How Do Resuscitation Teams at Top-Performing Hospitals for In-Hospital Cardiac Arrest Succeed?
A Qualitative Study
Background: In-hospital cardiac arrest (IHCA) is common, and outcomes vary substantially across US hospitals, but reasons for these differences are largely unknown. We set out to better understand how top-performing hospitals organize their resuscitation teams to achieve high survival rates for IHCA.
Methods: We calculated risk-standardized IHCA survival to discharge rates across American Heart Association Get With The Guidelines–Resuscitation registry hospitals between 2012 and 2014. We identified geographically and academically diverse hospitals in the top, middle, and bottom quartiles of survival for IHCA and performed a qualitative study that included site visits with in-depth interviews of clinical and administrative staff at 9 hospitals. With the use of thematic analysis, data were analyzed to identify salient themes of perceived performance by informants.
Results: Across 9 hospitals, we interviewed 158 individuals from multiple disciplines including physicians (17.1%), nurses (45.6%), other clinical staff (17.1%), and administration (20.3%). We identified 4 broad themes related to resuscitation teams: (1) team design, (2) team composition and roles, (3) communication and leadership during IHCA, and (4) training and education. Resuscitation teams at top-performing hospitals demonstrated the following features: dedicated or designated resuscitation teams; participation of diverse disciplines as team members during IHCA; clear roles and responsibilities of team members; better communication and leadership during IHCA; and in-depth mock codes.
Conclusions: Resuscitation teams at hospitals with high IHCA survival differ from non–top-performing hospitals. Our findings suggest core elements of successful resuscitation teams that are associated with better outcomes and form the basis for future work to improve IHCA.
- cardiac arrest, sudden
- cardiopulmonary resuscitation
- health services research
- quality improvement
- qualitative research
Editorial, see p 164
What Is New?
In-hospital cardiac arrest is common and outcomes are variable across US hospitals, but the reasons for these differences are largely unknown.
Through site visits that included in-depth interviews of 158 clinical and administrative staff at 9 hospitals, we used qualitative methods to identify 4 broad themes related to resuscitation teams at top-performing hospitals in in-hospital cardiac arrest that distinguished them from non–top-performing hospitals: (1) team design, (2) team composition and roles, (3) communication and leadership during in-hospital cardiac arrest, and (4) training and education.
What Are the Clinical Implications?
Our findings suggest core elements of successful resuscitation teams that are associated with better outcomes.
Although additional work is required to validate these findings in larger cohorts of hospitals, our results may assist individual hospitals to examine their own care processes for in-hospital cardiac arrest and provide specific, actionable recommendations to improve management of these high-risk patients.
Approximately 200 000 patients experience an in-hospital cardiac arrest (IHCA) each year in the United States.1 Early initiation of cardiopulmonary resuscitation and defibrillation are critical for improving survival, given that every minute of delayed treatment decreases survival by 10%.2 It is not surprising that hospitals have chosen to dedicate substantial resources to train healthcare providers in resuscitation and establish facility-wide emergency response systems to optimize their performance and improve outcomes of patients experiencing cardiac arrest. Despite these sizeable investments, however, overall rates of in-hospital survival after these events remain poor, with substantial variation noted across facilities.3,4
On the surface, this variation in survival following IHCA may seem surprising. Established guidelines provide logical, sequential algorithms for advanced cardiac life support (ACLS) that are widely accepted and used across much of the world; thus, providers at most hospitals attempt to deliver the same treatments for the same reasons after the same ACLS training.5,6 Yet, these algorithms largely focus on guiding individuals on technical tasks at a patient’s bedside. They have less frequently addressed complex issues surrounding the implementation of these algorithms in real-world settings and the fact that teams, and not individuals, are ultimately responsible for providing resuscitation at hospitals. This is changing because recent guidelines7 have begun to expand on these topics given that the structure of resuscitation teams is known to vary across hospitals.8 However, there remains little empirical information to guide facilities on the optimal configuration of resuscitation teams despite their universal presence in hospitals over the past 5 decades.
