Wastewater SARS-CoV-2 and COVID-19 Hospital Admission and Mortality - A Controlled Experimental Study

After the COVID-19 pandemic, wastewater monitoring is increasingly used for infectious disease surveillance. Using the data from a controlled experimental hospital setting, this paper examines the association wastewater SARS-CoV-2 with COVID-19 hospital admission and mortality, and whether this association varies by patients' characteristics. Weekly wastewater samples were collected from the University of Miami (UM) hospitals where COVID-19 patients were admitted from February 2020 to October 2022, and SARS-CoV-2 was quantified using qPCR. Data on hospital admissions and their mortality and demographic characteristics and comorbidities were acquired from the UM hospitals. Using factor analysis and hierarchical clustering, patients were stratified into four clusters. Frist, we examined cross-correlations between time-lagged COVID-19 hospital admission and mortality, and time-lagged SARS-CoV-2 to identify appropriate time-lags. Second, we modelled daily hospital COVID-19 cases and mortality with respect to time-lagged SARS-CoV-2, vaccine status and time-lagged COVID-19 hospital cases (as proxy of the risk factor for the transmission of the disease for each cluster separately and for all clusters together. 1,856 COVID-19 patients were admitted in the UM hospitals during the study period and 347 (18.7%) of them died. In cluster 4 that represented patients with preexisting chronic health conditions and intubation, the fatality rate was 59%. COVID-19 hospital admission showed strong (temporal) autocorrelation, suggesting that the preexisting cases can indicate the transmission rate of infection. Our analysis suggests that a 1% increase in SARS-CoV-2 was associated with a 0.28% increase in COVID-19 related hospital admission (β ∼ 0.275; 95 % CI = 0.18 to 0.37; p < 0.01). Both a week lagged auto-regressive COVID-19 cases and SARS-CoV-2 in wastewater together explained 89% of the total variation in hospital admission due to COVID-19. Among four clusters, the second cluster of minority communities showed the strongest association between time-lagged SARS-CoV-2 in wastewater and hospital admissions due to COVID-19 followed by cluster 1 of adult patients with low prevalence of preexisting health conditions. However, time-lagged wastewater SARS-CoV-2 did not show any significant association with COVID-19 hospital admission for patients with the pre-existing health conditions. A week lagged wastewater SARS-CoV-2 did not show any significant association with COVID-19 mortality. Our results indicate that the association between time-lagged wastewater SARS-CoV-2 and COVID-19 hospital admission varied by patients' characteristics, suggesting variations in SARS-CoV-2 shedding by patients' characteristics. These findings warrant to incorporate patient-specific demographic characteristics and comorbidities in modelling infectious diseases surveillance using wastewater monitoring of the infectious agents.