Internet search patterns reveal clinical course of COVID-19 disease progression and pandemic spread across 32 countries

Abstract

Background: Coronavirus disease 2019 is an emerging infectious disease.It was first reported in Wuhan, China, and then broke out on a large scale around the world.This study aimed to assess the clinical significance of two different nutritional indices in 245 patients with COVID-19. Methods:In this retrospective single-center study, we finally included 245 consecutive patients who confirmed COVID-19 in Wuhan University Zhongnan Hospital from January 1 to February 29. Cases were classified as either discharged or dead. Demographic, clinical and laboratory datas were registered, two different nutritional indices were calculated: (i)the Controlling nutritional status (CONUT) score;(ii) prognostic nutritional index (PNI). We used univariate and multivariate logistic regression analysis to explore the relationship between nutritional indices and hospital death .Results: 212 of them were discharged and 33 of them died. In-hospital mortality was signifcantly higher in the severe group of PNI than in the moderate and normal groups. It was also significantly worse in the severe-CONUT group than in the moderate-, mild-, and normal-CONUT groups.Multivariate logistic regression analysis showed the CONUT score (odds ratio3.371,95%CI(1.124-10.106), p = 0.030) and PNI(odds ratio 0.721,95% CI(0.581-0.896),P=0.003) were independent predictors of all-cause death at an early stage; Multivariate logistic regression analysis also showed that the severe group of PNI was the independent risk predictor of in-hospital death(odds ratio 24.225, 95% CI (2.147-273.327) , p=0.010).The CONUT score cutoff value was 5.5 (56.00 and 80.81%; AUC 0.753; 95% CI (0.644-0.862) ;respectively).The PNI cutoff value was 40.58 (81.80 and 66.20%; AUC 0.778; 95% CI(0.686-0.809); respectively).We use PNI and the COUNT score to assess malnutrition, which can have a prognosis effect of COVID-19patients. Conclusion:The CONUT score and PNI could be a reliable prognostic marker of all-cause deathin patients with COVID-19. (933-5129) <0·001 PCT,ng/ml 0.0(0.00-0.11) 0 (0.00-0.07) 0 (0.00-0.03) 0.06 (0.00-0.23) 0.43 (0.1-1.16) <0·001 All-cause death 25(16.13%) 1(5.6%) 6(8.3%) 12(20.69%) 6(85.71%) <0·001 COPD, chronic obstructive pulmonary disease; WBC, white blood cells; ALT, alanine aminotransferase;TC,Total cholesterol; HDL, high density lipoprotein PT, prothrombin time;PCT,Procalcitonin; CONUT, Controlling nutritional status; PNI, Prognostic nutritional index; BMI, Body mass index *Between evaluation of clinical indicators and COUNT score, 90 patients lacked records of lymphocyte counts or albumin or total cholesterolModel 1 :The multivariate analysis of PNI associated with in-hospital death adjusted for age, sex, BMI, history of hypertension, history of chronic obstructive pulmonary disease ， history of coronary heart disease ,respiratory rate, alanine transaminase, creatinine, D-dimer, white blood cell count, Lymphocyte, PCT, Model 2 :The multivariate analysis of COUNT score associated with in-hospital death adjusted for age, sex, BMI, history of hypertension, h...

An early warning approach to monitor COVID-19 activity with multiple digital traces in near real time

“…These approaches have exploited information from internet and clinicians’ search engines ( 13 – 19 ), news reports ( 20 – 22 ), crowd-sourced participatory disease surveillance systems ( 23 , 24 ), Twitter microblogs ( 25 , 26 ), electronic health records ( 27 , 28 ), Wikipedia traffic ( 29 ), wearable devices ( 30 ), smartphone-connected thermometers ( 31 ), and travel websites ( 32 ) to estimate disease prevalence in near real time. Several have already been used to track COVID-19 ( 33 , 34 ). These data sources are liable to bias, however; for example, Google Search activity is sensitive to the intensity of news coverage ( 15 , 35 – 37 ).…”

Internet Search Effort on Covid-19 and the Underlying Public Interventions and Epidemiological Status

“…Results derived here studying indicated that fear of, or interest in death was the variable with the highest explanatory power in predicting internet search effort, implying that the rst concern is either to stay alive, or interest and compassion to the individuals who passed way (Du et al 2020, Lu andReis 2021). This interest was higher than any nancial related variable, or personal freedom of movement and entertainment, or every day habits.…”

COVID-19 Symptom-Related Google Searches and Local COVID-19 Incidence in Spain: Correlational Study (Preprint)

“…Higgins et al [17] 18-22 days "Coronavirus Symptoms," "Coronavirus Test," "Fever," "Cough," "Coronavirus," "Runny Nose," "Dry Cough," "Sore Throat," "Chills," and "Shortness of Breath" 32 countries Google Trends Lu and Reis [18]…”