Shi, Andy; Kamariza, Mireille; Nova, Nicole; Pereta, Albert; Danford, Chris; Kamel, Amine; Gothe, Patrik; Milam, Evrhet; Aurambault, Jean; Primke, Thorben; Li, Claire; Inkenbrandt, Josh; Huynh, Tuan; Chen, Evan; Lee, Christina; Croatto, Michael; Bentley, Helen; Lu, Wendy; Murray, Robert P.; Travassos, Mark A.; Openshaw, John J.; Coull, Brent A.; Greene, Casey S.; Shalem, Ophir; King, Gary; Probasco, Ryan; Cheng, David; Silbermann, Ben; Zhang, Feng; Lin, Xihong

doi:10.1101/2020.06.09.20126813

Cited by 14 publications

(19 citation statements)

References 35 publications

(38 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although we experimented with more sophisticated machine learning models, for the purpose of this paper, we wanted to show that we could arrive at a reasonable estimate using a simple and easily reproducible method. A similar approach has been described in [ 6 ] for US and UK data.…”

Section: Resultsmentioning

confidence: 99%

“…From the answers to two of the questions (Q22: symptoms and Q8: contact with coronavirus-infected individual in the household) plus demographic information, we built a generalized linear model and reported SARS-CoV-2 prevalence estimations that are on par with those carried out by a seroprevalence study in Spain. As shown in related studies [ 6 ], when public policy decisions need to be made rapidly in a situation of data scarcity and limited testing capacity, a large-scale population survey might be of great value to make fast assessments of prevalence, as it can be deployed rapidly and enable the collection of results within hours.…”

Section: Discussionmentioning

confidence: 99%

“…Moreover, the resulting sample might be biased and difficult to reweight [ 5 ]. Given the limitations of these traditional methods and given the need for rapid data collection, large-scale online surveys can be a valuable method to quickly assess and longitudinally monitor the situation and perceptions of the population in the context of a pandemic [ 6 ]. Thus, to shed light on important, yet unknown, questions related to COVID-19, we designed a 24-question online survey, called the Covid19Impact survey, to be targeted to the Spanish population.…”

Section: Introductionmentioning

confidence: 99%

“…In Israel, the Weizmann Institute and the Ministry of Health are collecting data on basic demographics, health, and potential exposure [ 12 ]. The project aims to predict the location of COVID-19 outbreaks by analyzing information collected about the virus symptoms and public behavior in real time [ 6 , 13 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Assessing the Impact of the COVID-19 Pandemic in Spain: Large-Scale, Online, Self-Reported Population Survey

Oliver¹,

Barber²,

Roomp³

et al. 2020

J Med Internet Res

View full text Add to dashboard Cite

Background Spain has been one of the countries most impacted by the COVID-19 pandemic. Since the first confirmed case was reported on January 31, 2020, there have been over 405,000 cases and 28,000 deaths in Spain. The economic and social impact is without precedent. Thus, it is important to quickly assess the situation and perception of the population. Large-scale online surveys have been shown to be an effective tool for this purpose. Objective We aim to assess the situation and perception of the Spanish population in four key areas related to the COVID-19 pandemic: social contact behavior during confinement, personal economic impact, labor situation, and health status. Methods We obtained a large sample using an online survey with 24 questions related to COVID-19 in the week of March 28-April 2, 2020, during the peak of the first wave of COVID-19 in Spain. The self-selection online survey method of nonprobability sampling was used to recruit 156,614 participants via social media posts that targeted the general adult population (age >18 years). Given such a large sample, the 95% CI was ±0.843 for all reported proportions. Results Regarding social behavior during confinement, participants mainly left their homes to satisfy basic needs. We found several statistically significant differences in social behavior across genders and age groups. The population’s willingness to comply with the confinement measures is evident. From the survey answers, we identified a significant adverse economic impact of the pandemic on those working in small businesses and a negative correlation between economic damage and willingness to stay in confinement. The survey revealed that close contacts play an important role in the transmission of the disease, and 28% of the participants lacked the necessary resources to properly isolate themselves. We also identified a significant lack of testing, with only 1% of the population tested and 6% of respondents unable to be tested despite their doctor’s recommendation. We developed a generalized linear model to identify the variables that were correlated with a positive SARS-CoV-2 test result. Using this model, we estimated an average of 5% for SARS-CoV-2 prevalence in the Spanish population during the time of the study. A seroprevalence study carried out later by the Spanish Ministry of Health reported a similar level of disease prevalence (5%). Conclusions Large-scale online population surveys, distributed via social media and online messaging platforms, can be an effective, cheap, and fast tool to assess the impact and prevalence of an infectious disease in the context of a pandemic, particularly when there is a scarcity of official data and limited testing capacity.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Assessing the Impact of the COVID-19 Pandemic in Spain: Large-Scale, Online, Self-Reported Population Survey

