Académique Documents
Professionnel Documents
Culture Documents
isits to emergency departments continue to climb Recognizing that the ED is the front door to the
FIGURE 1
The Kotter Change Model. Adapted with permission from Dr. John P. Kotter.
(Acknowledge, Introduce, Duration, Explanation, Thank wins. In the final phase (steps 7 and 8), the team consolidates
you) interaction components associated with best prac- gains to produce more wins and anchor the change.
tices. 15,16 Based on literature and understanding that Based on phase 1 of the Kotter model, a coalition of
enhancing the patient experience is a priority, 8 our team staff nurses were identified by the nurse manager, assistant
believed that implementing the service nursing bundle nurse managers, and the director of nursing, who were
would have a positive impact on patients’ perceptions of asked to be the service champions leading this
care. Successful implementation of this QI initiative was patient-experience initiative. This coalition was responsible
defined as adapting the service bundle so it became a for weekly observational audits of staff, using the service
consistent part of the nursing staff’s daily activities. Monthly nursing bundle.
PRC metrics were monitored to assess patient experience Phase 2, the vision for change, need for action, and the
outcomes of the service nursing bundle intervention. elements of the service nursing bundle were incorporated
into a 1-hour class designed by the ED Director and
Clinical Educator. The Hourly Rounding component
Methods focused on addressing toileting needs, positioning, pain,
and location of patients’ personal belongings. BSR elements
SETTING included outgoing staff members introducing the oncoming
nurse and inviting the patient to participate in the report.
SOMC is a rural 232-bed acute care, nonprofit, teaching The AIDET interaction strategies included acknowl-
hospital with a 43-bed emergency department (50,000 edging the patient and family; staff introducing themselves
patient visits per year), serving 8 counties in southern Ohio and their roles; discussing duration of the visit; explaining
and northern Kentucky. The ED patient population ranges the visit process; and thanking the patient. The training also
in age from newborn to geriatric, and the majority of the linked the service bundle elements to the goal of raising
local community is Appalachian in culture. Because of the overall quality rankings by 5% and included tactics nurses
Appalachian influence, the family and extended family could use to integrate these EBPs into their daily work.
often accompany patients to the emergency department, Expectations that these practices would be ingrained into
magnifying the opportunities and need for meaningful the emergency department culture were discussed during
communication. the training, and staff members were introduced to the
observational audit processes and service champion roles.
PLANNING THE INTERVENTION Attendees were also provided with an opportunity to
role-play using the bundle elements. All nursing staff
Institutional review board (IRB) approval was obtained members were invited to attend one of the 23 classes taught
from both the organization and the primary author’s by the ED director. One hundred twenty-two staff
educational institution. One hundred forty ED nursing staff members attended the classes that were offered over a
members were invited to participate in the QI project. week, at various times, to accommodate day, evening, and
Written informed consent was obtained from the ED staff night shifts
before attending the 1-hour interactive EBP service nursing Reinforcement, the final phase, involved providing staff
bundle education. PRC obtained consent from patients with feedback to demonstrate the impact of the service
before completing the phone survey. bundle. Weekly observation audit compliance and PRC
The Kotter Change Model was selected to guide metrics were shared with staff during huddles and through
this project, as ED staff work in a constant state of e-mails. Staff members were also able to observe service
urgency (Figure 1). This model has been successful in bundle implementation in action, which allowed them to
driving organizational change and is easily applied because it see how these practices were linked to patient satisfaction.
is designed to accommodate quick change and a sense of
urgency, both of which are common states in emergency METHOD OF EVALUATION
departments. 18 This 8-step model can be organized into 3
phases. In the first phase (steps 1 to 3), the leader needs to PRC metrics and audit tool compliance were analyzed to
create the climate for change, develop a coalition of explore the impact of the service nursing bundle on patient
supporters, and create the vision for the change project. In experiences. Phone survey data from patients registered for
phase 2 (steps 4 to 6), it is imperative to engage and enable treatment who were discharged from the emergency
the organization to make a change. This phase includes department during May to September 2015, were included
sharing the vision, engaging the staff, and creating quick in this QI project. For patients aged 17 and younger, a
FIGURE 2
®
Performance Criteria for Bedside Shift Report from Observational Staff Audits. Adapted from Studer Hardwiring Excellence.
