Chronic obstructive pulmonary disease

Original article

Comparison of early warning scores in patients with

COPD exacerbation: DECAF and NEWS score
Carlos Echevarria,‍ ‍ 1,2 John Steer,1,3 Stephen C Bourke1,3

►► Additional material is Abstract

published online only. To view Background  The National Early Warning Score 2 Key messages
please visit the journal online
(http://​dx.​doi.​org/​10.​1136/​(NEWS2) includes two oxygen saturation scales; the
thoraxjnl-​2019-​213470). second adjusts target saturations to 88%–92% for those What is the key question?
with hypercapnic respiratory failure. Using this second ►► Does National Early Warning Score 2 (NEWS2)
1
Newcastle University, scale in all patients with COPD exacerbation (’NEWS2All reduce alert frequency, and offer superior
Newcastle upon Tyne, UK prognostic performance to other versions of
2 ’) would simplify practice, but the impact on alert
Respiratory Medicine, Royal COPD
frequency and prognostic performance is unknown. NEWS and the DECAF (dyspnoea, eosinopenia,
Victoria Infimrary, Newcastle
upon Tyne, UK Admission NEWS2 score has not been compared consolidation, acidaemia, atrial fibrillation)
3
Respiratory Medicine, North with DECAF (dyspnoea, eosinopenia, consolidation, score in 2645 patients admitted with COPD
Tyneside General Hospital, North acidaemia, atrial fibrillation) for inpatient mortality exacerbation who have spirometry-proven
Shields, UK COPD?
prediction.
Methods  NEWS, NEWS2 and NEWS2All COPD and DECAF
Correspondence to What is the bottom line?
Prof Stephen C Bourke, were calculated at admission in 2645 patients with
►► NEWS2 is superior to the original NEWS, but
Respiratory Medicine, North COPD exacerbation attending consecutively to one of
using the second oxygen scale within NEWS2
Tyneside General Hospital, North six UK hospitals, all of whom met spirometry criteria
Shields NE29 8NH, UK; (target oxygen saturations 88%–92%) in all
for COPD. Alert frequency and appropriateness were
​stephen.​bourke@​nhct.​nhs.​uk patients with COPD exacerbation substantially
assessed for all NEWS iterations. Prognostic performance
reduces false alerts without increased risk.
Received 12 April 2019 was compared using the area under the receiver
DECAF offers superior risk-stratification on
Revised 2 July 2019 operating characteristic (AUROC) curve. Missing data
Accepted 2 July 2019 admission, but does not replace early warning
were imputed using multiple imputation.
Published Online First scores to detect subsequent deterioration.
6 August 2019
Findings  Compared with NEWS, NEWS2 reclassified
3.1% patients as not requiring review by a senior Why read on?
clinician (score≥5). NEWS2All COPD reduced alerts by ►► Our data support monitoring all patients
12.6%, or 16.1% if scoring for injudicious use of oxygen with COPD exacerbation using the second
was exempted. Mortality was low in reclassified patients, NEWS2 oxygen scale, and this advice should
with no patients dying the same day as being identified override the current Royal College of Physicians
as low risk. NEWS2All COPD was a better prognostic score recommendation that the second scale should
than NEWS (AUROC 0.72 vs 0.65, p<0.001), with only be used in hypercapnic patients and at the
similar performance to NEWS2 (AUROC 0.72 vs 0.70, discretion of a clinician.
p=0.090). DECAF was superior to all scores (validation
cohort AUROC 0.82) and offered a more clinically useful
range of risk stratification (DECAF=1.2%–25.5%;
NEWS2=3.5%–15.4%). Although NEWS is widely used in the UK,
Conclusion  NEWS2All COPD safely reduces the alert two key limitations have been highlighted. First,
frequency compared with NEWS2. DECAF offers superior patients with COPD and chronic hypoxaemia raise
prognostic performance to guide clinical decision-making false alerts, which may lead to alert fatigue and
on admission, but does not replace repeated measures of complacency. Second, in severe COPD exacerba-