Accordingly, we performed a qualitative study with the explicit purpose of better understanding how top-performing hospitals organize their resuscitation teams to achieve high survival rates for IHCA. Through in-depth site visits and detailed interviews of staff at multiple hospitals, we identified common themes linked to better outcomes by using a methodological approach that has been applied in diverse areas ranging from infection prevention to heart attack care.9–11 Our findings help unpack IHCA as a complex human task involving multiple individuals and provide novel insights that complement a large and growing literature of quantitative studies detailing differences in resuscitation care across hospitals.
Study Design and Sampling
Data and study materials will not be made available to other researchers for purposes of reproducing the results given the sensitive nature of the interviews and the potential for reidentification of individuals. We conducted a qualitative study, which included in-depth interviews of clinical and administrative staff at top-, middle-, and bottom-performing hospitals in regard to risk-standardized survival for patients with IHCA. Using data from the national prospective Get With The Guidelines-Resuscitation registry, we identified hospitals with at least 20 patients with inhospital cardiac arrest between January 1, 2012, and December 31, 2014 (n=192). We then used a previously described and validated method to calculate risk-standardized rates of survival to discharge for each hospital for each calendar year.12 This method was developed to specifically facilitate hospital comparisons and allowed us to rank each hospital into quartiles of performance. We focused on risk-standardized rates of survival to discharge rather than other outcomes (eg, return of spontaneous circulation) to provide a global measure of performance for resuscitation teams.
We defined top-performing hospitals as those consistently in the highest quartile for each of 3 calendar years (2012–2014), middle-performing hospitals as those consistently in the middle 2 quartiles for each year, and bottom-performing hospitals as those consistently in the lowest quartile for each year. Next, we purposefully selected hospitals from each of these groups for site visits, taking into account several criteria determined a priori, including teaching status, number of staffed beds, and US census region. In selecting facilities, we used the concept of information power as suggested by Malterud et al.13 This approach provides a pragmatic method for sample selection based on certain key parameters (eg, study aim, sample specificity, use of theory) but with specific attention to identifying a sample that will provide robust information directly related to the research question. This led us to focus on a larger number of top-performing hospitals as the most informative cases for investigating factors that promote better success during cardiac arrest; however, we still selected a few middle- and bottom-performing hospitals to gain a comparative view.
We approached 12 hospitals for site visits and 9 hospitals agreed to participate. One top-performing hospital declined after initially agreeing because of unexpected personal issues that prevented its leader of resuscitation services from participating, whereas another top-performing hospital and 1 bottom-performing hospital were concerned about the workload required. At each participating institution, the individual linked to the hospital’s participation in the Get With The Guidelines-Resuscitation registry, typically the director of resuscitation, was asked for her (or his) assistance with identifying key staff members involved in IHCA for interviews. Key staff we suggested for interviews included individuals from the hospital’s resuscitation team, rapid response or emergency medical team, and resuscitation committee, as well as physician trainees, nurses, attending physicians and administrative leadership in emergency medicine, hospital medicine, critical care, quality improvement, and general administration, as well. Once specific names were provided, potential participants were recruited by research staff through e-mail and by phone. Participation was voluntary and interviews were confidential. We conducted site visits until additional interviews produced no new concepts (ie, thematic saturation was reached). The institutional review board at the University of Michigan Medical School approved the research procedures and all study subjects provided written informed consent.
Data Collection and Measures
Investigators with experience in qualitative research, medicine, and nursing conducted in-depth, semistructured interviews. In most instances, interviews were conducted in-person for 1 hour with individual participants with modest incentives provided (ie, $20 gift card). In rare situations, we conducted interviews in groups with 2 to 5 participants, if requested, and when the participants represented similar resuscitation roles at a given hospital (eg, house staff). Overall, we found that the types of comments and concepts discussed did not differ substantially between individual and group interviews. For each interview, we typically included 2 researchers (1 clinician scientist, 1 qualitative research expert) to balance content and methodological expertise. One researcher was primarily conducting the interview, whereas the other took notes and prompted the discussion with questions to clarify or elaborate on ideas.
We initially built the interview guide based on a clinical framework developed from expert opinion and conceptually using the Tuckman stages of group organization.14 This was subsequently enriched by using the empirical results of a national survey that we conducted within Get With The Guidelines-Resuscitation hospitals and previously published.15 We then specifically piloted this interview guide at 2 institutions (ie, the University of Michigan and Ann Arbor VA Medical Center) before applying it during the site visits. This pilot testing allowed us to revise the questions to improve clarity and understanding.
