barriers and facilitators towards implementing the sepsis

ORIGINAL RESEARCH Open Access

Barriers and facilitators towardsimplementing the Sepsis Six care bundle(BLISS-1): a mixed methods investigationusing the theoretical domains frameworkNeil Roberts1* , Guy Hooper1, Fabiana Lorencatto2, Wendell Storr1 and Michael Spivey1

Abstract

Background: The ‘Sepsis 6’, a care bundle of basic, but vital, measures (e.g. intravenous fluid, antibiotics) has beenimplemented to improve sepsis treatment. However, uptake has been variable. Tools from behavioral sciences, suchas the Theoretical Domains Framework (TDF) may be used to understand and address such implementation issues.This study used a behavioral science approach to identify barriers and facilitators towards Sepsis Six implementationat a case study hospital.

Methods: Semi-structured interviews based on the TDF were conducted with a sample group of consultants, juniordoctors and nurses from Emergency Department, Medical and Surgical Admissions, to explore barriers/facilitatorsto Sepsis Six performance. Transcripts were analyzed following the combined principles of content and frameworkanalysis. Emerging themes informed a questionnaire to explore generalizability and importance across a sample of261 stakeholders. Median importance and agreement ratings for each theme were calculated overall and for eachrole and clinical area. These were used to identify important barriers and important facilitators as targets forperformance improvement.

Results: No new belief statements were discovered and data saturation was deemed achieved after 10 interviews.1699 utterances were coded into 64 belief statements, then collated into a 51-item questionnaire. 113questionnaire responses were obtained (44.3% response rate). Important barriers included insufficient audit andfeedback, poor teamwork and communication, concerns about using the Sepsis Six in certain patients, insufficienttraining, and resource concerns. Facilitators included confidence in knowledge and skills, beliefs in overall benefitsof the bundle, beliefs that identification and management of septic patients fell within everyone’s role, and thatregular use of the bundle made it easier to remember. Some beliefs were applicable for the entire group, otherswere specific to particular staff groups.

Discussion and Conclusions: A range of barriers and facilitators towards Sepsis Six performance across differentstaff groups were systematically identified using a theoretically-informed approach. This can inform development oftargeted performance improvement interventions.

Keywords: Sepsis, Theoretical domains framework, Barriers, Facilitators, Quality improvement, Implementation

* Correspondence: [email protected] of Critical Care and Anaesthesia, Royal Cornwall Hospital,Truro, Cornwall TR1 3LJ, UKFull list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Roberts et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2017) 25:96 DOI 10.1186/s13049-017-0437-2

http://crossmark.crossref.org/dialog/?doi=10.1186/s13049-017-0437-2&domain=pdf

http://orcid.org/0000-0003-3518-6813

mailto:[email protected]

http://creativecommons.org/licenses/by/4.0/

http://creativecommons.org/publicdomain/zero/1.0/

BackgroundSepsis remains a global issue, with timely recognitionand treatment crucial to outcome. Recent internationalconsensus emphasizes importance of early identifica-tion of organ dysfunction in patients with infection [1].Timely antibiotic administration in septic patients hasbeen adopted as a national standard against which UKhospitals are measured, with performance linked to fi-nancial reward or punishment [2]. Indeed, modern carestandards, with early antibiotics and fluid resuscitation,show significant mortality benefit compared to previousresearch [3–7]. However, many patients still die fromsepsis around the world each year, with internationalcare standards rarely achieved in full [8–10]. One strat-egy adopted to improve this is using a simplified carebundle, the ‘Sepsis 6’ (Table 1), which demonstrates in-creased compliance and an association with reducedmortality compared to full Surviving Sepsis Campaigncare bundles [11].Performing the Sepsis Six requires a range of behav-

iors to be performed by multiple individuals at differentorganizational levels (e.g. nurse identifies unwell pa-tient, junior doctor diagnoses sepsis, prescribes bundle,escalates patient to consultant and performs blood cul-tures, nurse administers oxygen, fluids and antibiotics.).Previous research identifies it as a complex ‘trajectoryof workflow’, requiring prioritization and coordination,prone to operational failure [12]. A range of cultural,contextual and behavioral determinants are likely to influ-ence implementation and result in variation in practicewithin and across hospitals. It is thus critical that qualityimprovement initiatives consider and address the broadspectrum of potential influences on implementation[13, 14]. However, systematic reviews identify that qual-ity improvement initiatives in both emergency medicineand antimicrobial stewardship often fail to considersocio-behavioral factors influencing clinical decisionmaking and practice [15, 16].Clinical practice is a form of human behavior, and

may thus be understood using theory and tools frombehavioral sciences [17]. Theory provides a replicable,generalizable framework through which to understanddeterminants of behavior [18]. The complexity and var-iety of available behavioral theories has posed a barrierto their use by non-specialists [19]. The Theoretical

Domains Framework (TDF) synthesizes key constructsfrom 33 theories relevant to healthcare professional be-havior change into 14 theoretical ‘domains,’ representingthe range of possible behavioral determinants, from‘knowledge’ to ‘social influences’ and ‘environmentalcontext and resources’ [18, 20, 21]. The TDF has beenapplied across a range of clinical contexts (e.g. anti-biotic prescribing, transfusion) to systematically identifybarriers/facilitators to healthcare professional behaviorchange (i.e. to conduct a ‘behavioral diagnosis’ to iden-tify ‘what’ needs to change) [22–24]. To inform subse-quent intervention design, TDF Domains have alsobeen mapped to individual Behavioral Change Tech-niques (BCTs), enabling selection of BCTs that arelikely to target identified barriers/facilitators [25, 26].Recent work has used the TDF to both analyze and

refine a pre-existing quality improvement interventionaround Sepsis Six implementation in a single hospital.This identified several themes regarding barriers/facili-tators to Sepsis Six performance: ‘Knowing what to doand why,’ ‘risks/benefits,’ ‘working together,’ ‘empowermentand support,’ and ‘staffing levels’ [27, 28]. Ethnographicwork identifies the need for a systematic theory-based ap-proach towards analyzing the complex nature of thisprocess, for example differences in barriers and facilitatorsbetween roles or clinical areas [12]. The increasing eco-nomic squeeze on healthcare, and the large resource de-mands of a critically ill patient, dictate not only that basicearly interventions such as the Sepsis Six are performedwell, but that quality improvement is focused, effective andefficient. There have been a limited number of studiesadopting a behavioral and social science approach to under-standing implementation issues in sepsis care, and there isa need for further studies to explore generalizability andcontribute to the limited body of knowledge in this area.Our study therefore aims to apply the TDF to system-

atically identify key barriers and facilitators towardsSepsis Six performance, and conduct a ‘behavioral diagno-sis’, prior to intervention development, at a different singlehospital. More broadly, the study aims to demonstrate a‘worked example’ of a replicable method for usingbehavioral theory in quality improvement processeswithin emergency medicine, and to enable systematicanalysis of where problems lie for different roles andclinical areas.

MethodsDesign and settingMixed-methods, two-phased study: 1) Semi-structuredinterviews based on the TDF, conducted with a sub-sample of relevant healthcare professionals to identifykey barriers/facilitators to Sepsis Six performance; and2) Questionnaires to explore generalizability and import-ance of identified barriers/facilitators.

Table 1 Sepsis Six care bundle [9]

Within the first hour of recognition of sepsis:

- Measured lactate/hemoglobin- Urine output- Blood cultures- Antibiotics- Oxygen- Intravenous fluids

Roberts et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2017) 25:96 of 18

SettingAcute areas of a 760-bed district general hospital inEngland: Emergency Department (ED), adult medicaladmissions (MAU) and adult surgical admissions units(SAU). Unpublished local audits of Sepsis Six perform-ance highlight low compliance in this hospital (0–20%septic patients receiving all bundle components withinone hour) [29].

Ethical approvalLocal Research and Development department approvalobtained confirming Service Development Project status,therefore not requiring formal ethics committee review.

Semi-structured interviewsParticipantsIn order to investigate barriers/facilitators from the per-spective of the range of clinical staff involved in Sepsis Siximplementation, participants were purposively sampledfrom relevant stakeholders groups, including: registerednurses, junior-level doctors, or consultant-level doctorsworking in ED, MAU or SAU.Potentially eligible participants were identified by the

lead investigators (NR,GH), both doctors working on thecritical care unit at the study hospital. Potential partici-pants were approached over a 6-week timeframe, eitherin person or by e-mail. In line with previous studiesusing the TDF, a minimum initial sample of ten partici-pants was proposed for full data analysis. An eleventhparticipant was then analyzed, and if new beliefsemerged, sampling continued until data saturation wasachieved (i.e. no new themes identified) [24, 30]. Thir-teen participants were interviewed initially, with a strati-fied sample of one consultant, junior doctor and nurseeach from ED, MAU and SAU selected as part of the ini-tial ten analyzed transcripts, with further participants se-lected at random.

