Objectives: The objective of this study was to explore perceptions of communication quality at pediatric rapid response events and to determine whether these perceptions differed between rapid response team (RRT) members (RRTm) and floor providers (FP).
Methods: This survey study was conducted of clinical providers involved in RRT events at a tertiary care children’s hospital. Perceptions of RRT communication were assessed by using a 5-point Likert scale, and qualitative comments were collected. Responses were compared between RRTm (responder nurses and intensive care fellows) and FP (floor nurses and resident physicians).
Results: Survey response was 64% (18 of 28) for RRTm and 70% (194 of 278) for FP. RRTm gave lower ratings than FP for communication of: (1) the purpose of the call; (2) airway and breathing; (3) circulation; (4) background information; and (5) possible diagnosis and treatment. RRTm were more likely than FP to indicate that description of background information delayed communication of critical management problems (“often”: RRTm, 7 of 17 [41%]; FP, 23 of 175 [13%]; “always”: RRTm, 2 of 18 [12%]; FP, 19 of 175 [11%]; P = .001 for overall comparison). A structured approach for communication was generally supported, although less strongly among floor nurses. Themes from qualitative responses included role confusion, fractured room entry, and a dismissive attitude by RRTm.
Conclusions: A disconnect in perceived quality of communication was observed between RRTm and FP at pediatric rapid response events. A structured approach with well-defined roles may improve communication quality.
Rapid response teams (RRTs) have been increasingly implemented in hospitals as a means to quickly mobilize a multidisciplinary team to assess and manage patients whose condition is deteriorating. A meta-analysis of 5 studies including >400 000 pediatric patients found a 37.7% decrease in rates of cardiopulmonary arrest outside of the PICU and a 21.4% decrease in hospital mortality associated with use of RRTs.1 Implementation of an RRT has also been associated with a decrease in respiratory arrests outside of the PICU.2,3
Communication among providers is essential for providing safe and effective care. Communication errors seem to be especially common in emergency situations.4–6 RRTs (as well as code teams) face particular challenges for communication and teamwork due to the unpredictable and sporadic nature of the events, involving providers with different levels of experience from different environments,7,8 in situations which may or may not involve a true “hand-off” with transfer of ongoing primary responsibility. Surveys in the adult hospital setting have identified barriers to communication at rapid response events that reflect a disconnect between rapid response team members (RRTm) and floor providers (FP), including perceived lack of support of RRT activation,9–11 lack of a shared mental model due to disjointed care plans,7,11 inadequate involvement of bedside nurses,7 and lack of role definition. To our knowledge, no study to date has evaluated perceptions of communication quality at pediatric rapid response events and whether similar barriers to communication exist.
The purpose of the present study was to explore health care providers’ perceptions of communication at pediatric rapid response events. We hypothesized that barriers to communication exist and that perceptions of communication quality differ between RRTm and FP.
This survey study was conducted at a single tertiary care children’s hospital. An anonymous paper version of the survey was distributed in May and July 2011 at Johns Hopkins Children’s Center to all pediatric residents, pediatric floor nurses, PICU fellows, and pediatric responder nurses (PICU nurse responders and pediatric shift coordinator nurses). Shift coordinator nurses at our institution are experienced nurses who assist on all floors and manage beds. An identical electronic survey (www.surveymonkey.com) was sent in January 2012 to the floor nurses who had not initially received the paper survey or been able to complete it due to scheduling or clinical demands. No incentive was provided for completion.
The body of the survey consisted of questions regarding communication at in situ rapid response events that the respondent had previously attended at our institution (the survey instrument is presented in the Appendix). Questions were in multiple choice format on a 5-point Likert-type scale, from “poor” to “excellent” or “disagree strongly” to “agree strongly,” as appropriate. Background information regarding level of experience with rapid response events was also included. The survey instrument was developed through identification of key communication issues identified in previous studies,5,12 as well as the investigators’ first-hand experience at both simulated and in situ RRT debriefings. Nursing staff were queried regarding structured hand-off in general at rapid response events as part of the survey. Physician staff were specifically questioned regarding the mnemonic “ABC-SBAR” (airway, breathing, circulation followed by situation, background, assessment, and recommendation)5; all residents and PICU fellows were expected to have been exposed to this tool during their orientation and in simulation training sessions. Qualitative data were obtained from the following items: (1) “How would you describe patient hand-off at a rapid response, in general?” and (2) “Comments or observations about content/quality of patient hand-off at rapid responses? Have you ever observed or participated in a rapid response in which patient hand-off was particularly effective or ineffective?” Before dissemination, the survey was reviewed by a member of each group (eg, floor nurse, responder nurse, floor physician, responder physician), and minor changes were made for readability. Participants were not provided information regarding the planned grouping or analysis of responses (RRTm versus FP). All survey responses were included in the analysis for each particular question, regardless of completion of the rest of the survey. The study was approved by the Institutional Review Board of Johns Hopkins Medicine.