Interviews began with a question about an individual’s role at the hospital and in resuscitation. We then inquired about their perspectives on care of patients before, during, and after IHCA, often asking participants to describe their involvement in a recent or memorable IHCA as an example. We asked specifically about teamwork and leadership during resuscitation and quality efforts in IHCA and other areas. Finally, we asked about holistic processes around IHCA, including data collection, overall views about their hospital’s best practices, and their most needed areas of improvement. All interviews were conducted using a standard semistructured interview guide that included open-ended questions to elicit detailed accounts about participants’ experiences with IHCA, with subsequent probing questions based on the interviewee’s response (Table 1). All interviews were audio recorded, transcribed by independent, professional transcriptionists, and then deidentified for analysis.
We performed thematic analysis to develop and apply consistent and comprehensive coding to the open-ended, textual data from the interview transcripts.16–18 This approach to conducting qualitative analysis has been widely applied in health research19 and is well suited to answer questions around individual experiences, views, and opinions, and to evaluate practices and processes across organizations, as well. Iteratively developed codes, which reflect the emergent themes of these data, allow for verbatim quotations or observations to be catalogued into their essential concepts.
We followed a generally recommended process to construct a preliminary codebook.20 Each member of the research team read several interview transcripts to ensure content immersion. The research team then developed a codebook together using examples from the data. Next, transcripts from the initial 2 sites were coded and discussed by multiple team members to ensure intercoder agreement and further refine the codebook. Codes were defined within categories, structured initially on the timeline of IHCA (concepts related to before, during, and after the event), and reviewed after successive site visits. This allowed us to refine the properties and dimensions of existing codes, and to identify new codes to fit concepts as they developed from the data, as well. Multiple team members coded 25% of the same transcripts for the first 5 sites to enhance consistency.
All research team members reviewed the coded transcripts and collated data, as well, from code reports around single codes to identify broader patterns of meaning (ie, themes). In this phase, we evaluated the themes against the data to develop a detailed analysis of each and to decide on an informative name for each theme. The process of refining codes and describing themes continued after each site visit until we found no new concepts in remaining transcripts, thus ensuring thematic saturation. To enhance rigor, we triangulated in at least 3 ways: (1) having multiple investigators code data and participate in analysis and interpretation; (2) performing interviews with multiple informants; and (3) collecting hospital documents and protocols about resuscitation, and attending resuscitation committee meetings at sites whenever possible. The diversity of backgrounds and experiences among the research team promoted a more in-depth discussion and understanding of the conceptual content of the data. We involved all research team members in regular team meetings during the analytic phase with disagreements discussed openly and consensus reached through collaborative discussions.
We maintained documentation through minutes from team meetings of the construction of the code structures and definitions, and the principles, as well, that we used in defining and applying the codes to theme development. Using the coded data, we summarized key themes in tables that helped describe the hospitals’ experiences with IHCA and resuscitation teams. We entered all data in MAXQDA to facilitate review, analysis, and reporting. The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the article as written.
Study Hospitals and Staff
Tables 2 and 3 (including the Table in the online-only Data Supplement) display characteristics of the 9 hospitals and key staff that participated in the study. Overall, we interviewed between 12 and 29 participants per hospital for a total of 158 individuals across multiple disciplines, including physicians, nurses, other clinical staff, and administration. In total, there were 78 hours and 29 minutes of interviews, yielding 778 482 transcribed words.
Themes of IHCA Performance
Based on the perceptions of the participants, we identified 4 broad themes that distinguished resuscitation teams at top-performing hospitals in IHCA in comparison with non–top-performing hospitals (Table 4). These themes reflected (1) team design, (2) team composition and roles, (3) communication and leadership, and (4) training and education. Differences across these 4 themes are discussed in greater detail in the next sections, including representative quotations.
In general, top-performing hospitals had dedicated or designated resuscitation teams versus ad hoc resuscitation teams at non–top-performing hospitals. Dedicated teams were used by many top-performing hospitals and referred to the presence of a core group of nursing or respiratory therapy staff with no clinical responsibilities during a given shift that superseded their roles on a resuscitation team.