parent completed the phone survey. Patients who died in of treatment and tests, and total time spent in the ED. A
the emergency department, were admitted, or transferred to 5-point Likert scale accompanied each question: 1 = Poor, 2
another facility were excluded from the phone survey. = Fair, 3 = Good, 4 = Very Good, and 5 = Excellent. Using
The pre-intervention group included 100 randomly top-box methodology, only Excellent responses were used
chosen patients who received care before implementation of to compare survey rating and percentile rankings before and
the bundle. The postintervention group contained 97 after bundle implementation. 4 The PRC database includes
different randomly chosen patients who received care after 397 hospitals. 7
the bundle was implemented. The 197 subjects represent
1.3% of our ED monthly volume.
Based upon our emergency department’s monthly Results
volume, a priori power analysis revealed that 1,074 service
nursing bundle audits were required to assess statistical One hundred twenty-two staff members (92%) completed
differences. Twenty-three separate audits of each bundle an anonymous online survey containing demographic
element (AIDET use, Hourly Rounding, and BSR) were questions regarding age and years of service. Eighteen staff
obtained on both day and night shifts over an 8-week members were unable to participate in the service
period. These audits were conducted on different patients education. All nursing staff members were involved in the
and yielded 138 audits per week (all tools used are available random audits. Staff demographics revealed the age of the
from the author on request). An example of the BSR audit majority of the participants (N = 46) ranged from 20 to
tool (Figure 2) was adapted from The Studer Group. 14 29 years and (N = 60) had 5 years or less experience as ED
RNs (Table 1).
DATA ANALYSIS Our patient population ranged from age 6 months to
91 years old (Table 2), with a mean age of 45.85 years
Descriptive statistics, logistic regression, and odds ratios (standard deviation [SD] 22.72). Other patient demo-
were used to analyze the impact of the service nursing graphic trends indicated that the majority of patients
bundle implementation. 17 These analyses explored re- participating in the PRC survey were female (N = 119).
sponses to 5 PRC questions representing patient percep- Comparable to our facility ED patient volume arrival
tions of overall quality of care, overall quality of nursing trends, day shift represented 65% of our sample. During
care, nurses’ understanding and caring, nurses’ explanation June and July, 51% of the total surveys were collected
TABLE 1 TABLE 2
ED staff demographics Descriptive information of patients and their ED
Demographics Total Staff Completing visits
ED Staff Service Education Age (years) Mean: 45.85 ± 22.72
140 122 (92%) Range: 6 months–91 years
Age Total Percent Gender
20–29 46 38.66% Male 78 (39.6%)
30–39 28 23.53% Female 119 (60.4%)
40–49 20 16.81% Group
50–59 18 15.13% Pre-intervention Bundle 100 (50.8%)
60–69 7 5.88% Postintervention Bundle 97 (49.2%)
Experience Shift
0–5 years 60 50.85% Day 129 (65.5%)
6–10 years 19 16.10% Night 68 (34.5%)
11–15 years 19 16.10%
16–20 years 9 7.63% Length of Stay (hours) Mean: 2.39 ± 1.31
21–25 years 4 3.39% Range: 0.30–7.25
Greater than 25 years 7 5.93% Less than 1 hour 24 (12.2%)
1–2 hours 60 (30.5%)
2–3 hours 59 (29.9%)
3–4 hours 33 (16.8%)
Greater than 4 hours 21 (10.7%)
before bundle implementation. After bundle implementa-
tion, 49% of the total surveys were collected during August (Mean ± standard deviation).
and September.
