Abstract
BACKGROUND: Virtual wards are an NHS priority, designed to deliver acute care, monitoring and treatment to people at home, providing an alternative to hospital admissions or facilitating earlier hospital discharge. The aim of this study was to understand the clinical acuity, care pathways and outcomes of people admitted to virtual wards in Bristol, North Somerset and South Gloucestershire (BNSSG) who were remotely monitored. METHODS: A retrospective observational cohort study of all remotely monitored patients aged 16 + years admitted to a virtual ward in BNSSG between October 2023-February 2025. Clinical observations (respiratory rate, oxygen saturations, systolic blood pressure, pulse rate and temperature) were collected, and National Early Warning Scores (NEWS2) were calculated. The area under the curve (AUC) of NEWS2 to predict hospital (re)admissions is presented. RESULTS: 2,533 admissions across five care pathways were included: respiratory (41%), frailty (18%), outpatient parenteral antimicrobial therapy (19%), heart failure (7%) and general (18%; pathways not mutually exclusive). Median virtual ward length of stay was 10 days (interquartile range 6-14). During the study period, 177 (7%) virtual ward admissions resulted in a hospital (re)admission and 9 (<1%) died. First NEWS2 = 0-2 in 1,651/2,479 (67%) admissions, NEWS2 = 3-4 in 569 (23%), NEWS2 = 5-6 in 204 (8%) and NEWS = 7+ in 55 (2%) admissions. Maximum NEWS2 was reasonably low during most admissions, with 800/2,479 (32%) NEWS2 = 0-2, 887 (36%) NEWS2 = 3-4, 568 (23%) NEWS2 = 5-6, and 224 (9%) NEWS2 = 7 +. The clinical acuity of most patients remained stable, with 964/2,331 (41%) deteriorating by 1-2 points and 696 (30%) not deteriorating at all. Both first NEWS2 and maximum NEWS2 had poor ability to predict hospital (re)admission (AUC 0.55 [95% CI 0.51-0.60] and 0.55 [95% CI 0.50-0.59], respectively). CONCLUSION: Most remotely monitored patients had low clinical acuity on admission to virtual wards, however 10% had high clinical acuity with a NEWS2 value of at least 5. The distribution of NEWS2 on admission to virtual wards was very similar to the distribution of NEWS2 on admission to acute hospitals, as identified in the 2022 Society for Acute Medicine Benchmarking Audit. NEWS2 had poor predictive accuracy in this setting. However, hospital (re)admission rates were low (7%) so this should be interpreted with caution.