NEWS2 during hospitalisation to detect the deteriorating tions (ECOPD), excess oxygen is associated with
patient. increased need for ventilation and mortality,2–4 but
this risk is not identified by NEWS,5 6 which may
►► http://​dx.​doi.​org/​10.​1136/​
encourage unsafe oxygen prescribing to achieve a
thoraxjnl-​2019-​213788 lower score. The recently updated NEWS2 includes
Background a second oxygen saturation scale which is intended
The National Early Warning Score (NEWS) facil- for patients with hypercapnic respiratory failure
itates prompt recognition of clinical deterioration (table 1). New onset of confusion has also been
© Author(s) (or their
in patients hospitalised with a wide range of condi- added to the level of consciousness index.1 The new
employer(s)) 2019. Re-use
permitted under CC BY-NC. No tions.1 It is composed of bedside indices (respiratory oxygen saturation scale encourages the prescrip-
commercial re-use. See rights rate, oxygen saturations, systolic blood pressure, tion and delivery of oxygen to target saturations
and permissions. Published pulse, level of consciousness and temperature) with of 88%–92% in those with hypercapnia,7 while
by BMJ.
clear thresholds commanding the urgency of clin- the original oxygen saturation scale is retained
To cite: Echevarria C, Steer J, ical response. Early warning scores, such as NEWS, for patients without hypercapnia. This incorrectly
Bourke SC. Thorax are intended to be scored repeatedly during the assumes that patients with ECOPD without hyper-
2019;74:941–946. inpatient stay to monitor patients for deterioration. capnia are not at risk of harm from excess oxygen;
Echevarria C, et al. Thorax 2019;74:941–946. doi:10.1136/thoraxjnl-2019-213470    941
 Chronic obstructive pulmonary disease
88%–92%, but risk is not avoided as the target can be increased
Table 1  National Early Warning Score 2 (NEWS2) with risk score for
to 94%–98% following confirmation of normocapnia. In unse-
oxygen saturations alone shown for scale 1 (normocapnic patients)
lected patients with a clinical diagnosis of ECOPD, there is
and for scale 2 (hypercapnic patients)
randomised controlled trial (RCT) evidence that target satura-
NEWS2 scale 1 and original NEWS NEWS2 scale 2 tions of 88%–92% improve mortality.6 This favours adopting
(normocapnic patients) (hypercapnic patients)
the second NEWS2 oxygen scale (88%–92%) in all patients with
Oxygen saturations, % Score Oxygen saturations, % Score ECOPD, which would simplify the pathway facilitating imple-
91 or less 3 83 or less 3 mentation, and should greatly reduce the risk of excess oxygen
92–93 2 84–85 2 use.
The DECAF (dyspnoea, eosinopenia, consolidation, acidaemia,
94–95 1 86–87 1
atrial fibrillation) score is a risk stratification tool intended to be
96 or more 0 88–92 0 scored on admission and accurately predicts risk of death.11 12 It
    93 or more on air 0 can be easily calculated at the bedside to guide treatment, such
    93–94 on O2 1 as hospital at home for low-risk patients.13 DECAF and NEWS2
serve separate purposes in clinical practice: the former to risk
    95–96 on O2 2
assess patients accurately at admission and the latter to monitor
    97 or more on O2 3 patients for deterioration throughout admission. However,
Scale 1 is the same as the original NEWS oxygen saturation index and is admission NEWS2 has not been compared with DECAF. The UK
recommended for most patients including patients with COPD without hypercapnic
National Asthma and COPD Audit Programme currently relies
respiratory failure. Scale 2 is recommended for patients with hypercapnic respiratory
failure, whose target saturations are set lower (ie, 88%–92%). on admission NEWS2 to allow case-mix adjustment for mortality
risk; the superiority of DECAF would support its routine use
for this purpose.14 Furthermore, if admission NEWS2 offered
this can lead to CO2 retention5 and is associated with increased similar performance to DECAF in ECOPD, then the principle of
