The Journal of Emergency Medicine, Vol. -, No. -, pp.

1–13, 2019
Ó 2019 Elsevier Inc. All rights reserved.
0736-4679/$ - see front matter

https://doi.org/10.1016/j.jemermed.2019.10.035

Original
Contributions

TEN BEST PRACTICES FOR IMPROVING EMERGENCY MEDICINE

PROVIDER–NURSE COMMUNICATION

A. Zachary Hettinger, MD, MS,*† Natalie Benda, PHD,* Emilie Roth, PHD,‡ Daniel Hoffman, BS,* Akhila Iyer, MS,*
Ella Franklin, MSN, RN,* Shawna Perry, MD,§ R. J. Fairbanks, MD, MS,*†k and Ann M. Bisantz, PHDk
*National Center for Human Factors in Healthcare, MedStar Health, Washington, DC, †Department of Emergency Medicine, Georgetown
University School of Medicine, Washington, DC, ‡Roth Cognitive Engineering, Stanford, California, §Department of Emergency Medicine,
University of Florida, Jacksonville Medical Center, Jacksonville, Florida, and kDepartment of Industrial and Systems Engineering, University at
Buffalo, The State University of New York, Buffalo, New York
Reprint Address: Aaron Zachary Hettinger, MD, MS, National Center for Human Factors in Healthcare, MedStar Health, 3007 Tilden St NW,
Suite 6N, Washington, DC 20008

, Abstract—Background: The current state of scientific barriers; and factors affecting successful communication,
knowledge regarding communication between emergency and has implications for both system and training design.
medicine (EM) providers indicates that communication is Key implications for emergency nursing practice from this
critical to safe and effective patient care. Objectives: In research are distilled in 10 ‘best practice’ strategies for
this study, we identified communication needs of EM nurses improving EM nurse–physician communication. Ó 2019
and physicians; in particular, what information should be Elsevier Inc. All rights reserved.
conveyed, when, how, and to whom. Methods: Five semi-
structured focus groups and one interview were conducted , Keywords—information needs; communication
with nine nurses, eight attending physicians, and four resi- methods; communication strategies; shared awareness;
dents. Questions addressed how EM personnel use and share concept maps; nurse–physician communication
information about patients and clinical work, what informa-
tion tends to be exchanged, and what additional information INTRODUCTION
would be helpful to share. Sessions were audio recorded.
Transcripts were generated and analyzed using a concept Communication and coordination between medical pro-
mapping approach (a visual qualitative analysis technique viders play a critical role in delivering safe and effective
to represent and convey synthesized knowledge). Results: care to patients (1–5). Communication is particularly
Eleven concept maps were produced summarizing: informa- critical in the complex, dynamic environment of the
tion physicians needed from nurses and vice versa; methods emergency department (ED) (1–3,5). Studies have found
of communication that could be utilized; barriers or obsta- that emergency medicine (EM) clinicians spend between
cles to effective communication; strategies to enhance or
55% and 89% of their time engaging in communication
ensure effective communication; and environmental or situ-
activities, with multiple studies describing challenges
ational factors that impact communication. Conclusions:
Our main finding of this research is that communication en- related to consistent and effective communication among
sures shared awareness of patient health status, the care EM staff (6–11).
plan, status of plan steps and orders, and, especially, any From a patient safety perspective, the most problem-
critical changes or ‘‘surprises’’ regarding the health of a atic communication failures have been found to occur
patient. Additionally, the research identified shared among health care professionals (10,12). ED staff mem-
information needs; communication methods, strategies and bers report that the ‘tumultuous’ nature of the

RECEIVED: 18 June 2019; FINAL SUBMISSION RECEIVED: 21 October 2019;

ACCEPTED: 27 October 2019

1
2 A. Z. Hettinger et al.

environment is the largest barrier to interprofessional of a commercially available EHR, with minimal differ-
communication (11). Difficulties with communication ences between the two sites.
and shared awareness among health care professionals Clinicians were recruited via ED e-mail listservs via
have also been further exacerbated by the implementation hospital leadership staff on a voluntary basis and direct
of electronic health records (EHRs) (13). Specific in-person solicitation at departmental meetings or team
training for EM providers has been proposed for commu- huddles. Nine EM nurses, eight EM attending physicians,
nication and related nontechnical skills to help deal with and four EM resident physicians participated and were
information-sharing issues in EM (14,15). Often, educa- compensated $150 for their time. Procedures were
tion related to communication focuses on provider– approved by the health care system’s institutional review
patient communication, or competencies have been board; all participants provided verbal consent.
developed for EM provider training related to interper-
sonal communication skills (15–17). Interpersonal Study Protocol
communication competencies, however, do not inform
providers regarding what information should be The study utilized semi-structured focus groups and inter-
conveyed, when, how, and to whom. Successful patient views to examine information needs of EM clinicians.
care also requires a high degree of coordination and Each group contained unique roles: for example, nurses,
communication among health care professionals, attending physicians, or resident physicians. The semi-
especially within the unique environment of the ED, structured format includes a series of questions that are
which requires input from a wide variety of sources asked of the group that allow for free discussion of a topic
both inside and outside of the facility (4). but ensure that key topics of interest to the research team
This study utilized focus groups and interviews to are approached during the session. The questions were
concretely identify information-sharing needs of emer- created by a team of EM and human factors experts and
gency physicians and nurses. We focused on information were designed to understand how different personnel
that is helpful but may not typically be shared or is diffi- use and share information about patients, what informa-
cult to obtain. Participants identified types of information tion tends to be exchanged, and what information not
needed by emergency physicians and nurses, strategies routinely communicated between nurses and physicians
for communicating this information, and circumstances would be helpful to share. Table 1 summarizes the ques-
under which various types of information are needed. tions and follow-up probes.
Thus, this research also fills a research gap in identifying
the commonalities and differences in information needs Data Collection
by different types of clinicians. Findings were translated
into 10 best practices for improving emergency provider– Data were collected through semi-structured focus
nurse communication. The objective of this study was to groups or interviews in hospital conference rooms, led
identify communication needs of EM nurses and physi- by the same facilitator (NCB), with other researchers pre-
cians, in particular, what data should be conveyed, by sent (RMR and DJH). Additional researchers participated
whom, and the most appropriate time to convey the infor- in the focus groups via conference call (AMB, RLB,
mation based on the clinical scenario. NMM, and EMR). Sessions lasted approximately
90 min. Researchers took notes and audio was recorded
to supplement analysis.
METHODS
Study Design Data Analysis