I think what we have is again the fact that you have the ‘team’…You’re not waiting for someone to leave something that they’re doing. You have a dedicated team that’s this is all they’re doing, waiting for us, like having a fire service…You’re not looking who will respond today.
—Attending Critical Care Physician; Hospital A; Top-Performing
Frequently, these team members had specific clinical expertise used for other emergency services such as rapid-response teams or other acute conditions.
They do rapid response, they do code response, they are part of our sepsis team and our difficult airway response team…
—Critical Care Nurse; Hospital H; Top-Performing
Some top-performing hospitals had designated, rather than dedicated, teams. What distinguished designated teams was that their nursing and respiratory therapy staff may have other clinical responsibilities that involved direct patient care, but systems were in place for shedding these quickly when an IHCA occurred.
Our [ICU Floor Nurse] can’t get off the unit…she would let the supervisor know at the beginning of her shift, and say I’m not going to be able to go to a code tonight if it gets, if one gets called. So then they come up with a plan beforehand, on who’s going to assume that role so they’re not doing it in the moment, during the crisis.
—Nurse Supervisor; Hospital G; Top-Performing
In contrast, non–top-performing hospitals created resuscitation teams ad hoc in response to an IHCA.
We don’t assign code nurses anymore. As soon as we hear it called, you will see if there’s people in the hallway, or a head sticking out doors.—Critical Care Nurse; Hospital B; Bottom-Performing
So, we’ve got a broken process of who would respond to codes…And that’s another one of my wish lists is to either consolidate that in some form or fashion or, again, to have a dedicated team…
—Respiratory Therapist; Hospital F; Bottom-Performing
I know some hospitals have code teams…No, we don’t have that…we don’t have a code team.
—Resident Physician; Hospital I; Bottom-Performing
Team Composition and Roles
Team composition did not differ across hospitals in regard to inclusion of physicians, nurses, anesthesia, and respiratory therapists. Top-performing hospitals reported more support staff (eg, pharmacy, clerical, security, and spiritual staff) for their emergency response systems, and greater acknowledgment of the diversity and experience required to fully perform resuscitation care, as well. One top-performing hospital also described greater familiarity among team members as an advantage.
The team’s been working together for so many years that they’ve built up a really great rapport, and they know each other’s skill and their deficits…team work wise…and that’s the only reason why you have the outcome.
—Clinical Nurse Specialist and Educator; Hospital C; Top-Performing
Top-performing hospitals also tended to have clearly defined roles and responsibilities for team members during an IHCA.
Fifteen years ago when I started it was a free for all…. So when (Medical Director) took over and, and kind of structured everything…You just show up and you know what you’re supposed to do, and there’s no screaming and there’s no yelling.
—Critical Care Nurse; Hospital G; Top-Performing
At non–top-performing hospitals, specific roles and responsibilities of members were less clear, even well into the response, generating variable degrees of distress.
Very honestly things are more chaotic…I just feel a different level of anxiety when you come to these codes and roles are not always as clearly defined.
—Critical Care Attending Physician; Hospital E; Middle-Performing
It’s kind of situation by situation…Respiratory always goes right to the airway to start with the airway, which is intuitive, and then in terms of the other staff, it just kind of depends. They seem to be comfortable assigning their own roles…
—Emergency Medicine Physician; Hospital B; Bottom-Performing
We also saw major differences in the presence of resident physicians and their function during an IHCA. At non–top-performing hospitals, resident physicians were, at times, described as a weakness given the frequent turnover of trainees.
I don’t mean [residents] suck, but look at what we give ‘em. They come in as first years, they don’t know anything. They come in as second years, they sort of know what’s going on. By the third year, they’re starting their stride. They start to get good at what they do, and then they graduate and leave, and then we’re back to the people that are being fed through the PEZ container…
—ACLS Instructor; Hospital I; Bottom-Performing
At top-performing hospitals, on the other hand, specific back-up plans for resident physicians involved more experienced clinicians.
What happens a lot is the MICU resident will take ownership of the code and then the fellow will be as a supportive role. And depending on their code experience, you know, they might fly pretty much independently or they might need some guidance in terms of either fellow suggestions or nursing suggestions…
—Critical Care Nurse; Hospital H; Top-Performing
Top-performing hospitals also spent time empowering bedside nurses in their roles as first responders, including allowing them to defibrillate without the presence of physicians if indicated.
