M a t e r n a l E a r l y Wa r n i n g

Systems
Alexander M. Friedman, MD

KEYWORDS
 Maternal early warning system  Modified early obstetric warning system
 Modified early warning criteria  Maternal morbidity  Maternal mortality

KEY POINTS
 Maternal mortality case reviews find that severe vital sign abnormalities often precede
clinical recognition of critical illness.
 Early warning systems have been used in other specialties to identify patients at high risk
for clinical decompensation.
 Specific early warning criteria have been developed for obstetric patients.
 Maternal early warning systems are being advocated by obstetric leadership.
 Although early warning systems are a promising strategy for improving maternal out-
comes, research evidence is limited.

INTRODUCTION

The burden posed by severe morbidity and mortality during pregnancy and childbirth
in the developed world has long been overlooked, and prevention has been neglected
despite its vital importance in improving outcomes. As a result, recent data indicate
that maternal death and severe morbidity—key indicators of population health—not
only remain common1 but are actually increasing in the United States.2,3 Dramatic ad-
vances in neonatal and fetal care over the past 3 decades have not been matched by
improved maternal care, with the Centers for Disease Control and Prevention esti-
mating that 52,000 women suffer major morbidity annually.2 National organizations,
including the American Congress of Obstetricians and Gynecologists, American
Board of Obstetrics and Gynecology, the Society for Maternal-Fetal Medicine, the
Joint Commission, Amnesty International, and the Eunice Shriver Kennedy National
Institute of Child Health and Human development have all recently issued recommen-
dations to the obstetric community to increase awareness of maternal mortality and
promote improved care of the mother. The failure to prioritize maternal care—a

Financial Disclosure: The author does not report any potential conflicts of interest.
Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, College of
Physicians and Surgeons, Columbia University, 622 East 168th Street, New York, NY 10032, USA
E-mail address: [email protected]

Obstet Gynecol Clin N Am 42 (2015) 289–298

http://dx.doi.org/10.1016/j.ogc.2015.01.006 obgyn.theclinics.com
0889-8545/15/$ – see front matter Ó 2015 Elsevier Inc. All rights reserved.
290 Friedman

question of “Where is the ‘M’ in maternal-fetal medicine?”4—has led to educational,

clinical, and research initiatives to improving maternal outcomes.5 However, efforts
to decrease maternal death and severe morbidity have been limited, and the only vali-
dated strategy that has emerged to systematically improve maternal outcomes is
routine use of postcesarean thromboembolism prophylaxis.6,7
Many cases of major maternal morbidity and mortality may be preventable,7–9 and
obstetric early warning systems that alert care providers of abnormal physiologic pa-
rameters that may precede critical illness have been advocated,7 as they may repre-
sent a means of improving clinical outcomes. Other specialties have used early
warning systems to predict clinical deterioration with varying success.10 In its 2007
triennial report on maternal death, the United Kingdom’s Confidential Enquiry into
Maternal Death recommended adoption of the modified early obstetric warning sys-
tem (MEOWS)11 which utilizes a combination of physiologic and neurologic parame-
ters to identify obstetric inpatients who require urgent or emergent evaluation by a
care provider. The parameters in MEOWS seek to identify patients with hypertensive
disorders, hemorrhage, thromboembolism, sepsis, and cardiovascular and cerebro-
vascular complications, conditions that account for more than 50% of all maternal
deaths and disproportionate major morbidity.1,7,12 Given the need to implement stra-
tegies that will systematically improve maternal outcomes and the emerging interest
and research literature on maternal early warning systems, this review covers the
following topics:
1. The clinical rationale for early warning systems including the research literature on
early alerts in other specialties
2. Clinical parameters and recommended care in maternal early warning systems
3. Research evidence supporting maternal early warning systems
4. Future directions in optimizing and validating maternal early warning systems