This research used focus groups and interviews with Notes were collated to provide a chronological record of
emergency physicians and nurses to identify strategies the focus group conversations. Audio recordings were
and barriers associated with information sharing in EM. then reviewed in conjunction with the notes to validate
the notes and provide any missing detail. The resulting
Study Setting and Population data were segmented into discrete thoughts/categories
and used for further analysis using a ‘‘concept mapping’’
Participants were recruited from two EDs within a large approach. Concept maps are figures designed to represent
not-for-profit academic distributed health care delivery or convey knowledge: they have been used widely in
network, encompassing 10 hospitals and 280 ambulatory complex domains, including health care, weather fore-
sites. Both institutions are urban, tertiary care facilities casting, rocket propulsion, and space exploration (18).
that train resident physicians, although only one site Data were analyzed by author AB to create concept
trains EM residents. Both sites use the same installation maps describing nurse–physician communication
Ten Best Practices for Improving Emergency Medicine Provider–Nurse Communication 3

Table 1. Semi-Structured Questions and Follow-Up Probes

Question 1: What information do you typically need to get from the [physicians/nurses] treating the patient?
Follow-up probes:  What information do the [physicians/nurses] often know that you don’t easily know that would be helpful to you?
 How do you get or try to get this information?
 Do you always get this information? What can make this difficult?
Question 2: What information do you typically have that you think might be important for [physicians/nurses] to know?
Follow-up probes:  How do you share that information?
 Does this always get shared? What are the obstacles to sharing this information?

processes, with review and input from author ER. Maps segmented into sub-maps centered on the higher-level
were developed following a standard process outlined nodes, which included various communication methods,
in Figure 1 (19). After identifying the domain (ED com- communication-enhancing strategies; obstacles and bar-
munications between physicians and nurses) and focus riers to communication; and situational factors that
(communication methods) of the concept map (Step 1), impacted communication.
the segmented data transcripts were reviewed to identify The sub-maps were further refined (Step 4) for consis-
effective communication strategies, ineffective strategies, tency, to combine nodes, rephrase links, or rearrange con-
and obstacles and barriers to effective communication. cepts so that nodes did not appear more than once, and to
Context or situational contingencies related to the strate- identify additional relationships among nodes where
gies or barriers were also identified. appropriate (Step 5). Links between the sub-maps were
Concept node identification (Step 1), Concept also made explicit (e.g., interruptions are an example of
Arrangement (Step 2), and Linking (Step 3) were then an obstacle/barrier to communication that occurs when
done iteratively, working progressively through the phones are used—and therefore are shown in both the
interview segments, which were reviewed to identify phone sub-map, and the obstacles-barriers sub-map).
content for the concept map nodes and links. For This set of cross-linked sub-maps represents the knowl-
instance, the interview segment ‘‘verbal communication edge model (Step 6).
especially for time critical information–patient is septic;
patient will go to OR in 30 min’’ resulted in nodes ‘‘ver-
bal communication’’ linked via a ‘‘useful/better for’’ RESULTS
link to ‘‘time critical information (e.g., patient is sep-
tic).’’ Nodes were added or augmented (with similar or Data were collected across six sessions averaging 73 min
complementary information) as the concept map was in duration (range of 62–90 min). Five sessions included
populated. After all data segments were represented, four participants (two with nurses, two with attending
and to enhance comprehensibility, the larger map was physicians, and one with resident physicians) and one

1. Concept Node Iden fica on: 6. Build the Knowledge Model:

Select focus ques on(s) to serve as Combine cross-linked maps to create a
the guiding topic (ini al note) of knowledge model
the concept map

2. Concept Arrangement: 5. Iden fying Addi onal Rela onships:

Iden fy the concepts that are the Iden fying linkages in refined concept
most important, related to the map and between sub-maps
topic and inclusive of key themes

3. Linking: 4. Refinement:
Linking related concepts in a top- Add, subtract, edit, or subsume nodes
down manner, using short words and linkages to make the concept map
to describe linkage as necessary representa ve yet concise

Figure 1. Standard process concept map.

