14 V O L U M E 9 I S S U E 1 2 0 1 3 infant

U P D A T E I N P R A C T I C E © 2013 SNL All rights reserved

Necrotising enterocolitis (NEC), aninflammatory disease of the bowel, is a

major cause of morbidity and mortality ininfants born before 32 weeks of gestationor with a birthweight less than 1500g1.

Over the past 20 years, despite significantadvances in neonatal care, the incidence ofNEC in very low birthweight (VLBW)infants has not changed markedly, pres-umably in part due to improved survivalrates. The mortality rate in confirmed NECremains greater than 20%2,3.

The NEC care bundle was developed inresponse to concerns from clinicians in theEast of England about a perceived increasein the incidence of NEC in their localunits. With the exact aetiological factorscontributing to the development of NECstill widely debated, the Perinatal Networkdecided to develop a network-wide projectacross all 17 of the East of England’s neo-natal units (FIGURE 1), using care bundlemethodology. The primary aim was to red-uce the incidence of NEC across the region.

Project lead and dietetic time was fundedby the Perinatal Network and managersleading local changes were offeredprotected time, as resources allowed.

Care bundlesCare bundles are a group of evidence-based interventions related to a disease orcare process that, when executed together,result in better outcomes than whenimplemented individually4. They are aquality improvement tool focusing on a

The development and implementation of a care bundle aimed at reducing theincidence of NECA care bundle to reduce the incidence of necrotising enterocolitis (NEC) was designed andimplemented across the 17 neonatal units in the East of England Perinatal Network with theultimate aim of reducing the incidence of NEC. The project called for significant changes inpractice across a wide geographical area and involved neonatal, midwifery and allied healthcareprofessionals. This article describes the care bundle development process and implementationstrategies required to bring about the changes involved. Data collection is continuing todetermine whether implementation has been successful in reducing the incidence of NEC.

Lynne RadboneBSc, RDLead Neonatal Dietitian, East of EnglandPerinatal Network lynne.radbone@eoescg.nhs.uk

Jennifer BirchMBChB, MRCPCH, MMedSciConsultant Neonatologist, Luton andDunstable University Hospitaljennifer.birch@ldh.nhs.uk

Michele UptonRGN, RM, ENB405Innovation Lead, East of England PerinatalNetworksmichele.upton@eoescg.nhs.uk

Keywords

necrotising enterocolitis; care bundle;change management; audit; compliance

Key points

Radbone L., Birch J., Upton M. Thedevelopment and implementation of acare bundle aimed at reducing theincidence of NEC. Infant 2013; 9(1): 14-19.1. Feeding practices and delivery of care,

aimed at reducing NEC rates, werestandardised across the East of England.

2. Ambitious improvements require theefforts of effective change agents whouse small cycles of change and regularfeedback to bring changes to fruition.

3. Adequate resources, planning forsustainability and open, frequentcommunication are vital for long-termproject success.

particular patient group or disease process.Care bundles encourage collaborativeworking and behavioural changes andallow for the development of new insightson care processes, resulting in improvedclinical and often improved organisationaloutcomes. Care bundle elements must beevidence-based and widely regarded asbeing best practice, while at the same timebeing simple to implement and part ofday-to-day practice.

Data collection on NEC incidence At the beginning of 2010 a data review wascarried out across the East of EnglandPerinatal Network using the Standardised

FIGURE 1 The East of England PerinatalNetwork comprises the Eastern andBedfordshire and Hertfordshire networkswith two units also belonging to the NorthEast London network.

Bedfordshire

HertfordshireNorth Essex

South Essex

Suffolk

Norfolk

Cambridgeshire

U P D A T E I N P R A C T I C E

V O L U M E 9 I S S U E 1 2 0 1 3 15infant

Electronic Neonatal Database (SEND) toestablish the incidence of NEC between2007 and 2009. In the absence of a nationaldefinition for NEC, and in order to capturea broad understanding of the number ofbabies treated for suspected or confirmedNEC, SEND data was reviewed using thecrude criteria of any baby with a SENDrecorded diagnosis of:■ Suspected NEC■ Confirmed NEC■ Having received metronidazole at any

point of the neonatal admission.Results showed that between 2007 and

2008 there was an increase of more than3% of all neonatal admissions fulfilling theabove criteria.

