Search

Background To address the hospital bed demand for Delta and Omicron surges in Singapore, the National University Health System (NUHS) developed a COVID Virtual Ward to relieve bed pressures on its three acute hospitals—National University Hospital, Ng Teng Fong General Hospital, Alexandra Hospital. To serve a multilingual population, the COVID Virtual Ward featuring protocolized teleconsultation of high-risk patients, use of a vital signs chatbot, supplemented by home visits where necessary. This study aims to evaluate the safety, outcomes and utilisation of the Virtual Ward as a scalable response to COVID-19 surges. Methods This is a retrospective cohort study of all patients admitted to the COVID Virtual Ward between 23 September to 9 November 2021. Patients were defined as “early discharge” if they were referred from inpatient COVID-19 wards and “admission avoidance” if they were referred directly from primary care or emergency services. Patient demographics, utilisation measures and clinical outcomes were extracted from the electronic health record system. The primary outcomes were escalation to hospital and mortality. Use of the vital signs chatbot was evaluated by examining compliance levels, need for automated reminders and alerts triggered. Patient experience was evaluated using data extracted from a quality improvement feedback form. Results 238 patients were admitted to the COVID Virtual Ward from 23 September to 9 November, of whom 42% were male, 67.6% of Chinese ethnicity. 43.7% were over the age of 70, 20.5% were immunocompromised, and 36.6% were not fully vaccinated. 17.2% of patients were escalated to hospital and 2.1% of patients died. Patients who were escalated to hospital were more likely to be immunocompromised or to have a higher ISARIC 4C-Mortality Score. There were no missed deteriorations. All patients received teleconsults (median of 5 teleconsults per patient, IQR 3–7). 21.4% of patients received home visits. 77.7% of patients engaged with the vital signs chatbot, with a compliance rate of 84%. All patients would recommend the programme to others in their situation. Conclusions Virtual Wards are a scalable, safe and patient-centered strategy to care for high risk COVID-19 patients at home. Trial Registration NA.

Illness severity scores are regularly employed for quality improvement and benchmarking in the intensive care unit, but poor generalization performance, particularly with respect to probability calibration, has limited their use for decision support. These models tend to perform worse in patients at a high risk for mortality. We hypothesized that a sequential modeling approach wherein an initial regression model assigns risk and all patients deemed high risk then have their risk quantified by a second, high-risk-specific, regression model would result in a model with superior calibration across the risk spectrum. We compared this approach to a logistic regression model and a sophisticated machine learning approach, the gradient boosting machine. The sequential approach did not have an effect on the receiver operating characteristic curve or the precision-recall curve but resulted in improved reliability curves. The gradient boosting machine achieved a small improvement in discrimination performance and was similarly calibrated to the sequential models.

OBJECTIVES: Although one third or more of critically ill patients in the United States are obese, obesity is not incorporated as a contributing factor in any of the commonly used severity of illness scores. We hypothesize that selected severity of illness scores would perform differently if body mass index categorization was incorporated and that the performance of these score models would improve after consideration of body mass index as an additional model feature. DESIGN: Retrospective cohort analysis from a multicenter ICU database which contains deidentified data for more than 200,000 ICU admissions from 208 distinct ICUs across the United States between 2014 and 2015. SETTING: First ICU admission of patients with documented height and weight. PATIENTS: One-hundred eight-thousand four-hundred two patients from 189 different ICUs across United States were included in the analyses, of whom 4,661 (4%) were classified as underweight, 32,134 (30%) as normal weight, 32,278 (30%) as overweight, 30,259 (28%) as obese, and 9,070 (8%) as morbidly obese.None. MEASUREMENTS AND MAIN RESULTS: To assess the effect of adding body mass index as a risk adjustment element to the Acute Physiology and Chronic Health Evaluation IV and Oxford Acute Severity of Illness scoring systems, we examined the impact of this addition on both discrimination and calibration. We performed three assessments based upon 1) the original scoring systems, 2) a recalibrated version of the systems, and 3) a recalibrated version incorporating body mass index as a covariate. We also performed a subgroup analysis in groups defined using World Health Organization guidelines for obesity. Incorporating body mass index into the models provided a minor improvement in both discrimination and calibration. In a subgroup analysis, model discrimination was higher in groups with higher body mass index, but calibration worsened. CONCLUSIONS: The performance of ICU prognostic models utilizing body mass index category as a scoring element was inconsistent across body mass index categories. Overall, adding body mass index as a risk adjustment variable led only to a minor improvement in scoring system performance.

Introduction: Hospital-at-Home (HaH) programmes are well-established in Australia, Europe, and the United States. However, there is limited experience in Asia, where the hospital is traditionally seen as a safe and trusted space for healing. This cross-sectional study aimed to explore attitudes and perceptions among patients and caregivers in Singapore toward this care model.Methods: A quantitative study design was adopted to collect data among patients and their caregivers from medical wards within two acute hospitals in Singapore. Using a series of closed-ended and open-ended questions, the investigator-administered survey aimed to explore barriers and facilitators determining patients' and caregivers' responses. The study questionnaire was pretested and validated. Data were summarised using descriptive statistics, and logistic regression was performed to determine key factors influencing patients' decisions to enrol in such programmes.Results: Survey responses were collected from 120 participants (101 patients, 19 caregivers; response rate: 76%), of which 87 respondents (72.5%) expressed willingness to try HaH if offered. Many respondents valued non-quantifiable programme benefits, including perceived gains in quality of life. Among them, reasons cited for acceptance included preference for the comfort of their home environment, presence of family members, and confidence toward remote monitoring modalities. Among respondents who were unwilling to accept HaH, a common reason indicated was stronger confidence toward hospital care.Discussion: Most patients surveyed were open to having acute care delivered in their home environment, and concerns expressed may largely be addressed by operational considerations. The findings provide useful insights toward the planning of HaH programmes in Singapore.

