Clinical Research

Nutrition in Clinical Practice

Volume 0 Number 0
Screening Tool for the Assessment of Malnutrition in July 2020 1–7
© 2020 American Society for
Pediatrics (STAMP) in the Electronic Health Record: A Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10562
Validation Study [Link]

Michelle Reed, MS, RD, CSP, LDN, CDE1 ; Kathleen Mullaney, MSN, APRN,
ACCNS-P, CRNI2 ; Christy Ruhmann, RD, BS1 ; Peter March, MSN, RN, NEA-BC1 ;
Virginia H. Conte, MSN, RN, CPN1 ; Lore Noyes, MS, RD, CSP, LDN, CNSC1 ;
and Mark Bleazard, MSN, APRN, ACCNS-P, CRN1

Abstract
Background: The impact of malnutrition on pediatric patients in the acute care setting is significant. Hospitalized patients with
malnutrition have been shown to have poor clinical outcomes. Nutrition screening is the first critical step in identifying and treating
malnutrition. Although several pediatric nutrition screening tools exist, none incorporate both electronic health record (EHR)
compatibility and the recommended indicators of pediatric malnutrition, a gap recently identified in a systematic review by the
Academy of Nutrition and Dietetics. The aim of this study was to prove the validity of a new version of Screening Tool for the
Assessment of Malnutrition in Pediatrics (STAMP), EHR-STAMP, modified for incorporation into the EHR and inclusion of
updated pediatric malnutrition indicators. Methods: An interprofessional team modified the existing STAMP for integration into
the EHR. Audits were performed by the research dietitian to assess accuracy and provide feedback for continuous improvement
of the tool design. Results: A total of 3553 pediatric inpatients were studied from August 2017 to May 2019. Accuracy, sensitivity,
and specificity improved with each modification to the EHR-STAMP. The final version of the EHR-STAMP found 85% accuracy,
89% sensitivity, and 97% specificity, with a positive predictive value of 60% and a negative predictive value of 94%. Conclusion:
The EHR-STAMP is a highly reliable tool in the screening of nutrition risk for pediatric hospitalized patients. The tool is easy to
use, EHR compatible, and incorporates the current indicators recommended for assessing pediatric malnutrition. (Nutr Clin Pract.
2020;0:1–7)

Keywords
child; electronic health record; malnutrition; nutrition screening; nutrition status; pediatrics

Background chomotor and mental development, which can eventually

lead to behavioral problems in childhood and adolescence,
Malnutrition in neonatal and pediatric patients imposes such as attention deficit hyperactivity disorder, aggressive
significant negative impacts, especially for those hospital- behavior, and externalizing behaviors.12
ized in acute care settings. Children with malnutrition tend Children with chronic illness, such as asthma and di-
to have poor clinical outcomes with higher morbidity and abetes, are at higher risk for malnutrition, as are those
mortality, increased length of hospital stay, and delayed
wound healing.1-10 According to the World Health Orga-
nization (WHO), malnutrition is defined as the “cellular
imbalance between the supply of nutrients and energy and From the 1 Nemours/Alfred I. duPont Hospital for Children,
Wilmington, Delaware, USA; and the 2 Nemours/Alfred I. duPont
the body’s demand for them to ensure growth, maintenance,
Hospital for Children, The Children’s Hospital of Philadelphia,
and specific function.”11 Philadelphia, Pennsylvania, USA.
Protein and calorie deficiencies result in impaired im-
Financial disclosure: None declared.
mune response and alter cytokine and chemokine produc-
Conflicts of interest: None declared.
tion, contributing significantly to increased inflammation
and infection during stress or illness. Infants and children Corresponding Author:
Michelle Reed, MS, RD, CSP, LDN, CDE, Department of Clinical
who present to the hospital with malnutrition have an even
Nutrition, Nemours/Alfred I. duPont Hospital for Children,
higher risk of complications from acute illness. Long-term Wilmington, DE, USA.
consequences of infant malnutrition include delayed psy- Email: [Link]@[Link]
2 Nutrition in Clinical Practice 0(0)