A multidisciplinary team (MDT) withneonatal, surgical, dietetic and allied healthprofessional representation from mostneonatal units within the network wasbrought together. As the care bundle wasdeveloped, additional expertise was soughtfrom relevant stakeholders includingmidwifery teams, lactation specialists anddepartmental leaders.

An initial scoping exercise was under-taken to understand the range in practicesrelating to the management and treatmentof NEC, the existence, scope and use of unitfeeding guidelines and milk preparation.

The findings of the scoping exerciseshowed significant variation across thenetwork in resources relating to dieteticsupport, the content and use of enteralfeeding guidelines as well as milk kitchenfacilities, milk preparation standards andfeeding practices. Variation in feedingpractice focused particularly on the timingof commencement of feeds, volume andfrequency of trophic feeds, the intervalbetween feeds, the rate of increasing feedvolume in the at-risk population and thetype of milk used for first feed5.

The care bundle development processcontinued with a robust literature reviewencompassing 15 factors implicated in thedevelopment of NEC (TABLE 1). Althoughhigh-level evidence was sought on eachtheme, only three were identified aselements that would meet the care bundlecriteria of being simple to implement,widely regarded as best practice and day-to-day interventions. These three elementswere: the use of maternal breast milk toinitiate feeding; the prevention ofinfection; and the use of standardisedfeeding strategies.

A retrospective case note review ofbabies with suspected or confirmed NEC

term and on discharge is more challenging.Therefore it was decided that a two-pronged approach, ensuring both early andlong-term provision of breast milk, shouldbe adopted. The final care bundle elementswere therefore:■ Facilitating the use of maternal breast

milk to initiate feeding through the com-mencement of early expression

■ Promoting the long-term use of maternalbreast milk through ongoing support forexpression and breastfeeding

■ Following a standardised enteral feedingguideline

■ Prevention of infection through the useof an aseptic non-touch technique(ANTT) when preparing milk feeds.

Strategies to support theimplementation of the care bundleUnsuccessful mobilisation or engagementof leaders is a key factor in the failure ofchange projects7. A recognised modeldescribed by Bevan presents change as twodivisions – anatomical change andphysiological change7.

Anatomical change can be described asthe structures (processes) employed torealise change, eg clear definition of theproject, a top down approach, correct scaleand pace of change and ensuringsustainability from the start. Physiologicalchange can be described as processes thatcreate a higher purpose and deepermeaning for staff, a connection with theirvalues and a creation of hope andoptimism about the future.

One or two ‘champions’ drawn predom-inantly from the neonatal nursing teamswere identified as key links for theteaching, implementation and change cycleprocesses in each unit. Lactation specialistsand infant feeding co-ordinators as well asmedical and midwifery colleaguessupported the champions in their role. Tofacilitate engagement and allow formultidisciplinary discussion of theevidence behind the chosen interventions,the project leader and neonatal dietitianresponsible for writing the enteral feedingguideline, undertook teaching sessionsacross the network units.

Nursing care plans and ‘top tips’ forexpressing breast milk were key tools tosupport implementation of the twoelements relating to expressing andbreastfeeding support. An ANTT milkpreparation prompt was designed to guidestaff through the milk preparation process

in 2009 was undertaken to look for themesfrom the East of England population. Thecriteria for inclusion in the case notereview were refined to any baby who had aminimum of both seven days intravenousantibiotics and seven days of being nil bymouth for suspected or confirmed NEC.Details of babies meeting these criteriawere identified from SEND and confirmedby each unit against their own SENDdatabase and local admission book. Adetailed case note review was then carriedout for each baby meeting the criteria.

