National Early Warning Scores Following Emergency Hospital Transfer: Implications for Care Home Residents

Objective

home residents. NEWS2 may have a role as an adjunct to acute care decision making for hospitalized residents. Ó 2023 Published by Elsevier Inc. on behalf of AMDA e The Society for Post-Acute and Long-Term Care Medicine. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/ 4.0/).
Care homes deliver care for residents living with complex care needs, [1][2][3][4] representing a population with high rates of emergency transfer to hospital. 5 Care home is an umbrella term for long-term care facilities with and without on-site nursing staff, termed nursing homes and residential homes respectively in the United Kingdom.
Older adult residents often do not display overt signs of acute (short-duration) illness, 6 otherwise termed deterioration. The trajectory of deterioration among care home residents may be unpredictable, 7 posing challenges for health care teams in care home and hospital settings. The National Early Warning Score (NEWS2) is a "track and trigger" system used across UK hospitals to support clinical judgment in identifying patients at risk of further deterioration, 8 in order to facilitate prompt and appropriate clinical responses. NEWS2 requires measurement of 6 parameters: temperature, pulse, systolic blood pressure, respiratory rate, oxygen saturation, and consciousness. 8 The overall NEWS2 triggers a response, ranging from repeating observations within a specific time period, to initiating an emergency medical response. 8 NEWS2 is advocated for use across the UK health care system, 8 and is increasingly being adopted in care homes. 9,10 The ability of NEWS2 to predict further deterioration and adverse health outcomes, such as death or critical care admission, is well-evidenced for hospital patients overall, 8,11 but not specifically for care home residents. The evidence to support NEWS2 use in care homes is sparse. 9,12 The value of NEWS2 may be dependent on the population to which it is applied 13,14 and concerns have been expressed that it may be less applicable to care home residents because of their age, frailty, and multimorbidity, 12,15 all of which may influence NEWS2 and its ability to identify residents at risk of adverse outcomes.
This study aimed to explore the ability of NEWS2, on emergency admission to hospital, to identify care home residents at risk of admission and discharge within 7 days, prolonged hospitalization (>7 days), and death (within 7 days).

Methods
The sample of care home residents was drawn from the Society for Acute Medicine Benchmarking Audit (SAMBA), [16][17][18][19] a national audit (United Kingdom) of acute medical care. Participation is open to all hospitals receiving acutely unwell (nonelective, adult) medical patients, excluding nonacute and community hospitals.
Two waves of SAMBA data collection, each conducted over a 24hour period, were analyzed: June 27, 2019, 16,19 and January 30, 2020. 17 Patients not living in care homes and those younger than 60 years were excluded. This defined the study population as older adults living in care homes who were admitted to hospital for acute care, either via the emergency department or directly to acute medical units.
-Same-day discharge (no overnight stay). -Hospital admission followed by discharge within 7 days. -Ongoing inpatient care at 7 days (ward-level/intensive care or readmission). -Died within 7 days of admission. STATA 15 was used to conduct the Kruskal-Wallis test to investigate the null hypothesis of no difference in median NEWS2 across outcome groups. Pearson's c 2 was used to test the null hypothesis of no association between NEWS 2 category and resident outcome. Multinomial logistic regression modeled the relationship between NEWS2 category and outcome, using low NEWS2 and same-day discharge as reference categories, and adjusted for age, gender, and SAMBA wave. This dataset did not include contextual information about resident comorbidity, dependency, or frailty.

Results
Data from 676 acute hospital admissions across 160 UK hospital sites across 2 waves of SAMBA were analyzed. It is possible, but improbable, that this does not represent 676 unique residents; for the same resident to be represented twice, this would require the same individual to be admitted to hospital in the two 24-hour periods of data collection 6 months apart. Approximately 70% (486 of 676) of residents were older than 80; 70% were women.
The median NEWS2 was 1 for same-day discharge, 2 for discharge within 7 days and admission >7 days, and 6 for residents who died. Overall, the differences observed between the median NEWS2 across different outcome groups are significant (P < .001).
In summary, the risk of hospitalization for up to 7 days, prolonged hospital admission, and death (compared with same-day discharge) is higher for residents with intermediate, high, or critically high NEWS2 (compared with the low NEWS2 category). Although the risk of these outcomes increases with progressively high NEWS2 category, statistically significant differences are observed only when comparing high and critically high NEWS2 (but not intermediate scores) with low NEWS2 categories.