with congenital heart defects or malabsorption conditions pediatric nutrition screening tools were identified: PYMS,
(eg, cystic fibrosis) or patients receiving chemotherapy.3 STRONG(kids), and STAMP.13,17,18 It was determined that
Additionally, certain acute conditions, such as burns and STAMP was the ideal screening tool to adopt, as it was
severe trauma, can negatively affect a child’s growth and concise, objective, and easy to use for the clinical nursing
nutrition status for up to 2 years.12 Early assessment of staff who routinely complete a nutrition screen on admis-
nutrition status and provision of appropriate nutrition sup- sion. However, the clinical nutrition department recognized
port are extremely important and associated with decreased that incorporating STAMP into the Epic (Verona, WI) EHR
morbidity and mortality.1 To prevent malnutrition in the would require modifications to the tool. Thus, following a
hospitalized child, nutrition risk needs to be evaluated at framework of evidence-based practice, an interprofessional
the time of admission and then at regular intervals until work group was created to modify and adopt the tool, using
discharge.12,13 quality improvement methodologies.
Although it is widely recognized that early identification The STAMP tool assigns nutrition risk based on 3
of malnutrition risk is important in the pediatric acute parameters: nutrition implication of the patient’s medical
care setting, standardization in screening practices does diagnoses, recent nutrition intake, and anthropometrics.13
not exist. Malnutrition is often unrecognized by clini- To be compatible with the EHR, the greatest modifications
cians, and its clinical impact is largely underestimated.14 had to be made to the anthropometrics parameter, as the
In fact, up to half of the pediatric patients admitted to original tool used the STAMP child weight and height
the hospital setting have some degree of malnutrition, perentile tables to identify extremes in proportionality (un-
suggesting the need for accurate and timely evaluation.1,15 derweight or overweight).13 Although a percentile indicates
Several pediatric nutrition screening tools have been de- an individual’s growth parameter rank among a population,
veloped to facilitate nutrition risk assessment and care it does not identify deviation from population norm, and
allocation. In February 2020, the Academy of Nutrition it can not quantify extreme values. Conversely, a z-score
and Dietetics (Academy) published a systematic review is a statistical measurement quantifying standard deviation
of pediatric nutrition screening tools.16 Three well-studied from the norm, either in the positive or in the negative
inpatient tools with moderate validity and moderate-to-high direction, and it is a more sensitive indicator of growth
reliability were identified: Paediatric Yorkhill Malnutrition failure compared with percentiles.
Score (PYMS), Screening Tool for the Assessment of Mal- The interprofessional work group used the
nutrition in Pediatrics (STAMP), and Screening Tool for Academy/ASPEN malnutrition indicators to modify
Risk on Nutritional status and Growth (STRONGkids).16 the anthropometrics section, using weight-for-length z-
Although STAMP was the only tool supported by grade scores in patients aged 29 days to 24 months and body
I (good/strong) evidence, authors noted that the tool used mass index–for-age z-scores in patients aged 2–20 years
percentiles for anthropometric data, which may present a old.1 For neonates (aged ≤28 days), WHO definitions
barrier to electronic health record (EHR) compatibility. for low, very low, and extremely low birth weights were
Authors concluded it may be favorable to update existing used to assign risk.19 The work group piloted the original
validated tools, such as STAMP, to integrate them into EHR-STAMP tool on a subset of patients in January 2017.
the EHR.16 The present study aimed to assess the validity In August 2017, the tool was implemented throughout the
of EHR-STAMP, an updated version of STAMP designed entire institution. Education for clinical nurses included
for EHR integration, and incorporation of the pediatric presentations within each practice area. Department-wide
malnutrition indicators identified by the Academy and the education was disseminated using the “nursing shared
American Society for Parenteral and Enteral Nutrition governance” structure. Over 2 years of testing, a series
(ASPEN).1 of modifications were made with feedback from clinical
nurses and dietitians using “plan-do-check-act” cycles
(PDCA). The first major EHR-STAMP modification was
Methods made in April 2018, to automate the anthropometric
In 2016, the clinical nutrition department at a 200-bed risk assignment. Design modifications were completed to
children’s hospital, located in the mid-Atlantic region of the allow the tool to automatically retrieve anthropometric
United States, identified a need to update the institution’s data included in routine nursing documentation, calculate
nutrition screening process. At the time, nutrition screening z-score, and assign a risk score. Subsequent modifications
used resources from both clinical nursing and clinical nutri- to the tool focused on refining verbiage around nutrition
tion departments, relied heavily on subjective information intake (November 2018) and diagnosis implication (April
and anthropometric data with poor sensitivity for malnu- 2019) to reduce user error. The finalized version of the
trition, and neglected to assign nutrition risk. A review of EHR-STAMP tool (Figure 1) was implemented into
pediatric nutrition screening tools was conducted as part of practice in April 2019, replacing the institution’s previous
a monthly clinical nutrition journal club, and 3 validated nutrition screening process.
Reed et al 3