Using these criteria there were a total of112 episodes of NEC in 102 babies in 2009,an incidence of 1.16 per 100 admissions6.Known and suspected predisposing factorsfor NEC were included within the review,eg gender, presence of a patent ductusarteriosis (PDA), presence of umbilicallines, the type of milk being fed at theonset of NEC, timing of blood trans-fusions, stool history and other factors. Noclear trends in these factors were identifiedin the East of England population.

The working group recognised that,while mothers of preterm babies dofrequently start expressing breast milk,maintaining the use of expressed breastmilk (EBM) or breastfeeding in the longer

■ Use of donor breast milk in prevention

of NEC

■ Use of maternal milk in prevention of NEC

■ Early vs late feeding

■ Trophic feeding

■ Fortifiers and milk additives

■ Blood transfusions

■ Enteral feeding regimens in prevention

of NEC

■ Reflux and the use of anti-reflux

medication

■ Supplementation of feeds

■ Use of probiotics

■ Feeding with a UAC/UVC in place

■ Kangaroo care

■ Resistant infection outbreak leading to

clusters of NEC in affected units

■ Oral lactoferrin for prevention of NEC

■ Role of calprotectin in diagnosing NEC

TABLE 1 Factors implicated in the develop-ment of NEC, the basis for a literature review.Terms: UAC = umbilical artery catheter, UVC = umbilical venous catheter.

for use in milk preparation areas. To aidsimplicity of use, the standardised enteralfeeding guidelines included two algor-ithms (APPENDIX 1). The standardisedenteral feeding guideline was developed bya specialist neonatal dietitian with thesupport of a MDT that included represent-ation from across the region. Guidance wasevidence-based as far as possible, butwhere evidence was insufficient or poorquality, guidance was based on regionalconsensus and national best practice. Thefirst algorithm detailed feed volume andadvancement rates based on the assessedrisk of NEC (high, moderate or standardbased on defined clinical criteria). Thesecond provided guidance on milk choiceand breast milk fortification.

To further support the change process,information folders, posters, electronicpresentations and monthly electronicnewsletters were produced. A letter wassent to chief executives, medical directorsand commissioners to outline the projectaims and to secure their support. The useof parent stories and a focus on reducingthe incidence of a devastating diseasethrough the provision of optimal,evidence-based feeding practices weresome of the ‘physiological’ aspects of thestrategies used.

Implementation of the care bundle tookplace during January and February of 2011following a formal regional launch.

Care bundle deliveryIt is widely accepted that active staffinvolvement is essential for qualityimprovement in an organisational setting8.All members of the MDT were involved inthe delivery of the care pathway. Thepathway commenced prior to delivery fora woman in threatened preterm labourand ended at discharge. Support forbreastfeeding continued throughout theinpatient stay and, in units wherecommunity teams existed, into the home environment.

Partnership working between neonataland maternity teams was apparent earlyon in the project. Encouragement of earlybreast milk expression was increasinglyinitiated by neonatologists, either at orprior to delivery, and was no longer seenas only a nursing or midwifery role.Neonatal nursing and medical teams alongwith dietetic colleagues agreed the initialfeeding plan based on clinical criteria,which was regularly re-assessed throughclinical review, and nursing observation of

feed tolerance. Lactation consultants havebeen key drivers for the implementation ofthe care bundle and in some units have led the initiative across neonatal andmaternity services.

Aims of the NEC care bundleThe primary aim of the project was toreduce the incidence of NEC in the East ofEngland. With the small number of casesinvolved, it was recognised thatdemonstrating this would be a statisticalchallenge. Due to the nature of theinterventions, a number of secondary aimswere identified which included: ■ To achieve high rates of compliance with

the four care bundle elements■ To measure the impact of the care bundle

elements on preterm breastfeeding rates■ To measure the impact of the bundle on

length of stay■ To measure parent satisfaction9.

