Syndromic Detection of Upper Respiratory Infections in Primary Healthcare as a Candidate for COVID-19 Early Warning in Brazil: A Retrospective Ecological Study

Background: Syndromic Surveillance of mild cases of respiratory viruses are a valuable tool for public health, able to warn health authorities of the onset of an outbreak 2 to 4 weeks before more traditional methods such as severe acute respiratory syndrome (SARS) and confirmed cases notifications to health surveillance systems.Methods: We did a retrospective ecological study using secondary data on upper respiratory infections (URI) in primary healthcare (PHC) and SARS in Brazil from 2017 to April 2020. We ran time series analysis to check whether URI in PHC was synchronized or anticipated SARS hospitalizations using Spearman Coefficient.Findings: We identified 589,767,324 patient visits to PHC facilities in 97% Brazilian municipalities, and 24,994,522 of them were coded as URI. In the same period 166,967 SARS hospitalizations were notified. We observed a strong correlation between URI and SARS (0.830), a lower correlation for other acute symptoms in PHC (0.331), and even lower for general PHC encounters (0.228). The correlation between URI and SARS is moderate with a month of advance (0.547). Similar patterns were found in all five Brazilian regions.Interpretation: This is the first report of a study using routinely collected administrative data from PHC covering an entire developing country for early warning of infectious disease outbreak. Our data strongly suggest that the rise in patient consultation in PHC encounters for URI anticipates the increase in SARS hospitalisation and, therefore, is a candidate for outbreak alerts and prediction of SARS cases.Funding: Rockefeller Foundation, JBS SA, FAPESB, CNPq.Declaration of Interest: None.Ethical Approval: The study was approved by the Ethical Review Board of Oswaldo Cruz Foundation - Brasília Regional Office, CAAE 31735320.10000.8027.