Parameter Score 0 Score 1 Score 2 Score 3

Anthropometrics

Neonates aged ≤ 28 ≥2500 1500 to 2499 1000 to 1499 <1000

days birth weight (in
grams)

Infants and children > -1.00 z-score -1.00 to -1.99 -2.00 to -2.99 -3.00 or greater
aged <24 months,
weight-for-length
z-score

Children and > -1.00 -1.00 to -1.99 -2.00 to -2.99 -3.00 or greater
adolescents, aged
2–20 years, BMI-for-
age z-score
Nutrional Intake Usual intake or N/A Decreased intake Minimal or no
decreased intake for greater than 3 intake for >3 days
for <2 days prior to days prior to prior to admission
admission admission
Diagnosis Nutrion No nutrion risk N/A Possible Nutrion Definite Nutrion
Implicaon associated with Risk Risk
diagnosis
Biliary atresia,
Cle lip/palate, bowel obstrucon,
celiac disease, burn,
diabetes, dysphagia, Crohn’s/ulcerave
heart disease, colis, cysc fibrosis,
mulple food eang disorder,
allergies, failure to thrive,
neuromuscular intesnal failure,
conditions (cerebral ketogenic diet,
palsy, Duchenne kidney disease or
muscular dystrophy, failure, liver disease
spinal muscular or failure, major
atrophy), pressure trauma/surgery,
ulcer stage 1, metabolic disorders,
prematurity (< 37 oncology/bone
weeks, up to 1 year marrow transplant,
old), tube feeding pressure ulcer
stages 2-4, TPN
dependence

Total Score 0–1 2–3 4–9

Implicaon Low Risk Moderate Risk High Risk

Rescreened by Rescreened by Seen by diean

diean in 7 days diean in 3 days within 48 hours

Figure 1. EHR-STAMP tool. BMI, body mass index; PN, parenteral nutrition.

Data Collection patient’s growth chart. A single dietitian performed the

review of records to maintain reliability. From the dietitian’s
Audits were routinely performed by the project’s lead dieti- chart review, a verified score was calculated and compared
tian to assess for screening accuracy, provide feedback to with the score completed by the clinical nurse on admission.
clinical practice areas, and identify opportunities for nursing Scoring accuracy was determined based on whether the
education. The lead dietitian reviewed charts of recently nursing nutrition risk score on admission matched the dieti-
admitted patients to identify nutrition risk of diagnoses and tian’s verified score upon chart review (ie, a patient identified
intake prior to admission (based on admitting physician as low risk by the lead dietitian was also screened low risk by
documentation) and confirm anthropometric scoring by the nurse). Screening errors were further assessed to identify
assessing accuracy of anthropometric measurements in the opportunities for improvement in the tool and informed
4 Nutrition in Clinical Practice 0(0)

Figure 2. Patient distribution across medical/surgical areas.

the series of tool modifications previously discussed. patients from each series of the modifications (Table 1).
Errors included erroneous anthropometric measurements, The mean age was 7.14 ± 6.12 years; 46.7% (n = 1660)
anthropometric z-score or birth-weight risk assignment, were female. Patients were included across 5 different med-
intake risk assignment, and diagnosis risk assignment. ical/surgical areas (Figure 2). The distribution of patients
The decision to conduct a formal chart audit was made, in this study was reflective of the typical distribution of
and an application to the organization’s institutional review patients regularly admitted to the inpatient setting in the
board (IRB) determined the study to be exempt from organization. A total of 387 patients were included in the
IRB oversight. For the purposes of the IRB-approved tool study of the finalized EHR-STAMP tool: mean age 7.75 ±
analysis, authors limited the study sample to those patients 6.14 years; 45.5% (n = 176) were female. In the final sample,
admitted to a pediatric medical/surgical inpatient practice 71.3% (n = 276) of patients were categorized as low risk,
area (see details of practice areas in Figure 2). Patients 19.9% (n = 77) as moderate risk, and 8.8% (n = 34) as high
admitted to intensive care units (cardiac, neonatal, and risk.
pediatric) and cardiac stepdown units were excluded. Accuracy, sensitivity, and specificity data are presented
in Table 1, with data for the total sample in the first
Statistical Analysis column, followed by each iteration of the modification
process in chronological order, from the original tool to
Accuracy, sensitivity, and specificity were analyzed by com-
the finalized EHR-STAMP. Results over time show stepwise
paring the nursing nutrition risk score on admission with
improvements in accuracy, sensitivity, and specificity with
the verified nutrition risk score, as calculated by the lead di-
each modification to EHR-STAMP. Positive and negative
etitian. Positive predictive value (PPV) was analyzed by as-
predictive values were calculated for the finalized EHR-
sessing the presence of malnutrition among patients scored
STAMP tool as 0.60 and 0.94, respectively.
as high risk on admission. Negative predictive value (NPV)
In the total patient sample (N = 3553), errors for high-
was analyzed by assessing the absence of malnutrition
and moderate-risk patients were further assessed because
among patients who scored moderate and low risk on ad-
of concern for delays in patient care when under scoring
mission. Malnutrition was assessed using Academy/ASPEN
patients. Among the high-risk patients who were under
malnutrition indicators by the lead dietitian.1
scored on admission (n = 79), 35.4% had a nutrition consult
or referral ordered, prompting dietitian intervention. Of
Results the moderate-risk patients under scored on admission (n
A total of 3553 patients were included in the study, ranging = 335), 66.6% were discharged to home before a nutrition
in age from <1 month to 20 years. The total sample included rescreen would have been prompted. Of the patients who
Reed et al 5

