ABSTRACT
The transmission risk of SARS-CoV-2 within hospitals can exceed that in the general community because of more frequent close proximity interactions (CPIs). Heterogeneity of risk across wards is still poorly described. We measured CPIs in 15 clinical wards across three hospitals using wearable sensors over 36 hours in spring 2020. This data was combined with a transmission model to estimate and compare transmission risks across wards. We found a four-fold range of epidemic risk between wards, with patients frequently presenting high risk to patients and healthcare workers (HCWs). Using a simulation study, we then assessed the potential impact on global risk of targeting individuals for prevention based on their contact patterns. We found that targeting individuals with the highest cumulative contact hours was most impactful. This study reveals patterns of interactions between individuals in hospital during a pandemic and opens new routes for research into airborne nosocomial risk. One Sentence SummaryWe measured contacts between staff, patients and visitors in 15 hospital wards, and used models to predict epidemic risk and evaluate interventions.
ABSTRACT
IntroductionA new respiratory virus, SARS-CoV-2, has emerged and spread worldwide since late 2019. This study aims at analyzing clinical presentation on admission and the determinants associated with direct admission or transfer to intensive care units (ICUs) in hospitalized COVID-19 patients. Patients and MethodsIn this prospective hospital-based study, socio-demographic, clinical and biological characteristics, on admission, of adult COVID-19 hospitalized patients were prospectively collected and analyzed. The outcome was admission/transfer to intensive care units compared with total hospital stay in medical wards according to patient characteristics. ResultsOf the 412 patients included, 325 were discharged and 87 died in hospital. Multivariable regression showed increasing odds of admission/transfer to ICUs with male gender (OR, 1.99 [95%CI, 1.07-3.73]), temperature (OR, 1.37 [95% CI, 1.01-1.88] per degree Celsius increase), abnormal lung auscultation on admission (OR, 2.62 [95% CI, 1.40-4.90]), elevated level of CRP (OR, 6.96 [95% CI, 1.45-33.35 for CRP>100mg/L vs CRP<10mg/L). Increased time was observed between symptom onset and hospital admission (OR, 4.82 [95% CI, 1.61-14.43] for time >10 days vs time <3 days) and monocytopenia (OR, 2.49 [95% CI, 1.29-4.82]). Monocytosis was associated with lower risk of admission/transfer to ICUs (OR, 0.25 [95% CI, 0.05-1.13]). ConclusionsClinical and biological features on admission and time until admission were associated with admission to ICUs. Signs to predict worsening on admission could be partially associated with the time until admission. This finding reinforces the need for appropriate guidelines to manage COVID-19 patients in this time window.
