For this qualitative evidence synthesis, there were 18 eligible studies, which focused on why HCPs may fail to escalate as per the EWS protocols (see Figure 1, Study Flow Diagram). The findings of these 18 qualitative studies are synthesised and presented in this paper.
Characteristics of included studies
Eighteen qualitative studies were eligible for inclusion with three conducted in Australia,(17-19) six in the UK,(20-25) five in the USA,(26-30) and one each in Ireland,(31) Norway,(32) Denmark,(9) and Singapore.(33) Ten studies included nurses only (registered, unregistered),(9, 19, 23, 25, 26, 28-30, 32, 33) three studies includes nurses and doctors only,(18, 24, 31) and five studies included a mixture of HCPs and staff [nurses, physicians, administrators, respiratory technicians, health care assistants, safety leads and managers].(17, 20-22, 27) A total of 599 participants were interviewed across the studies with sample sizes ranging from six participants(25) to 218 participants.(17) To gain an understanding of the barriers and facilitators to escalation, eight studies used face-to-face interviews,(19, 20, 23, 24, 26, 27, 31, 33) and seven studies used focus groups.(9, 17, 18, 25, 28, 29, 32) Three studies (21, 22, 30) used a combination of methods including interviews, observations of interactions, and documentary evidence [protocols and audit data], two of which were conducted in the same hospital and sample.(21, 22) The first study by Mackintosh (2012)(21) contained 150 hours of observations and used thematic analysis while the second study (Mackintosh, 2014)(22) contained 180 hours of observation and the analysis focused on the structural conditions that shape delivery of the rapid response drawing on Bourdieu's logic of practice. Data from both were extracted for this thematic analysis. The key study characteristics are outlined in Table 1.
Thematic synthesis produced five overarching themes and 22 sub-themes with multiple interdependencies. These are categorised into barriers (twelve sub-themes) and facilitators (ten sub-themes) of escalation. These are described for each of the five overarching themes: Governance, Rapid Response Team (RRT) Response, Professional Boundaries, Clinical Experience, and EWS Parameters (see Figure 2).
Synthesis of results
Barriers to escalation
Quotations from either primary study participants (in italics) or study authors relating to the ‘barriers’ for each key theme (n=5) and sub-theme (n=10) are presented in Table 2.
Governance
‘Governance’ refers to the overall organisational or institutional specific factors affecting why HCPs fail to escalate, or barriers to escalation. Fourteen papers described governance issues as factors contributing to a failure to escalate care.(9, 17-22, 24-27, 30, 31, 33) Three sub-themes including Standardisation, Resources and Lack of accountability were identified.
‘Standardisation’ was an issue reported in twelve studies.(17-21, 25, 26, 30, 31, 33) Standardisation included a lack of clear policies or protocols for action which was reported in four studies(20, 21, 25, 30) and this led to inaction or confusion amongst staff as to who to call or when. In addition to a lack of clear policies or protocols, ‘standardisation’ included a lack of knowledge of policies or protocols by staff, reported in six studies.(17-20, 26, 30) Where staff were not familiar with the correct protocol for escalation this was a barrier to escalation. Lack of education or training was reported in six studies by participants with no standardised, or regular training in place.(18, 19, 25, 26, 31, 33)
‘Resources’ were reported as barriers in nine studies(9, 17, 20, 24, 26, 27, 30, 31, 33) whereby staffing shortages, particularly in conducting the required monitoring of patients, (eight studies),(9, 17, 20, 24, 27, 30, 31, 33) poor communication systems/protocols (three studies)(17, 20, 31) and the perceived workload of the RRT (six studies)(17, 20, 24, 26, 31, 33) were all reported as barriers to escalation: “Perceived busyness of the ICU nurses discouraged participants from RRT activation. Participants noted that responding RRT members occasionally talked about how busy they were.”
‘Lack of accountability’ and a blame culture was a reported sub-theme in three papers.(21, 22, 33) This was particularly the case in settings where health care assistants (HCAs) or equivalent staff were involved in documenting patient vital signs. HCAs believed there was often blame put on them by more senior staff when something went wrong. For example, junior staff described situations where a patient deteriorated and they informed senior staff, but the senior staff did not escalate care, and then when the patient collapsed or deteriorated the blame was put on the junior staff member.(21, 22, 33) This lack of accountability of senior staff was a barrier to these staff in raising concerns about deterioration.