4 A. Z. Hettinger et al.

Table 2. Participant Details They also wanted to be alerted, ideally verbally, to

changes or additions to the patient workflow (e.g., new or-
Characteristic
ders or a need to recheck vital signs) and any holdups
N = 23 Nurse Resident Attending from the physician’s end (e.g., the physician is waiting
on results of consult to decide on disposition). Likewise,
Gender, count (%)
Male 0 (0) 3 (75) 2 (25) physicians expressed a desire for updates on patient prog-
Female 9 (100) 1 (25) 6 (75) ress, particularly the status of items on the critical path to
Experience, years* disposition or the next stage of care (e.g., whether the pa-
<1 1 (11.1) 0 (0) 0 (0)
1–5 2 (22.2) 4 (100) 2 (25) tient has eaten, has started to walk), including psychoso-
6–10 3 (33.3) 0 (0) 3 (37.5) cial factors that might impact disposition or timing of
11–15 2 (22.2) 0 (0) 2 (25) disposition (e.g., the availability of a ride home). They
>15 1 (11.1) 0 (0) 1 (12.5)
Age, years also wanted to be alerted to any holdups to progress and
18–29 1 (11.1) 4 (100) 0 (0) associated reasons (e.g., delays in an ordered intravenous
30–39 6 (66.7) 0 (0) 4 (50) line due to difficulty in putting in the line), and the status
40–49 0 (0) 0 (0) 4 (50)
50–59 2 (22.2) 0 (0) 0 (0) of nursing actions, particularly ones that might impact
>59 0 (0) 0 (0) 0 (0) what orders the physicians needed to enter (e.g., alerted
to medication administration to avoid unnecessary medi-
* Experience refers to the total years in which a provider has prac-
ticed under their reported role (nurse, attending, or resident). cation order duplication). Finally, physicians indicated a
desire for a ‘virtual poke’—a reminder when all pending
steps have been completed for patient disposition.
session had one participant (a nurse). Participant details Both groups wanted notification when a patient was
are summarized in Table 2. particularly sick or critical, or when there was other sur-
prising information requiring attention or specific precau-
Knowledge Model tions. For example, nurses wanted to be alerted if a patient
was contagious, requiring staff to don protective gear.
Eleven maps were developed. One map describes cate- They also wanted to be informed when a physician was
gories of information that physicians felt they needed concerned about test results or possible test results. In
from nurses, and vice versa (Figure 2). Five maps turn, physicians wanted to be informed of information
describe methods of communication: (Verbal or Face to from triage and initial nursing assessment that might
Face, Phones/Walkie-talkies/Text, Electronic Ordering impact treatment priority (e.g., an overall assessment of
[General], Electronic Free Text [Electronic Status Board how sick or stable the patients were).
Comments Field], and Nursing Notes. One map describes Physicians also wanted to be alerted of any changes
barriers or obstacles to effective communication, and one observed during nurses’ ongoing patient assessments or
describes strategies to enhance or ensure effective re-assessments after interventions, such as changes in pa-
communication. Three maps describe environmental or tient vital signs or symptoms (including level of pain) that
situational factors that could affect communication (Staff might signal a change in patient status and shift diagnosis
Workload; Level of Nursing Experience; Physical Layout or disposition (especially for the worse—a patient
of the ED). Figures 3–5 provide examples of the methods, becoming a ‘surprise sick’). They also wanted to be
strategies/barriers, and situational maps, respectively. immediately informed—ideally, verbally—of any critical
laboratory values and any critical information about the
Types of information to be shared. The first concept map patient’s status or changes in status relative to being
shows information that nurses indicated they needed from particularly ‘sick’ that the nurse might be first to recog-
physicians and vice versa (Figure 2). Examination of the nize due to their more regular contact with the patient.
map reveals symmetry between the information that Both groups indicated a need to know critical informa-
nurses needed from physicians and the information that tion that may have been communicated by the patient or
physicians needed from nurses. For example, they both the patient’s family to only one of them. Nurses indicated
indicated a desire to understand the status of the patient that physicians sometimes obtained information directly
workflow, holdups to progress (with reasons), and an indi- from the patient, knowledge that might be different or
cation of when all steps for disposition were complete. broader than what the nurse might have been told (e.g.,
Nurses wanted to be informed of special circum- information the patient deemed to be sensitive). Simi-
stances that would indicate a change in urgency, type of larly, physicians wanted information that might have
care, or deviations from standard protocol, and factors been told to nurses by the patient or family prior to the pa-
that might indicate a shift in priorities for orders or a tient being seen by the physician, or during patient care,
need for a particular sequencing of orders or activities. that provided nurses with a more holistic view of the
Ten Best Practices for Improving Emergency Medicine Provider–Nurse Communication 5

A
Figure 2. (A) Physician-to-nurse communication needs. (B) Nurse-to-physician communication needs. CT = computed tomogra-
phy; ECG = electrocardiogram; ED = emergency department; EHR = electronic health record; ICU = intensive care unit; OR = oper-
ating room.

patient, and thus, is valuable to communicate. This in- know about when consults have taken place and the re-
cludes information that can impact initial assessment or sults of these consults, even when the information is
the priority of seeing the patients, or is important for pa- transmitted to only one of them. Similarly, they both
tient treatment or disposition/timing of disposition (e.g., need information about test results that may have been
social history, problems patients are experiencing with communicated via phone to only one of them. Physicians
orders or meds, questions from the patient’s family, and indicated an interest in critical patient information that
the background urgency of the request). might be entered in nursing notes that they don’t currently
Other information that needs to be shared included the have easy access to. Nurses also asked for information
substance and outcomes of conversations with other care that physicians may obtain from external information
teams (e.g., consultants), especially if they impact the sources that nurses do not have access to. For instance,
plan of care or disposition, for example, the reason for discharge summaries from other hospitals (available
an admitting order. Nurses and physicians both need to only to physicians) may contain information about safety
6 A. Z. Hettinger et al.