mortality risk.8 parsimony would favour reliance on NEWS2 alone and render
Despite clear guidelines by the British Thoracic Society (BTS) the DECAF score redundant.
promoting target saturations of 88%–92% in most patients with The primary aim of this study was to assess the effect of modi-
COPD,7 this is frequently not achieved in clinical practice.9 10 By fying NEWS2 to use the second oxygen saturation scale in all
assigning risk points to those with excess oxygen, NEWS2 will patients with ECOPD (target saturations of 88%–92%) with
encourage the correct titration of oxygen in patients identified respect to inpatient alert frequency and appropriateness, and
as at risk. Unfortunately, this is limited to patients confirmed mortality prediction. We refer to this adapted version of NEWS2
to have hypercapnic respiratory failure on blood gas analysis as ‘NEWS2All COPD’. Furthermore, we compare the performances
and with clinician approval. Patients with hypercapnia will be at of admission NEWS2All COPD, NEWS2, NEWS and the DECAF
risk until blood gas analysis and clinical review, while those with score.
normocapnia will remain at risk throughout. The BTS guide-
lines limit risk by recommending initial target saturations of
Methods
The DECAF derivation and validation cohorts are composed of
2645 consecutive admissions of unique patients with ECOPD
Table 2  Indices that comprise news and DECAF scores, and rates of
missing data in each cohort to six UK hospitals with preadmission obstructive spirometry,
age of 35 or older, and smoking history of 10 or more cigarette
Derivation cohort Validation cohort pack-years (ISRCTN13946813 and ISRCTN29082260).11 12
n=920 n=1725 Collection of the DECAF and NEWS score indices, including
NEWS and NEWS2 confusion, was pre-specified in the original study protocols. The
 Respiratory rate 0.33% 1.8%
first set of NEWS indices was used, either from the accident and
emergency (A+E) department, or the medical admissions unit
 Oxygen saturation 0% 1.8%
if the patient bypassed A+E. Missing data were handled using
 Supplemental oxygen 1.5% 0% multiple imputation as described previously.15 16 NEWS2 was
 Systolic blood pressure 0% 1.7% initially calculated using scale 1 and scale 2 for normocapnic and
 Pulse 0% 1.5% hypercapnic patients, respectively (‘NEWS2’). NEWS2 was then
re-calculated with the assumption that all patients with COPD
 Level of consciousness 0% 1.8%
should have oxygen saturations of 88%–92% (‘NEWS2All COPD’).
 Confusion (NEWS2 only) 0% 2.6% Patients were assigned to standard risk groups for each early
 Temperature 0.54% 2.6% warning score. For NEWS, NEWS2 and NEWS2All COPD, a score
DECAF score of 0 to 4 is low risk, 5 or 6 is moderate risk, and 7 or more is
high risk. A patient with a NEWS of 5 or more should have at
 eMRCD score 0% 0%
least hourly observations and be reviewed by a clinician with
 Eosinopenia 0% 0% competencies in dealing with acute illness, such as a ward doctor.
 Chest X-ray consolidation 0% 0% A score of 7 or more should prompt an emergency review, with
 Acidaemia (pH <7.3) 0%* 0%* continuous monitoring of observations.1 For the DECAF score,
0 to 1 equates to a low in-hospital mortality risk, 2 is moderate
 Atrial fibrillation 0% 0%
risk, and 3 or more is high risk. Alert frequencies (NEWS,
*In patients without an arterial blood gas, it was assumed that none had an
acidaemia of <7.30 provided oxygen saturations were 93% or more while NEWS2, NEWS2All COPD score ≥5) and other proportions were
breathing room air. This was based on data from the derivation cohort: of 118 compared with Fisher’s exact test. The performance of NEWS,
patients meeting these criteria, none had an arterial pH of <7.30. NEWS2, NEWS2All COPD and DECAF for the prediction of
942 Echevarria C, et al. Thorax 2019;74:941–946. doi:10.1136/thoraxjnl-2019-213470
 Chronic obstructive pulmonary disease