RRT Response
‘RRT Response’ refers to how the RRT responded when a call for help was made. This key theme was apparent in ten papers.(9, 18-20, 25-27, 29, 30, 32) Two sub-themes including RRT behaviours and Fear were identified.
‘RRT behaviours’ were a barrier to escalation or subsequent escalation calls when a ‘lack of professionalism’ was shown by the RRT to the staff who made the call. This was reported in eight papers.(9, 18-20, 26, 29, 30, 32) A ‘negative response’ or a total ‘lack of response’ (i.e. the RRT did not come) was also a barrier to escalation or subsequent escalation reported in eight papers.(9, 18-20, 25, 26, 29, 32) Participants reported being questioned as to whether the call to the RRT was necessary, they often reported feeling belittled or criticised and the experience of this negative response was a barrier to subsequent escalation.
Participants reported ‘fear’ was a barrier to escalation in seven papers.(9, 18-20, 26, 27, 30) ‘Fear of reprimand’ by members of the RRT for activating a call was reported by participants as well as ‘fear of looking stupid or dumb’ to colleagues, both of which were significant barriers to escalation.
Professional Boundaries
‘Professional boundaries’ refers to the barriers to escalation that were apparent in the included studies surrounding hierarchy, power, and jurisdictional control. Ten papers described professional boundaries as core contributing factors to not escalating.(18-22, 24, 26, 27, 30, 31) Two sub-themes including Hierarchy and Increased conflict were identified.
Participants described having to negotiate hierarchical boundaries in order to escalate care in eight papers.(18, 20-22, 25-27, 30) In some instances, participants described being reprimanded by the patient’s primary ward physician for calling the RRT. The primary ward physician often felt it was “their patient and their job to look after them” and that the junior staff had “gone over their head” in calling the RRT.(27, 34) This in turn led to an increase in conflict between nurses and ward physicians. Calling for help (escalation) also led to increased conflict among other staff.(18, 19, 21, 24, 26, 27, 31) In particular, the use of the RRT was often viewed as a jurisdictional shift in responsibility for acutely ill patients by members of the RRT who felt some nurses “washed their hands” of the situation. This may contribute to the negative responses of RRT, as described above.
Clinical Experience
‘Clinical experience’ refers to the barriers to escalation specifically related to individual staff and their level of confidence and ability to detect deterioration, which was reported in six studies.(9, 19, 20, 24, 26, 30) Two sub-themes including Clinical over confidence and Lack of clinical confidence were identified.
‘Clinical over confidence’ reported in five papers, (9, 20, 24, 26, 30) was characterised by participants being overly confident in their clinical ability. Participants expressed confidence that their clinical judgement was a better gauge of when to escalate care, irrespective of the EWS, and also that they were better placed to care for their own patient rather than the RRT.
In contrast, ‘lack of clinical confidence’, was reported in three studies.(19, 26, 30) Here it was participant’s inability to detect deterioration or doubting their own skills and ability to detect deterioration that led to a delay in escalation or to no escalation.
Early Warning System Parameters
‘EWS Parameters’ refers to the system specific barriers to escalation, which were reported in eight studies.(9, 17, 21-23, 28, 31, 33) One sub-theme, Patient variability was identified.
‘Patient variability’ that is the existence of specific groups of patients, for example, those with chronic obstructive pulmonary disease, was reported as a barrier. For these patients, who by default fall outside the normal range for the various vital signs, participants reported either excessive triggering of the EWS or else staff simply ignored the EWS for these patients. “The inability of the MEWS to tailor alarm settings and limits to accommodate patients whose vital sign measurements normally fell outside predetermined thresholds was cited by focus group participants as a major barrier to effective use of the system”.(28) The need for parameter adjustment was also cited within the patient variability sub-theme: participants reported that parameters were rarely reviewed or adjusted and that this was a continual problem for interns and nurses "If parameters aren’t charted you're expected to check the observation and inform the intern more than is necessary".(31)
The themes of ‘governance’, ‘professional boundaries’, ‘RRT Response’, ‘Clinical Experience’, and ‘Early Warning System Parameters’ are individual but inter-related barriers to escalation of care. Each theme may be its own barrier, but when taken together they create an environment in which escalation of care may occur too late or not occur at all. For example, a lack of governance such as a lack of clear policies or protocols, or lack of knowledge of policies or protocols by all staff creates the potential for conflicts in professional boundaries. This may create a level of ‘fear’ for junior staff, particular those with less clinical confidence, who experience negative attitudes from both the RRT and primary ward physicians which contribute to a reluctance to activate the RRT in the future.