Audit of complianceCompliance with the care bundle wasaudited throughout the first year ofimplementation. Ten randomly selectedbabies born at less than 35 weeks’ gestationwere audited each month by localchampions or network personnel using aconsistent network-wide audit form. An‘all or none’ approach was adopted, withthe expectation that every aspect of eachelement had to be met to be consideredfully compliant.

Units progressed to a three-monthlyaudit when 100% compliance was achievedin all four elements for two consecutivemonths. It was generally considered that

the bundle had been well embedded oncethis level of compliance had been reached.

At this point audit criteria were changedand compliance requirements heightenedwith the expectation that breast milkexpression would commence within sixhours of delivery to achieve fullcompliance along with the previouscriteria. This was part of a staged approachto improving practice and bringing it inline with best evidence6, while recognisingthe number of changes the initialimplementation criteria demanded. Asanticipated, the percentage compliancewith all four elements dipped while unitsworked towards achieving the strictercriteria. Three units were able to maintainthe high compliance rates despite theseincreased expectations.

The audit tool was used to measure unitcompliance with the bundle and toevaluate use of the network enteral feedingguideline. The audit tool ascertainedaccuracy on classification of babies into thethree NEC risk categories and adherence toguidance on feed management followingthis classification. Deviations from theguidance, both justified and non-justifiedwere recorded. The tool also supported theidentification of non-audited feedingtrends across the network, including timeto first feed and milk used for first feed.

The audit captured information on theuse of the specific breast milk expressionlog designed for the project and ondissemination of the regional breastfeedinginformation leaflet. Reasons for non-compliance with the breast milk expressionelements were also recorded.

U P D A T E I N P R A C T I C E

16 V O L U M E 9 I S S U E 1 2 0 1 3 infant

Perc

enta

ge c

om

plia

nce 70

80

90

100

60

50

40

30

20

10

0

Feb March April May June July Aug Sept Oct Nov Dec

FIGURE 2 Network compliance with all four care bundle elements, 2011.

The milk preparation audit toolcaptured information on whether thespecified ANTT was used to prepare thefeeds. This was self-audited by staff as feedswere prepared.

FIGURE 2 shows network percentage com-pliance with the care bundle (comprised ofall care bundle elements) in the first 11months of the implementation period. Thedecline in compliance after August reflectsthe transition of some units to the stricteraudit criteria for early expression.Throughout this time period, compliancewith the ANTT guidance element wasconsistently high while cross-networkcompliance with the standardised enteralfeeding guideline element increased steadilymonth-on-month throughout 2011.

Each unit received an individualisedend-of-year report detailing feedback onlocal compliance with the NEC care bundle since implementation in January2011 and providing specific feedback andpoints for practice on use of the enteralfeeding guideline.

The future of the care bundleThe care bundle has become a formalelement of practice across the region withelements incorporated into a user-friendlynutrition pathway (APPENDIX 1),incorporating all documentation relatingto the processes involved and providing aneasy auditable trail. This document istransferable from unit to unit and supportsa consistent approach to nutritional careacross the East of England. Early outcomesfor NEC rates, mortality and morbidity arecurrently being measured along withfeeding practice trends identified in year

one. These and the results of the secondaryaims of the care bundle will be publishedat a later date.

All four care bundle elements havebecome embedded practice within theneonatal units in the East of England.Wider health professional groups withinthe MDT are now undertaking rolestraditionally undertaken by specific groupsof staff. Implementation of the standard-ised care pathway and feeding guidelinesappear to have resulted in earlier, safer,more consistent feeding practices forpreterm infants. It has removed guessworkand enhanced staff confidence whenmaking decisions about feeding and hasled to the increased use of breast milk forthe initiation of feeding in moderate andhigh risk babies (FIGURE 3).