EARLY WARNING SYSTEMS IN OTHER SPECIALTIES

Early warning systems have been used in several specialties, primarily with the goal of
identifying patients who may become critically ill and improving outcomes with early
intervention. These systems are classified as either triggering systems, in which a pa-
tient is at risk based on one positive parameter, or scoring systems, in which different
parameters contribute to a single numerical value and score cutoff levels predict
risk.13,14 The Pediatric Early Warning Score created by Duncan and colleagues15 to
predict actual or impending cardiopulmonary arrest in hospitalized children is shown
in Table 1. In their cohort of more than 32,000 patients, a cutoff score of 5 was 78%
sensitive and 95% specific in predicting actual or impending cardiopulmonary arrest
with an area under the receiver operating characteristic (AUROC) curve of 0.90,
yielding 68 true-positive and 1763 false-positive cases. Findings from their analysis
show an important consideration in early warning systems: in a population with low
risk for critical illness, even alert systems with test characteristics that perform rela-
tively well may result in large numbers of false-positive cases.
Early warning systems and scores have been studied in several clinical settings
including pediatrics,14,16 general medical and surgical admission populations,17–19
and medical20 and surgical21 specialties. The quality of data supporting the use of indi-
vidual systems is generally poor.22 The number of alert systems has proliferated, and
studies generally lack appropriate methodology and adequate statistical powering
given the relative infrequency of critical illness that may occur even in a large population.
A systematic review of pediatric early warning systems found that the validity, reliability,
and utility of pediatric alert criteria were weak.13 A systematic review of early warning
 Maternal Early Warning Systems 291

Table 1
The Pediatric Early Warning Score system

Score 2 1 0 1 2
Age-specific items
<3 mo
HR <90 90–109 110–150 151–180 >180
RR <20 20–29 30–60 61–80 >80
SBP <50 50–59 60–80 81–100 >100
3–12 mo
HR <80 80–99 100–150 151–170 >170
RR <20 20–24 25–50 51–70 >70
SBP <70 70–79 80–100 99–120 >120
1–4 y
HR <70 70–89 90–120 121–150 >150
RR <15 15–19 20–40 41–60 >60
SBP <75 75–89 90–110 111–125 >125
4–12 y
HR <60 60–69 70–110 111–130 >130
RR <12 12–19 20–30 31–40 >40
SBP <80 80–90 90–120 120–130 >130
>12 y
HR <50 50–59 60–100 101–120 >120
RR <8 8–12 12–16 15–24 >24
SBP <86 85–101 100–130 131–150 >150
General items
Pulses Absent Doppler Present Bounding
O2 saturation (%) <85 85–95 >95
Capillary refill CRT >3 2–3 CRT <2
LOC <7 7–11 12–15
Oxygen therapy >50% or >4 L/min Any <50% or <4 L/min None
Bolus fluid Any None
Temperature <35 35–<36 36 >38.5–<40 >40

The score is calculated by adding the demographic and medication subscores. Patients received 1
point for each of the following: abnormal airway (not tracheostomy), home oxygen, any previous
admission to an ICU, central venous line in situ, transplant recipient, severe cerebral palsy,
gastrostomy tube, and greater than 3 medical specialties involved in care. The medication sub-
score is from the number of medication administered in 24 hours. 3 5 0, 4–6 5 1, 7–9 5 2,
9–12 5 3, 12–15 5 4, 16 5 5.
Abbreviations: HR, heart rate (beats per minute); LOC, level of consciousness measured with
the Glasgow Coma Scale; RR, respiratory rate (breaths per minute); SBP, systolic blood pressure
(mm Hg).
From Duncan H, Hutchison J, Parshuram CS. The pediatric early warning system score: a severity
of illness score to predict urgent medical need in hospitalized children. J Crit Care 2006;21:275;
with permission.

systems for adults admitted to medical or surgical wards found that although predictive
values for death (AUROC curve, 0.88–0.93) and cardiac arrest within 48 hours (AUROC
curve, 0.74–0.86) were fairly high, the overall impact on health outcomes and resource
utilization was unclear.17 A Cochrane review of early warning systems and critical care
292 Friedman