Johns Hopkins Children’s Center is a tertiary care center; over the time period of the study, it had 180 beds and ∼200 calls for the RRT per year. RRT calls at this center can be initiated by nurses, physicians, other allied health professionals, or families. The RRT may be activated whenever additional personnel are urgently desired for assistance at the bedside, including an evaluation by PICU staff or a severe clinical deterioration (including cardiac arrest), and regardless of whether transfer to a higher level of care is desired. At this institution, there is no separate process or team composition for different types of calls (ie, the same team is called for cardiac arrest versus other types of emergencies).
Calls for the RRT were activated by an operator and sent via text page to the team members. The pages included the location of the emergency, the telephone number of the caller, and a brief description of the emergency (eg, cardiac arrest, respiratory distress, seizure, hypotension) if this information had been quickly and clearly described by the caller. Given that the RRTm do not always know if they are responding to a cardiac arrest or a less urgent issue, they are trained to always assume the worst and respond immediately. Thus, the expected arrival of RRT clinical providers was typically <5 minutes.
The RRT includes 1 PICU fellow, 1 PICU responder nurse (experienced PICU nurses who have undergone additional orientation to the rapid response system), 1 shift coordinator nurse, and 3 pediatric residents (1 from each level of training), as well as representatives from respiratory therapy, pharmacy, security, and chaplaincy if available. Pediatric residents were considered members of the floor team for the study, as the focus was to evaluate communication at the time of arrival of providers with more specialized training in resuscitation. Combined fellows (PICU with pediatric anesthesia or pediatric cardiology) were included as RRTm for the study because they function in the same role as other PICU fellows within the RRT system at our institution.
The data analysis for the present article was generated by using SAS version 9.2 (SAS Institute, Inc, Cary, NC). Continuous data were evaluated by using Student’s t test for parametric data and the Wilcoxon rank-sum test for nonparametric data. Categorical data were evaluated by using the χ or Fisher’s exact test, as appropriate. The Wilcoxon rank-sum test was used for comparison of the Likert-style survey results. A 2-sided P value <.05 was considered statistically significant.
Qualitative data were initially reviewed and organized into themes by 2 authors (M.C.M. and E.A.H.); all authors then reviewed quotations within these themes for further description in the table and text. Any disagreements among authors regarding presentation of themes or inclusion of quotations thereof were resolved by discussion among all authors.
Survey responses were submitted by 18 (64%) of 28 RRTm and 194 (70%) of 278 FP (Table 1), for an overall response rate of 69% (212 of 306). Of those who started the survey, 17 (94%) of 18 RRTm and 177 (91%) of 194 FP completed it. Six of the FP who did not complete the survey indicated that they had not attended at least 1 rapid response event.
The item “Patient hand-off from bedside clinicians to the rapid response team is important for patient care” was rated as agree strongly by nearly all respondents (RRTm: 17 of 18 [94%]; FP: 177 of 185 [96%]; P = .57). Hand-off communication at pediatric rapid response events was rated as generally good by both groups; however, communication of specific elements of hand-off was rated significantly lower by RRTm compared with FP (Table 2).
Ratings of organization of critical information during hand-off (as assessed by critical information being lost or delayed or a desire for increased structure) are presented in Table 3. RRTm were significantly more likely than FP to indicate that description of background information delayed communication of critical management issues (“often”: RRTm, 7 of 17 [41%]; FP, 23 of 175 [13%]; “always”: RRTm, 2 of 18 [12%]; FP, 19 of 175 [11%]; P for overall comparison = .001).
Nurses were asked whether a structured approach (eg, a mnemonic) would help with rapid response hand-off, with RRT nurses responding significantly more positively than floor nurses. Physicians were queried regarding the specific mnemonic “ABC-SBAR” (airway, breathing, circulation followed by situation, background, assessment, and recommendation),5 which is taught during intern orientation at the institution. Although only 29% (20 of 70) of resident physicians reported familiarity with this mnemonic, resident and RRT physicians supported its use for rapid response hand-off (Table 3).
Qualitative responses regarding hand-off at pediatric rapid response events were provided by 16 (89%) of 18 RRTm and 137 (71%) of 194 FP. Themes are summarized in Table 4. Barriers to communication that were repeatedly mentioned by providers included chaos, role confusion, fractured room entry, and dismissiveness by RRTm.