What has the care bundle projecttaught staff?Implementation of standardised practiceacross a network is possible. There iswillingness within individual units toaccept, embrace and contribute positivelyto changes in practice. Where cliniciansand local teams have identified this, thechange process is made considerably easier.However, to be successful, there has to berecognition that network projects need tobe led by individuals with dedicated time.They have to be thoroughly planned,include appropriate representation and becarried out within realistic time frames.Where appropriate, professional contactgroups for key disciplines should beestablished to harness the specificapproaches they offer a project, addingdepth to the work. Communication and

ongoing engagement are crucial aspects ofproject management with particularattention being given to keeping staffinformed of progress and developmentswithout a communication overload.

Training was delivered by the projectleaders and then disseminated across theunit by a locally-based champion. Thechampions were able to directly influencethe success of the project in their own unitby embracing and leading on the proposedchange strategy. This process offersmultiple rewards including: changebrought about by local individuals whounderstand the culture of their teams;opportunities for career development forleaders; recognition for individuals fortheir successes; and opportunities for unitsto benchmark themselves against eachother and share good practice.

Those who take on the rewardingchallenge of leading network change needpatience, flexibility and a supportive andpositive approach. Considerable time willbe spent advising local change agents andproviding essential and timely support tomaintain engagement. Adequate resources,planning for sustainability and open,frequent communication are vital for long-term project success.

References1. Maheshwari A., Corbin L., Schelonka R. Neonatal

necrotizing enterocolitis. Res Reps Neonatol

2011;1:39-53.

2. Chauhan M., Henderson G., McGuire W. Enteral

feeding for very low birth weight infants: reducing

the risk of necrotising enterocolitis. Arch Dis Child

Fetal Neonatal Ed 2008;93:F162-66.

3. Rees C.M., Eaton S., Pierro A. Trends in infant

mortality from necrotising enterocolitis in England

and Wales and the USA. Arch Dis Child Fetal

Neonatal Ed 2008;93:F395-96.

4. Haraden C. What is a bundle? Institute for

Healthcare Improvement; 2011. [Online]. Available

from: www.ihi.org/knowledge/Pages/

ImprovementStories/WhatIsaBundle.aspx

[Accessed: 6 Sept 2012].

5. Mitra S., Birch J., Upton M. Current neonatal

practice in relevance to the management of

necrotising enterocolitis in East of England. Acta

Paediatr 2010;99(Suppl 462):47-120.

6. Mitra S., Birch J., Upton M. Outcomes following NEC

in the east of England. Arch Dis Child Fetal Neonatal

Ed 2011;96:Fa26.

7. Bevan H., Plesk P., Winstanley L. Leading Large Scale

Change, A Practical Guide. NHS Institute for

Innovation and Improvement; 2011.

8. Wilkinson J., Powell A., Davies H. Are Clinicians

Engaged in Quality Improvement? London: Health

Foundation; 2011.

9. POPPY Steering Group. Family-centred Care in

Neonatal Units. A Summary of Research Results and

Recommendations from the POPPY Project. London:

NCT; 2009.

U P D A T E I N P R A C T I C E

V O L U M E 9 I S S U E 1 2 0 1 3 17infant

FIGURE 3 The use of expressed breast milk (EBM) for initiation of feeding: percentage of babiesreceiving EBM as first feed.

Unit level and risk category Four months postimplementation

12 months postimplementation

Level 3 – High 100 100

Level 3 – Moderate 76 91

Level 3 – Standard 35 60

Level 2 – High 67 70

Level 2 – Moderate 58 67

Level 2 – Standard 38 19

Level 1 – High 73 100

Level 1 – Moderate 69 83

Level 1 – Standard 37 32

U P D A T E I N P R A C T I C E

18 V O L U M E 9 I S S U E 1 2 0 1 3 infant

APPENDIX 1

development and implementation of a care bundle_layout 09/01/2013 10:00 Page 18

U P D A T E I N P R A C T I C E

V O L U M E 9 I S S U E 1 2 0 1 3 19infant

development and implementation of a care bundle_layout 09/01/2013 10:00 Page 19