Table 3  Numbers of deaths by risk group for NEWS2, NEWS2 88%–92% and NEWS
Low-risk groups (0–4) Moderate-risk groups (5–6) High-risk groups (7+)
All cohorts Died All % Died All % Died All %
 NEWS2All COPD 35 1003 3.5 39 628 6.2 154 1014 15.2
 NEWS2 23 752 3.1 45 671 6.7 160 1222 13.1
 NEWS 25 670 3.7 57 730 7.8 146 1245 11.7
NEWS, NEWS2 and NEWS2 88%–92%: a score of 0 to 4 is low risk, 5 or 6 is moderate risk, and 7 or more is high risk.

inpatient mortality was assessed and compared using the area NEWS2All COPD resulted in a shift of total scores towards the
under the receiver operating characteristic (AUROC) curve.17 lower risk range compared with NEWS2 (figure 2). NEWS2All
Calibration was assessed using the Hosmer-Lemeshow (HL) test COPD
identified fewer patients requiring medical review relative
and by visual assessment in calibration plots.18 19 Analyses were to NEWS2 (9.5% absolute reduction), but the risk of death in
performed on IBM SPSS Statistics V.22 and SigmaPlot V.12.3. the low-risk group was similar (3.5% and 3.1%, p=0.686).
A patient receiving excess oxygen may not necessarily require
Results an urgent clinical review and should first have their oxygen
Population and missing data titrated down to target range. For NEWS2, if excess oxygen
In 2645 patients, the mean (SD) age was 73.1 (10.2), 54% were scores were discounted, there was a significant reduction in the
women and the mean (SD) FEV1 was 45.1% (SD 18.2) predicted. frequency of alerts for patients with a score of 5 or more (71.6%
Almost half were unable to leave the house unassisted (Extended to 68.0%, p=0.005). For NEWS2All COPD, discounting excess
Medical Research Council Dyspnoea score 5a or 5b=47%), and oxygen reduced reviews in this group from 62.1% to 58.6%
29.8% had consolidation on the admission chest X-ray. Further- (p=0.010), with little difference in the percentage of deaths in
more, 17.9% of patients were acidaemic on the admission arte- the low-risk patient group (counting excess oxygen=3.5% vs
rial blood gas (ABG) analysis, and 10.1% had a pH of less than discounting excess oxygen=3.6%, p=0.905). Of those patients
7.3. Comorbid conditions were common; 55.4% had a Charlson who were assigned a low-risk NEWS2 or NEWS2All COPD score,
index of 2 or more, and 28.9% scored 3 or more. Overall inpa- none died that same day. This suggests that all of these patients
tient mortality was 8.6%; 96 in 920 (10.4%) in the derivation were correctly categorised as low risk at the time; early warning
cohort and 132 in 1725 (7.7%) in the validation cohort. scores are repeated with each set of observations and are likely
A more detailed breakdown of patient characteristics by site to have risen in those who subsequently died.
and cohort is published elsewhere.11 12 There were no missing
data for the primary outcome, inpatient mortality. Rates of Performance of admission DECAF and early warning scores
missing data for indices were low and are shown in table 2. for inpatient mortality
For all patients (n=2645), NEWS2All COPD was a stronger
Risk groups and alerts mortality predictor than NEWS (AUROC NEWS2All COPD=0.72,
Inpatient mortality by risk group is shown in table 3. The 95% CI 0.68 to 0.76, vs NEWS=0.65, 95% CI 0.61 to 0.68;
percentage of patients classified as requiring an urgent review p<0.001). NEWS2All COPD showed a trend towards superiority
(score of 5 or more) for NEWS2All COPD, NEWS2 and NEWS was compared with NEWS2 (AUROC NEWS2All COPD=0.72, 95% CI
62.1%, 71.6% and 74.7% (middle column figure 1). 0.68 to 0.76, vs NEWS2=0.70, 95% CI 0.67 to 0.74; p=0.090).
Of interest, the addition of confusion to the ‘level of conscious-
ness’ component of NEWS2 accounted for part of the improved
performance in NEWS2 compared with NEWS. For example,

Figure 1  Frequency of alerts for NEWS2All COPD, NEWS2 and NEWS.

Figure shows the percentage of patients in each risk category, grouped
together by early warning score. The first column in each group is
NEWS2All COPD, the second is NEWS2 and the third is NEWS. P values were
calculated using Fisher’s exact test. NEWS 5–6=medium risk, prompting Figure 2  Histogram of NEWS scores. Figure shows all patients from
urgent response by clinician/clinical team and minimum hourly the derivation and validation cohorts. The number of individuals is
observations. NEWS 7=high risk. Urgent response by clinician/clinical shown in the x-axis (‘count’) and the total NEWS2 score is shown on the
team, which may include critical care, and continuous monitoring or y-axis. The grey lines separate low-risk (0–4), moderate-risk (5–6) and
vital signs. high-risk (7 or more) groups.
Echevarria C, et al. Thorax 2019;74:941–946. doi:10.1136/thoraxjnl-2019-213470 943
 Chronic obstructive pulmonary disease