outreach noted poor methodologic quality of most studies reviewed and a found a lack
of evidence regarding the benefit of critical care outreach in studies included in the
analysis.10 For example, the medical emergency response improvement team (MERIT)
study randomly selected 23 hospitals to a medical emergency team system with call pa-
rameters to see if unplanned intensive care unit (ICU) admissions, cardiac arrests, and
deaths could be prevented, and found that although emergency team calling was
greatly increased, there was no improvement in outcomes.23
Varying criteria across scoring systems and a lack of consistency in detecting deteri-
oration of patients’ conditions have been identified as major concerns24 and led the Royal
College of Physicians to design a national standard for nonpregnant adults for the United
Kingdom called the National Early Warning Score (NEWS).25 NEWS was designed to
address specific concerns identified with the proliferation of different systems including:
(1) varying parameters and weighting leading to unfamiliarity across hospitals or in
different clinical settings within hospitals, (2) poor validation of early warning systems in
detecting a broad range of acute severe illness across different clinical settings, (3)
lack of clear definitions for an appropriate clinical response in the setting of a positive
alert, and (4) an absence of uniform criteria to base postgraduate and undergraduate
training. The NEWS parameters and scoring system are shown in Table 2. Based on
the severity of abnormalities present, evaluation of the patient by a nurse, provider, or crit-
ical care specialist is recommended.
Although research literature on NEWS suggests it performs favorably compared with
other alert systems,26,27 its authors note that although a nationally standardized alert
system may standardize care, simplify clinical management and communication, and
be important for research validation, cut-points and scoring algorithms may be revised
in the future.25 Optimizing alert system performance is an important goal. A warning
system that results in a large number of false-positive results relative to true cases
detected early may potentially worsen clinical care, function as a nuisance alarm, and
contribute to alarm fatigue. Alarm fatigue—wherein clinical providers becoming

Table 2
NEWS

Physiologic Parameters 3 2 1 0 1 2 3
Respiration rate 8 9–11 12–20 21–24 25
Oxygen saturation 91 92–93 94–95 96
Any supplemental oxygen Yes No
Temperature 35.0 35.1–36.0 36.1–38.0 38.1–39.0 39.1
Systolic blood pressure 90 91–100 101–110 111–219 220
Heart rate 40 41–50 51–90 91–110 111–130 131
Level of consciousness A V, P, or U

Respiration rate (breaths per minute); Oxygen saturation (%); Temperature (degrees Celsius); Sys-
tolic blood pressure (mm Hg); Heart rate (beats per minute). Level of consciousness is based on the
Alert Voice Pain Unresponsive scale, which assesses 4 possible outcomes to measure and record a
patient’s level of consciousness. A low score (NEWS of 1–4) should prompt an assessment by a regis-
tered nurse. A medium score (NEWS of 5–6 or any single parameter of 3) should prompt an urgent
review by a clinician such as a ward-based physician or acute-team nurse. A high score (NEWS of 7
or more) should prompt emergency assessment by a critical care team with likely transfer of patient
to higher acuity setting.
From National Early Warning Score (NEWS). Standardising the assessment of acute-illness
severity in the NHS. London: Royal College of Physicians; 2012. Available at: https://www.
rcplondon.ac.uk/sites/default/files/documents/national-early-warning-score-standardising-
assessment-acute-illness-severity-nhs.pdf. Accessed December 1, 2014; with permission.
 Maternal Early Warning Systems 293

overwhelmed and desensitized to alerts of little to no clinical usefulness (nuisance

alarms) that can occur several hundred per day per patient depending on the hospital
unit—is a recognized source of medical errors.28,29 To establish the validity of an early
warning system in any population (particularly in an obstetric population in which the
rate of major morbidity and critical illness is low), a careful assessment of test charac-
teristics is necessary.

MATERNAL EARLY WARNING SYSTEMS

Early warning systems to detect critical illness in obstetric patients have been specif-
ically designed for this population because of (1) the physiologic changes that occur
during pregnancy and (2) the small number of conditions responsible for most maternal
severe morbidity and mortality. Recommendations for use of the NEWS include the re-
striction that it is not applicable to pregnant patients.25 Adoption of maternal alert sys-
tems was strongly advocated in the 2007 Saving Mothers’ Lives report from the
Confidential Enquiries into Maternal and Child Health (CEMACH) in the United
Kingdom. The report made hospital-based adoption of MEOWS a “top ten” recommen-
dation, an urgent priority “which every commissioner, provider, policy maker and other
stakeholder involved in providing maternity services should plan to introduce, and
audit, as soon as possible.”30 The CEMACH report includes case reviews of maternal
deaths and found that “in many cases in this Report, the early warning signs of impend-
ing maternal collapse went unrecognized.”30
A MEOWS scoring system is shown in Table 3. In this system, 2 moderately abnormal
parameters (yellow alerts) or 1 severely abnormal parameter (red alert) triggers a clinical
response to urgently assess the patient’s status and make a follow-up surveillance plan.
The parameters are designed to detect patients suffering from conditions that may lead
to severe maternal morbidity and mortality. In the United States, the conditions respon-
sible for most adverse maternal outcomes include hemorrhage, venous thromboembo-
lism, hypertensive diseases of pregnancy, sepsis, and cardiovascular causes as
demonstrated in Table 4, which presents data on deaths from the Centers for Disease