MaterialsA topic guide consisting of 29 questions to elicit beliefsabout Sepsis Six performance was designed based on theTDF. It included at least one question relating to eachdomain. Table 2 lists sample questions from this studyfor each domain. The topic guide was developed by 3critical care physicians with expertise in sepsis and ahealth psychologist with TDF experience. The question-naire was piloted with nurses, junior doctors and consul-tants from other clinical areas within the study hospital.The topic guide was revised to clarify wording and isavailable as Additional file 1.

ProcedureParticipants were invited either in person or by email.Written consent was obtained after a briefing regarding

the study’s purpose. Interviews took place either in per-son in a private location, or by telephone, by a trainedinterviewer (NR, GH), and were digitally recorded. Re-cordings were transcribed verbatim and anonymized.

AnalysisInterviews were coded and analyzed in 3 discrete steps,using content analysis following a combined frameworkanalysis approach. These are standard methods fromother TDF-based studies using semi-structured inter-views [31–36].

Pilot coding A coding framework was developed to pro-mote consistent coding of utterances into appropriate do-mains. This was adapted from previous Sepsis Six TDFresearch [27, 28]. To promote greater coding consistency,a pilot interview was transcribed and coded jointly by twomembers of the research team (NR,GH) [36]. Discrepan-cies were resolved through discussion.

Coding of participant responses into TDF domainsParticipant responses were split into individual ‘utterances’and coded into to the TDF domain it was judged to bestrepresent. For example, “[The Sepsis Six is] a package ofcare which has been shown to improve mortality inpatients with sepsis.” was allocated to “Knowledge”.Utterances corresponding to more than one domain wereallocated as such, for example “We've got really good nurs-ing staff who can do bloods and blood cultures as well.”was allocated to both ‘Skills’ and ‘Social and ProfessionalRole’.

Thematic synthesis and generation of belief statementsUtterances within domains were compared acrosstranscripts, and those expressing similar views weregrouped together. Belief statements were then generatedsummarizing each group of similar utterances. Beliefstatements are defined as ‘a statement that provides de-tail about the role of the domain in influencing behavior’[36]. For example “It’s easy [to remember the steps]. It’sthree in and three out.” and “Relatively easy [to remem-ber] actually. The fact I can remember them for thisinterview has proven to me that they’re relatively easy toremember.” were represented in the belief statement ‘It’seasy/difficult to remember the six steps in clinicalpractice’. Thematic synthesis was conducted by the leadresearchers (NR,GH). Each generated belief statementwas independently reviewed by a health psychologist(FL) to promote robust and defensible coding accordingto the TDF, and ensure the generated belief statementprovided a valid representation of the constituent utter-ances [37]. Regular consensus discussions were held toresolve disagreement [31].


QuestionnaireParticipantsAll staff currently eligible from the stakeholder staffgroups (consultants, junior doctors and nurses fromMAU, SAU and ED) were identified. Staff on long-termsick or maternity leave were excluded from the distri-bution list.

MaterialsA questionnaire containing 54 two-part questions wassynthesized using belief statements identified in the in-terviews. Belief statements concerning importance, ordifferent perspectives on the same topic which would beanswered through individual questionnaire responses (eg“my colleagues do not have the skills.../I do not have theskills...” were collapsed and combined. Practical usefulnessof the resultant data was also factored in, for example ageneral ‘resource availability’ belief statement was sepa-rated into questions on ‘staff ’ or ‘equipment’ in order tofocus on which resources were most in need of interven-tion. After entering basic demographic information,

participants were asked to rank, on a 5-point Likert scale,agreement with two opposing statements constructedfrom each belief statement. For example, “performing theSepsis Six IS part of my role” scoring 1 and “performing theSepsis Six is NOT part of my role” scoring 5. They werethen asked to rate this statement in terms of importanceto their delivery of the Sepsis Six, ranging from 1 “very un-important” to 5 “very important”. Each domain had atleast one questionnaire item associated with it. The ques-tionnaire was available in paper format or electronicallyusing SurveyMonkey. It was piloted with nursing andmedical staff from other hospital areas (so not included inthe study population), and revised to simplify formattingand wording of the Likert scale. A final version of thequestionnaire is available in Additional file 2.

ProcedureData collection took place over six weeks from June–July2016. Participants were sent an invitation email. Consul-tants and lead nurses were asked to promote the ques-tionnaire to staff. Paper questionnaires were accessible

Table 2 Theoretical domains framework with sample questions [20]

Domain Content Sample question as applied to this study

Knowledge An awareness of something What do you understand by the Sepsis Six?

Skills Ability or proficiency acquired through practice Can you think of any ways in which your own skills forperforming the steps in the Sepsis Six could be improved?

Social/professional roleand identity

Set of behaviors and qualities of an individual insocial or work setting

To what extent do you consider performing the steps inthe Sepsis Six a part of your role?

Beliefs about capabilities Views about one’s ability/talent/capability toperform the target behavior(s)

Are there any particular steps that you are more or lessconfident about performing?

Optimism Confidence that things will happen for the best orthat desired goals will be attained

How optimistic or pessimistic are you that improvingperformance of the Sepsis Six holds the potential toimprove patient care in the future?

Beliefs about consequences Acceptance of the truth, reality or validity aboutoutcomes of a behavior in a given situation

To what extent do you believe that performing the stepsin the Sepsis Six can affect patient outcomes?

Reinforcement Increasing the likelihood of a behavior beingperformed by establishing an association betweenperforming a behavior and a given stimulus or cue

Are you aware of any ways in which performing the SepsisSix is rewarded?

Intentions Conscious decision to perform a behavior orresolve to act in a certain way

To what extent do you intend to (continue to) perform theSepsis Six in daily clinical practice?

Motivation and Goals Mental representation of outcomes or states thatan individual wants to achieve

Do you have any specific goals for performing the Sepsis Six?

Memory, attention anddecision processes

The ability to retain information, focus selectivelyon aspects of the environment and choosebetween two or more alternatives

How easy or difficult is it to remember the steps involved inthe Sepsis Six when you are performing it in daily clinicalpractice?

Environmental contextand resources

Circumstances of a person’s situation/environmentthat affect behavior

To what extent does your working environment have sufficientlevels of resources needed to allow performance of the SepsisSix within one hour of recognition?

Social influences Interpersonal processes that can cause individualsto change thoughts/feelings/behaviors

Are there any conflicting beliefs amongst your colleagues aboutthe Sepsis Six?

Emotions Complex reaction pattern by which individualattempts to deal with a personally significantmatter or event

To what extent do you feel that your emotional state affectsyour performance of the Sepsis Six?

Behavioral regulation Anything aimed at managing or changingobjectively observed or measured actions

Do you ever receive feedback on your performance of theSepsis Six on septic patients?


in clinical areas to further optimize response rate fromthose who may not check email. Potential participantswere followed up by weekly emails, and reminders inperson.

AnalysisAnalysis was performed on Microsoft Excel. Questionswere reverse scored if necessary such that all facilitatorstatements were associated with positive agreementscores, and all barriers associated with negative agree-ment scores. A median score (and interquartile range)was calculated for each part (agreement/importance) ofthe 54 questions, for each of 16 analysis groups, as fol-lows: 1) overall median for the hospital (as a whole), me-dian for 2) each role, 3) each clinical area and 4) eachrole within each clinical area.Likert scales produce ordinal data, with no guarantee

that each participant has the same baseline, or the sameintervals between descriptors. Therefore, median agree-ment scores and importance scores for the questionnaireas a whole were then calculated overall for each of these16 analysis groups, in order to establish a ‘baseline’ levelof agreement and importance for each group.

Selection of important belief statementsAgreement and importance were assessed separately foreach question. Therefore, selection of important barriersrequired dichotomizing the answers to both of these as-pects. For each analysis group, median agreement andimportance scores for each question were compared tothe group’s overall questionnaire baseline agreement andimportance scores. Statements were then dichotomized,being classified as barriers for an analysis group if themedian agreement score for the individual question wasless than the group’s baseline agreement score for thequestionnaire; otherwise they were classified as facilita-tors. Statements were classified as ‘important’ for ananalysis group if the median importance score for the in-dividual question was equal to or above the group’sbaseline importance score for the questionnaire; other-wise they were classified as ‘unimportant’. This approachallowed identification of relatively important, or rela-tively unimportant, barriers or facilitators, for each ana-lysis group.Once important barriers and facilitators were identi-

fied for each analysis group, they were compared to lookfor areas of ‘discord’ between roles or clinical areas,where an important barrier for some (for example,‘Nurses’ or ‘Surgery’) was an important facilitator forothers (for example, ‘Junior doctors’ or ‘EmergencyDepartment’). Identification of discordant beliefs allowsfor further tailoring of subsequent quality improvementinterventions.

ResultsSemi-structured interviewsParticipant characteristicsNo new belief statements were identified after analysisof the initial ten participant interviews; therefore, the-matic data saturation was deemed achieved and no fur-ther interviews were conducted or analyzed (Datasaturation table in Additional file 3). Interviews lasted amean 35.8 min (range 18–48). 70% participants weremale. Three participants were consultants, three werejunior doctors and four were nurses. Three participantscame from MAU, three from SAU and four from ED.Participants had worked at the study hospital for a mean6.5 years (range 0.8–17).