Table 4  Sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV) at NEWS thresholds of 5 and 7
Low-risk group Low and moderate risk group combined
(cut-off less than 5 for NEWS) (cut-off less than 7 for NEWS)
Sensitivity Specificity NPV Sensitivity Specificity PPV
NEWS2All COPD 0.85 0.40 0.97 0.68 0.64 0.15
NEWS2 0.90 0.30 0.97 0.71 0.55 0.13
NEWS 0.89 0.27 0.96 0.64 0.55 0.12
DECAF 0.93 0.52 0.99 0.73 0.80 0.25

when scored without confusion, the AUROC for NEWS2All COPD Calibration curves are shown in figure 4. A good prognostic
was only 0.68 (95% CI 0.64 to 0.71) which is significantly worse tool should cover a range of risks, have an intercept close to zero
than NEWS2All COPD scored with confusion (p<0.001). and a slope of one, and have the intersection of predicted and
DECAF provided optimal prediction of inpatient mortality, observed risk close to the line (shown by the dots). NEWS2All
and was superior to NEWS2 and NEWS2All COPD in terms of COPD
was well calibrated and covered a larger range of risks than
performance as shown in figure 3 (further data in online supple- other early warning scores.
mentary table 1).
Table 4 shows the performance of early warning scores at Discussion
low-risk and high-risk NEWS thresholds (5 and 7), with the Patients admitted with ECOPD have a modest in-hospital
DECAF score the gold standard. The positive predictive value mortality rate (3.9%–4.3%),10 20 but frequently trigger alerts
(PPV) and negative predictive value (NPV) most inform the clini- on the National Early Warning Scale (NEWS). This can place a
cian: the PPV is the chance of a patient dying if they score above substantial burden on healthcare professionals, create compla-
the threshold, while the NPV is the chance of a patient surviving cency and promote excess oxygen use in those who are at partic-
if they score below the threshold. Sensitivity is the proportion of ular risk of its harmful effects.6 In this study, NEWS2 showed
those who died that were correctly identified as at risk (above the superior discrimination for mortality to NEWS, but only
threshold), while the specificity looks at the proportion of those
who survived, identified as not at risk (below the threshold).
In the low-risk group, a higher sensitivity and NPV is desir-
able. With regards to sensitivity, NEWS2 performed optimally,
while both NEWS2 scores performed equally well for NPV. In
the high-risk group, a high specificity and PPV is desirable, for
which NEWS2All COPD performed optimally.
Calibration describes how well individuals’ predicted risks
from a logistic regression model agree with the observed risk in
terms of the proportion of individuals affected. The goodness of
fit can be measured with the HL test, with poor fit indicated by
a small calculated p value. DECAF, NEWS2All COPD and NEWS
were well calibrated with HL statistics of p=0.209, p=0.237
and p=0.736. NEWS2 was not well calibrated, with an HL
statistic of p=0.048.