Table 3
A modified early obstetric warning system

Physiologic Parameters Yellow Alert Red Alert

Respiration rate 21–30 <10 or >30
Oxygen saturation <95
Temperature 35–36 <35 or >38
Systolic blood pressure 150–160 or 90–100 <90 or >160
Diastolic blood pressure 90–100 >100
Heart rate 100–120 or 40–50 >120 or <40
Pain score 2–3
Neurologic response Voice Unresponsive, pain

Respiration rate (breaths per minute); Oxygen saturation (%); Temperature (degrees Celsius); Sys-
tolic blood pressure (mm Hg); Heart rate (beats per minute). Level of consciousness is based on the
Alert Voice Pain Unresponsive scale, which assesses 4 possible outcomes to measure and record a
patient’s level of consciousness. Pain scores are as follows: (0 5 no pain, 1 5 slight pain on move-
ment, 2 5 intermittent pain at rest/moderate pain on movement). A single red score or 2 yellow
scores triggers an evaluation.
From Singh S, McGlennan A, England A, et al. A validation study of the CEMACH recommended
modified early obstetric warning system (MEOWS). Anaesthesia 2012;67:12–8; with permission.
294 Friedman

Table 4
Causes of pregnancy-related death resulting in live births in the United States, 1998–2005

Condition Deaths (%)

Pulmonary embolism 9.7
Hemorrhage 9.7
Amniotic fluid embolism 9.0
Hypertensive disorders 15.0
Infection 9.2
Anesthesia 1.2
Cardiomyopathy 13.3
Cerebrovascular accident 7.0
Cardiovascular conditions 12.5
Other/unknown 13.5

Data derived from the Centers for Disease Control and Prevention’s Pregnancy Mortality Surveil-
lance System. Maternal deaths (n 5 2856) in this table resulted in a live birth. Maternal deaths asso-
ciated with a stillbirth (n 5 243) or with an undelivered pregnancy (n 5 589) were not included.
From Berg CJ, Callaghan WM, Syverson C, et al. Pregnancy-related mortality in the United States,
1998 to 2005. Obstet Gynecol 2010;116:1302–9; with permission.

Control and Prevention’s Pregnancy Mortality Surveillance System.2,12,31 Data from

state mortality reviews suggest that deaths from all of these causes may be reduced
by improved care. For example, a California review found that 70% of hemorrhage
deaths, 60% of preeclampsia/eclampsia deaths, and 63% of sepsis/infection deaths
may have been preventable.22 The MEOWS alert parameters may lead to detection
of the following unrecognized conditions: hemorrhage (as demonstrated by hypoten-
sion and tachycardia), sepsis (fever, hypotension, tachycardia, hypoxia), venous throm-
boembolism (tachycardia, tachypnea, hypoxia), preeclampsia (hypertension, hypoxia),
and cardiovascular complications (tachycardia, bradycardia, hypoxia, hypotension).
The subsequent 2011 Saving Mothers’ Lives report found that for many patients
critically ill secondary to diagnoses such as sepsis and hemorrhage, MEOWS alerts
had the potential to improve outcomes by facilitating early recognition.7
In the United States, the National Partnership for Maternal Safety, a national leadership
group,32 proposed a simplified early warning system adapted from MEOWS, the
Maternal Early Warning Criteria (MEWC).33 While MEOWS represents a simple scoring
system, MEWC represents a trigger system. If a patient has any single abnormal param-
eter (Table 5) a prompt bedside assessment by a provider is required. The simplified
MEWC parameters were chosen to minimize the rate of false alarms, facilitate implemen-
tation, and retain sensitivity, given that case reviews of maternal deaths frequently show
frank vital sign abnormalities preceding recognized critical illness. Given the range of
possible clinical care settings, from small nonteaching community hospitals to large qua-
ternary referral academic centers, determining the optimal personnel and protocol
response for alerts is the responsibility of the individual center. Based on the care setting,
an alert response may be led by obstetric providers, anesthesiologists, hospitals, inten-
sivists, emergency physicians, or a rapid response team.34 Likewise, because of the
varying clinical settings, outcome variables and process measures for this system have
not been defined. The American Congress of Obstetricians and Gynecologists District
II’s Safe Motherhood Initiative, a collaboration to improve maternal outcomes in New
York, a state with one of the highest mortality rates in the country, has endorsed
MEWC for use in all hospitals providing obstetric services.
 Maternal Early Warning Systems 295