B
Figure 2. (continued).

issues (e.g., past aggressive behavior) or medical history sition (e.g., planned trajectory in terms of admission vs.
that could impact care. Overall, both groups expressed a discharge home), and importantly, any changes to the
need for the nurse and physician to share awareness of the plan or disposition.
diagnosis, plan of care, and (especially) disposition. This This shared awareness is necessary to allow nurses to
requires that the physician communicate to the nurse their better prioritize tasks (understanding better what is on the
working diagnosis that is motivating orders, their plan for critical path, and to be more proactive in beginning tasks
patient care, and their expectations with respect to dispo- on the critical path to disposition). Shared awareness of

Figure 3. Verbal communication map. ED = emergency department; OR = operating room.

Ten Best Practices for Improving Emergency Medicine Provider–Nurse Communication 7

Figure 4. Communication-enhancing strategies map.

the care plan also supports nurses in providing the physi- awareness of the care plan enables nurses to answer pa-
cian with relevant information and reminders—some- tient questions and manage patient expectations. In
thing physicians welcomed. Experienced nurses, in contrast, without information sharing, nurses have to
particular, serve as a double check, posing diplomatic interpret the orders and figure out the plan, making it
questions that enable the physician to catch and correct difficult to answer patient and family member questions.
potential errors or miscommunication. For example, Finally, both nurses and physicians indicated that phy-
such questions could direct attention to information in sicians often possessed information that would be useful
the EHR that the physician may otherwise have missed. to communicate to nurses to better support overall ED
As one physician noted, physicians are interested in resource management. For example, physicians should
knowing more generally ‘what are you seeing that I am communicate which patients can be put in hallway beds
not seeing’ that cause the nurse and physician to be on to allow rooms to be freed up, and also let nursing team
a different page and may help redirect the physician’s leads know when tasks aren’t getting done (e.g., because
thinking or plan. Questioning orders if unsure of their nurses have become overloaded), especially in the case of
appropriateness, and asking for clarification on the plan critical patients.
of care or disposition could encourage the physician to
more fully think through their line of reasoning and Communication methods. The next set of maps focus on
potentially catch and correct errors. Finally, shared communication methods. The verbal communication

Figure 5. Staff workload communication map.

8 A. Z. Hettinger et al.

Table 3. Design Recommendations for Communication-Enhancing EHR Features

Goals Requirements

Shared, patient-centered view that ensures that the 1) Methods for physician to communicate the assessment plan, the working
physician and nurse are on the same page with diagnosis, the plan of care and the disposition plan
respect to working diagnosis, assessment plan, 2) A method to provide physicians with nurses’ ongoing assessments, and
tasks and priorities, status of workflow items and especially ’after action’ assessment (impact of drugs, impact of pain
holdups if any, and patient disposition medication) that may be currently documented in the nursing notes
3) A representation of the workflow—the items required to be completed for
patient disposition, their status, and whether there are any holdups
4) A representation of information communicated by the patient to any provider
5) A method for tracking status of consults (that they have been requested, that
they have come, what they have concluded and ordered)
Accessible, transparent documentation that allows 1) Methods to more effectively structure and highlight information that is currently
notes and charts to be more useful as shared in nursing charts (observations and nursing assessments)
information sources 2) Methods to more effectively structure and highlight physician’s charts
(physician’s diagnosis)
Multi-way, real-time communication that can 1) Method such as an instantaneous message board where questions can be
overcome barriers for verbal communication posed and answered and key information that impacts disposition can be
(e.g., interruptions, potential for shared
misunderstanding) while still supporting rapid 2) Method to post reminders to self and others
information exchange 3) Method to generate and share to-do lists