Figure 4  Calibration curves for DECAF, NEWS2All COPD, NEWS2 and

NEWS in all cohorts. Figure shows calibration curves for NEWS2All COPD,
NEWS2 and NEWS in all patients. DECAF is shown in the validation
cohort, as inclusion of the derivation cohort would unfairly favour its
performance. Predicted risk is shown on the y-axis which was calculated
using logistic regression analysis to give individual’s mortality risk as a
percentage. The x-axis shows the observed risk, which is a proportion
Figure 3  Receiver operator curves for the DECAF score, NEWS2All COPD of the number of patients who died within an assigned risk range. The
and NEWS all cohorts. Figure shows the performance of the DECAF, slope of the equation should be near 1 (eg, for DECAF, the slope is
NEWS2All COPD, NEWS2 and NEWS scores. A higher area under the receiver 1.038), and the intercept should be near 0 (for DECAF, this is 0.0022).
operating characteristic curve (‘A’) shows better prediction. DECAF is The R2 represents the correlation between the predicted risk and
not included in the derivation cohort, as this would unfairly favour its observed risk using Pearson’s correlation, with scores closest to 1
performance. showing maximum correlation.
944 Echevarria C, et al. Thorax 2019;74:941–946. doi:10.1136/thoraxjnl-2019-213470
 Chronic obstructive pulmonary disease
reclassified 3.1% as not requiring a review, based on an alert risk of hypercapnia based on ICD-10 codes, ‘NEWS2-without
trigger of 5 or more. NEWS2All COPD, adopting target saturations confusion’ was slightly inferior to NEWS for identifying death
of 88%–92% and scale 2 of the NEWS2 tool for all patients, within 24 hours. In a subgroup of 1394 patients with confirmed
led to an absolute reduction in alert frequency of 12.6%. In hypercapnia, ‘NEWS2-without confusion’ offered similar
clinical practice, titrating down oxygen prior to alerts may be discrimination and a higher PPV than NEWS. In common with
appropriate; this approach decreased alerts by 16.1%. The our study, ‘NEWS2-without confusion’ generated fewer alerts,
performance of NEWS2All COPD was similar to NEWS2, with no and mortality rates were low on the same day as patients having
increased risk of death in the low-risk group. Of key importance, low-risk assessments. Our study differs in that we only included
there were no deaths on the same day that patients were classi- patients with confirmed COPD, NEWS2 was correctly scored
fied as not requiring review. and case ascertainment was robust.
DECAF was prognostically superior to all scores, showed the While no patient should receive a diagnosis of COPD without
most consistent performance and was well calibrated. Accu- obstructive spirometry, we acknowledge that this does occur
rate mortality prediction informs clinical decision-making; the and that our results may not be generalisable to such patients.
low-risk DECAF group was larger with a lower risk of death However, we highlight that the Austin RCT showed that target
than seen with other scores, which makes DECAF suitable for SpO2 88%–92% in all patients with a presumed diagnosis
the identification of patients for early discharge or hospital at of ECOPD at ambulance pick-up was associated with lower
home services. Additionally, in the National Asthma and COPD mortality, and this included patients without a spirometry-con-
Audit, baseline mortality risk is being quantified by the NEWS2 firmed diagnosis of COPD.6
score, but our results show NEWS2 should not be used in place An important limitation of this and other studies of NEWS2 is
of the DECAF score for this purpose. that we did not observe NEWS2 in practice. Real-time recording
Eligibility criteria were broad and rates of missing data were of NEWS2 may result in changes in clinical behaviour. For
low, with robust data analyses, including methods to deal with example, while the performance of NEWS2All COPD and NEWS2
missing data. The assessment of the NEWS score was pre-spec- was similar in our study, the reduction in false alerts with
ified, as was the assessment of confusion; therefore, there is no NEWS2All COPD could result in the more timely clinical assessment
bias in the subsequent re-coding of NEWS2. Furthermore, in of other patients and hence a positive clinical outcome. Prospec-
both cohorts, the inclusion of NEWS indices was pre-specified, tive randomised studies are required to address these points,25
and data were collected prior to and/or without knowledge of and to see if the use of NEWS2All COPD leads to improvements in