Table 1. EHR-STAMP Validation Results Across PDCA Cycles.

Original Final EHR-

Pooled data tool Tool, edit 1 Tool, edit 2 STAMP
Measure (N = 3553) (n = 1725) (n = 1096) (n = 351) (n = 381)

Accuracy 0.77 0.71 0.81 0.81 0.85

Sensitivity 0.74 0.69 0.76 0.79 0.89
Specificity 0.95 0.93 0.97 0.97 0.97
Positive 0.60
predictive
Negative 0.94
predictive

EHR, electronic health record; STAMP, Screening Tool for the Assessment of Malnutrition in Pediatrics.
Bolded values are results for the finalized EHR-STAMP.

remained admitted for >3 days (n = 112), 25.9% had a An additional strength of the current study is its unique
nutrition consult or referral ordered within the first 3 days analysis of the relationship between screening error and
of admission. Therefore, of the 3553 patients included in the delays in patient care. All widely used and studied pediatric
study, only 3.8% (n = 134) were at risk of delayed inpatient nutrition screening tools have inherent error risk because
nutrition care because of screening errors. of the necessity of a healthcare provider to perform the
screen. This is the first study to assess how many patients are
negatively affected by screening error. In combination with
provider-ordered nutrition consults and referrals, the EHR-
Discussion STAMP effectively allocated resources, with minimal risk
The STAMP is a quick, simple, and reliable tool for assess- for delayed patient care. It is important to note that the tool’s
ing and reassessing nutrition risk in the acute care pediatric sensitivity was not at the expense of its specificity. Although
setting. The aim of this study was to prove the validity of a the EHR-STAMP was able to identify patients at high
new version, EHR-STAMP, modified to be integrated into nutrition risk, its high specificity and negative predictive
the EHR and to incorporate updated recommendations on value prove that the tool does not falsely overassign risk to
pediatric malnutrition indicators. do so.
The EHR-STAMP showed higher sensitivity (0.89) and Several factors contributed to the EHR-STAMP work
specificity (0.97) than the original STAMP (0.63–0.81 and group’s success. The interprofessional work group included
0.36–0.91, respectively), with positive and negative predic- representation from clinical nutrition, nursing, and infor-
tive values comparable with those of the original STAMP matics. This allowed the team to cohesively address the
validation study (PPV, 0.60 vs 0.55; NPV, 0.94 vs 0.95, concerns of clinical nurses while ensuring that the tool was
respectively).13,17,20 Using the validity algorithm published effectively meeting clinical nutrition screening needs in an
in the Academy’s recent systematic review of pediatric evidence-based approach. Recruitment of an informatics
nutrition screening tools, the EHR-STAMP tool has a senior analyst on the team allowed for in-the-moment dis-
moderate degree of overall validity, comparable with that cussions about EHR capabilities and innovative problem-
of STAMP, as well as PYMS and STRONG(kids).16 solving. In addition to creating a highly effective interpro-
A strength of the current tool is that it is based on the fessional work group, this project was initiated at the same
well-designed, previously validated STAMP but modified to time as a hospital-wide initiative to improve the process
be relevant in the era of widespread EHR adoption. The of obtaining accurate heights and weights on admission.
STAMP was designed to incorporate information clinical From 2016 to 2019, completion of weights upon admission
nurses would routinely obtain on admission into the nutri- increased from 77.7% to 95.2%, and completion of heights
tion screening process, thus making the tool quick and easy upon admission increased from 69.3% to 91.3%. This had
to perform. The interprofessional work group expanded on a direct effect on the success of EHR-STAMP implemen-
the original design by creating a tool that automatically tation, as the tool is dependent on timely and accurate
retrieved anthropometric data from the patient’s chart and documentation of anthropometric data on admission.
assigned it a score, thus automating one-third of the tool. To the authors’ knowledge, this is one of the largest
Another strength of the tool is its creation in the Epic plat- study cohorts for a pediatric nutrition screening validation
form. Epic is one of the leading EHR providers, controlling study. The cohort (n = 387) used to assess the validity of
28% of the US acute care hospital market share in 2018.21 the finalized EHR-STAMP was larger than the original
6 Nutrition in Clinical Practice 0(0)