Table 5
MEWC

Systolic BP, mm Hg <90 or >160

Diastolic BP, mm Hg >100
Heart rate, beats per minute <50 or >120
Respiratory rate, breaths per minute <10 or >30
Oxygen saturation, % on room air <95
Oliguria, milliliters per hour for 2 h <35
Neurologic: Maternal agitation, confusion, or unresponsiveness; Patient with preeclampsia
reporting a nonremitting headache or shortness of breath

The presence of any of the abnormal parameters above necessitates the prompt evaluation of the
patient by a provider.
From Mhyre JM, D’Oria R, Hameed AB, et al. The maternal early warning criteria: a proposal from
the national partnership for maternal safety. Obstet Gynecol 2014;124:782–6; with permission.

RESEARCH EVIDENCE SUPPORTING MATERNAL EARLY WARNING SYSTEMS

Emerging research literature has sought to characterize the potential benefits of

maternal early warning systems. Singh and colleagues35 assessed MEOWS parame-
ters in an obstetric population to determine test characteristics (see Table 3) for
screening for severe maternal morbidity. Their definition of morbidity included obstet-
ric hemorrhage, severe preeclampsia, infection, and thromboembolism among other
diagnoses. Of 673 obstetric admissions, 86 women (13%) suffered morbidity, and
200 women (30%) had a positive MEOWS screen (either 1 red alert or 2 yellow alerts).
Morbidity was driven by 3 conditions—hemorrhage, pre-eclampsia, and infection—
which accounted for 94% of morbidity cases. MEOWS was 89% sensitive for
predicting morbidity (95% confidence interval [CI], 81%–95%) and 79% specific
(95% CI, 76–82) with a positive predictive value of 39% (95% CI, 96%–99%). Although
this trial demonstrated that MEOWS was reasonably sensitive for detecting morbidity,
it was not designed to determine MEOWS efficacy in (1) identifying otherwise unde-
tected acute or impending critical illness, (2) optimizing process measures related to
managing morbidity (eg, time to administration of antihypertensives), or (3) improving
clinically meaningful outcomes.
An analysis by Carle and colleagues36 used early warning parameters to predict
death for patients with obstetric diagnoses admitted to ICUs. This analysis found
that the study model and the early warning system presented in the 2003–2005 Report
on Confidential Enquiries into Maternal Deaths were highly predictive of death for ICU
patients with AUROC curves of 0.96 (95% CI, 0.92–0.99) and 0.94 (95% CI, 0.88–0.99),
respectively. Although these data lend support to the general validity of maternal early
warning parameters being highly sensitive for critical illness and risk of death, deter-
mining the clinical benefit of early detection and intervention was outside the scope
of the analysis.
Austin and colleagues37 performed a retrospective cohort analysis of severely ill ob-
stetric patients admitted to high-acuity units at a single referral center. Of 64 patients
that met study inclusion criteria, 5 (7.6%) may have had a condition detectable by an
early warning system before clinical deterioration and admission to a high-acuity
setting. Given the small number of patients and the single-center design, interpretation
of validity and generalizability is limited. Although the lack of data on validated out-
comes improvement may lead to the clinicians questioning the value of dedicating
296 Friedman

resources to maternal early warning systems,38 leadership has shown enthusiasm

with survey research showing high rates of hospital adoption of MEOWS within the
United Kingdom.39