Facilitators to escalation
Illustrative quotations from primary study participants or study authors relating to the facilitators of escalation for each key theme (n=5) and sub-theme (n=12) are presented in Table 3.
Governance
‘Governance’ was a key theme within ten papers.(17, 20-24, 26, 28, 30, 31) Three sub-themes of ‘Accountability’, ‘Standardisation’ and ‘Resources’ as facilitators of escalation among the study participants were identified.
Accountability was a motivating factor in four studies, whereby staff activated the RRT in order to ‘cover their own backs’ in case something went wrong.(20-22, 31) In this respect, the RRT was viewed as a safety net by the nurses and they valued the extra support it provided.
In addition, ‘standardisation’ was reported in seven studies, where clear policies or protocols for action (17, 21-23, 26, 30, 31) and participant knowledge of these policies or protocols for escalation (21, 22, 31) was a key facilitator of escalation. A clear outline of when to call and who to call, that was communicated to and understood by all staff members, was a facilitator of escalation.
Resources (that is sufficient staffing levels and good communication such as use of handover tools) was a key facilitator of escalation in seven studies,(20, 21, 23, 24, 26, 28, 31) as exemplified here: "There is now a single resident who covers the ward for the week and twice daily attending ward rounds. I think this has made things better for juniors because they have a single point of contact who is not going to be off site or in theatre".(20)
RRT Response
The behaviours of RRTs were reported as facilitators of escalation within this key theme in ten studies.(9, 18, 20, 21, 24, 26, 27, 29, 30, 32) Three sub-themes of ‘RRT behaviours’ (including professionalism, decision-makers and collaborative), ‘Expertise’ and ‘Additional support’ were identified.
In terms of RRT behaviours, where there was a ‘professional response’ or a ‘positive response’ from the RRT, this encouraged staff to escalate in subsequent events.(9, 20, 26, 30, 32) The RRT were seen as ‘decision-makers’ and ‘doers’ in emergency situations and these were both facilitators of escalation.(20, 26, 32) The RRT were viewed as collaborative but also of facilitating collaboration between staff, and this was another facilitator of escalation within three studies.(24, 29, 32)
In addition to how the RRT behaved, they were also described as being ‘experts’(9, 24, 26, 27) with specific specialised skills and expertise necessary when a patient deteriorated.
They were also seen as providing ‘additional support’ (18, 21, 24, 26, 27, 29, 30, 32) in emergency situations and this was a source of comfort reported by participants.
Professional Boundaries
Professional boundaries as a key theme was included in nine studies.(17-21, 24, 27, 30, 31) This included the sub-themes of a ‘Licence to escalate’ and a ‘Bridge across boundaries’.