STAMP validation cohort (n = 238).13 The study cohort whether these results are replicable in other pediatric acute
included patients in a wide age range and with diverse care hospital settings and whether the tool can be validated
medical needs, supporting its applicability to other acute in other settings (eg, intensive care, ambulatory care).
care hospital settings serving medical and surgical pediatric
patients. Acknowledgments
Catherine Haut, DNP, CPNP-AC, FAANP, Helen McCarthy,
Limitations PhD, RD, Elizabeth Froh, PhD, RN.

Although the EHR-STAMP tool demonstrated comparable

Statement of Authorship
validity and improved accuracy, sensitivity, and specificity
when compared with the original STAMP tool, limita- M. Reed, K. Mullaney, C. Ruhmann, P. March, V. H.
tions to the generalization of these findings do exist. This Conte, and L. Noyes contributed to the conception and
manuscript contains the first published findings of the new design of the research. M. Reed, K. Mullaney, C. Ruhmann,
EHR-STAMP, which incorporates significant modifications V. H. Conte, L. Noyes, and M. Bleazard drafted and
from the original validated tool. Further research is needed critically revised the manuscript. P. March critically revised
to provide additional validity and prove reproducibility of the manuscript. All authors contributed to the acquisition,
this study’s results. This study presents findings from a analysis, and interpretation of the data, agree to be fully
large, single-center, pediatric institution in the mid-Atlantic accountable for ensuring the integrity and accuracy of the
region within the United States. The single-center sample work, and read and approved the final manuscript.
could represent bias, although authors did make an effort to
include a broad patient population over a large time frame References
to minimize bias. 1. Becker P, Carney LN, Corkins MR, et al. Consensus statement of the
Additionally, our study did not include patients in the Academy of Nutrition and Dietetics/American Society for Parenteral
and Enteral Nutrition: indicators recommended for the identification
critical care setting. Future research should focus on val-
and documentation of pediatric malnutrition (undernutrition). Nutr
idating the EHR-STAMP in this population. As with the Clin Pract. 2015;30(1):147-161.
original STAMP, validity of the EHR-STAMP is contingent 2. Beer SS, Juarez MD, Vega MW, Canada NL. Pediatric malnutrition:
upon proficiency in completing each component of the tool, putting the new definition and standards into practice. Nutr Clin Pract.
and the unavoidable risk of user error presents an inherent 2015;30(5):609-624.
3. Mehta NM, Corkins MR, Lyman B, et al. Defining pediatric malnu-
limitation. Institutions looking to adopt the EHR-STAMP
trition: a paradigm shift toward etiology-related definitions. JPEN J
should establish educational resources to confirm and main- Parenter Enteral Nutr. 2013;37(4):460-481.
tain ongoing competencies to ensure validity. Furthermore, 4. Wonoputri N, Djais JT, Rosalina I. Validity of nutritional screening
institutions adopting the tool should consider performing tools for hospitalized children. J Nutr Metab. 2014;2014:143649.
audits and providing feedback on specific incidences of tool 5. Moeeni V, Walls T, Day AS. The STRONGkids nutritional risk
screening tool can be used by paediatric nurses to identify hospitalised
completion error, especially when it applies to those patients
children at risk. Acta Paediatr. 2014;103(12):e528-e531.
who were underscored and thus put at risk of delayed or 6. Bohl DD, Shen MR, Mayo BC, et al. Malnutrition predicts infectious
missed patient care. and wound complications following posterior lumbar spinal fusion.
Spine (Phila Pa 1976). 2016;41(21):1693-1699.
Conclusions 7. Galera-Martínez R, Moráis-López A, Rivero de la Rosa MD, et al.
Reproducibility and inter-rater reliability of 2 paediatric nutritional
The results of this study reveal that EHR-STAMP is a screening tools. J Pediatr Gastroenterol Nutr. 2017;64(3):e65-e70.
reliable, sensitive, and specific nutrition screening tool in 8. Rub G, Marderfeld L, Poraz I, et al. Validation of a nutritional
screening tool for ambulatory use in pediatrics. J Pediatr Gastroenterol
the hospitalized pediatric population. The EHR-STAMP
Nutr. 2016;62(5):771-775.
directly answers the Academy’s recent call for pediatric 9. Sermet-Gaudelus I, Poisson-Salomon AS, Colomb V, et al. Simple pe-
screening tools that are not only EHR integrated but also diatric nutritional risk score to identify children at risk of malnutrition.
include the current recommended indicators for identifying Am J Clin Nutr. 2000;72(1):64-70.
pediatric malnutrition.16 To the authors’ knowledge, this is 10. Bunting J, Weaver LT. Anthropometry in a children’s hospital: a study
of staff knowledge, use and quality of equipment. J Hum Nutr Diet.
the first pediatric nutrition screening tool to incorporate
1997;10(1):17-23.
the Academy/ASPEN pediatric malnutrition indicators and 11. de Onís M, Monteiro C, Akré J, Glugston G. The worldwide
publish validation results. The EHR-STAMP is easy to per- magnitude of protein-energy malnutrition: an overview from the
form and readily integrated into nursing workflow and ap- WHO Global Database on Child Growth. Bull World Health Organ.
propriately allocates clinical nutrition resources to prioritize 1993;71(6):703-712.
12. Joosten KF, Hulst JM. Malnutrition in pediatric hospital patients:
those patients at highest nutrition risk. Embedding the tool
current issues. Nutrition. 2011;27(2):133-137.
in Epic facilitates easy adoption of EHR-STAMP across 13. McCarthy H, Dixon M, Crabtree I, Eaton-Evans MJ, McNulty H. The
a wide range of pediatric institutions, as Epic is a leading development and evaluation of the Screening Tool for the Assessment
EHR provider. Further studies are needed to determine of Malnutrition in Paediatrics (STAMP©) for use by healthcare staff.
J Hum Nutr Diet. 2012;25(4):311-318.
Reed et al 7