DISCUSSION

Maternal early warning systems represent a promising strategy for reducing severe
maternal morbidity and mortality. For a maternal early warning system to contribute
to improved health outcomes, it must (1) identify patients at risk for critical illness
and who benefit from timely intervention and (2) not result in such a high number of
false-positive alerts that patient care is otherwise compromised. The clinical rationale
for maternal early warning systems is largely based on case reviews of maternal mor-
tality that show delayed response to abnormal vital sign parameters and other findings
suggestive of acute decompensation.7,8,30 Given the rare occurrence of maternal
death in the developed world, well-powered data to assess the benefits of trends
and interventions in improving safety are often limited to national vital statistics.
Limited current research evidence supports that early warning parameters may be
clinically useful in identifying patients that may become critically ill or at high risk for
mortality. However, these data are primarily from single centers, and further data
are needed to improve generalizability and validity. Currently, no data show what
the optimal responses are in particular settings to improve maternal care once an alert
has been initiated. Hospitals will likely require different response protocols based on
physician staffing, teaching status, bed size, nursing expertise, consultant services,
and critical care availability. Furthermore, the identification of an at-risk patient does
not ensure that (1) the correct diagnostic workup and evaluation will be performed
in setting of a positive screening result or (2) that with the correct diagnosis, subse-
quent interventions will be appropriate. Given that severe maternal morbidity and mor-
tality are rare, clinical decision support tools may be a necessary component for
providing optimal responses, particularly at smaller, nonteaching centers.
Maternal early warning systems are a promising surveillance strategy designed to
address the finding of multiple mortality reviews that clinical responses to acute dete-
rioration in obstetric patients are often delayed or inadequate. These systems are
increasingly being embraced and implemented by obstetric safety leadership. Future
work needs to focus on refining alert parameters, optimal response strategies across
clinical settings, and creating provider support tools for managing high-risk patients.

REFERENCES

1. Khan KS, Wojdyla D, Say L, et al. WHO analysis of causes of maternal death: a
systematic review. Lancet 2006;367:1066–74.
2. Callaghan WM, Creanga AA, Kuklina EV. Severe maternal morbidity among deliv-
ery and postpartum hospitalizations in the United States. Obstet Gynecol 2012;
120:1029–36.
3. Berg CJ, Callaghan WM, Syverson C, et al. Pregnancy-related mortality in the
United States, 1998 to 2005. Obstet Gynecol 2010;116:1302–9.
4. D’Alton ME. Where is the “M” in maternal-fetal medicine? Obstet Gynecol 2010;
116:1401–4.
5. D’Alton ME, Bonanno CA, Berkowitz RL, et al. Putting the “M” back in maternal-
fetal medicine. Am J Obstet Gynecol 2012;208:442–8.
6. Clark SL, Belfort MA, Dildy GA, et al. Maternal death in the 21st century: causes,
prevention, and relationship to cesarean delivery. Am J Obstet Gynecol 2008;
199:36.e1–5 [discussion 91–2.e7–11].
 Maternal Early Warning Systems 297

7. Cantwell R, Clutton-Brock T, Cooper G, et al. Saving mothers’ lives: reviewing

maternal deaths to make motherhood safer: 2006-2008. The eighth report of
the confidential enquiries into maternal deaths in the United Kingdom. BJOG
2011;118(Suppl 1):1–203.
8. The California Pregnancy-Associated Mortality Review. Report from 2002 and
2003 maternal death reviews. Sacremento (CA): California Department of Public
Health, Maternal Child and Adolescent Health Division; 2011.
9. Clark SL. Strategies for reducing maternal mortality. Semin Perinatol 2012;36:
42–7.
10. McGaughey J, Alderdice F, Fowler R, et al. Outreach and early warning systems
(EWS) for the prevention of intensive care admission and death of critically ill
adult patients on general hospital wards. Cochrane Database Syst Rev
2007;(3):CD005529.
11. The Confidential Enquiry Into Maternal and Child Health (CEMACH). Saving
mother’s lives: reviewing maternal deaths to make motherhood safer -
2003-2005. London: CEMACH; 2007.
12. Callaghan WM. Overview of maternal mortality in the United States. Semin Peri-
natol 2012;36:2–6.
13. Chapman SM, Grocott MP, Franck LS. Systematic review of paediatric alert
criteria for identifying hospitalised children at risk of critical deterioration. Inten-
sive Care Med 2010;36:600–11.
14. Seiger N, Maconochie I, Oostenbrink R, et al. Validity of different pediatric early
warning scores in the emergency department. Pediatrics 2013;132:e841–50.
15. Duncan H, Hutchison J, Parshuram CS. The pediatric early warning system
score: a severity of illness score to predict urgent medical need in hospitalized
children. J Crit Care 2006;21:271–8.
16. Gold DL, Mihalov LK, Cohen DM. Evaluating the pediatric early warning score
(PEWS) system for admitted patients in the pediatric emergency department.
Acad Emerg Med 2014;21:1249–56.
17. Smith ME, Chiovaro JC, O’Neil M, et al. Early warning system scores for clinical
deterioration in hospitalized patients: a systematic review. Ann Am Thorac Soc
2014;11:1454–65.
18. Subbe CP, Davies RG, Williams E, et al. Effect of introducing the Modified Early
Warning score on clinical outcomes, cardio-pulmonary arrests and intensive
care utilisation in acute medical admissions. Anaesthesia 2003;58:797–802.
19. Groarke JD, Gallagher J, Stack J, et al. Use of an admission early warning score
to predict patient morbidity and mortality and treatment success. Emerg Med J
2008;25:803–6.
20. Bokhari SW, Munir T, Memon S, et al. Impact of critical care reconfiguration and
track-and-trigger outreach team intervention on outcomes of haematology pa-
tients requiring intensive care admission. Ann Hematol 2010;89:505–12.
21. Patel MS, Jones MA, Jiggins M, et al. Does the use of a “track and trigger” warn-
ing system reduce mortality in trauma patients? Injury 2011;42:1455–9.
22. Gao H, McDonnell A, Harrison DA, et al. Systematic review and evaluation of
physiological track and trigger warning systems for identifying at-risk patients
on the ward. Intensive Care Med 2007;33:667–79.
23. Hillman K, Chen J, Cretikos M, et al. Introduction of the medical emergency team
(MET) system: a cluster-randomised controlled trial. Lancet 2005;365:2091–7.
24. Patterson C, Maclean F, Bell C, et al. Early warning systems in the UK: variation in
content and implementation strategy has implications for a NHS early warning
system. Clin Med 2011;11:424–7.
298 Friedman