Coding of responses into TDF domainsIn total, 1699 utterances were coded into 14 TDFdomains.Extracted utterances were synthesized into 64 belief

statements. These are presented in Table 3 with examplequotes. The most populated domains were Social andProfessional role (9 belief statements) and Intentions (8belief statements).

QuestionnaireParticipant characteristicsTwo hundred fifty-five potential participants were invitedto complete the questionnaire. 54 of these were consul-tants (36 medical, 10 surgical, 8 ED), 82 junior doctors(32 medical, 27 surgical, 23 ED), and the remaining 119were nurses (38 medical, 18 surgical, 63 ED). Responserates are given in Table 4.Length of time participants had worked in their

current role ranged from 4 months to 30 years.

Belief statementsForty-six belief statements were ‘important facilitators’for at least one analyzed participant group. 30 beliefstatements were ‘important barriers’ for least one ana-lyzed participant group.Status of each belief statement amongst the overall

sample is presented in Table 5, grouped by domain. Dis-cordance is illustrated here by the number of groups forwhom a belief statement was an important barrier,against those for whom it was an important facilitator.The complete results table, with median agreement andimportance scores and interquartile ranges for each par-ticipant group for each belief statement, is presented inAdditional file 4. Questionnaire results are discussed bydomain in the following text.

Behavioral regulationBelief within this domain focused on audit, feedback,improvement plans and the discussion of sepsis in


Table 3 Results of interview analysis

Domain Belief Statement Example Utterance Frequency(number ofinterviews)

Knowledge I know/do not know what the Sepsis Sixinvolves

[The Sepsis Six is] a package of care which has beenshown to improve mortality in patients with sepsis.(Consultant 1)

10

My colleagues do/do not know what theSepsis Six involves

the more senior sort of colleagues weren’t familiar withsystemic inflammatory response syndrome, recogniseall the sort of markers. (Nurse 4)

10

I am aware/not aware of the evidencebehind the Sepsis Six

There have been obviously clinical trials which I can’tremember the names of. (Junior doctor 2)

10

My colleagues are aware/not aware of theevidence behind the Sepsis Six

If I’m honest then no. [My colleagues and I are notaware of the evidence behind the Sepsis Six] (Nurse 3)

2

People would give better Sepsis Sixperformance if they were more aware ofthe later complications of poorly managedsepsis

I think if they had perhaps more awareness about how,like, poor sepsis management could affect a patientlong term, they might be, they might have moreurgency in carrying it all, sort of out. About what wouldhappen in the long term. (Nurse 4)

1

Having knowledge and understanding ofthe Sepsis Six does/does not influence thelikelihood of it being performed

And it’s always, [giving antibiotics] tends to get done Ithink, because I think everyone understands theurgency. (Junior doctor 1)

9

Skills I do/do not have the skills to perform theSepsis Six

I mean I haven’t given antibiotics myself. (Junior doctor 1) 9

My colleagues do/do not have the skills toperform the Sepsis Six

maybe if [nurses are] newly qualified, not being able togive the IV antibiotics because they would then have,because they haven’t done their IV pack (Nurse 1)

9

There is/is insufficient provision of trainingand assessment in the skills required toperform the Sepsis Six

I think we’ve recognised that, and we’ve trained ournurses to deliver antibiotics, fluids, take blood culturesand lactates, put in urinary catheters, so we know ournurses can do all of this, we train them to do all of this.(Consultant 1)

9

Memory, Attention andDecisions

It’s easy/difficult to remember the 6 stepsin clinical practice

Give 3, take 3 away. And that we have it written downon our proformas. (Consultant 1)

10

The decision to start the Sepsis Six is notmade because sepsis is not recognised

So I think I’ve got a reasonable understanding ofrecognising sepsis. (Junior doctor 3)

8

Regular use of the Sepsis Six makes iteasier to remember the steps

I mean when doing on a daily basis pretty easy toremember. But I guess if you’re not doing it on a dailybasis you might forget (Junior doctor 2)

7

Behavioral Regulation Sepsis Six performance is (not) monitoredor audited regularly in my department

[Sepsis 6 is audited on a] weekly basis and the resultsare published weekly. (Junior doctor 3)

10

I/we get insufficient feedback on ourSepsis Six performance

Yes it would be helpful to have more monitoringsystems in place. And individual feedback to clinicians.(Consultant 2)

9

There are sufficient tools in place to helpguide and track Sepsis Six performance inindividual patients

Yes, we’ve got [a Sepsis 6 tool], it’s at the back of thepro forma and the BUFALO stickers. So there’s a lot ofguidance. (Consultant 3)

10

Improving sepsis care and Sepsis Sixperformance is (not) discussed in regularmeetings in my department

What we’re doing in surgery is auditing this sort ofthing on a monthly basis, and that’s going to bepresented at governance meetings. (Consultant 2)

5

Sepsis Six performance improves if we areinvolved in the quality improvementprocess

Yeah, I think so, because I think people would ownthings more if they felt it was, they were included in it(Nurse 2)

3

There are (no) action plans to improveSepsis Six performance

there are other things that we’re doing such asimplementing junior doctor training, nurse training onsepsis, through educational sessions, through induction.(Consultant 1)

4


Table 3 Results of interview analysis (Continued)

Social Influences My colleagues opinions do/do not affectmy performance of the Sepsis Six

I don’t think the opinions of my colleagues does affectwhether I do the Sepsis Six actually. (Junior doctor 3)

10

My Colleagues do/do not believe that theSepsis Six is beneficial to patient care

Yeah I think it’s generally believed that these stepsbenefit patient outcomes, so I think everyone’s kind ofin favour of them. (Junior doctor 2)

9

Departmental culture facilitates/hindersperformance of the Sepsis Six

I think that’s because there isn’t a culture of doing fluidcharts on every patient that comes through us(Consultant 1)

4

There is insufficient leadership to improveSepsis Six performance

But if there was a clear strategy, a clearer kind of teamrole, leadership role for the patients, the benefits of it,can’t see why it wouldn’t be used and why it couldn’timprove. (Consultant 3)

7

Healthcare workers do/do not feel able toescalate up the hierarchy

if you have it on a care pathway, that gives themallowance, permission almost to phone the consultantand escalate it, so they’re allowed to do that, ratherthan feeling I shouldn’t do this. (Consultant 2)

4

Having a Sepsis “Champion” would/wouldnot improve performance of the Sepsis Six

it might be beneficial if, other wards as well to have adesignated sepsis champion or link nurse as such, sothat we can perhaps hold regular meetings everycouple of months, to see how we can make changesto sepsis care. (Nurse 4)

3

Social and Professional Role Performing the steps in the Sepsis Six is(not) my role

I think they’re all part of it. (Consultant 2) 10

Performing all steps in the Sepsis Six is(not) my colleagues’ role

I think it should be everyone’s responsibility and role todo it. (Consultant 3)

3

It is my/my colleagues’ role (doctor/nurse/HCA) to identify septic patients

our nurses are very good at identifying sick patients.(Junior doctor 3)

5

It is my role to decide when to performthe Sepsis Six

we usually don’t give oxygen to somebody unless theirsats are low, but in this instance, occasionally I’ve beentold by a surgeon, I want them to have 2 l of oxygen.(Nurse 2)

4

There is high turnover of medical/nursingstaff in areas looking after septic patients

Our medical staff, so half of them are transient, half ofthem are permanent. (Consultant 1)

3

My role is to improve Sepsis Sixperformance through non-clinical factors(leadership, support, supervision)

My role is that even if the patient is stable to ensure allthe steps had been followed, and to reinforce andeducate. (Consultant 3)

4

There are some steps in the Sepsis Sixwhich I/my colleagues do not/are notallowed to perform

I don’t know, as a trust I don’t think the nurses usually takeblood cultures, it seems to be a doctor role. (Nurse 2)

6

Non-clinical staff (eg bed management)put pressure on clinical staff to prioritisetasks other than Sepsis Six

And there’s such a drive for discharging them, andgetting patients out, and often the bed manager putsso much pressure on the nursing staff on the wards.(Consultant 2)

1

Staff should be empowered to improvetheir role in Sepsis Six performance

So I think being involved in the audit kind of made us kindof aware of what needs to be done. (Junior doctor 3)

2

Environment, Context andResources

I do (not) have sufficient resources (staff;time; equipment; medicines; bed) toperform the Sepsis Six in one hour.