ObjectivesThe COVID-19 outbreak in Singapore has largely centred around migrant worker dormitories, comprising over 90% of all cases in the country. Dormitories are home to a culturally and linguistically distinct, low-income population, without on-site healthcare after-hours. The primary objective of this study was to assess the engagement and utilisation of a simple, low-cost, accessible, mobile health solution for remote self-reporting of vital parameters in dormitory residents with COVID-19.DesignRetrospective review of medical care.SettingTwo large migrant worker dormitories with a combined population of 31 546.ParticipantsAll COVID-19-affected residents housed in dormitories during the study period.InterventionAll residents were taught to use a chat assistant to self-report their temperature, heart rate and oxygen saturations. Results flowed into a dashboard, which alerted clinicians of abnormal results.OutcomesThe primary outcome measure was engagement rate. This was derived from the total number of residents who registered on the platform over the total number of COVID-19-affected residents in the dormitories during the study period. Secondary outcome measures included outcomes of the alerts and subsequent escalations of care.Results800 of the 931 COVID-19-affected residents (85.9%) engaged with the platform to log a total of 12 511 discrete episodes of vital signs. Among 372 abnormal readings, 96 teleconsultations were initiated, of which 7 (1.8%) were escalated to emergency services and 18 (4.9%) were triaged to earlier physical medical review on-site.ConclusionsA chat-assistant-based self-reporting platform is an effective and safe community-based intervention to monitor marginalised populations with distinct cultural and linguistic backgrounds, living communally and affected by COVID-19. Lessons learnt from this approach may be applied to develop safe and cost-effective telemedicine solutions across similar settings.

Objective: Severity of illness scores rest on the assumption that patients have normal physiologic values at baseline and that patients with similar severity of illness scores have the same degree of deviation from their usual state. Prior studies have reported differences in baseline physiology, including laboratory markers, between obese and normal weight individuals, but these differences have not been analyzed in the ICU. We compared deviation from baseline of pertinent ICU laboratory test results between obese and normal weight patients, adjusted for the severity of illness. Design: Retrospective cohort study in a large ICU database. Setting: Tertiary teaching hospital. Patients: Obese and normal weight patients who had laboratory results documented between 3 days and 1 year prior to hospital admission. Interventions: None. Measurements and Main Results: Seven hundred sixty-nine normal weight patients were compared with 1,258 obese patients. After adjusting for the severity of illness score, age, comorbidity index, baseline laboratory result, and ICU type, the following deviations were found to be statistically significant: WBC 0.80 (95% CI, 0.27–1.33) × 109/L; p = 0.003; log (blood urea nitrogen) 0.01 (95% CI, 0.00–0.02); p = 0.014; log (creatinine) 0.03 (95% CI, 0.02–0.05), p < 0.001; with all deviations higher in obese patients. A logistic regression analysis suggested that after adjusting for age and severity of illness at least one of these deviations had a statistically significant effect on hospital mortality (p = 0.009). Conclusions: Among patients with the same severity of illness score, we detected clinically small but significant deviations in WBC, creatinine, and blood urea nitrogen from baseline in obese compared with normal weight patients. These small deviations are likely to be increasingly important as bigger data are analyzed in increasingly precise ways. Recognition of the extent to which all critically ill patients may deviate from their own baseline may improve the objectivity, precision, and generalizability of ICU mortality prediction and severity adjustment models.

Objective Severity of illness scores used in critical care for benchmarking, quality assurance and risk stratification have been mainly created in high-income countries. In low and middle-income countries (LMICs), they cannot be widely utilized due to the demand for large amounts of data that may not be available (e.g. laboratory results). We attempt to create a new severity prognostication model using fewer variables that are easier to collect in an LMIC. Setting Two intensive care units, one private and one public, from Sao Paulo, Brazil Patients An ICU for the first time. Interventions None. Measurements and Mains results The dataset from the private ICU was used as a training set for model development to predict in-hospital mortality. Three different machine learning models were applied to five different blocks of candidate variables. The resulting 15 models were then validated on a separate dataset from the public ICU, and discrimination and calibration compared to identify the best model. The best performing model used logistic regression on a small set of 10 variables: highest respiratory rate, lowest systolic blood pressure, highest body temperature and Glasgow Coma Scale during the first hour of ICU admission; age; prior functional capacity; type of ICU admission; source of ICU admission; and length of hospital stay prior to ICU admission. On the validation dataset, our new score, named SEVERITAS, had an area under the receiver operating curve of 0.84 (0.82 – 0.86) and standardized mortality ratio of 1.00 (0.91–1.08). Moreover, SEVERITAS had similar discrimination compared to SAPS-3 and better discrimination than the simplified TropICS and R-MPM. Conclusions Our study proposes a new ICU mortality prediction model using simple logistic regression on a small set of easily collected variables may be better suited than currently available models for use in low and middle-income countries.