map (Figure 3) describes situations in which verbal and updates that keep the team on track. The utility of the
communication is useful or preferred—such as time- comments field is hindered by a lack of consistency and
critical information, information related to critically ill conventions in use, lack of use by consultants, being
patients, changes in patient status or orders, or when bar- ‘‘cryptic’’ or hard to decipher, and the fact that comments
riers to the care plan arose. Verbal communication is also do not transfer into the patient’s chart. Use of comments
preferred when sharing the physician’s care plan for the is dependent on physician preferences to look for infor-
patient, learning about information obtained outside of mation there, and also whether the ED layout and work-
the EHR (e.g., over the phone or from other systems), flow supported frequent checking of the EHR.
or providing information about less critical patients, Participants also made more general comments
perhaps because it was more efficient than using the regarding the utility of the EHR. In general, the utility
EHR. One specific form of verbal communication is the of these systems as a communication method was either
formal verbal order. Participants associated it with the enhanced, or hampered by, usability-related system
challenge of needing to remember and enter the verbal or- design features, including features that made important
der into the computer. It also requires additional checking information (e.g., significant changes, vital signs, order
or can result in errors if the verbal orders are unusual, priorities) salient; supported documenting interventions
outside of a nurse’s scope of knowledge, or delivered in and outcomes; and allowed treatment or process out-
high-tempo stressful situations. Institutional policies, cul- comes to be evaluated. As with the comments field,
ture, and preferences can affect whether verbal communi- communication through the EHR requires individuals to
cation occurs between providers. For instance, some explicitly access the computer system (and therefore
physicians prefer to use the EHR comments field, and was most useful for less critical, more straightforward pa-
nurses may or may not choose to interrupt a physician tients). Finally, participants expressed concerns about the
with a low-priority question (e.g., if a patient can eat or degree to which information was up to date, noting that
drink). Finally, face-to-face verbal communication is relevant information may not have been entered in a
facilitated by ED layouts that allow nurses and physicians timely fashion, or that physicians may be documenting
to cross paths. on a paper chart.
The concept map for the EHR comments field is orga- Interesting patterns emerged regarding the use of mo-
nized similarly to the verbal communication map. Partic- bile technologies that were available for communicating,
ipants provided a number of examples of information specifically phones, walkie-talkies, and texting. All three
types for which the comments field was most useful; in forms allow person-to-person communication without
some cases, information that was NOT seen as appro- requiring crossing paths or co-location; however, the
priate for verbal communication (e.g., orders regarding three modes differ in critical ways. Most importantly,
patient diet or restrictions; information about straightfor- both phones and walkie-talkies have the potential to inter-
ward or routine patients). Comments were not seen as rupt and thus, negatively impact caregiving (an obstacle
appropriate for atypical information. Participants did to communication). Texting does not have that disadvan-
note that the comments field was useful for lists of tasks tage, but is not a ‘‘verbal’’ conversation. Phones and
Ten Best Practices for Improving Emergency Medicine Provider–Nurse Communication 9

Table 4. Strategies to Support Effective Nurse-Physician Communication

1. Recommendation: Communicate diagnostic assessment, plan of care and, especially, disposition plan to other team members as early
as possible. Update the team of any changes to the plan.
 Example: After initial assessment of a 55-year-old man with chest pain, the physician leaves the room and discusses the diagnostic
plan with the nurse and notes that even if the testing is negative, the patient is high risk enough to warrant an overnight observation.
 Why: Early intention for admission or discharge can allow the nurse to start planning for disposition, including completing paperwork
or transportation planning, ultimately reducing length of stay.
2. Recommendation: Communicate pending tasks/steps in the patient’s care as well as information regarding changes or holdups to
tasks or orders.
 Example: The physician documents in the EHR patient comment field that they are waiting on an x-ray study and urine pregnancy
test. After 10 min the nurse notices that the urine test was never ordered and requests the physician to place the order to facilitate
obtaining the x-ray study.
 Why: Having a shared understanding of what the next steps for the patient are can aid in detecting errors of omission or proactive
monitoring of the patient’s condition.
3. Recommendation: Communicate details regarding proactive diagnostic testing and therapeutic interventions.
 Example: The nurse initiates a standing abdominal pain protocol that includes obtaining a urine and urine pregnancy test on women
of childbearing age and confirms with the physician that they would like blood drawn and a line placed.
 Why: Differences in clinical practice exist between providers and must account for the unique condition of each patient. However,
available treatment protocols should be initiated when possible to prevent delays in care and should be coordinated with the
provider that will likely be caring for the patient when door-to-provider times will lead to significant delays.
4. Recommendation: Don’t assume everyone has a shared understanding: recognize that you might have unique access to information
and make sure that it is shared in a timely manner.
 Example: An orthopedic physician consults on a patient in the ED and informs the resident physician that the patient will be booked
for surgery within the hour. One hour later there is a nursing and clerical scramble to get the patient ready for the operating room
because they were unaware of the new plan.
 Why: Having the most up-to-date information will allow all of the members of the team to prioritize their own workload. Assuming that
the other team members will eventually get the plan, see the electronic order, or find out from the patient will not only cause delays in
care, but will lead to significant frustration by other team members and potential patient safety concerns.
5. Recommendation: Notify providers of any critical or unexpected changes in vital signs or patient status
 Example: After obtaining discharge vital signs the nurse informs the physician, who expresses concern regarding the patient’s
renewed tachycardia that will now require further diagnostics and possible inpatient evaluation.
 Why: Even if vital signs and nursing notes are entered without delay in the EHR, it is likely that the provider will not see the results
immediately or at all, depending on the specific configuration of the EHR and where the information is placed.
6. Recommendation: Do not assume electronic orders substitute for verbal communication
 Example: While placing an order in the EHR for broad-spectrum antibiotics to cover sepsis, the physician contacts the nurse to
inform them of the order and the need to meet the 3-h sepsis treatment window.
 Why: There can be significant delay for a nurse to log into the EHR and view outstanding orders if they are busy with bedside care for
other patients. Unless the nurses are using mobile workstations or handheld devices providing task monitoring/notifications, there is
likely to be delays during busy times when they are not at an EHR workstation.
7. Recommendation: Use asynchronous communication for lower priority items to aid in prioritization.
 Example: The nurse leaves a ‘‘paper towel’’ note on the physician’s keyboard requesting electrolyte orders for a patient with mild
abnormalities on their laboratory values.
 Why: The judicious use of asynchronous communication methods like notes, orders, and approved electronic communication can
reduce interruptions while providing better situation awareness. Text messages have facilitated this process in the inpatient setting,
but more compliant technology with greater integration in the EHR could facilitate future communication.
8. Recommendation: Adapt communication strategies based on team members’ experience level and existing relationships.
 Example: Performing a huddle at the beginning of the shift allows all team members to know each other’s names, roles, and
experience level (e.g., experienced nurse, but new to facility), especially in the presence of medical students and residents that may
be new members of well-established teams.
 Why: Mis-estimating experience level and comfort level of novice nurses and providers can lead to misunderstandings and delays,
including delays in obtaining intravenous access from a novice nurse or waiting on orders from a medical student.
9. Recommendation: Adapt communication strategies to the physical layout of the ED, especially in those facilities where nurses and
physicians may have workstations out of sight from one another or where it is not obvious which staff members are on different care
teams.
 Example: Periodic rounding on shared patients between a nurse and physician can proactively address issues including delays in
diagnostic testing and requests for additional therapeutic measures (e.g., repeat doses of pain medications).
 Why: Lack of physical proximity reduces the situation awareness of team members regarding each other’s workflow and potential
issues. The use of large private rooms for patients may mean the staff do not hear changes in condition or requests for help from both
the patient and medical staff. Furthermore, less experienced staff may be reluctant to interrupt others; rounding can provide a natural
break for concerns to be discussed.
10. Recommendation: Use strategies that exploit provider experience level regardless of role hierarchy.
 Example: An experienced charge nurse recommends that a recently graduated resident upgrades their patient to the intensive care
unit based on the patient’s acuity level and anticipated nursing requirements.
 Why: Institutional knowledge that often lies with the most experienced staff members can be critical to caring for patients and should
be considered whenever offered regardless of role, including clerical staff and environmental services.