Not enough beds (Consultant 2) 10

The equipment I have does/doesn’t work our gas machine is down a lot of the time (Juniordoctor 3)

8

The layout of the hospital hinders/helpsmy performance of the Sepsis six in onehour (patient location, equipment,medicine).

trying to get a patient seen and then treated within thattime, and then if they’re coming up to 4 h of being inthe department are they moved to another ward beforetheir treatment sort of is delivered, (Nurse 4)

8

Belief in Consequences Performing the steps in the Sepsis Siximproves patient outcomes

Yeah, I believe it’s vital. There’s evidence out therewhich supports, supports it, so, yeah (Consultant 2)

10

The benefits of performing the Sepsis Sixoutweigh the risks

I think generally the advantages should outweigh therisks. (Junior doctor 2)

8



The benefits vs risks of performing theSepsis Six (or some parts of it) are (not)different in certain patient groups

I think any patient with known heart problems, I’d be alittle bit more careful. (Nurse 3)

10

The quicker the steps can be delivered, themore impact they have

Well, I believe if it’s carried out promptly within the 1 hthen it can definitely improve the patient’s outcome.(Nurse 3)

5

Early and regular reassessment of patientsrequiring the Sepsis Six gives the bestoutcomes

I think if you keep doing them without reassessmentthen that would lead to problems, but in the first hourI don’t think it’s an issue. (Consultant 2)

1

Belief in Capabilties I am (not) confident performing the stepsin the Sepsis Six

I think, you know, I’ve got the skill to perform these 6steps, there’s no doubt about it. The training has beenthere, I have the skill to do it (Consultant 1)

9

My colleagues are (not) confidentperforming the steps in the Sepsis Six

I mean I would hope most people were... but yeah, Ithink most people are confident that I’ve seen (Juniordoctor 2)

7

Some of the Sepsis Six steps are moredifficult than others to achieve (urineoutput, cultures, antibiotics)

it’s just the urine output measurement which causesongoing difficulties, (Consultant 1)

9

There is good/poor communication andteamwork between members of the teamlooking after septic patients

It’s definitely a team priority to be able to carry it allout, so if we can sort of work together, I believe that itcan be done a lot quicker, rather than doing it single-handedly. (Nurse 4)

10

We provide good sepsis care at thishospital

I think for the most part our septic patients isreasonably well recognised via the acute care bundle,because they come in, they have the acute carepathway, filled out for every patient (Consultant 2)

1

I am confident looking after sick septicpatients

I’m very good at dealing with a crisis and just gettingon with it (Nurse 2)

1

Intentions I (don’t) prioritise performing the Sepsis Sixon a septic patient over other tasks

I think unless somebody was having a cardiac arrest Iwould prioritise this probably above most other things.(Nurse 2)

10

I intend to improve my knowledge of theSepsis Six

I think I do need to know a bit more about it, so Imight try and educate myself before I go back to work(Nurse 2)

2

I intend to continue to perform the SepsisSix on septic patients

I guess, well I’ll carry on carrying it out until it’s, unlessthere’s anything else new that comes up that improvessepsis care (Nurse 4)

8

I am more likely to complete all steps ofthe Sepsis Six if I think the patient is sick/less likely if they are well

if we are concerned someone really is poorly, then theyoften will become catheterised (Nurse 2)

5

Sometimes I choose (not) to complete thefull Sepsis Six because the risks andbenefits are different for that patient/situation.

your octogenarian who’s got sepsis, you might not gochucking in 2 l immediately. (Consultant 2)

7

My colleagues (don’t) prioritise performingthe Sepsis Six on a septic patients overother tasks

but it’s usually the other pressures that we have on, likeprioritising other patients for example, and how big ourcaseload is at that time (Nurse 4)

2

Some steps in the Sepsis Six are more/lessimportant than others

I like having the fluids here quickly. That’s one of thebetter ones, I think (Nurse 3)

6

I (don’t) perform the Sepsis Six despite nothaving a confirmed diagnosis because I(don’t) believe the risks of undertreatingsepsis outweigh the risks of performing theSepsis Six

they’re not septic, but they just got a big SIRS responseand they looked septic when they came in, so havingantibiotics in that situation is not the wrong thing todo, as a one off, but a patient’s presenting withperitonitis or what’s not, they need to have early sepsissource control. (Consultant 2)

3


governance meetings. There were clear differences be-tween departments and roles, with a trend towards be-liefs in this domain being barriers for participants.Particularly the surgical department showed importantbarriers, along with the junior doctors of MAU. Overallthere was a general perception of ‘not enough feedback’(8 analysis groups), with the notable exception of EDnurses, for whom the detailed feedback they receivedwas an important facilitator. The strength of beliefamongst MAU juniors was sufficient to classify lack ofan improvement plan as an important barrier for the

overall sample (2 groups), despite the converse being animportant facilitator for eleven other groups. An exceptionto the trend for this domain was a widespread belief thatinvolving clinical staff in performance improvement pro-cesses would lead to greater improvements (15 groups).

Belief in capabilitiesThough all 16 groups believed their confidence in theSepsis Six itself were an important facilitator, non-technical skills such as teamwork and communicationwere noted as important barriers for most groups (10/16and 11/16 respectively). Discord was also observed withthese beliefs, whereby confidence in such skills were im-portant facilitators for groups within the surgical depart-ment, and for ED juniors (6/16 and 5/16 respectively).

Belief in consequencesOverall, this domain included important facilitators acrossall 16 groups, with important beliefs in the beneficial con-sequences of timely Sepsis Six performance and of bene-fits versus risks of the bundle. Notable exceptions here area belief amongst MAU and ED doctors that the risks ofthe Sepsis Six outweigh the benefits in certain patientgroups (e.g. oxygen in chronic obstructive pulmonary dis-ease or intravenous fluids in cardiac patients). This wasnot reflected in surgical participants, who believed theSepsis Six was always of benefit in their patients.


Goals I work towards a goal that the Sepsis Sixshould be completed and documentedwithin an hour on all septic patients.

We should do it on all, it should be done within anhour. (Consultant 3)

10

The hospital has/does not have a goal ofimproving Sepsis Six compliance

I know that the Trust is starting a BUFALO[sic] to helppeople remember how to deliver the Sepsis Six (Nurse 4)

10

Optimism Sepsis Six compliance at this hospital will(not) improve

Knowing how well in general all the care bundles areused, unless there’s a clear strategy on how to improveit, my worry would be that it might not improvesignificantly (Consultant 3)

4

Increasing Sepsis Six compliance willimprove patient care

I’m very optimistic that if we can push this forward thatit will hugely improve patient care, and their outcome.(Nurse 2)

9

Reinforcement Individuals are not formally rewarded orpunished for (failing to) complete theSepsis Six

Not punished, obviously it’s audited, and thedepartments are fed back (Junior doctor 2)

10

The department or hospital is (not)formally rewarded or punished for (failingto) complete the Sepsis Six

No, I’m not aware of any ways in which we as, do youmean as a trust are punished? (Nurse 4)

6

Emotions I get emotionally affected negatively/positively by managing septic patients

I mean obviously you do [get affected emotionally bylooking after septic patients], if they’re unwell (Juniordoctor 2)

9

If we are affected emotionally (eg stressed,excited, fatigued) it leads to better/worseclinical performance when looking afterseptic patients

Well if anything it makes me go, try and make, do itfaster because I recognise that they’re quite sick. (Juniordoctor 1)

9

I feel good if I deliver the Sepsis Six/bad if Idon’t deliver the Sepsis Six to a septicpatient

I try to carry it out within the 1 h, and when it hasn’thappened I kind of feel, like frustrated (Nurse 4)

4

Table 4 Survey response rates

Area and Role Numberinvited

Numbercompleted

Responserate

MAU Consultants 36 15 41.7%

MAU Junior Doctors 32 15 46.9%

MAU Nurses 38 14 36.8%

SAU Consultants 10 7 70.0%

SAU Junior Doctors 27 12 44.4%

SAU Nurses 18 12 66.7%

ED Consultants 9 7 77.8%

ED Junior Doctors 22 7 31.8%

ED Nurses 63 24 38.1%

Total 255 113 44.3%


Table

5Classificatio

nof

beliefstatem

entsaccordingto

questio

nnaire

results,p

resented

bydo

main

Dom

ain

Barrierbe

lief

Facilitator

belief

Overall

sample

grou

p

Impo

rtantbarrier

(num

berof

analysisgrou

ps)

Unimpo

rtant

barrier(num

ber

ofanalysisgrou

ps)

Unimpo

rtant

facilitator

(num

ber

ofanalysisgrou

ps)

Impo

rtant

facilitator(num

ber

ofanalysisgrou

ps)

