Characterizing COVID-19 and Influenza Illnesses in the Real World via Person-Generated Health Data

Allison Shapiro,Nicole Marinsek,Ieuan Clay,Benjamin Bradshaw,Ernesto Ramirez,Jae Min,Andrew Trister,Yuedong Wang,Tim Althoff,Luca Foschini

doi:10.1016/j.patter.2020.100188

Abstract

SummaryThe fight against COVID-19 is hindered by similarly presenting viral infections that may confound detection and monitoring. We examined person-generated health data (PGHD), consisting of survey and commercial wearable data from individuals' everyday lives, for 230 people who reported a COVID-19 diagnosis between March 30, 2020, and April 27, 2020 (n = 41 with wearable data). Compared with self-reported diagnosed flu cases from the same time frame (n = 426, 85 with wearable data) or pre-pandemic (n = 6,270, 1,265 with wearable data), COVID-19 patients reported a distinct symptom constellation that lasted longer (median of 12 versus 9 and 7 days, respectively) and peaked later after illness onset. Wearable data showed significant changes in daily steps and prevalence of anomalous resting heart rate measurements, of similar magnitudes for both the flu and COVID-19 cohorts. Our findings highlight the need to include flu comparator arms when evaluating PGHD applications aimed to be highly specific for COVID-19.

Highlights

The emergence of the novel SARS-CoV-2 (COVID-19) pandemic necessitates an understanding of symptom prevalence and progression among individuals with COVID-19, as well as how COVID-19 symptoms compare with those of other infectious diseases
Our contributions are 2-fold: first, we examine the presentation of COVID-19 symptoms outside of strictly clinical settings both in terms of constellation and time course, and contextualize them with comparisons with seasonal influenza; second, by analyzing wearable data around symptoms onset we show that physiological signals, such as resting heart rate (RHR) change significantly near symptom onset, as do physical activity measures, such as step counts, these changes appear to be similar in timing and magnitude across influenza-like illnesses (ILI) and COVID-19 cohorts
Data Collection and Cohort Definitions We compare a cohort of self-reported diagnosed COVID-19 cases (n = 230) to two groups of diagnosed flu cases: nonCOVID-19 flu cases (n = 426), which occurred in the same time frame as the COVID-19 cases, and pre-COVID-19 flu (n = 6,270), which occurred earlier in the 2019–2020 flu season before the outbreak of COVID-19

Summary

Introduction

The emergence of the novel SARS-CoV-2 (COVID-19) pandemic necessitates an understanding of symptom prevalence and progression among individuals with COVID-19, as well as how COVID-19 symptoms compare with those of other infectious diseases. In addition to self-report, data from commercial sensors may be used for large-scale surveillance of influenza-like illnesses (ILI), given that resting heart rate (RHR)[6,7,8,9,10] and temperature[11] change in the presence of an infection. A hotspot detection system, including smart thermometers and internet searches, has been shown to provide accurate early-warning indicators of increasing or decreasing state-level US COVID-19.12 Syndromic surveillance based on symptom self-report has recently been shown to scale to tens of thousands of responses per day,[5] and wearables sensors, being worn by one in five Americans,[13] could further increase the volume of daily feeds of person-generated health data (PGHD) used at the aggregate level for syndromic surveillance and hotspot detection

Results

Discussion

Conclusion