EHR = electronic health record; ED = emergency department.

10 A. Z. Hettinger et al.

walkie-talkies were both seen as useful for types of verbal shared knowledge across the team. Lack of knowledge,
communication; however, walkie-talkies may be less or shared knowledge, is also a barrier. For instance, if
acceptable to nurses due to interruptions; whereas some physicians and nurses see patients separately, they may
phone systems require a method to update and dissemi- not understand what (sometimes subtle) information to
nate phone numbers each shift. share with one another. Lack of knowledge can also
Participants expressed interesting perspectives include knowledge on the part of a patient (e.g., not
regarding the role of the nursing note as a communication knowing their home medications); knowledge about a pa-
method. In particular, although the nursing note serves to tient (e.g., not having formulated the plan); or knowledge
document nursing actions with respect to patient care, about processes (e.g., the optimal sequencing of diag-
and (from the nursing perspective) could serve as a source nostic testing and therapeutic interventions). Finally,
of information for physicians, the overall belief was that there are team-dynamic-related barriers, based on
the utility of the nursing note for communication is perceived status differences between physicians and
hampered by the fact that physicians do not view this as nurses. Nurses might not feel empowered to question or
a method of communication because it is not two-way interrupt physicians.
(i.e., it only facilitates communication from nurses to The last three concept maps organized situational or
physicians and not vice versa), and also that the structure environmental factors that affected communication. The
(or in some cases, perceived lack of structure) of the note physical layout of the ED can either facilitate or impede
makes it difficult for physicians to find specific informa- communication (Figure 5). Communication is facilitated
tion of interest or be alerted to new information or critical when individuals work in close physical proximity (e.g.,
changes. Thus, the note cannot be used to communicate co-locating nurse and physician computer stations, seeing
aspects of the care plan or changes in patient status patients together): individuals can find one another, might
from the physician’s perspective. Despite the fact that shadow or overhear relevant in-person or phone conversa-
nurses might enter key information regarding changes tions, and are more likely to cross paths and exchange
in patient status or plans, or information related to patient quick updates. In contrast, layouts that separate nurses
family or other social concerns, these are unlikely to be and physicians make these serendipitous encounters
seen by physicians. less likely and require effortful action to locate each
other. More broadly, work processes combined with
Communication strategies and barriers. Participants workstation locations can lead to a situation where physi-
described a number of communication-enhancing strate- cians are at the physician workstations entering informa-
gies (Figure 4): physicians communicating the care plan tion while nurses are ‘‘acting out the care plan’’ with
directly to nurses and, in some cases, patients (in the pres- patients and not using the computer system. Thus, nurses
ence of nurses); using formal methods such as a ‘‘team are not immediately aware of plan or care changes or new
huddle’’; using a point person (e.g., charge nurse or orders that were entered. Workload also can affect
team leader) to facilitate communication between the communication. Higher workload makes communication
physician and nursing team; and having nurses explicitly of important changes in patient status less likely and
remind physicians about alerts generated within the EHR. serves as a barrier to some communication strategies,
ED culture is also important: communication is enhanced such as use of a ‘‘team huddle’’ and serendipitous
when physicians explicitly encourage encounters, are communication arising from physical proximity/crossing
approachable, and seem open to interruptions (even paths. In addition to the number or complexity of patients,
when information was less critical). In contrast, the obsta- staff workload is also tied to the number of learners (e.g.,
cles and barriers concept map represents factors that interns, new residents) and when nursing team leaders are
impede communication. Broadly, these factors include also assigned their own patients. Finally, the relative
system design factors (e.g., poor usability or design of experience level of staff, particularly nurses, is an impor-
the EHR and related systems); ED layout or logistics- tant factor in successful communication. Less-
related factors (e.g., noise levels, temporary nurses experienced nurses may be less likely to proactively
without ED-issued phones, layout factors—described in communicate about important changes to patient health
detail below); and workflow and staffing issues (e.g., rela- or delays in carrying out the care plan (perhaps because
tive experience of nurses and physicians, high workload, they were less confident), are less able to differentiate
tasks and conversations being interrupted). The number critical from noncritical information, have less experi-
of learners (e.g., residents, interns) is also a staffing- ence predicting physician information needs or desired
related communication barrier, resulting in more compli- communication methods, are less able to anticipate next
cated communication patterns (e.g., resident to nurse; steps or orders (and thus, do not look out for them),
resident to physicians), which can lead to fragmented and, in general, are less aware of the importance of their
communication between pairs of individuals rather than role in providing critical information.
Ten Best Practices for Improving Emergency Medicine Provider–Nurse Communication 11