Beliefin

Capabilities

ThereisPO

ORteam

workwhe

nlooking

aftersepticpatients

ThereisGOODteam

workwhe

nlooking

aftersepticpatients

Impo

rtant

barrier

100

06

Iam

NOTconfiden

tpe

rform

ingthe

SepsisSix

IAM

confiden

tpe

rform

ingtheSepsis

Six

Impo

rtant

enabler

00

016

Someof

thestep

sin

theSepsisSixare

MORE

DIFFICULT

tope

rform

than

othe

rsThestep

sin

theSepsisSixareEQ

UALLY

EASY

ORDIFFICULT

tope

rform

Unimpo

rtant

barrier

94

03

ThereisPO

ORcommun

icationbe

tween

mem

bersof

theteam

lookingafterseptic

patients

ThereisGOODcommun

ication

betw

eenmem

bersof

theteam

looking

aftersepticpatients

Impo

rtant

barrier

110

05

WeprovidePO

ORsepsiscare

atthis

hospital

WeprovideGOODsepsiscare

atthis

hospital

Unimpo

rtant

barrier

83

05

Beliefin

Con

sequ

ences

DeliveringtheSepsisSixqu

icklydo

esNOTincrease

how

muchbe

nefit

ithas

DeliveringtheSepsisSixqu

icklyDOES

increase

thebe

nefit

ithas

Impo

rtant

enabler

00

016

Perfo

rmingthestep

sin

theSepsisSix

does

NOTim

provepatient

outcom

esPerfo

rmingthestep

sin

theSepsisSix

DOES

improvepatient

outcom

esIm

portant

enabler

00

016

Overall,theRISKSof

perfo

rmingthe

SepsisSixou

tweigh

thebe

nefits

Overall,theBENEFITSof

perfo

rmingthe

SepsisSixou

tweigh

therisks

Impo

rtant

enabler

10

015

TheRISKSof

perfo

rmingtheSepsisSix

outw

eigh

thebe

nefitsin

CERTA

INpatient

grou

ps

TheBENEFITSof

perfo

rmingtheSepsis

Sixou

tweigh

therisks

inALL

patient

grou

ps

Unimpo

rtant

barrier

63

16

Early

andregu

larreassessmen

tof

patientsrequ

iring

theSepsisSixhasNO

effect

onou

tcom

es

Early

andregu

larreassessmen

tof

patientsrequ

iring

theSepsisSixgives

theBEST

outcom

es

Impo

rtant

enabler

00

016

Behavioral

Regu

latio

nSepsisSixpe

rform

ance

isNOTaudited

regu

larly

inmyde

partmen

tSepsisSixpe

rform

ance

ISaudited

regu

larly

inmyde

partmen

tUnimpo

rtant

barrier

25

45

ThereareINSU

FFICIENTtoolsin

useto

guide&trackSepsisSixpe

rform

ance

inindividu

alpatients

ThereareSU

FFICIENTtoolsin

useto

guide&trackSepsisSixpe

rform

ance

inindividu

alpatients

Unimpo

rtant

barrier

54

25

Wege

tINSU

FFICIENTfeed

back

onou

rSepsisSixpe

rform

ance

Wege

tSU

FFICIENTfeed

back

onou

rSepsisSixpe

rform

ance

Unimpo

rtant

barrier

87

01

SepsisSixpe

rform

ance

isNOTdiscussed

inmeetin

gsin

myde

partmen

tSepsisSixpe

rform

ance

ISdiscussedin

meetin

gsin

myde

partmen

tUnimpo

rtant

barrier

42

46

Involvingclinicalstaffin

SepsisSix

perfo

rmance

improvem

entwillNOTlead

togreaterim

provem

ent

Involvingclinicalstaffin

SepsisSix

perfo

rmance

improvem

entWILLlead

togreaterim

provem

ents

Impo

rtant

enabler

00

115

ThereareNOplansin

placeto

improve

SepsisSixpe

rform

ance

atmyho

spital

ThereARE

plansin

placeto

improve

SepsisSixpe

rform

ance

atmyho

spital

Impo

rtant

barrier

21

211


Table

5Classificatio

nof

beliefstatem

entsaccordingto

questio

nnaire

results,p

resented

bydo

main(Con

tinued)

Environm

ent,

Con

text

and

Resources

ThereisINSU

FFICIENTstaffingto

perfo

rmtheSepsisSix

ThereisSU

FFICIENTstaffingto

perfo

rmtheSepsisSix

Impo

rtant

barrier

130

03

ThereisINSU

FFICIENTtim

eto

perfo

rmtheSepsisSix

ThereisSU

FFICIENTtim

eto

perfo

rmtheSepsisSix

Impo

rtant

barrier

110

05

ThereisINSU

FFICIENTeq

uipm

ent/

med

icationto

perfo

rmtheSepsisSix

ThereisSU

FFICIENTeq

uipm

ent/

med

icationto

perfo

rmtheSepsisSix

Impo

rtant

barrier

50

011

ThereareINSU

FFICIENTbe

dsavailablein

myde

partmen

tto

look

afterseptic

patients

ThereareSU

FFICIENTbe

dsavailablein

myde

partmen

tto

look

afterseptic

patients

Impo

rtant

barrier

150

01

Theeq

uipm

entIn

eedto

perfo

rmthe

SepsisSixdo

esNOTworkor

works

poorly

Theeq

uipm

entIn

eedto

perfo

rmthe

SepsisSixDOES

workwell

Impo

rtant

enabler

30

013

SepticpatientsareRA

RELY

managed

inan

approp

riate

locatio

nSepticpatientsareALW

AYS

managed

inan

approp

riate

locatio

nIm

portant

barrier

150

01

Emotions

IdoNOTfeelbadifId

ono

tde

liver

the

SepsisSixto

asepticpatient

IDOfeelbadifId

ono

tde

liver

the

SepsisSixto

asepticpatient

Impo

rtant

enabler

00

115

IdoNOTfeelanxiou

s/stressed

whe

ntreatin

gsepticpatients

IDOfeelanxiou

s/stressed

whe

ntreatin

gsepticpatients

Unimpo

rtant

barrier

313

00

Goals

IdoNOThave

atim

e-basedgo

alfor

completingtheSepsisSixon

septic

patients

Mygo

alisto

completetheSepsisSix

with

inan

HOURon

allsep

ticpatients

Impo

rtant

enabler

00

016

Intentions

IdoNOTintend

toim

provemy

know

ledg

eof

theSepsisSix

IINTENDto

improvemyknow

ledg

eof

theSepsisSix

Unimpo

rtant

barrier

16

27

Whe

nun

certainabou

tdiagno

sisIW

AIT

FORCONFIRM

ATIONof

sepsisbe

fore

perfo

rmingtheSepsisSix

Whe

nun

certainabou

tdiagno

sisI

PERFORM

theSepsisSixrather

than

misstreatin

gpo

tentialsep

sis

Impo

rtant

enabler

21

013

IdoNOTintend

tocontinue

tope

rform

theSepsisSixon

septicpatients

IDOintend

tocontinue

tope

rform

the

SepsisSixon

septicpatients

Impo

rtant

enabler

00

016

Iam

UNLIKELY

tocompleteallsteps

oftheSepsisSixifIthink

thepatient

iswell

Iam

LIKELY

tocompleteallsteps

ofthe

SepsisSixeven

ifIthink

thepatient

iswell

Unimpo

rtant

barrier

78

10

IdoNOTprioritisepe

rform

ingtheSepsis

Sixon

asepticpatient

over

othe

rtasks

IDOprioritisepe

rform

ingtheSepsisSix

onasepticpatient

over

othe

rtasks

Impo

rtant

enabler

20

113

SOMEstep

sin

theSepsisSixaremoreor

less

impo

rtantthan

othe

rsALL

step

sin

theSepsisSixareeq

ually

impo

rtant

Unimpo

rtant

barrier

64

15

Know

ledg

eIam

NOTaw

areof

whattheSepsisSix

involves

IAM

awareof

whattheSepsisSix

involves

Impo

rtant

enabler

00

016

Iam

NOTaw

areof

theeviden

cesupp

ortin

gtheSepsisSix

IAM

awareof

theeviden

cesupp

ortin

gtheSepsisSix

Unimpo

rtant

enabler

01

213


Table

5Classificatio

nof

beliefstatem

entsaccordingto

questio

nnaire

results,p

resented

bydo

main(Con

tinued)