the (objective) outcome. For the AUROC curve comparison, mortality by preventing excess oxygen prescribing, but would be
DECAF was compared with NEWS iterations only in the vali- challenging to perform. Mortality was high in this study, though
dation cohort; comparing performance in the derivation cohort appropriate given the higher levels of comorbidity, pneumonia
would have introduced bias in favour of the DECAF score. A key and frailty in our cohort compared with UK national audit data
strength is that our population was well described, with respi- collected at a similar time.
ratory specialist confirmation of exacerbation and spirometric The current BTS oxygen guidelines recommend initial target
confirmation of COPD. These results may not apply to those saturations of 88%–92% in patients with COPD. NEWS2
with presumed ECOPD without obstructive spirometry (patients requires confirmation of hypercapnia before the alternative
misdiagnosed with COPD). In our study, considerable efforts oxygen saturation scale is activated and should only be instigated
were made to obtain previous lung function results: we checked by a clinician. Of concern, the NEWS2 implementation guidance
primary and secondary paper and electronic records before and NEWS2 chart may result in many patients receiving excess
concluding spirometry was unavailable. oxygen prior to ABG analysis or clinician review. RCT data show
Our study has the following limitations. Scores were compared a mortality benefit in those treated with oxygen target saturations
using admission data only. For the comparison with the DECAF 88%–92% at the point of ambulance pick-up. These patients did
score, using admission data is appropriate as we are assessing the not have baseline ABG analysis prior to oxygen treatment, and
tools’ performance as a guide to inform clinical decision-making this approach was beneficial even though a substantial propor-
at the point of admission. However, for NEWS the inclusion of tion of patients were subsequently described as not having
multiple measures would have added further useful data. COPD.6 RCTs have identified increases in mortality with excess
‘NEWS2-without confusion’ has been compared with NEWS oxygen in patients with a range of acute medical presentations
by Hodgson and others, in a large cohort of patients.21 They even where there is no risk of hypercapnia.26 Using NEWS2All
reported concerns that high-risk patients were inappropriately COPD
in all patients with ECOPD could offer improved prognostic
downgraded by NEWS2. However, NEWS2 was calculated from performance and calibration, with significantly fewer alerts, and
NEWS, which does not include confusion. This is an important lead to a reduction in harm from excess oxygen delivery in this
omission as we have shown that confusion offers additional at-risk population.
prognostic benefit, and inclusion of confusion may have resulted
in fewer patients being downgraded. Furthermore, time to death Contributors  The study was conceived by SCB and designed by CE, JS and SCB.
was not reported: in our study, no low-risk patients died the CE performed the statistical analyses. All authors contributed to data analysis and
same day, which supports correct risk categorisation. In common interpretation, and all were involved in drafting the manuscript and approving the
final version. The corresponding author attests that all listed authors meet authorship
with other studies assessing similar early warning scores (such as criteria and that no others meeting the criteria have been omitted.
Salford NEWS),22 Hodgson et al23 used ICD-10 codes to iden-
Funding  Department of Health, Breathe North appeal, Northumbria NHS
tify patients, which are known to be inaccurate. Foundation Trust Teaching and Research Fellowship programme and Novartis
A recent retrospective study by Pimentel et al24 offers the Pharmaceuticals UK.
advantage of capture of sequential early warning scores in Disclaimer  The funders of the study had no role in study design, data collection,
251 266 acute adult admissions. Again, reliance on coding limits data analysis, data interpretation or writing of the report.
data quality in terms of case ascertainment. Again, confusion was Competing interests  All authors have completed the ICMJE uniform disclosure
not included, and therefore NEWS2 was not correctly scored. form at www.​icmje.​org/​coi_​disclosure.​pdf and declare: JS has no conflicts of
In 48 898 patients with conditions that were associated with interest to declare. CE reports grants from National Institute of Health Research,