When I came here, the policy was nurses couldn’t press the button to defibrillate. They could charge the monitor, they could recognize Vfib, they couldn’t press the button…it looked like the ceiling of the Sistine Chapel where one physician’s finger was reaching across and trying to make contact with the button. And (Nurse Leader) and (Physician Leader) and others realized that that was ridiculous. We couldn’t get a physician in the room in 2 minutes reliably, and they certainly wouldn’t be up to speed on what needed to be done, so nurses defibrillate now….
—Nursing Educator; Hospital A; Top-Performing
One thing that has frustrated me or did frustrate me as a bedside nurse, was that I was ACLS trained, I was ACLS trained to administer medications…But the culture was “never,” you know, you never push epi, you never do anything without a doctor even though we are being trained…We did make a decision about 2014…our shocks are now delivered within 1 minute…CPR is generally delivered or started in under 1 minute for a witnessed cardiac arrest for certain.
—Critical Care Nurse; Hospital H; Top-Performing
Hospitals perceived essential skills like chest compressions quite differently when performed by different individuals in these roles. In 1 top-performing hospital, chest compressions were highly valued and their performance was limited to specific individuals (eg, a critical care nurse):
I mean the days of, you know, letting the nursing student or the paramedic student do the CPR, they just…they don’t really happen anymore.
—Emergency Medicine Nurse; Hospital G; Top-Performing
In contrast, a bottom-performing hospital valued this skill less and suggested it was easily performed by less experienced providers.
We teach the techs that they can do compressions in code situations. If we have nursing students, we’ll let them do compressions. We have plenty of people to rotate through in general so we’ll rotate anyone through to do compressions.
—Nursing Education; Hospital F; Bottom-Performing
Last, crowd control was universally considered a problem because of the arrival of nonteam members.
Our biggest problem is too many people show up…We have codes where there’s 17 to 22 people who respond…So sometimes it’s difficult determining who’s in charge of this code…
—Nursing Supervisor; Hospital I; Bottom-Performing
Communication and Leadership
Top-performing hospitals described different patterns of communication that encouraged multidisciplinary discussion and mutual respect across team members. This did not mean that communication was universally perfect or that tension never arose, but that specific mechanisms to address breakdowns were available.
I think we have really good codes, and then we have codes that don’t run well. On really good codes, the communication is fantastic…If I had to do a percentage, probably 75 are good communication, and there’s 25 that aren’t.
—Critical Care Nurse; Hospital H; Top-Performing
We’re very, very fortunate in that my great partner is our senior medical director here at the hospital, so he and I work very closely together. So I’ll hear it in fact more from him…He’ll say, hey, did you hear about [the resuscitation team] had a call last night, there was a concern between [team nurse] and the physician…So that lets me know that I didn’t even have to be part of that process. It got escalated appropriately and [the senior medical director]…will take whatever feedback and then do whatever investigating needs to happen.
—Senior Nursing Officer; Hospital C; Top-Performing
At non–top-performing hospitals, communication during an IHCA was described more frequently as chaotic, confusing, or unsatisfactory with gaps in practices endorsed by guidelines like closed-loop communication where team members explicitly acknowledge and verbalize requests as they are done.
Communication just needs to get better. There are some residents who are really good at giving direct orders or finding roles, closing the loop, all that stuff. But, there are some who aren’t trained on that and they don’t know how to do it and so, will talk softly or they won’t give a complete order and things kind of get lost.
—Critical Care Nurse; Hospital I; Bottom-Performing
Interviewer: Would you be able to identify one or two things that you really think—as it pertains to resuscitation care specifically—where you would like to see things improved further…? Participant: Communication. Closed-loop communication.
—Nurse Coordinator; Hospital E; Middle-Performing
Another key part of communication was the presence of strong leadership during the IHCA.
Intensivists usually will take over… you role model off of him. So if he is calm and, you know, everything seems to be calm. So I have seen a lot of codes where if they get real anxious, then it makes the whole room anxious.
—Nurse Manager; Hospital A; Top-Performing
At 1 top-performing hospital, good leadership was highly valued, regardless of whether the team leader was a physician or nurse. Indeed, a coleadership model with nurses who were specialized members of the resuscitation team was highlighted.