Mem

ory,Atten

tion

andDecisions

It’sDIFFICULT

toremem

berallthe

step

sof

theSepsisSixin

day-to-day

clinical

practice

It’sEA

SYto

remem

berallthe

step

sof

theSepsisSixin

day-to-day

clinical

practice

Impo

rtant

enabler

00

016

IOFTEN

misssepsis

IRARELY

misssepsis

Impo

rtant

barrier

70

09

Regu

laruseof

theSepsisSixdo

esNOT

makeiteasier

toremem

berthestep

sinvolved

Regu

laruseof

theSepsisSixDOES

makeiteasier

toremem

berthestep

sinvolved

Impo

rtant

enabler

00

313

Optim

ism

SepsisSixpe

rform

ance

atthisho

spital

willNOTim

prove

SepsisSixpe

rform

ance

thisthisho

spital

WILLim

prove

Impo

rtant

barrier

31

012

Increasing

SepsisSixpe

rform

ance

will

NOTim

provepatient

care

Increasing

SepsisSixpe

rform

ance

WILL

improvepatient

care

Impo

rtant

enabler

00

016

Reinforcem

ent

Theho

spitalisNOTform

allyrewarde

dfor

good

SepsisSixpe

rform

ance

Theho

spitalISform

allyrewarde

dfor

good

SepsisSixpe

rform

ance

Unimpo

rtant

barrier

015

10

Individu

alsareNOTform

allyrewarde

dfor

good

SepsisSixpe

rform

ance

Individu

alsARE

form

allyrewarde

dfor

good

SepsisSixpe

rform

ance

Unimpo

rtant

barrier

214

00

SocialInfluen

ces

Thecultu

rewith

inmyde

partmen

tHINDERSpe

rform

ance

oftheSepsisSix

Thecultu

rewith

inmyde

partmen

tHELPS

perfo

rmance

oftheSepsisSix

Impo

rtant

barrier

60

010

ThereisINSU

FFICIENTleadership

for

improvingSepsisSixpe

rform

ance

ThereisSU

FFICIENTleadership

for

improvingSepsisSixpe

rform

ance

Unimpo

rtant

barrier

61

18

Mycolleagues’op

inions

abou

ttheSepsis

Sixdo

NOTaffect

whe

ther

Iperform

itMycolleagues’op

inions

abou

tthe

SepsisSixDOaffect

whe

ther

Iperform

it

Unimpo

rtant

barrier

510

01

Mycolleaguesdo

NOTbe

lieve

that

the

SepsisSixisbe

neficialtopatients

MycolleaguesDObe

lieve

that

the

SepsisSixisbe

neficialtopatients

Unimpo

rtant

enabler

00

313

IdoNOTfeelableto

escalate

whe

nIam

concerne

dabou

tapatient

who

may

need

theSepsisSix

IDOfeelableto

escalate

whe

nIam

concerne

dabou

tapatient

who

may

need

theSepsisSix

Impo

rtant

enabler

00

016

Havingalocalsep

sis‘champion

’wou

ldNOTim

provepe

rform

ance

oftheSepsis

Six

Havingalocalsep

sis‘champion

’WOULD

improvepe

rform

ance

ofthe

SepsisSix

Unimpo

rtant

barrier

012

22

Skills

IdoNOThave

thene

cessaryskillsto

perfo

rmtheSepsisSix

IHAVE

thene

cessaryskillsto

perfo

rmtheSepsisSix

Impo

rtant

enabler

00

115

ThereisINSU

FFICIENTprovisionof

training

requ

iredto

perfo

rmtheSepsis

Six

ThereisSU

FFICIENTprovisionof

training

requ

iredto

perfo

rmtheSepsis

Six

Impo

rtant

barrier

90

07

Socialand

Profession

alRo

leItisNOTpartof

myroleto

decide

whe

nto

perfo

rmtheSepsisSix

ItISpartof

myroleto

decide

whe

nto

perfo

rmtheSepsisSix

Impo

rtant

enabler

00

016

Perfo

rmingtheSepsisSixisNOTpartof

myrole

Perfo

rmingtheSepsisSixISpartof

my

role

Impo

rtant

enabler

00

016


Table

5Classificatio

nof

beliefstatem

entsaccordingto

questio

nnaire

results,p

resented

bydo

main(Con

tinued)

ItisNOTpartof

myroleto

iden

tifyseptic

patients

ItISpartof

myroleto

iden

tifyseptic

patients

Impo

rtant

enabler

00

016

ThereisRA

PIDturnover

ofmed

ical/

nursingstaffin

areaslookingafterseptic

patients

ThereisSLOW

turnover

ofmed

ical/

nursingstaffin

areaslookingafter

septicpatients

Unimpo

rtant

barrier

106

00

ItisNOTpartof

myroleto

improve

SepsisSixpe

rform

ance

throug

hleadership

&supp

ort

ItISpartof

myroleto

improveSepsis

Sixpe

rform

ance

throug

hleadership

&supp

ort

Unimpo

rtant

enabler

00

214

Therearesomestep

sin

theSepsisSix

which

Iam

NOTALLOWED

tope

rform

Iam

ALLOWED

tope

rform

allsteps

intheSepsisSix

Impo

rtant

enabler

21

211

Sixteenan

alysisgrou

psin

total:Overallsample(1),3de

partmen

ts(M

AU,ED,Surge

ry),3roles(Con

sulta

nts,nu

rses,jun

iordo

ctors)an

d9roleswith

inde

partmen

ts(e.g.EDNurses)


EmotionsThis was a mixed domain, with 15 groups expressingregret if they failed to deliver the Sepsis Six as an im-portant facilitator, but unimportant barrier beliefsamongst 13 groups about feeling too calm and relaxedwhen treating septic patients. Interview participantsexpressed that staff should be more worried and morestressed about sepsis, as it can be so indolent when com-pared to more obvious emergencies like bleeding.

Environment, context and resourcesThis domain included numerous barriers, particularly forED and MAU. Reported important barriers includedinsufficient staff (13 groups), time (11 groups), equipment(5 groups), and beds (15 groups) to adequately deliver theSepsis Six. Participants from ED reported having sufficientnecessary equipment, however, encountered difficultiesusing it (e.g. gas machine in ED resus is often faulty). Incontrast, equipment on MAU worked well but there wereoften insufficient levels (e.g. lack of drip stands forfluids). There were also concerns regarding locationof care (15 groups), for example with surgically septicpatients coming in via ED and MAU.

GoalsA time-based goal of completing the Sepsis Six care bun-dle within one hour was an important facilitator endorsedby all 16 groups.

IntentionsFacilitators within this domain included prioritizationof septic patients (13 groups), intending to continueperforming the bundle (16 groups), performing it if un-certain rather than waiting for confirmation of sepsis(13 groups), and intending to improve knowledge ofthe bundle (7 groups). Barriers included participantsbelieving that some steps were more important thanothers (6 groups), and being unlikely to complete the fullbundle if they believed the patient to be well (6 groups).Discord throughout this domain was evident in ED staffexpressing strong competing priorities compared to othergroups.

KnowledgeThis was an important facilitator domain – most partici-pants knew about the Sepsis Six bundle (16 groups), andthe evidence supporting it (13 groups).

Memory, attention and decisionsOverall this domain had various facilitators towardsSepsis Six performance, with participants finding thebundle easy to remember (16 groups), and that using itregularly helps this (13 groups). There was one stronglydiscordant belief, with about half of participant groups

reporting that missing sepsis was an important barrierfor them compared to the other groups who believedthat for them, rarely missing sepsis was an importantfacilitator.

OptimismAll sixteen groups believed that improving performance ofthe bundle would improve patient care. There was astrongly discordant belief about whether performance ofthe bundle at the hospital would improve or not, withpessimism amongst ED and MAU juniors making this abarrier for the overall sample (3 groups), despite theoptimistic facilitator belief being true for many others (12groups).

ReinforcementThis domain featured two barrier beliefs, with neither in-dividuals nor the hospital being rewarded for performingthe bundle. Despite this, most participants reported thesebarriers as unimportant barriers to their ongoing SepsisSix performance.

SkillsMost participants believed their skills in the Sepsis Sixitself to be an important facilitator (15 groups). Therewas, however, a strongly discordant belief regardingwhether training in the bundle was sufficient or not, witharound half of the groups seeing lack of training as an im-portant barrier (7 groups) – compared to the remaining 9groups who believed the training they received was an im-portant facilitator.

Social and professional roleOverall this was a domain of important facilitators – allsixteen groups expressed as important facilitators thatmost aspects of Sepsis Six performance were part oftheir role, from identification of septic patients, decid-ing to perform the bundle to performing it themselves.One common important barrier was the high turnover ofboth medical and nursing staff in their roles (10 groups).A discordant theme was how much of the Sepsis Six aparticipant was actually allowed to perform. The majorityof groups found this as a strong facilitator (11 groups), butmedical and especially surgical nurses (2 groups) feltrestricted and found this limitation of role to be an im-portant barrier. Of note, ED nurses, who regularly do ven-ous gases, blood cultures and have recently had protocolsimplemented to allow them to give the first dose of antibi-otics whilst waiting for medical review, expressed this be-lief as a strong facilitator.

Social influencesThis domain had both important facilitators and barriers.Most participants (13 groups) expressed as a strong


facilitator that they could escalate septic patients, and thatthey felt that their colleagues believe the Sepsis Six to beof benefit (16 groups). There were many areas of discord -departmental culture was an important barrier for sixgroups but a facilitator for the other ten, for example therewas a culture of not measuring a urine output in ED. Sixgroups expressed lack of leadership in improving bundleperformance as an important barrier, compared to eightgroups expressing good leadership in bundle performanceas an important facilitator.

DiscussionThis study illustrates a structured and replicable theory-based approach to identify multiple barriers and facilita-tors towards performing the Sepsis Six in admissions areasof a large UK hospital. This addresses a recently identifiedgap in theory-based implementation research in emer-gency care, and illustrates a method which could be usedto improve implementation of other similar interventions[15]. There are multiple barriers, including insufficient re-sources, insufficient training, poor communication andteamwork and lack of audit or feedback on performance.However, participants were confident in their knowledgeand skills when performing the bundle and believed it tobe beneficial to their patients. Whilst some beliefs arecommon between participant groups, there are manyareas of discord identified where beliefs that are relativebarriers for one group are relative facilitators for another.