Discussion
Higher NEWS2 measurements on emergency presentation to hospital are associated with an increased risk of admission for up to 7 days, prolonged hospitalization (>7 days), and mortality for care home residents. The main differences are observed for residents with high or critically high NEWS2 (with reference to the low NEWS2 category), with no (statistically) significant difference for intermediate readings. The findings suggest that, in conjunction with clinical judgment, the NEWS2 on arrival to hospital following emergency conveyance has a role for hospital teams in identifying residents who are at highest risk of further deterioration. This is especially pertinent for residents with the highest NEWS2 readings (5), who are the most likely to experience adverse health outcomes, and who require urgent decisions about what level of medical intervention is required and is consistent with their care preferences. It is possible that high (5) NEWS2 may help clinical teams to identify residents on arrival to hospital who may be experiencing an end-of-life event, to facilitate discussion with families and carers about treatment intensity and palliative care decisions. Our findings do not suggest that NEWS2 should replace clinical judgment or override residents' wishes, which are paramount.

Comparison With Other Work
Care home residents in this study have higher mortality rates across all NEWS2 categories compared with the total population in the corresponding SAMBA cohorts, 1.7% in 2019 19 and 2.3% in 2020. 17 This is likely to reflect the high levels of frailty and multimorbidity 1,3 in the care home population, leading to susceptibility to adverse outcomes. It is also possible, because frailty is associated with blunted homeostatic response, that residents are less likely to manifest physiological derangement during acute illness, meaning that even those with lower NEWS2 are at higher risk of deterioration than people with similar scores who do not live in a care home and are not as frail. This hypothesis would require empirical testing.
NEWS2 is increasingly being adopted in care homes across the United Kingdom. 9,10 Previously published work demonstrated that only 7% of residents in care homes had a high or critically high NEWS2 at baseline, rising to 18% if care home staff were concerned about resident deterioration. 9 The cohort of residents in this study have a higher proportion (28%) of high/critically high scores on admission to hospital. This indicates higher levels of physiological derangement in residents admitted to hospital than observed in care homes, and is an expected finding for residents requiring acute hospital care. One of the reasons that NEWS2 may perform differently in care home settings is because the prevalence of severe acute illness would be substantially lower than in hospital settings. 12 This is one of the reasons that specific validation in the care home setting is required.
The risk of mortality (23%) in this study was significantly higher for residents with critically high NEWS2 (7), compared with previously reported (30-day) mortality rates for adult patients across the age spectrum (13%). 14 This suggests that the highest NEWS2 readings are associated with a particularly high risk of mortality for care home residents, compared with the general population.

Strengths and Limitations
To our knowledge, this is the first study to explore the association between NEWS2 on arrival at hospital and health outcomes in a large, nationally representative dataset. As participation in SAMBA is voluntary, there may be differences between participating and nonparticipating hospitals, although the size of the hospitals that participate is comparable to acute hospital services nationally, 20 and covers urban and rural locations across the United Kingdom. 17 It is recognized that NEWS2 trajectory is particularly important, 21 but neither subsequent nor baseline (when not acutely unwell) NEWS2 were available, with outcomes measured at 7 days in this study. It is important to acknowledge that NEWS2 is designed to identify people at imminent risk of deterioration, and loses discriminatory value over longer time periods. 22 The reference NEWS2 group was low and the reference outcome category was same-day discharge. *Denotes statistical significance at the .05 level.  (13) 104 (16) In this study, we had no access to contextual information such as resident preferences/frailty, community outcomes, or primary diagnosis. Frailty may be a more important variable than age when it comes to understanding the appropriateness of NEWS2 in vulnerable groups of older adults.
Estimates of relative risk for the least commonly observed outcomes (such as death) within the smallest NEWS2 categories (such as critically high), reached statistical significance, but are imprecise.

Implications for Future Research
Further research is required to evaluate the performance of the NEWS2 in care home residents, a population characterized by frailty and complex care needs, 1 to ascertain whether it enhances the delivery of resident-centered care.
The use of NEWS2 in care homes is becoming widespread despite a limited evidence base. 9,12,23 Further research is required to establish the association between care home NEWS2 measures and (both primary and secondary care) health outcomes.

Conclusion and Implications
Higher NEWS2 measurements on emergency presentation to hospital are associated with an increased risk of hospitalization for up to 7 days, prolonged admission, and mortality for care home residents. NEWS2 may have a role as an adjunct to acute care decision making for hospitalized residents.