Oliver¹,

Barber²,

Roomp³

et al. 2020

J Med Internet Res

View full text Add to dashboard Cite

show abstract

“…However these studies either looked at only a single health care institution or system 9,10 , used national voluntary surveys that are subject to selection bias 11 , or small scale ethnicity specific opinion surveys. 12 In the absence of large scale nationally representative individual-level data in the US to investigate this matter, analysis of ecologically aggregated county-level data across the US can provide valuable insights.…”

Section: Main Textmentioning

confidence: 99%

Identifying US Counties with High Cumulative COVID-19 Burden and Their Characteristics

Gaynor

Quick

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Identifying areas with high COVID-19 burden and their characteristics can help improve vaccine distribution and uptake, reduce burdens on health care systems, and allow for better allocation of public health intervention resources. Synthesizing data from various government and nonprofit institutions of 3,142 United States (US) counties as of 12/21/2020, we studied county-level characteristics that are associated with cumulative case and death rates using regression analyses. Our results showed counties that are more rural, counties with more White/non-White segregation, and counties with higher percentages of people of color, in poverty, with no high school diploma, and with medical comorbidities such as diabetes and hypertension are associated with higher cumulative COVID-19 case and death rates. We identify the hardest hit counties in US using model-estimated case and death rates, which provide more reliable estimates of cumulative COVID-19 burdens than those using raw observed county-specific rates. Identification of counties with high disease burdens and understanding the characteristics of these counties can help inform policies to improve vaccine distribution, deployment and uptake, prevent overwhelming health care systems, and enhance testing access, personal protection equipment access, and other resource allocation efforts, all of which can help save more lives for vulnerable communities.Significance statementWe found counties that are more rural, counties with more White/non-White segregation, and counties with higher percentages of people of color, in poverty, with no high school diploma, and with medical comorbidities such as diabetes and hypertension are associated with higher cumulative COVID-19 case and death rates. We also identified individual counties with high cumulative COVID-19 burden. Identification of counties with high disease burdens and understanding the characteristics of these counties can help inform policies to improve vaccine distribution, deployment and uptake, prevent overwhelming health care systems, and enhance testing access, personal protection equipment access, and other resource allocation efforts, all of which can help save more lives for vulnerable communities.

show abstract

Pulmonary adverse drug event data in hypertension with implications on COVID-19 morbidity

Jaberi-Douraki

Meyer

Riviere

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Hypertension is a recognized comorbidity for COVID-19. The association of antihypertensive medications with outcomes in patients with hypertension is not fully described. However, angiotensin-converting enzyme 2 (ACE2), responsible for host entry of the novel coronavirus (SARS-CoV-2) leading to COVID-19, is postulated to be upregulated in patients taking angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin II receptor blockers (ARBs). Here, we evaluated the occurrence of pulmonary adverse drug events (ADEs) in patients with hypertension receiving ACEIs/ARBs to determine if disparities exist between individual drugs within the respective classes using data from the FDA Spontaneous Reporting Systems. For this purpose, we proposed the proportional reporting ratio to provide a statistical summary for the commonality of an ADE for a specific drug as compared to the entire database for drugs in the same or other classes. In addition, a statistical procedure, multiple logistic regression analysis, was employed to correct hidden confounders when causative covariates are underreported or untrusted to correct analyses of drug-ADE combinations. To date, analyses have been focused on drug classes rather than individual drugs which may have different ADE profiles depending on the underlying diseases present. A retrospective analysis of thirteen pulmonary ADEs showed significant differences associated with quinapril and trandolapril, compared to other ACEIs and ARBs. Specifically, quinapril and trandolapril were found to have a statistically significantly higher incidence of pulmonary ADEs compared with other ACEIs as well as ARBs (P < 0.0001) for group comparison (i.e., ACEIs vs. ARBs vs. quinapril vs. trandolapril) and (P ≤ 0.0007) for pairwise comparison (i.e., ACEIs vs. quinapril, ACEIs vs. trandolapril, ARBs vs. quinapril, or ARBs vs. trandolapril). This study suggests that specific members of the ACEI antihypertensive class (quinapril and trandolapril) have a significantly higher cluster of pulmonary ADEs.

show abstract

Population-scale Longitudinal Mapping of COVID-19 Symptoms, Behavior, and Testing Identifies Contributors to Continued Disease Spread in the United States

Cited by 14 publications

References 35 publications

Assessing the Impact of the COVID-19 Pandemic in Spain: Large-Scale, Online, Self-Reported Population Survey

Assessing the Impact of the COVID-19 Pandemic in Spain: Large-Scale, Online, Self-Reported Population Survey

Identifying US Counties with High Cumulative COVID-19 Burden and Their Characteristics

Pulmonary adverse drug event data in hypertension with implications on COVID-19 morbidity

Contact Info

Product

Resources

About