Interviewer: So who’s running the code?
Participant: Again, if there’s a physician there, then usually…most of the time it’s going to be the [resuscitation team] nurse because she is ACLS certified, so she’s going to be calling the drugs and things like that. The physician may come in and sort of weigh in on what’s happening, but they’re looking to the [resuscitation team] nurse a lot of times.
—Respiratory Therapist; Hospital C; Top-Performing
There is just a very wonderful collaboration that is felt between the hospitalists and the [resuscitation team]. When they’re…a hospitalist responds to a code, they respond, but they defer to the [resuscitation team]. Meaning, not that the hospitalist is being lazy but that they know that…their expertise, and they defer to them because of that expertise.
—Clinical Nurse and Educator; Hospital C; Top-Performing
Training and Educational Efforts
All hospitals discussed training and educational efforts around resuscitation, but varied in their implementation of these efforts. A notable example was mock codes, which were described as more in-depth at top-performing hospitals. Specifically, mock codes were: (1) unplanned and held on a regular basis, (2) conducted in actual patient rooms rather than simulated environments, (3) multidisciplinary, and (4) included structured postdebriefing.
One top-performing hospital described mock codes in high-risk areas targeted for improvement because of low traffic or a higher incidence of codes. They also focused on teamwork and communication during the mock codes.
So the mock codes will typically actually take place in areas that are maybe a low traffic area or someplace not everybody knows where it’s located so that it helps facilitate people finding their way through the building.
—Pharmacist; Hospital D; Top-Performing
If there’s codes happening outside of the ICU or CCU, then we target those areas for mock codes.
—Critical Care Nurse Educator; Hospital D; Top-Performing
“And it’s all about team training, less clinical, more team…
—Cardiology Physician; Hospital D; Top-Performing
Non–top-performing hospitals did not describe the same type of processes for mock codes. When present at these hospitals, mock codes were generally perfunctory, not multidisciplinary and conducted in simulated environments.
There’s too many people who…well, I’m busy with other things. You can’t be pulling me away for this stuff [mock codes], you know, that kind of an attitude. And to me, that’s really the only way you get a true mock code, you know.
—Clinical Nurse and ACLS Instructor; Hospital I; Bottom-Performing
On the mock code side, generally physicians don’t participate in the mock code…
—Nurse Educator; Hospital F; Bottom-Performing
Beyond mock codes, top-performing hospitals often enhanced their educational opportunities by introducing newer teaching approaches that often emphasized multidisciplinary care.
The ACLS is more about teamwork in allowing everyone to experience each of the roles so that they have a deeper understanding of how the team works. We call it a sport, you know, it’s because…it requires communication, it requires performance. And you don’t learn to play football by yourself.
—ACLS Instructor; Hospital A; Top-Performing
We describe the findings from a national qualitative study to understand how top-performing hospitals organize their resuscitation teams to achieve high survival rates in patients with IHCA. Based on the perceptions of the participants we interviewed, we discovered 4 broad themes that distinguish top-performing hospitals: team design, team composition and roles, communication and leadership, and training and educational efforts. In general, we noted that top-performing hospitals had formally organized teams composed of members from diverse disciplines with delineated roles and responsibilities. We also found top-performing hospitals more commonly encouraged strong communication and leadership during an IHCA while supporting training and educational efforts like comprehensive mock codes. To our knowledge, this is the first multicenter evaluation of the performance of resuscitation teams during IHCA using qualitative methods, and we believe our results will be valuable to hospitals interested in improving outcomes for these events.