Where does this fit with previous TDF-based research?Previous TDF-based research into Sepsis Six perform-ance [27, 28] identified several broad themes affectingSepsis Six performance which this study supports:“Knowing what to do and why” is supported by thisstudy’s facilitator beliefs around knowledge of the SepsisSix and its supporting evidence, and facilitator beliefsaround skill levels. “Risks and benefits” is supported bythis study’s important facilitator beliefs in the ‘Beliefsabout consequences’ domain, with widespread beliefs thatthe bundle is of benefit, with a few concerns about someelements in certain patient groups, for example fluids inheart failure. “Working together” was an important dis-cordant theme identified in ‘Beliefs about capabilities’ and‘Social Influences’ domains, and was variably an importantbarrier or facilitator, as were beliefs in the ‘Social Influ-ences’ and ‘Behavioral Regulation’ domains based around“Empowerment and Support”. “Staffing levels” is a themewidely addressed in the important barrier beliefs from thisstudy’s ‘Environment, Context and Resources’ domain.However, this study builds on previous research throughidentification of specific belief statements, and areas ofdiscord between clinical areas and staff groups. This isvitally important when considering intervention design.For example, ED nurses had enough equipment like

antibiotics or access to oxygen, but had problems withsome of its function – such as the blood gas machine,whereas medical nurses had functional equipment but notenough of it. Being able to systematically identify detailsuch as this allows appropriate focus of interventions – inthis case, on putting a 24-h repair service in place for theED gas machine, and reviewing MAU equipment to allowpurchase of shortage items such as drip stands.

Where does this fit in with previous ethnographicanalysis?Previous ethnographic research in Scotland identifies awide spectrum of complex systematic barriers and facili-tators towards Sepsis Six performance, far beyond thesix simple steps the bundle is intended to be. Many ofthese themes are also identified in the site studied here,for example the importance of teamwork and communi-cation, problems when prioritizing competing tasks, andmaintaining a purpose in completing the whole bundlewhen acting in a system under pressure [12].More pertinently, ethnographic analysis of ongoing Sep-

sis Six quality improvement work suggested a benefit inapplying systematic methods when analyzing elements ofcomplex task performance to identify where problems lie.This is supported by the presence of discordant themesseen in this study, with some beliefs acting as barriers forone role but facilitators for another. Additionally, thisstudy has identified the pressures brought about by thesystem, such as extreme pressure of resources or difficul-ties prioritizing, looking beyond the usual implementationapproach of individual behavior change [12]. As suggestedin the ethnographic analysis, this study presents a methodthat uses a theory-based approach that looks at both indi-viduals and systems when analyzing barriers and facilita-tors to implementation.

Strengths and limitations of using the TDF to addresssepsis six implementationThe Sepsis Six is intended to be simple yet is poorlyperformed in day-to-day clinical practice. The systematicmethods used in this paper provide an in-depth insightinto factors affecting this performance in staff groups atthe study hospital. Whilst many identified beliefs mayhave been intuitive, such as those identifying resourceshortages, the TDF-based method identified a broadspectrum of contextualized beliefs beyond this, such asleadership of the Sepsis Six quality-improvement process.The frequency and importance of these beliefs are likelyto differ between and across individuals and groups, andthus the use of questionnaires to assess generalizabilityand importance of beliefs allowed identification of whichstaff groups and clinical areas perceive them as barriers orfacilitators. Furthermore, identification of discordant be-liefs between staff groups, facilitates more targeted use of


behavior change techniques and hence better-designed in-terventions to improve performance of clinical behaviorssuch as the Sepsis Six – for example, a technique to im-prove teamwork could be focused on MAU or ED ratherthan surgery given findings that poor teamwork was animportant barrier for these departments but an importantfacilitator for surgery.TDF domains have also been linked to Behavior Change

Techniques (BCTs) in order to create a ‘treatment’ inter-vention for the ‘behavioral diagnosis’. Previous TDF-basedSepsis Six research has then been used to systematicallyguide the design of an intervention at the study hospital[26, 27]. The data collected in this study could be used todo similar.There were several aspects of the method designed to

reduce subjectivity during the analysis, including regular,iterative, multidisciplinary consensus discussions to cri-tique the data and emerging analysis.There may have been a degree of self-selection bias

whereby those who completed the questionnaire weremore likely to be engaged in quality improvement or in-terested in sepsis. The response rate, out of all eligiblestaff, is not unreasonable when compared to similar cross-sectional TDF studies and therefore the authors believethis to be a fair and representative sample, but those whodid not take part may be less motivated or knowledgeableabout sepsis care, biasing the results [38]. Overall, theresults had a positive skew, towards the ‘facilitator’ andthe ‘important’ end of the Likert scale used. This did notaffect the usefulness of the results, since a belief scoring asan important weak facilitator within an analysis group, ifall other beliefs were important strong facilitators, wasjudged to be as relevant to address in an intervention as abelief actually scoring as a barrier. Furthermore, our ana-lysis method allows for each participant to have a different‘baseline’ to their Likert scale, hence the positive skewshould not have affected the results in terms of producingresults which are practically useful in intervention design.In addition, the results of larger participant subgroups (forexample, ‘Surgery’) may have been influenced by theircomposition (for example if made up mostly of nurses).The pre-planned subgroup analysis aims this should notcompromise the results for individual staff groups. Com-bining quantitative and qualitative data in this way inorder to identify important beliefs, has been performedpreviously using the TDF [22]. The method used in thisstudy enabled identification of pragmatic and useful re-sults through in-depth qualitative exploration of a prob-lem then larger scale quantitative research.

Further researchAs a single hospital study, the generalizability of resultsmay be limited. Different hospitals have different pressuresand cultures as demonstrated by differences between roles

and departments in this study. However, the concord-ance seen between this study and the previous singlecenter and multi-center studies suggest there may becommon themes in factors affecting Sepsis Six deliverybetween hospitals. If identified, these could be targetedfor behavior change intervention at a national level.Completion of the questionnaire in other hospitals toassess for generalizability of results would systematicallyidentify any targets for larger scale quality improvement,or reinforce the need for other hospitals to undertake asimilar process through identification of discordance be-tween participant groups. A multicenter study is currentlyunderway using the questionnaire produced in this study.

ConclusionsOverall, this study describes a systematic, replicablemethod for identifying targets for behavior change inorder to improve performance of an intervention, in thiscase the Sepsis Six. There were common barriers relatingto lack of staffing and other resources, variable audit andfeedback, variable communication, leadership and team-work amongst staff groups. However there was a goodknowledge and skill-base relating to the Sepsis Six, firmbelief in the positive consequences of the bundle, andstrong beliefs in it being part of each participant’s role toidentify septic patients and perform the bundle. Hospitalscould use theory-based methods such as this to systemat-ically identify barriers and facilitators towards their ownsepsis bundle performance. They could then use thesedata to systematically guide design of an intervention.Conducting a behavioral diagnosis using this methodcould facilitate better and more efficient performance im-provement not only in sepsis but throughout acute care.

Additional files

Additional file 1: Barriers and FaciLitators to Implementing the SepsisSix (BLISS) Topic Guide. (DOC 88 kb)

Additional file 2: Barriers and Levers to Implementing the Sepsis Six(BLISS) Questionnaire. (DOCX 67 kb)

Additional file 3: Table demonstrating belief statement occurrence anddata saturation in analyzed interviews (Participants 1–10) with absence ofnew themes in Participant 11. (DOCX 30 kb)

Additional file 4: Complete results table with median scores andinterquartile ranges (IQR). (DOCX 225 kb)

AbbreviationsBCT: Behavior Change Technique; ED: Emergency Department; MAU: MedicalAdmissions Unit; SAU: Surgical Admissions Unit; TDF: Theoretical DomainsFramework

AcknowledgementsMs. Mandy Gorton for her assistance creating the online survey.

FundingSociety of Devon Intensive Therapists (SODIT) provided funding for interviewtranscription from its annual research bursary. There was no other involvementof outside agencies and no influence on any aspect of the study.


dx.doi.org/10.1186/s13049-017-0437-2

dx.doi.org/10.1186/s13049-017-0437-2

dx.doi.org/10.1186/s13049-017-0437-2

dx.doi.org/10.1186/s13049-017-0437-2

Availability of data and materialsSee Supplementary Digital Content for full results table, data saturation table,questionnaire and interview topic guide.

Authors’ contributionsNR participated in study conception, design and coordination, performedand coded the interviews, collected survey data and analyzed the data, andparticipated in drafting and editing the manuscript. GH also participated instudy conception, design and coordination, performed and coded theinterviews, collected survey data and analyzed the data, and participated indrafting and editing the manuscript. FL participated in study conception anddesign, provided expert guidance in implementation and health psychologyduring design, data collection and data analysis, and participated in draftingand editing the manuscript. WS participated in study design, collectedsurvey data, and participated in drafting and editing the manuscript. MSparticipated in study conception, design and coordination, and in draftingand editing the manuscript. All read and approved the final manuscript.

Ethics approval and consent to participateLocal Research and Development department approval obtained confirmingService Development Project status and waiving need for formal ethicscommittee review, which was not sought. Interview participants providedinformed written consent. Questionnaire participants were provided with aninformation sheet and informed consent was given as part of completion ofquestionnaire.