Our study found that RRTm and FP in a tertiary children’s hospital have discordant perceptions of communication quality at pediatric rapid response events and that barriers to communication may be present. RRTm reported that critical information is often poorly communicated or delayed during a rapid response event, whereas FP were significantly less likely to perceive such loss or delay of information. Similar discrepancies in perceived effectiveness of communication have been reported elsewhere in various settings. In 1 study, a researcher directly observed RRT events in a large teaching hospital and found several examples of ineffective communication despite high teamwork ratings by providers.7 Similarly, a study of hand-off by pediatric interns after call shifts found that the most important piece of information was not communicated 60% of the time despite high ratings of hand-off by participants.13
Rapid response communication is distinct from other examples of “hand-off” in health care in that there is an intense initial collaboration of care, which may or may not be followed by a true transfer of primary responsibility if the patient is transferred to the ICU. The short-term goal may be more accurately described as a “briefing” directed at integration of the resuscitation expertise of RRTm with the patient familiarity of FP. In our study, RRTm acknowledged that communication at rapid response events was generally “good” but rated inclusion and organization of critical information less favorably than FP, with 1 PICU fellow noting “if the patient is really sick I focus on ABCs [airway, breathing, circulation] and may not ‘hear’ what they are telling me.” This particular comment reflects a common theme that is highly illustrative of the frame of mind of an ICU-trained rescuer and should be the basis for any structured communication tools that are developed.
Frameworks for communication from the aviation and military communities,14 auto racing,15 and health care16 have emphasized that shared understanding or “situational awareness” must be directed proactively toward actionable items during communication in high-intensity settings to maximize effectiveness. For example, if the FP started with “I am concerned his airway is not open, I am not seeing consistent chest rise…,” he or she will have immediately focused the responders’ attention on an essential action item. From an ICU-trained provider’s perspective, once it is established that the airway is open, the patient is breathing or being ventilated adequately, and has a pulse and adequate perfusion, the provider can then focus on the situation that preceded the current physical state. Alternatively, if the FP starts with language such as “I called you because I entered the room and found the child…” and the RRTm noted inadequate chest rise and possible obstructed airway when glancing at the child on entry to the room, it would be nearly impossible and arguably inappropriate for them to stop and listen to the FP rather than focusing on airway management at that time.
Traditional teaching of oral presentations in medical curricula may instill a framework (eg, chief complaint, followed by history of present illness, medical history) that is insensitive to contextual elements that would indicate what content is relevant to therapeutic intervention.17 In the resuscitation setting, information related to airway, breathing, and circulation is most likely to guide immediate therapy18; these elements are often delayed or omitted in communication during a rapid response event, however.5,8 RRTm may assign increased relative importance to these pieces of critical information relevant to their expertise and anticipated actions, whereas FP may perceive a high-quality “briefing” or “hand-off” as 1 that is delivered quickly and completely in the traditional structure. Furthermore, because pediatric rapid response events only infrequently involve cardiac arrest,2 disorganized communication may not be evident until critical events occur. In our study, 1 responder made the concerning comment: “The more ill the patient, the worse it [communication] usually is.”
According to our study of provider accounts from in situ pediatric rapid response events, communication may not optimally coordinate FP and RRTm expertise. Structured communication holds promise in this setting; however, in our study, floor nurses were less interested in a communication tool (eg, a mnemonic) compared with RRT nurses and physicians, perhaps because of a perception that communication was already adequate. Previous studies have identified that critical care–trained physicians often use a more free-form and interactive approach to routine hand-off to create a shared mental model (rather than using a list- or mnemonic-based approach).18,19 However, in the rapid response setting, our data suggest that critical care–trained physicians and nurses prefer increased structure to direct urgent intervention. Whether ABC-SBAR or another tool is used, team training with RRTm and FP in simulated rapid response events with a strategy to improve goal-directed communication may be useful to improve the perceived deficiencies observed in our study.
The recurring theme of “chaos” reported in the qualitative comments in our study is likely closely interlaced with the observations of role confusion and fractured room entry. Role confusion in the rapid response setting, especially with respect to a clearly defined leader and the bedside nurse, has been commonly cited in previous reports.7,8 An evaluation of hand-off strategies in industries (including spaceflight, nuclear power, and railroad dispatching) cited “unambiguous transfer of responsibility” as a key element of safe hand-off.20 Ambiguity regarding whether a “briefing” or true “hand-off” with transfer of primary leadership is occurring upon RRTm arrival may provide a particular challenge in the rapid response setting. Qualitative comments from our study suggest that staggered room entry by staff can exacerbate confusion, as physicians of various levels of experience and familiarity with the patient arrive at different times and may have divergent comments or instructions. A priori designation of team member roles has been recommended for emergency teams21 and is often practiced in simulated settings when groups enter a scenario concurrently. Our study suggests that to more accurately simulate true rapid response events, fragmented entry of providers should occur so that teams can practice communication in this setting, including periodic reassessments of leadership (ie, questioning of “who is in charge?”).