25. National Early Warning Score (NEWS). Standardising the assessment of acute-
illness severity in the NHS. London: Royal College of Physicians; 2012. Accessed
December 1, 2014.
26. Smith GB, Prytherch DR, Meredith P, et al. The ability of the National Early Warn-
ing Score (NEWS) to discriminate patients at risk of early cardiac arrest, unantic-
ipated intensive care unit admission, and death. Resuscitation 2013;84:465–70.
27. Badriyah T, Briggs JS, Meredith P, et al. Decision-tree early warning score
(DTEWS) validates the design of the National Early Warning Score (NEWS).
Resuscitation 2014;85:418–23.
28. Patient Safety Advisory Group. Medical device alarm safety in hospitals. Sentinel
Event Alert 2013;(50):1–3.
29. ECRI Institute releases top 10 health technology hazards report for 2014. Available
at: http://www.ecri.org/Press/Pages/2014_Top_Ten_Hazards.aspx. Accessed
December 1, 2014.
30. Lewis G, editor. Saving mothers’ lives: reviewing maternal deaths to make moth-
erhood safer (2003-2005), the seventh confidential enquiry into maternal deaths
in the United Kingdom. London: Confidential Enquiry into Maternal and Child
Health; 2007.
31. Berg CJ, Chang J, Callaghan WM, et al. Pregnancy-related mortality in the United
States, 1991-1997. Obstet Gynecol 2003;101:289–96.
32. D’Alton ME, Main EK, Menard MK, et al. The national partnership for maternal
safety. Obstet Gynecol 2014;123:973–7.
33. Mhyre JM, D’Oria R, Hameed AB, et al. The maternal early warning criteria: a pro-
posal from the national partnership for maternal safety. Obstet Gynecol 2014;124:
782–6.
34. ACOG II Safe Motherhood Initiative. Maternal Early Warning Systems. 2014.
Available at: http://www.acog.org/-/media/Districts/District-II/PDFs/SMI/v2/OR-
MEWS.pdf. Accessed December 1, 2014.
35. Singh S, McGlennan A, England A, et al. A validation study of the CEMACH rec-
ommended modified early obstetric warning system (MEOWS). Anaesthesia
2012;67:12–8.
36. Carle C, Alexander P, Columb M, et al. Design and internal validation of an ob-
stetric early warning score: secondary analysis of the intensive care national audit
and research centre case mix programme database. Anaesthesia 2013;68:
354–67.
37. Austin DM, Sadler L, McLintock C, et al. Early detection of severe maternal
morbidity: a retrospective assessment of the role of an early warning score sys-
tem. Aust N Z J Obstet Gynaecol 2014;54:152–5.
38. Mackintosh N, Watson K, Rance S, et al. Value of a modified early obstetric warn-
ing system (MEOWS) in managing maternal complications in the peripartum
period: an ethnographic study. BMJ Qual Saf 2014;23:26–34.
39. Isaacs RA, Wee MY, Bick DE, et al. A national survey of obstetric early warning
systems in the United Kingdom: five years on. Anaesthesia 2014;69:687–92.