Measurement of SARS-CoV-2 RNA in wastewater tracks community infection dynamics

Abstract

We measured severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA concentrations in primary sewage sludge in the New Haven, Connecticut, USA, metropolitan area during the Coronavirus Disease 2019 (COVID-19) outbreak in Spring 2020. SARS-CoV-2 RNA was detected throughout the more than 10-week study and, when adjusted for time lags, tracked the rise and fall of cases seen in SARS-CoV-2 clinical test results and local COVID-19 hospital admissions. Relative to these indicators, SARS-CoV-2 RNA concentrations in sludge were 0-2 d ahead of SARS-CoV-2 positive test results by date of specimen collection, 0-2 d ahead of the percentage of positive tests by date of specimen collection, 1-4 d ahead of local hospital admissions and 6-8 d ahead of SARS-CoV-2 positive test results by reporting date. Our data show the utility of viral RNA monitoring in municipal wastewater for SARS-CoV-2 infection surveillance at a population-wide level. In communities facing a delay between specimen collection and the reporting of test results, immediate wastewater results can provide considerable advance notice of infection dynamics.

Extended Data Fig. 1 |. Estimated daily distributed lag parameters describing the association between viral RNA in sludge and positive COVID-19 tests, based on date of report to Ct Department of Health.

The top row figures are adjusted for testing volume, bottom row figures are unadjusted for testing volume. a, daily lags of sludge data at longer time lags (0 to 4 days in the past) best correlate with the time series of reported positive tests (adjusted). b, cumulative relationship between viral RNA in sludge and number of positive tests (adjusted). c, daily lags of sludge data at longer time lags (0 to 4 days in the past) best correlate with the time series of reported cases (not adjusted for number of tests). d, cumulative relationship between viral RNA in sludge and number of positive tests (not adjusted). The posterior means at the center of each data point and 90% credible intervals for error bars are displayed. For each lag, n=73 daily virus RNA samples, n=75 daily reported adjusted positive test values, n=75 daily reported unadjusted positive test values.

Fig. 1 |. Replicate RNA extraction and analyses for SARS-CoV-2 RNA.

a, Comparison of SARS-CoV-2 RNA concentration between two replicates (Rep 1 and Rep 2) using the N1 primer set. b, Comparison of SARS-CoV-2 RNA concentration between two replicates using the N2 primer set.

Fig. 2 |. Sludge SARS-CoV-2 RNA concentration time course and other COVID-19 outbreak indicators on linear (left) and log (right) scales.

All data represent the cities of New Haven, Hamden, East Haven and Woodbridge, Connecticut, which are served by the ESWPAF. The blue vertical dashed lines indicate the first week of analysis, March 19–25, 2020. a,b, Number of positive SARS-CoV-2 test results, reported by date of specimen collection. c,d, Percentage of positive SARS-CoV-2 test results, reported by date of specimen collection. e,f, Number of COVID-19 admissions to Yale New Haven Hospital for residents of the four cities. g,h, Number of positive SARS-CoV-2 test results by public reporting date. i,j, Primary sludge SARS-CoV-2 RNA concentration (virus RNA gene copies per ml of sludge).

Fig. 3 |. Estimated daily distributed lag parameters describing the association between viral RNA in sludge and COVID-19 epidemiological parameters.

a, Daily lags of 0–2 d and leads of 1 d are associated with the number of positive tests based on specimen collection date. b, Cumulative relationship between viral RNA in sludge and the number of positive tests based on specimen collection date. c, Daily lags of 0–2 d and leads of 1 d are associated with the percentage of positive tests based on specimen collection date. d, Cumulative relationship between viral RNA in sludge and the percentage of positive tests based on specimen collection date. e, Daily lags of 1–4 d are associated with hospitalization. f, Cumulative relationship between viral RNA in sludge and hospital admissions. g, Daily lags of sludge virus RNA data at longer time lags (6–8 d in the past) best correlate with the time series of publicly reported positive tests. h, Cumulative beta relationship between viral RNA in sludge and reported number of positive tests. Posterior means at the center of each data point and 90% credible intervals for error bars are displayed. For each lag, n = 75 daily values for positive tests by date of specimen collection (a,b), 75 daily values for percentage of positive tests by date of specimen collection (c,d), 75 daily values for hospital admission (e,f) and 75 daily values for publicly reported positive tests (g,h).