DISCUSSION ever, that the research team members with EM experience

all had experience in multiple hospitals and health care sys-
The results of this study, which identified issues, barriers, tems, from different U.S. regions, helping to mitigate this
and potential strategies surrounding nurse–physician limitation and supporting generalization of the findings. Re-
communication in hospital EDs, provide important in- sults were based on discussions with 23 individuals: more
sights in how to improve communication and patient care. participants may have led to additional examples and
On the whole, the information needed by physicians more experienced physicians (only one physician had
and nurses was similar. Discrepancies in knowledge arose more than 15 years’ experience). Note that the sample
from timing (who spoke with patients or family when); size is consistent with recommendations for focus group
differential access times to the EHR (e.g., for order status); research and that the analysis indicated thematic saturation
complexities in information sharing among trainees, (19). Additionally, participants only represented nurses,
nurses, and physicians; or because each role may have attending physicians, and resident physicians. Future
had unique opportunities to access information. Resolving research could expand the research questions to include
these discrepancies and maintaining shared knowledge is other types of providers, such as mid-level providers, tech-
recognized by both groups as critical to patient care. nicians, or consultants. Additional types of providers could
Shared awareness supports care, allowing physicians and have led to an expansion of findings but does not invalidate
nurses to prioritize patients and tasks; allowing nurses to those described here. Finally, this study was conducted
take proactive steps to alleviate holdups and allow comple- before the two EDs converted to full electronic physician
tion of the care plan; supporting communication with pa- documentation. However, it is unlikely that this change
tients; and providing nurses with a basis for alerting or would have significantly affected the results, as most physi-
reminding physicians about relevant changes or next steps. cians do not complete their documentation until after the pa-
It is necessary, therefore, for nurses and physicians to tient has been disposed from the ED, and the note is
realize that they may NOT, in fact, have a complete typically only available once it has been completed and
shared understanding with one another, and, therefore, signed—and, therefore, is not available as a means of
take active and continued steps to update one another. communication during the patient stay. Future areas of
In an attempt to do just that, nurses and physicians research should focus on how best to implement these rec-
make use of a variety of communication methods and ommendations, including training and workflow changes to
strategies. Key findings are that both nurses and physi- improve communication in the ED.
cians choose methods for communication differently
based on (often subtle) differences in information content CONCLUSIONS
and degree of urgency. Local work culture and beliefs
regarding preference and hierarchy also play a role. Addi- Successful EM nurse–physician communication is required
tionally, situational and ED design factors facilitate or to ensure that all care team members share awareness of the
lead to obstacles in communication. Awareness of the patient health status, care plan, status of plan steps and or-
impact of these factors is also key in communicating ders, and especially, any critical changes or ‘‘surprises’’
and successfully developing a shared understanding. regarding the health of the patient. This research used focus
These findings have implications for both system design groups and concept-mapping analysis to identify shared in-
and work practice. Table 3 outlines design recommenda- formation needs, communication methods, communication
tions for communication-enhancing EHR features. strategies and barriers, and factors affecting successful
communication. Ten take-away tips for enhancing EM
Recommendations for success: 10 strategies to support nurse-physician communication are provided.
effecting EM communication
Acknowledgments—The authors acknowledge the contributions
Additionally, we have distilled a set of recommendations of Robert Wears, MD, PhD, to this work. In addition, the authors
in support of effective nurse–physician EM communica- would like to thank Nicolette McGeorge, PhD for her assistance
tion (Table 4). with data collection.
This work was supported by the Agency for Healthcare
Limitations Research and Quality, United States (R01HS022542).

There are several limitations with this research. First, partic-

ipants were recruited from two hospitals within the same REFERENCES
health care system. Recruiting from other systems or
1. Leonard M, Graham S, Bonacum D. The human factor: the critical
geographic regions may have revealed additional communi- importance of effective teamwork and communication in providing
cation needs, strategies, barriers, and methods. Note, how- safe care. Qual Saf Health Care 2004;13(suppl 1):i85–90.
12 A. Z. Hettinger et al.