Providing high-quality resuscitation can be difficult for hospitals. IHCAs are sudden and unexpected events that can stress the resources of even the most resilient organizations. After the initial publication of seminal studies of the potential life-saving benefit of chest compressions, noninvasive ventilation, and defibrillation in the early 1960s,21–23 reports began to surface about different hospitals’ approaches to providing resuscitation.24–26 Given widespread concerns about the best processes for applying these treatments, numerous professional organizations then developed protocols to standardize cardiopulmonary resuscitation, basic life support, and ACLS. In the 1970s, these protocols were standardized by groups like the American Heart Association into ACLS, a focus of resuscitation care that has continued largely uninterrupted over almost 5 decades.27 Regulatory organizations have also been interested in ensuring adequate provision of resuscitation. For example, The Joint Commission mandates standardized policies and procedures for providing resuscitation services within healthcare organizations, including the availability of emergency equipment and adequate training of individual providers.28
Despite this considerable attention to IHCA, surprisingly few recommendations have been made about how resuscitation teams should be ideally constructed. National data suggest up to a quarter of hospitals may not even have resuscitation teams and the variability around practices in the design, composition, and roles and responsibilities of team members is believed to be substantial.8 Furthermore, it is uncertain whether differences in many of these features lead to differences in IHCA performance. It is in this context that our findings are particularly striking and useful because we noted substantial differences in the ways that top-performing hospitals distinguished themselves. We summarize these findings in Table 4 as a set of potential recommendations for clinicians and hospital administrators to consider in designing their systems for resuscitation.
Our findings highlight the importance of dedicated or designated resuscitation teams at hospitals. Top-performing hospitals described dedicated teams as a powerful way of achieving the consistent clinical expertise and teamwork required for effectively managing these high-stakes events. However, dedicated teams are not likely to be feasible at many hospitals, as they can require resource investments that may be barriers for some facilities. This constraint led other top-performing hospitals to use designated teams where resuscitation team members are identified a priori as predetermined responders to an IHCA. We also found that some top-performing hospitals sought ways to leverage the clinical expertise of resuscitation teams in different ways across their hospitals to offset resource investments. An analogy raised by 1 hospital was the current trend to still have dedicated firefighters available for fires while expanding their involvement with other critical emergencies.29
Our findings also indicate that hospitals need to carefully craft the composition of their teams and the roles and responsibilities of the providers on these teams. Ideally, inclusion of members of the team should consider the cognitive and technical skills that are required to perform a successful resuscitation, not simply titles. One top-performing hospital seemed to find great success by being very specific about assigning roles and responsibilities so that individual responders knew what to do immediately on arrival at an IHCA. At other top-performing hospitals, there was a focus on delineating team member roles and responsibilities as early as possible during an IHCA.
Improving communication and leadership during an IHCA should also be a top priority for facilities, although our findings also suggest that these elements are influenced by factors related to other themes. For example, hospitals with a dedicated team were likely to face fewer challenges with communication and leadership by virtue of a smaller cohort of providers being responsible for all acute resuscitation care. Finally, our data revealed the benefits of incorporating mock codes into training and educational efforts. Key features of a robust mock code program to more accurately simulate real-world IHCAs included conducting events that were nonscheduled, held in actual patient rooms and at all times of day, and multidisciplinary. Mock codes should also be efficient to achieve staff buy-in and include multidisciplinary post–mock code debriefing.
Although we used established techniques to improve the rigor of our findings, our study has limitations. First, we visited hospitals at a single point in time. The non–top-performing hospitals could have been on a trajectory toward improvement that was not captured in our data. To minimize this possibility, we calculated risk-standardized survival rates over a 3-year period from the cohort of hospitals we used for selection and focused on those with consistent performance in each of the 3 years for IHCA survival. Second, our interviews were based on the perceptions of the informants, and there is always the possibility of social desirability response bias in this type of work, which would occur if participants misrepresented their experiences to provide desirable answers. In a similar vein, we also could not account for whether providers at hospitals were aware of their organization’s Get With The Guidelines-Resuscitation scores that are provided to participants in the registry. To address these concerns, we interviewed several key staff in each hospital to obtain a comprehensive picture of care and used scripted probes during our semistructured interviews to elicit specific details that would be difficult to misrepresent (eg, how are resuscitation teams designed). We also encouraged respondents to share both positive and negative experiences. Third, we focused only on hospitals enrolled in a large national registry and were limited in our ability to evaluate very small hospitals (<200 beds) with few cases of IHCA. Although our findings are specific to these sites, the detailed nature of the data allows an assessment of the degree to which similar issues might occur at other institutions supporting transferability of key processes. Additional work will need to confirm the generalizability of our findings to other organizations. Finally, our study identified key themes that we hypothesize influence performance in IHCA. Additional work is needed to develop specific tools for measuring underlying features related to these themes and to quantitatively test the impact of these constructs among a larger sample of hospitals. Thus, this work requires confirmation.