Echevarria C, et al. Thorax 2019;74:941–946. doi:10.1136/thoraxjnl-2019-213470 945

 Chronic obstructive pulmonary disease
outside of the submitted work. SCB reports grants from National Institute of Health 11 Steer J, Gibson J, Bourke SC. The DECAF score: predicting hospital mortality in
Research, Philips Respironics and from Pfizer Open Air, personal fees from Pfizer, exacerbations of chronic obstructive pulmonary disease. Thorax 2012;67:970–6.
AstraZeneca and ResMed, and non-financial support from Boehringer Ingelheim and 12 Echevarria C, Steer J, Heslop-Marshall K, et al. Validation of the DECAF
GlaxoSmithKline outside the submitted work. No author has financial relationships score to predict hospital mortality in acute exacerbations of COPD. Thorax
with any organisation that might have an interest in the submitted work. 2016;71:133–40.
Patient consent for publication  Not required. 13 Echevarria C, Gray J, Hartley T, et al. Home treatment of COPD exacerbation selected
by DECAF score: a non-inferiority, randomised controlled trial and economic
Ethics approval  The study was approved by NRES Committee North East, UK (REC evaluation. Thorax 2018;73:713–22.
reference: 12/NE/0379). 14 Royal College of Physicians. National Asthma and COPD Audit Programme (NACAP):
Provenance and peer review  Not commissioned; externally peer reviewed. secondary care workstream—COPD resources, 2018. Available: https://www.​
rcplondon.​ac.​uk/​projects/​outputs/​national-​asthma-​and-​copd-​audit-​programme-​
Data availability statement  Data are available on reasonable request. nacap-​secondary-​care-​workstream-​copd [Accessed 8 Nov 2018].
Open access  This is an open access article distributed in accordance with the 15 Echevarria C, Steer J, Heslop-Marshall K, et al. The PEARL score predicts 90-day
Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which readmission or death after hospitalisation for acute exacerbation of COPD. Thorax
permits others to distribute, remix, adapt, build upon this work non-commercially, 2017;72:686–93.
and license their derivative works on different terms, provided the original work is 16 Rubin DB. Multiple imputation after 18+ years. J Am Stat Assoc 1996;91:473–89.
properly cited, appropriate credit is given, any changes made indicated, and the use 17 DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or
is non-commercial. See: http://​creativecommons.​org/​licenses/​by-​nc/​4.​0/. more correlated receiver operating characteristic curves: a nonparametric approach.
Biometrics 1988;44:837–45.
18 Hosmer DW, Lemesbow S. Goodness of fit tests for the multiple logistic regression
References model. Commun Stat Theory Methods 1980;9:1043–69.
1 Royal College of Physicians. National Early Warning Score (NEWS) 2: standardising the
19 Moons KGM, Altman DG, Reitsma JB, et al. Transparent reporting of a multivariable
assessment of acute-illness severity in the NHS. Updated report of a working party.
prediction model for individual prognosis or diagnosis (TRIPOD): explanation and
London: RCP, 2017.
elaboration. Ann Intern Med 2015;162:W1–73.
2 Wedzicha JA, Seemungal TAR. COPD exacerbations: defining their cause and
prevention. Lancet 2007;370:786–96. 20 Stone RA MV, Mortier K, Holzhauer-Barrie J, et al. COPD: working together. National
3 Inspiring change. The national confidential enquiry into patient outcome and death. chronic obstructive pulmonary disease (COPD) audit programme: clinical audit
Inspiring change, 2017. of COPD exacerbations admitted to acute hospitals in England and Wales 2017.
4 Alrawi YA, Potter JF, Myint PK. UK National COPD Resources and Outcomes Project National clinical audit report. London: RCP, 2018.
(NCROP): 2008 national audit data presents an opportunity to highlight the areas for 21 Hodgson LE, Congleton J, Venn R, et al. NEWS 2—too little evidence to implement?
improvement in COPD care in the ageing population. COPD 2010;7:360–5. Clin Med 2018;18:371–3.
5 O’Driscoll BR, Grant K, Green D, et al. The national early warning score gives 22 Pedersen NE, Rasmussen LS, Petersen JA, et al. Modifications of the national early
misleading scores for oxygen saturation in patients at risk of hypercapnia. Clin Med warning score for patients with chronic respiratory disease. Acta Anaesthesiol Scand
2014;14:695–6. 2018;62:242–52.
6 Austin MA, Wills KE, Blizzard L, et al. Effect of high flow oxygen on mortality in 23 Hodgson LE, Dimitrov BD, Congleton J, et al. A validation of the national early
chronic obstructive pulmonary disease patients in prehospital setting: randomised warning score to predict outcome in patients with COPD exacerbation. Thorax
controlled trial. BMJ 2010;341:c5462. 2017;72:23–30.
7 O’Driscoll BR, Howard LS, Earis J, et al. BTS guideline for oxygen use in adults in 24 Pimentel MAF, Redfern OC, Gerry S, et al. A comparison of the ability of the
healthcare and emergency settings. Thorax 2017;72(Suppl 1):ii1–90. National Early Warning Score and the National Early Warning Score 2 to identify
8 Robinson TD, Freiberg DB, Regnis JA, et al. The role of hypoventilation and ventilation- patients at risk of in-hospital mortality: a multi-centre database study. Resuscitation
perfusion redistribution in oxygen-induced hypercapnia during acute exacerbations of 2019;134:147–56.
chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2000;161:1524–9. 25 O’Driscoll R, Bakerly N, Murphy P, et al. NEWS2 needs to be tested in prospective
9 O’Driscoll R.BTS emergency oxygen audit report; 2015. trials involving patients with confirmed hypercapnia. Resuscitation 2019;139:369–70.
10 Stone RA, Holzhauer-Barrie J, Lowe D, et al. OPD: who cares matters. National chronic 26 Chu DK, Kim LH-Y, Young PJ, et al. Mortality and morbidity in acutely ill adults treated
obstructive pulmonary disease (COPD) audit programme: clinical audit of COPD with liberal versus conservative oxygen therapy (IOTA): a systematic review and meta-
exacerbations admitted to acute units in England and Wales 2014, 2015. analysis. Lancet 2018;391:1693–705.

946 Echevarria C, et al. Thorax 2019;74:941–946. doi:10.1136/thoraxjnl-2019-213470