Licence to escalate was where the staff perceived the EWS as tool to enable escalation across hierarchical and occupational boundaries and was apparent in nine studies, (17-21, 24, 27, 30, 31) as exemplified from the following extracts: “Across both sites the score provided staff with the licence to escalate care across hierarchical and occupational boundaries”.(21) "The nurses actually have something they can do about it versus just kind of watching the patient flounder".(27) The EWS was used as tool by nurses to establish a legitimate reason for escalating care to more senior staff without having to seek permission. This licence created a ‘Bridge across boundaries’. This refers to the view that the EWS facilitates cross-profession communication and teamwork and is a workaround and means of getting something done, i.e. getting a patient seen to, and was referenced in four studies. (18, 20, 21, 27) "We used to actually use them as a way of getting round a resident or whoever who really wasn't doing what you know you needed for your patient".(18)
Clinical Experience
Clinical experience was a key theme within 12 studies and included sub-themes of ‘Clinical confidence’ (to recognise deterioration, confidence in own ability and skills), ‘Empowerment/validation’ and ‘Clinical judgement’.(9, 17-21, 23, 24, 26-31)
Where a staff member had clinical confidence in their own skills and ability and were able to recognise deterioration, this was a facilitator of escalation. Staff were confident enough to activate the RRT.(17, 23, 26, 28, 30)
Staff also felt ‘empowered’ by the EWS and the EWS ‘validated’ their reasons for escalation and calling for help from the RRT and seniors.(17, 18, 20, 21, 23, 27, 29, 31)
‘Clinical judgement’ was a facilitator of escalation in seven studies where staff referred to the importance of using clinical judgement when a patient deteriorates and not relying on a score or system alone.(9, 17, 18, 20, 26, 30, 31)
Early Warning System Parameters
The fifth key theme of EWS Parameters included the subtheme of ‘Triage mechanism’ and a ‘Tool for detecting deterioration’.(9, 19, 21, 23, 26-28, 31)
Staff described using the EWS as a mechanism for triage, to get a patient a higher level of care and to ensure patient safety. In addition, the EWS was seen as a valuable tool for picking up patient deterioration by staff and optimising patient outcomes. Doctors described using the system to gauge the severity of a patient's condition for triaging: "When I'm contacted to review a patient, I use 'NEWS' to prioritise the urgency in which they need to be reviewed.(31)
Just as the themes of ‘governance’, ‘professional boundaries’, ‘RRT Response’, ‘Clinical Experience’, and ‘Early Warning System Parameters’ were inter-related in the generation of barriers to escalation of care, the themes are inter-related in creating facilitators to the escalation of care. For example, clear governance in terms of policies or protocols, and knowledge of policies or protocols by all staff decreases the potential for conflicts in professional boundaries and increases role clarity. This in turn may create a more collaborative team-based approach that provides reassurance and confidence as opposed to engendering a level of ‘fear’ in junior staff, particularly those with less clinical confidence. All of which combines to create a climate within which activation of the RRT is more likely to happen.
Quality Appraisal
The CASP tool was used for quality appraisal (see Additional file 3). All 18 studies reported a clear statement of the aims. All 18 studies were judged to have used an appropriate qualitative methodology [e.g. focus groups or interviews], and were judged to have employed appropriate data collection methods (e.g. interviews or focus groups or observation techniques or document review). All studies had a clear statement of the findings and the research was deemed valuable. Seven out the 18 studies were judged to have a research design appropriate to the study aims, (18, 19, 22, 24, 30, 31, 33) while in 11 of 18 studies there was insufficient information on the rationale for the chosen qualitative methodology.(9, 17, 20, 21, 23, 25-29, 32) Thirteen out of the 18 studies were judged to have a recruitment strategy appropriate to the study aims (e.g. convenience sampling or purposeful sampling), (9, 17-24, 26, 27, 30, 32) in four studies there was insufficient information provided.(25, 28, 29, 33) In one study the recruitment strategy was deemed inappropriate (the study authors used ‘their judgement’ and snowball techniques).(31) Six out 18 studies considered the researcher and participant relationship within the study, (9, 19, 21, 25, 28, 29) while 11 out the 18 studies did not consider the researcher-participant relationship and the potential for bias this may introduce.(17, 18, 20, 22-24, 27, 30-33) In one study insufficient information was provided.(26) Seventeen out of the 18 studies reported having ethical approval while in one study it was unclear.(28) Fifteen out of 18 studies were judged to have rigorous data analysis (e.g. inductively and deductively coded, content analysis), (9, 18, 19, 21-25, 27-31, 33) while in two studies there was insufficient detail provided within in the study (the authors mentioned triangulation but provided no other details and in the second study no coding framework was provided).(20, 26) In one study, the analysis was deemed insufficient as there were missing observations which were not reported.(32)
Certainty of the evidence
Confidence in the review findings was assessed using GRADE-CERQual,(14) Overall the certainty of the evidence ranged from moderate to high (see Additional file 4) and there was strong consistency in the findings across studies. Confidence in review findings were downgraded primarily due to methodological limitations.