Consent for publicationInterview participants were consented for use of anonymized quotation oftheir data. No other consent to publish is applicable.

Competing interestsThe authors declare that they have no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Author details1Departments of Critical Care and Anaesthesia, Royal Cornwall Hospital,Truro, Cornwall TR1 3LJ, UK. 2Centre for Health Services Research andManagement, School of Health Sciences, City University of London, London,UK.

Received: 30 March 2017 Accepted: 6 September 2017

References1. Singer M, Deutschman C, Seymour C, Shankar-Hari M, Annane D, Bauer M,

et al. The third international consensus definitions for sepsis and septicshock (Sepsis-3). JAMA. 2016;315(8):801–10.

2. NHS England Commissioning for Quality and Innovation (CQUIN) Guidancefor 2015–16 (2015) https://www.england.nhs.uk/wp-content/uploads/2015/03/9-cquin-guid-2015-16.pdf (Last accessed 26th June 2016).

3. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in thetreatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368–77.

4. Kaukonen KM, Bailey M, Suzuki S, Pilcher D, Bellomo R. Mortality related tosevere sepsis and septic shock among critically ill patients in Australia andNew Zealand, 2000-2012. JAMA. 2014;311:1308–16.

5. The ProCESS Investigators. A randomized trial of protocol-based care forearly septic shock. N Engl J Med. 2014;370:1683–93.

6. The ARISE Investigators and the ANZICS Clinical Trials Group. Goal-directedresuscitation for patients with early septic shock. N Engl J Med. 2014;371:1496–506.

7. The ProMISe Trial Investigators. Trial of early, goal-directed resuscitation forseptic shock. N Engl J Med. 2015;372:1301–11.

8. Cronshaw HL. Impact of the surviving sepsis campaign on the recognitionand management of severe sepsis in the emergency department: are wefailing? Emerg Med J. 2011;28:670–5.

9. Kissoon N. Sepsis guideline implementation: benefits, pitfalls and possiblesolutions. Crit Care. 2014;18:207.

10. Kuo J, Chang H, Wu P, et al. Compliance and barriers to implementing thesepsis resuscitation bundle for patients developing septic shock in thegeneral medical wards. J Formos Med Assoc. 2012;111:77–82.

11. Daniels R, Nutbeam T, McNamara G, Galvin C. The sepsis six and the severesepsis resuscitation bundle: a prospective observational cohort study. EmergMed J. 2011;28:507–12.

12. Tarrant C, O’Donnell B, Martin G, et al. A complex endeavour: anethnographic study of the implementation of the sepsis six clinical carebundle. Implementation Sci. 2016;11:149.

13. Charani E. E. Castro-Sanchez, N. Sevdalis, Y. Kyratsis, L. Drumright, N. Shah,and A. Holmes. "Understanding the determinants of antimicrobialprescribing within hospitals: the role of “prescribing etiquette”.". Clin InfectDis. 2013;57(2):188–96.

14. Davey P, et al. Time for action—improving the design and reportingof behavior change interventions for antimicrobial stewardship inhospitals: early findings from a systematic review. Int J AntimicrobAgents. 2015;45(3):203–12.

15. Tavender EJ, Bosch M, Fiander M, Knott JC, Gruen RL, O’Connor D.Implementation research in emergency medicine: a systematic scopingreview. EMJ 2015;(ePublished ahead of print).

16. Charani E, Edwards R, Sevdalis N, Alexandrou B, Sibley E, Mullett D, FranklinB, et al. Behavior change strategies to influence antimicrobial prescribing inacute care: a systematic review. Clin Infect Dis. 2011;53(7):651–62.

17. Eccles M, Grimshaw J, Walker A, et al. Changing the behavior of healthcareprofessionals: the use of theory in promoting the uptake of researchfindings. J Clin Epidemiol. 2005;58(2):107–12.

18. Michie S, Prestwich A. Are interventions theory-based? Development of atheory coding scheme. Health Psych. 2010;29(1):1.

19. Davidoff F, Dixon-Woods M, Leviton L, et al. Demystifying theory and its usein improvement. BMJ Qual Saf Published Online First: 23 January 2015.doi: https://doi.org/10.1136/bmjqs-2014-003627

20. Cane J, O’Connor D, Michie S. Validation of the theoretical domainsframework for use in behavior change and implementation research.Implement Sci. 2012;7:37.

21. Francis JJ, O’Connor D, Curran J. Theories of behavior change synthesisedinto a set of theoretical groupings: introducing a thematic series on thetheoretical domains framework. Implement Sci. 2012;7:35.

22. Cuthbertson BH, Campbell MK, MacLennan G, et al. Clinical stakeholders’opinions on the use of selective decontamination of the digestive tract incritically ill patients in intensive care units: an international Delphi study.Crit Care. 2013;17:R266.

23. Gould NJ, Lorencatto F, Stanworth SJ, Michie S, Prior ME, Glidewell L, et al.Application of theory to enhance audit and feedback interventions toincrease the uptake of evidence-based transfusion practice: an interventiondevelopment protocol. Implement Sci. 2014;9(1):92.

24. Roberts N, Lorencatto F, Manson J, Brundage S, Jansen J. What helpsor hinders the transformation from a major tertiary center to a majortrauma center? Identifying barriers and enablers using the TheoreticalDomains Framework. Scand J Trauma Resusc Emerge Med.2016;24:30.

25. Cane J, Richardson M, Johnston M, et al. From lists of behavior changetechniques (BCTs) to structured hierarchies: comparison of two methods ofdeveloping a hierarchy of BCTs. British J Health Psych. 2015;20(1):130–50.

26. Michie S, Johnston M, Francis J, et al. From theory to intervention: mappingtheoretically derived behavioral determinants to behavior changetechniques. Appl Psychol. 2008;57(4):660–80.

27. Steinmo S, Fuller C, Stone S, Michie S. Characterising an implementationintervention in terms of behavior change techniques and theory: the ‘sepsissix’ clinical care bundle. Implement Sci. 2015;10:111.

28. Steinmo S, Michie S, Fuller C, Stanley S, Stapleton C, Stone S. Bridging thegap between pragmatic intervention design and theory: using behavioralscience tools to modify an existing quality improvement programme toimplement “sepsis six”. Implement Sci. 2016;11:14.

29. Personal communication, J Stratton, Governance Lead for Medicine, RoyalCornwall Hospital, 2015.

30. Francis JJ, Johnston M, Robertson C, et al. What is an adequate sample size?Operationalising data saturation for theory-based interview studies. PsycholHealth. 2010;25:1229–45.

31. McSherry LA, Dombrowski SU, Francis JJ, et al. ‘It’s a can of worms’:understanding primary care practitioners’ behaviors in relation to HPV usingthe theoretical domains framework. Implement Sci. 2012;7(1):73.


https://www.england.nhs.uk/wp-content/uploads/2015/03/9-cquin-guid-2015-16.pdf

https://www.england.nhs.uk/wp-content/uploads/2015/03/9-cquin-guid-2015-16.pdf

http://dx.doi.org/10.1136/bmjqs-2014-003627

32. Pope C, Ziebland S, Mays N. Qualitative research in health care. Analysingqualitative data. BMJ. 2000;320(7227):114–6.

33. Ritchie J, Lewis J. Qualitative research practice. London: Sage publications; 2003.34. Duncan EM, Francis JJ, Johnston M, et al. Learning curves, taking

instructions, and patient safety: using a theoretical domains framework inan interview study to investigate prescribing errors among trainee doctors.Implement Sci. 2012;7(1):86.

35. Curran JA, Brehaut J, Patey AM, Osmond M, et al. Understanding theCanadian adult CT head rule trial: use of the theoretical domains frameworkfor process evaluation. Implement Sci. 2013;8(25):10.

36. Islam R, Tinmouth AT, Francis JJ, et al. A cross-country comparison of intensivecare physicians’ beliefs about their transfusion behavior: a qualitative studyusing the theoretical domains framework. Implement Sci. 2012;7(1):93.

37. Bussieres AE, Patey AM, Francis JJ, et al. Identifying factors likely to influencecompliance with diagnostic imaging guideline recommendations for spinedisorders among chiropractors in North America: a focus group study usingthe theoretical domains framework. Implement Sci. 2012;7:82.

38. Beenstock J, Sniehotta F, White M, Bell R, Milne E, Araujo-Soares V. Whathelps and hinders midwives in engaging with pregnant women aboutstopping smoking? A cross-sectional survey of perceived implementationdifficulties among midwives in the North East of England. ImplementationSci. 2012;7:36.

• We accept pre-submission inquiries

• Our selector tool helps you to find the most relevant journal

• We provide round the clock customer support

• Convenient online submission

• Thorough peer review

• Inclusion in PubMed and all major indexing services

• Maximum visibility for your research

Submit your manuscript atwww.biomedcentral.com/submit

Submit your next manuscript to BioMed Central and we will help you at every step:


barriers and facilitators towards implementing the sepsis

Documents