For nearly 5 decades, resuscitation teams have been deployed in hospitals using variable approaches largely based on anecdote and convenience, rather than evidence. Results of this study move us forward from simply describing types of hospitals with consistently higher survival rates after IHCA to understanding potential keys to their success. Although additional work is required to validate these findings in larger cohorts of hospitals, our results will assist individual hospitals in examining their own care processes for IHCA and provide specific, actionable recommendations to improve management of these high-risk patients.
This article does not necessarily represent the position of the US government or the Department of Veterans Affairs.
Sources of Funding
This study was supported by the National Institutes of Health (R01HL123980). Dr Nallamothu also received funding from Veterans Affairs Health Services Research and Development (IIR 13–079) during this period and receives honoraria from the American Heart Association for editorial work. Dr Krein is supported by a Veterans Affairs Health Services Research and Development Service research career scientist award (RCS 11–222). Dr Chan receives funding from the National Institutes of Health and has received consultant funding from the American Heart Association and Optum Rx.
The online-only Data Supplement, podcast, and transcript are available as an online-only Data Supplement at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.118.033674/-/DC1.
- Received January 11, 2018.
- Accepted April 12, 2018.
- © 2018 American Heart Association, Inc.
- Merchant RM,
- Yang L,
- Becker LB,
- Berg RA,
- Nadkarni V,
- Nichol G,
- Carr BG,
- Mitra N,
- Bradley SM,
- Abella BS,
- Groeneveld PW
- Neumar RW,
- Shuster M,
- Callaway CW,
- Gent LM,
- Atkins DL,
- Bhanji F,
- Brooks SC,
- de Caen AR,
- Donnino MW,
- Ferrer JM,
- Kleinman ME,
- Kronick SL,
- Lavonas EJ,
- Link MS,
- Mancini ME,
- Morrison LJ,
- O’Connor RE,
- Samson RA,
- Schexnayder SM,
- Singletary EM,
- Sinz EH,
- Travers AH,
- Wyckoff MH,
- Hazinski MF
- Link MS,
- Berkow LC,
- Kudenchuk PJ,
- Halperin HR,
- Hess EP,
- Moitra VK,
- Neumar RW,
- O’Neil BJ,
- Paxton JH,
- Silvers SM,
- White RD,
- Yannopoulos D,
- Donnino MW
- Bhanji F,
- Finn JC,
- Lockey A,
- Monsieurs K,
- Frengley R,
- Iwami T,
- Lang E,
- Ma MH,
- Mancini ME,
- McNeil MA,
- Greif R,
- Billi JE,
- Nadkarni VM,
- Bigham B
- Bradley EH,
- Curry LA,
- Webster TR,
- Mattera JA,
- Roumanis SA,
- Radford MJ,
- McNamara RL,
- Barton BA,
- Berg DN,
- Krumholz HM
- Curry LA,
- Spatz E,
- Cherlin E,
- Thompson JW,
- Berg D,
- Ting HH,
- Decker C,
- Krumholz HM,
- Bradley EH
- Krumholz HM,
- Bradley EH,
- Curry LA
- Chan PS,
- Berg RA,
- Spertus JA,
- Schwamm LH,
- Bhatt DL,
- Fonarow GC,
- Heidenreich PA,
- Nallamothu BK,
- Tang F,
- Merchant RM
- Malterud K,
- Siersma VD,
- Guassora AD
- Chan PS,
- Krein SL,
- Tang F,
- Iwashyna TJ,
- Harrod M,
- Kennedy M,
- Lehrich J,
- Kronick S,
- Nallamothu BK
- Lyons E,
- Coyle A
- Terry G
- Creswell JW
- Kuckartz U
- Braun V,
- Clarke V
- Curry LA,
- Nembhard IM,
- Bradley EH
- 28.↵Joint Commission and Resuscitation–Resuscitation Central. http://www.resuscitationcentral.com/documentation/joint-commision-and-resuscitation. Accessed October 18, 2017.
- 29.↵Firefighters Train to Fight Terrorists, Too. NPR.org. https://www.npr.org/templates/story/story.php?storyId=127303858. Accessed December 14, 2017.