2. Risser DT, Rice MM, Salisbury ML, Simon R, Jay GD, Berns SD. working in the Emergency Department. Int Emerg Nurs
The potential for improved teamwork to reduce medical errors in 2013;21:168–72.
the emergency department. Ann Emerg Med 1999;34:373–83. 11. Jafari Varjoshani N, Hosseini MA, Khankeh HR, Ahmadi F. Tumul-
3. Slade D, Manidis M, McGregor J, et al. Communicating in hospital tuous atmosphere (physical, mental), the main barrier to emergency
emergency departments. Heidelberg, Germany: Springer; 2015. department inter-professional communication. Glob J Health Sci
4. Eisenberg EM. The social construction of health care teams. In: 2015;7:144–53.
Nemeth C, ed. Improving healthcare team communication: building 12. Apker J, Mallak LA, Gibson SC. Communicating in the ‘‘gray
on lessons from aviation and aerospace. Farnham, UK: Ashgate zone’’: perceptions about emergency physician-hospitalist handoffs
Publishing; 2008. and patient safety. Acad Emerg Med 2007;14:884–94.
5. Guarrera T, McGeorge N, Clark LN, et al. Cognitive engineering 13. Taylor SP, Ledford R, Palmer V, Abel E. We need to talk: an obser-
design of an emergency department information system. In: vational study of the impact of electronic medical record implemen-
Bisantz AM, Burns C, Fairbanks RJ, eds. Cognitive systems engi- tation on hospital communication. BMJ Qual Saf 2014;23:584–8.
neering in health care. 1st edn. Boca Raton, FL: CRC/Taylor & 14. Reisdorff EJ, Hughes MJ, Castaneda C, et al. Developing a valid
Francis; 2015. evaluation for interpersonal and communication skills. Acad Emerg
6. Shang N, Maddow C, Kannampallil TG, King B, Franklin A. Impor- Med 2006;13:1056–61.
tance of verbal communication in the electronic age [Abstract]. Ann 15. Hobgood CD, Riviello RJ, Jouriles N, Hamilton G. Assessment of
Emerg Med 2012;60:S90–1. communication and interpersonal skills competencies. Acad Emerg
7. Spencer R, Coiera E, Logan P. Variation in communication loads on Med 2002;9:1257–69.
clinical staff in the emergency department. Ann Emerg Med 2004; 16. Knopp R, Rosenzweig S, Bernstein E, Totten V. Physician-patient
44:268–73. communication in the emergency department, Part 1. Acad Emerg
8. Morgan SR, Kawar M, Rahman S, Gatewood JS, Fairbanks RJ. Med 1996;3:1065–9.
Physician and nurse perceptions of non-urgent communication in 17. Totten VL, Knopp R, Helpcm K, et al. Physician-patient communi-
the emergency department. Ann Emerg Med 2011;58:S189. cation in the emergency department, Part 2: communication strate-
9. Cameron KA, Engel KG, McCarthy DM, et al. Examining emer- gies for specific situations. Acad Emerg Med 1996;3:1146–53.
gency department communication through a staff-based participa- 18. Crandall B, Klein GA, Hoffman RR. Working minds: a practi-
tory research method: identifying barriers and solutions to tioner’s guide to cognitive task analysis. Cambridge, MA: The
meaningful change. Ann Emerg Med 2010;56:614–22. MIT Press; 2006.
10. Bagnasco A, Tubino B, Piccotti E, et al. Identifying and cor- 19. Glaser B, Strauss A. The discovery of grounded theory: strategies
recting communication failures among health professionals for qualitative research. New York, NY: Routledge; 1967.
Ten Best Practices for Improving Emergency Medicine Provider–Nurse Communication 13

ARTICLE SUMMARY
1. Why is this topic important?
Communication between nurses and emergency medi-
cine (EM) providers is critical to the safe and effective
care of patients in the emergency department. Under-
standing interactions and information needs among clin-
ical team members can not only aid in communication,
but can also provide a framework for training and the
design of workflow and health information technology
systems.
2. What does this study attempt to show?
The study used semi-structured focus groups with
emergency physicians and nurses to discover different
communication needs and strategies. The study attempts
to show the diversity in information needs and communi-
cation styles that differ by role and experience level.
3. What are the key findings?
The 10 ‘‘best practices’’ involving information needs
and communication strategies of emergency physicians
and nurses when collaborating in the care of patients,
including:
Communicate diagnostic assessment, plan of care, and
especially, disposition plan to other team members as
early as possible and do not assume there is a shared
team understanding of the patient’s care.
Electronic orders are not a substitute for verbal commu-
nication. However, clinicians should leverage asynchro-
nous communication for lower priority items.
Adapt communication strategies based on team mem-
bers’ experience level and existing relationships, while
avoiding hierarchy as a barrier to communication.
4. How is patient care impacted?
Adapting communication to the needs of other team
members can impact the safe and efficient care of patients.
Differences in communication needs and styles by team
member experience level should be taken into consider-
ation and should be part of any training of team members.
Team communication strategies should be integrated
into health information technology systems to reduce bar-
riers between nurses and physician and avoid potentially
negative impact on patient care.