ABSTRACT
Background: Accurately predicting patient outcomes in Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) could aid patient management and allocation of healthcare resources. There are a variety of methods which can be used to develop prognostic models, ranging from logistic regression and survival analysis to more complex machine learning algorithms and deep learning. Despite several models having been created for SARS-CoV-2, most of these have been found to be highly susceptible to bias. We aimed to develop and compare two separate predictive models for death during admission with SARS-CoV-2.MethodBetween March 1 - April 24, 2020, 398 patients were identified with laboratory confirmed SARS-CoV-2 in a London teaching hospital. Data from electronic health records were extracted and used to create two predictive models using: 1) a Cox regression model and 2) an artificial neural network (ANN). Model performance profiles were assessed by validation, discrimination, and calibration.Results Both the Cox regression and ANN models achieved high accuracy (83.8%, 95% confidence interval (CI): 73.8 - 91.1 and 90.0%, 95% CI: 81.2 - 95.6, respectively). The area under the receiver operator curve (AUROC) for the ANN (92.6%, 95% CI: 91.1 - 94.1) was significantly greater than that of the Cox regression model (86.9%, 95% CI: 85.7 - 88.2), p=0.0136. Both models achieved acceptable calibration with Brier scores of 0.13 and 0.11 for the Cox model and ANN, respectively. ConclusionWe demonstrate an ANN which is non-inferior to a Cox regression model but with potential for further development such that it can learn as new data becomes available. Deep learning techniques are particularly suited to complex datasets with non-linear solutions, which make them appropriate for use in conditions with a paucity of prior knowledge. Accurate prognostic models for SARS-CoV-2 can provide benefits at the patient, departmental and organisational level.
Subject(s)
Severe Acute Respiratory SyndromeABSTRACT
Background: The outbreak of SARS-CoV-2 (coronavirus, COVID-19), declared a pandemic by the World Health Organisation (WHO) is global health problem with ever-increasing attributed deaths. Vital sign trends are routinely used to monitor patients with changes in these parameters often preceding an adverse event. Wearable sensors can measure vital signs continuously (e.g. heart rate, respiratory rate, temperature) remotely and can be utilised to recognise early clinical deterioration. MethodsWe describe the protocol for a pilot, proof-of-concept, observational study to be conducted in an engineered hotel near London airports, United Kingdom. The study is set to continue for the duration of the pandemic. Individuals arriving to London with mild symptoms suggestive of COVID-19 or returning from high risk areas requiring quarantine, as recommended by Public Health England, or healthcare professionals with symptoms suggestive of COVID-19 unable to isolate at home will be eligible for a wearable patch to be applied for the duration of their stay. Notifications will be generated should deterioration be detected through the sensor and displayed on a central monitoring hub viewed by nursing staff, allowing for trend deterioration to be noted. The primary objective is to determine the feasibility of remote monitoring systems in detecting clinical deterioration for quarantined individuals in a hotel. DiscussionThis trial should prove the feasibility of a rapidly implemented model of healthcare delivery through remote monitoring during a global pandemic at a hotel, acting as an extension to a healthcare trust. Potential benefits would include reducing infection risk of COVID-19 to healthcare staff, with earlier recognition of clinical deterioration through ambulatory, continuous, remote monitoring using a discrete wearable sensor. We hope our results can power future, robust future randomised trials.Trial registrationClinicalTrials.gov Identifier: NCT04337489
Subject(s)
COVID-19ABSTRACT
Background Access to rapid diagnosis is key to the control and management of SARS-CoV-2. Reverse Transcriptase- Polymerase Chain Reaction (RT-PCR) testing usually requires a centralised laboratory and significant infrastructure. We describe the development and diagnostic accuracy assessment of a novel, rapid point-of-care RT-PCR test, the DnaNudge platform CovidNudge test, which requires no laboratory handling or sample pre-processing. Methods Nasopharyngeal swabs are inserted directly into a cartridge which contains all reagents and components required for RT-PCR reactions, including multiple technical replicates of seven SARS-CoV-2 gene targets (rdrp1, rdrp2, e-gene, n-gene, n1, n2 and n3) and human ribonuclease P (RNaseP) as a positive control. Between April and May 2020, swab samples were tested in parallel using the CovidNudge direct-to-cartridge platform and standard laboratory RT-PCR using swabs in viral transport medium. Samples were collected from three groups: self-referred healthcare workers with suspected COVID-19 (Group 1, n=280/386; 73%); patients attending the emergency department with suspected COVID-19 (Group 2, n=15/386; 4%) and hospital inpatient admissions with or without suspected COVID-19 (Group 3, n=91/386; 23%). Results Of 386 paired samples tested across all groups, 67 tested positive on the CovidNudge platform and 71 with standard laboratory RT-PCR. The sensitivity of the test varied by group (Group 1 93% [84-98%], Group 2 100% [48-100%] and Group 3 100% [29-100%], giving an average sensitivity of 94.4% (95% confidence interval 86-98%) and an overall specificity of 100% (95%CI 99-100%; Group 1 100% [98-100%]; Group 2 100% [69-100%] and Group 3 100% [96-100%]). Point of care testing performance was comparable during a period of high (25%) and low (3%) background prevalence. Amplification of the viral nucleocapsid (n1, n2, n3) targets were most sensitive for detection of SARS-CoV2, with the assay able to detect 1x104 viral particles in a single swab. Conclusions The CovidNudge platform offers a sensitive, specific and rapid point of care test for the presence of SARS-CoV-2 without laboratory handling or sample pre-processing. The implementation of such a device could be used to enable rapid decisions for clinical care and testing programs.
Subject(s)
COVID-19ABSTRACT
Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2) has become a global pandemic. Case identification is currently made by real-time polymerase chain reaction (PCR) during the acute phase and largely restricted to healthcare laboratories. Serological assays are emerging but independent validation is urgently required to assess their utility.We evaluated five different point-of-care (POC) SARS-CoV-2 antibody test kits against PCR, finding concordance across the assays (n=15). We subsequently tested 200 patients using the OrientGene COVID-19 IgG/IgM Rapid Test Cassette and find a sensitivity of 74% in the early infection period (day 5-9 post symptom onset), with 100% sensitivity not seen until day 13. Specificity was 96%, but in validating the serological tests uncovered potential false-negatives from PCR testing late-presenting cases. A positive predictive value (PPV) of 37% in the general population precludes any use for general screening. Where a case definition is applied however, the PPV is substantially improved (95·4%), supporting use of serology testing in carefully targeted populations. Larger studies in specific patient cohorts, including those with mild infection are urgently required to inform on the applicability of POC serological assays to help control the spread of SARS-CoV-2 and improve case finding of patients that may experience late complications.
Subject(s)
COVID-19 , Severe Acute Respiratory Syndrome , InfectionsABSTRACT
Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2) has become a global pandemic, causing over 3,600,000 reported cases and 250,000 deaths worldwide.1 Case identification has predominantly been made by real-time polymerase chain reaction (PCR) during the acute phase and largely restricted to healthcare laboratories. Serological assays are emerging but independent validation is urgently required to assess their utility. Where a plurality of point-of-care (POC) SARS-CoV-2 antibody test kits have become available, we will therefore aim to evaluate a range of kits against the current available gold-standard diagnostic test of PCR in an initial, exploratory study. We will then proceed to carry out testing with 200 hospital inpatients using the OrientGene COVID-19 IgG/IgM Rapid Test Cassette against PCR in order to undergo a preliminary evaluation of POC serological test performance characteristics within a hospital inpatient cohort.