At our institution, observations from the present study and others have helped us iteratively redefine roles for staff members who take part in RRT calls. One example evolved from the observation that in the case of the most critical events, the bedside nurse consistently spent the first few minutes of an emergency preparing the room for the RRT (eg, drawing up medications) rather than assessing the airway and pulse and starting chest compressions for the pulseless patient, which is considered a “loss of first responder instincts.”4 We subsequently have specified that the bedside nurse calls for help and starts chest compressions, while the second floor nurse to arrive gathers the emergency equipment and assists in performing basic life support. With arrival of RRTm, the pharmacist prepares epinephrine and the shift coordinator nurse aids with communication by functioning as a “translator” between the bedside and PICU nurse and the pharmacist (eg, takes the medications from the pharmacist and hands them to the PICU nurse to administer at the code leader’s request). The bedside nurse and PICU nurse work together to rapidly prepare the child for transport to the PICU if needed.
The perceived dismissiveness of RRTm toward FP was noted in our study, as well as in multiple other studies, and has been shown to be a barrier to calling the RRT as well as to optimal teamwork upon RRT arrival.8–11,12 The disconnect in perceived quality of communication between RRTm and FP demonstrated in our study could lead to RRTm appearing dismissive toward FP upon arrival to the event. As the FP gives information in a format that they perceive to be helpful, the RRTm may move to the bedside to directly observe the information (ie, airway, breathing, circulation) of most immediate concern. In addition, the lack of familiarity between RRTm and FP likely contributes to such conflict.8 A consensus conference on rapid response systems cited issues of resistance related to perceived hierarchies and territorialism from working in clinical “silos” as a key barrier to implementing rapid response systems.23 Positive responses or behaviors by RRTm toward FP (ie, displaying appreciation for the call) have been described as a strong motivation for future RRT calls, as well as a source of increased confidence of FP in their own actions and the RRT.9,10 Team training using simulation to promote familiarity, as well as optimized integration of RRTm and FP expertise, may assist in promoting a culture of mutual respect and improved collaboration.24
Strengths of our study include a high response rate (69% overall) and a high completion rate among those who started the survey. Qualitative comments were provided by the majority of responders (71% of FP and 89% of RRTm completing the survey), allowing several themes to be identified from all groups of participants.
A limitation of our study is its single-center nature; however, the similarity of themes identified here compared with those described elsewhere7,9–11 support its generalizability. RRTm in particular comprised a smaller group (18 responses), which may limit the ability to generalize these responses. Pediatric residents were classified as FP for this study; however, some may function similarly to RRTm at more advanced stages of training, and our study did not differentiate between communication at the time of resident versus PICU team arrival. The survey was distributed over a relatively small time period, with the electronic survey being sent several months after the paper survey, and perceptions and processes may vary at different times in a pediatric hospital due to variation in patient problems, trainee experience, and other factors. Recall bias is a concern due to the administration of the surveys at 1 point in time, requesting perceptions of all rapid responses attended by the participant at the institution. The participants may have been more likely to recall particularly negative or more recent experiences. Surveys distributed closer in time to the events and direct observations could further evaluate these findings in future studies. The term “hand-off” was used in several questions in the survey instrument, which may have been confusing in this setting, as providers are arriving at different times and the communication may not constitute a standard hand-off with transfer of primary responsibility. Our survey requested ratings of communication of certain elements at rapid response events; however, when poor ratings were given, we were not always able to determine whether providers perceived these items as omitted, inaccurate, or otherwise unclear. In addition, the different questions used regarding standardized communication between the physician and nurse surveys due to differing training at the time of sampling limits interpretability regarding this issue. The low report of familiarity by resident physicians (29%) regarding the ABC-SBAR mnemonic (despite its inclusion in educational sessions) limits our ability to evaluate use of this particular tool and reinforces the need for regular practice to maintain usefulness of any communication tool.
Numerous barriers to communication were observed at pediatric rapid response events, and RRTm and FP had differing perceptions of communication quality in this study. Methods to improve communication may include a structured goal-directed tool, improved role delineation (particularly of the team leader and bedside nurse), and building a culture of collaboration between RRTm and FP. Future studies may evaluate whether improved communication at pediatric rapid response events results in improved patient care.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- floor provider
- rapid response team
- rapid response team member
- Hunt EA,
- Walker AR,
- Shaffner DH,
- Miller MR,
- Pronovost PJ
- Leach LS,
- Mayo AM
- Benin AL,
- Borgstrom CP,
- Jenq GY,
- Roumanis SA,
- Horwitz LI
- Chang VY,
- Arora VM,
- Lev-Ari S,
- D’Arcy M,
- Keysar B
- Burke CS,
- Salas E,
- Wilson-Donnelly K,
- Priest H
- Brady PW,
- Muething S,
- Kotagal U,
- et al
- Patterson ES,
- Roth EM,
- Woods DD,
- Chow R,
- Gomes JO
- Shearer B,
- Marshall S,
- Buist MD,
- et al
- Copyright © 2015 by the American Academy of Pediatrics