Abstract Background Accurate methods of identifying patients with COVID-19 who are at high risk of poor outcomes has become especially important with the advent of limited-availability therapies such as monoclonal antibodies. Here we describe development and validation of a simple but accurate scoring tool to classify risk of hospitalization and mortality. Methods All consecutive patients testing positive for SARS-CoV-2 from March 25-October 1, 2020 within the Intermountain Healthcare system were included. The cohort was randomly divided into 70% derivation and 30% validation cohorts. A multivariable logistic regression model was fitted for 14-day hospitalization. The optimal model was then adapted to a simple, probabilistic score and applied to the validation cohort and evaluated for prediction of hospitalization and 28-day mortality. Results 22,816 patients were included; mean age was 40 years, 50.1% were female and 44% identified as non-white race or Hispanic/Latinx ethnicity. 6.2% required hospitalization and 0.4% died. Criteria in the simple model included: age (0.5 points per decade); high-risk comorbidities (2 points each): diabetes mellitus, severe immunocompromised status and obesity (body mass index≥30); non-white race/Hispanic or Latinx ethnicity (2 points), and 1 point each for: male sex, dyspnea, hypertension, coronary artery disease, cardiac arrythmia, congestive heart failure, chronic kidney disease, chronic pulmonary disease, chronic liver disease, cerebrovascular disease, and chronic neurologic disease. In the derivation cohort (n=16,030) area under the receiver-operator characteristic curve (AUROC) was 0.82 (95% CI 0.81-0.84) for hospitalization and 0.91 (0.83-0.94) for 28-day mortality; in the validation cohort (n=6,786) AUROC for hospitalization was 0.8 (CI 0.78-0.82) and for mortality 0.8 (CI 0.69-0.9). Conclusion A prediction score based on widely available patient attributes accurately risk stratifies patients with COVID-19 at the time of testing. Applications include patient selection for therapies targeted at preventing disease progression in non-hospitalized patients, including monoclonal antibodies. External validation in independent healthcare environments is needed.
Introduction: The implementation of the Stanford Emergency Critical Care Program (ECCP) was associated with a mortality benefit. We aimed to identify common critical care diagnoses associated with an increase in early Emergency Department (ED) downgrades by ECCP. Methods: This single center, retrospective cohort analysis included adult ED patients with an admission order to the Medical Intensive Care Unit (MICU) or ECCP service within 12 hours of ED arrival. The primary outcome was the proportion of patients downgraded to non-ICU status while in the ED within 6 hours of ICU admission order (ED downgrade < 6h), adjusted by severity of illness. For the 10 most frequent diagnoses, a difference-in-differences (DiD) analysis compared the change in outcomes for patients who received critical care admission orders during ECCP hours vs. non-ECCP hours. Secondary outcomes included the proportion of patients who never occupied an ICU bed, length of stay, and mortality. Results: 1882 patients were included. The top ten critical care diagnoses were respiratory failure, sepsis/septic shock, diabetic ketoacidosis (DKA), altered mental status, gastrointestinal bleed, cardiac arrest, other forms of shock, hyponatremia, airway monitoring, and renal failure/hyperkalemia. Admission during ECCP hours was associated with a severity-adjusted increase in ED downgrade < 6h (DiD 19% [CI: 13% - 25%]), as well as an increase in patients who never occupied an ICU bed (DiD 13% [5.9% - 21%]). A significant increase in ED downgrade < 6h was observed among patients with respiratory failure and sepsis/septic shock (DiD 25.9%, [13.9% - 37.9%], and DiD 13.0%, [1.9% - 25%], respectively). An increase in the proportion of patients who never occupied an ICU bed was observed among the respiratory failure and DKA groups (DiD 18.8%, [2.7% - 35%] and DiD 48.0%, [17% - 78%], respectively). There was no statistically significant difference in outcomes for other diagnoses. Conclusions: Admission during ECCP hours was associated with an increase in early ED downgrade for patients with respiratory failure and sepsis/septic shock, and an increase in proportion of patients who never occupied an ICU bed for respiratory failure and DKA. This work highlights a target population for ED-based interventions focused on ICU resource optimization.
Introduction: Intravenous fluids (IVF) are widely utilized in emergency departments (ED) and Intensive Care Units (ICUs) to aid in resuscitation and the two most common types are lactated Ringer's (LR) and normal saline (NS). In patients presenting with elevated serum lactate levels, there remains a paucity of data on the association between these two fluid types and lactate clearance. Methods: We conducted a secondary analysis of a large hybrid implementation and effectiveness study involving 22 hospitals in Idaho and Utah between November 2018 and February 2020. The intervention involved combining electronic order set modifications and alerts, and sequential clinical education to encourage prescribing of lactated Ringer's. Inclusion criteria for this sub-analysis were ED patients that had an initial lactate of 2 mmol/L or greater, were administered IVFs and had a second lactate drawn between 30 minutes and 24 hours later. We conducted linear regression analysis with covariates of patient demographics, presenting vital signs, symptoms and comorbidities to assess the change in serial lactate levels within and between each implementation period. Results: Our analysis included 2,761 patients, 1,543 in the pre-implementation cohort and 1,218 in the post-implementation cohort. During the pre-implementation period, NS accounted for median [interquartile range (IQR)] 90% (50%-100%) of the IVF prescribed and only 9% (0%-42%) post-implementation. Median (IQR) for LR was 0% (0%-43%) and 89% (50%-99%) for the same respective time periods. A median of 3,000 mL (IQR, 2,000-5,546mL) of IVF were administered and median time between lactate levels was 2.57 hours (1.85-3.68). Pre-implementation the median first measured lactate was 2.8 (IQR, 2.3-3.7) and second median lactate was 3.0 (2.4-4.2). Post-implementation the respective values were 2.8 (2.3-3.7) and 3.0 (2.4-4.4). The difference between the change in rise of lactate levels was observed to not be statistically significant (Mann-Whitney U, p-value=0.761). Conclusions: Following this multifaceted implementation encouraging lactated Ringer's use within a large integrated healthcare system, we observed a significant increase in the proportion of IVF that was LR and no association with lactate clearance.