14. Aadal L, Mortensen J, Nielsen JF. Weight reduction after severe brain 18. Huysentruyt K, Devreker T, Dejonckheere J, De Schepper J, Van-
injury: a challenge during the rehabilitation course. J Neurosci Nurs. denplas Y, Cools F. Accuracy of nutritional screening tools in
2015;47(2):85-90. assessing the risk of undernutrition in hospitalized children. J Pediatr
15. Campanozzi A, Russo M, Catucci A, et al. Hospital-acquired Gastroenterol Nutr. 2015;61(2):159-166.
malnutrition in children with mild clinical conditions. Nutrition. 19. World Health Organization. International Classification of Diseases
2009;25(5):540-547. for Mortality and Morbidity Statistics (11th Revision). 2018. Accessed
16. Becker PJ, Gunnell Bellini S, Wong Vega M, et al. Validity and reli- March 4, 2020. [Link]
ability of pediatric nutrition screening tools for hospital, outpatient, 20. Thomas PC, Marino LV, Williams SA, Beattie RM. Outcome of
and community settings: a 2018 Evidence Analysis Center systematic nutritional screening in the acute paediatric setting. Arch Dis Child.
review. J Acad Nutr Diet. 2020;120(2):288-318.e2. 2016;101(12):1119-1124.
17. Gerasimidis K, Keane O, Macleod I, Flynn DM, Wright CM. A four- 21. Coray T, Warburton P. US Hospital EMR Market Share 2019, Signif-
stage evaluation of the Paediatric Yorkhill Malnutrition Score in a icant Movement in Every Market Sector. 2019. Accessed March 16,
tertiary paediatric hospital and a district general hospital. Br J Nutr. 2020. [Link]
2010;104(5):751-756. 2019/1454.