In regard to the patient experience variables, length of
stay (LOS) was categorized into 5 numeric groups (Table 2), overall quality of care as excellent compared with the 48%
ranging from less than 1-hour (N = 24) to more than 4-hour of the surveyed patients in the pre-bundle implementation
(N = 21) visits. The majority of patients in our sample group (Table 4). Although this 11.8% increase was not
(N = 60, 30.5%) had a LOS of 1 to 2 hours. The 2- to statistically significant (pre = 48.0%, post = 59.8%, X 2 (1) =
3-hour range (N = 59) accounted for 29.9% of our visits, and 2756, P = 0.097), remarkably, our PRC rankings increased
our mean LOS was 2.39 hours (SD 1.31). from the 36th to the 85th percentile.
Table 3 includes the results from 1,104 audits Table 4 also displays the differences in the percentage of
conducted by 12 service champions. These audits represent patients who responded “Excellent” in the pre- and
13.8% of the total ED volume during the 8-week data postintervention groups. Analyses using the average of the
collection period in 2015. The first week of auditing bundle pre-June to July groups (35%) and the postintervention
implementation indicated that use of these elements ranged August to September groups (49.5%), comparing “excellent”
from 52% (BSR days) to 87% (AIDET communication ratings for the LOS variable between the pre- and postinter-
nights). As time progressed, additional audits demonstrated vention groups yielded statistically significant differences
that bundle element use steadily increased, with an (pre = 35.0%, post = 49.5%, X 2 (1) = 4.24, P b 0.05).
occasional backslide (BSR 65%). The last week of the Although analyses comparing excellent ratings between the
audit results indicated 100% use of all 3 service nursing pre- and postintervention groups on the other 4 questions
bundle elements across both shifts. were nonsignificant, all percentile rankings increased after
Service bundle impact was demonstrated by an increase service bundle implementation.
in post-bundle patients rating their overall quality of care Thirty-six out of 97 patients (37%) responded with
as excellent. The analysis involved combining and using excellent ratings on all 5 questions after the intervention of
the average percent of excellent responses in the pre- the EBP service-nursing bundle. Further exploration of the
intervention group compared with the postintervention pre- and post-bundle group differences indicated that the
group; 59.8% of the postintervention group rated their percentage of patients in the pre-bundle group responding
TABLE 3
Service nursing bundle: staff audits*
Date Week Communication Rounding Bedside Audits
Shift (N =)
Report
Yes No Yes No Yes No
08/03/15 Week 1 – Days 15 (65%) 8 15 (65%) 8 12 (52%) 11
Week 1 – Nights 20 (87%) 3 18 (78%) 5 16 (70%) 7 138
08/10/15 Week 2 – Days 20 (87%) 3 22 (96%) 1 18 (78%) 5
Week 2 – Nights 21 (91%) 2 16 (70%) 7 16 (70%) 7 138
08/17/15 Week 3 – Days 21 (91%) 2 23 (100%) 0 20 (87%) 3
Week 3 – Nights 22 (96%) 1 23 (100%) 0 23 (100%) 0 138
08/24/15 Week 4 – Days 18 (78%) 5 21 (91%) 2 23 (100%) 0
Week 4 – Nights 23 (100%) 0 22 (96%) 1 15 (65%) 8 138
08/31/15 Week 5 – Days 23 (100%) 0 22 (96%) 1 21(91%) 2
Week 5 – Nights 23 (100%) 0 23 (100%) 0 19 (83%) 4 138
09/07/15 Week 6 – Days 23 (100%) 0 23 (100%) 0 23 (100%) 0
Week 6 – Nights 22 (96%) 1 20 (87%) 3 22 (96%) 1 138
09/14/15 Week 7 – Days 23 (100%) 0 23 (100%) 0 21 (91%) 2
Week 7 – Nights 23 (100%) 0 23 (100%) 0 23 (100%) 0 138
09/21/15 Week 8 – Days 23 (100%) 0 23 (100%) 0 23 (100%) 0
Week 8 – Nights 23 (100%) 0 23 (100%) 0 23 (100%) 0 138
Total number of audits completed 1104
**23 weekly audits per bundle element per shift for total of 138 audits per week.
“excellent” to all 5 survey questions: (1)overall quality of and tests (t [161.97] = 2.65, P b 0.01), the total time spent
care, (2) overall quality of nursing care, (3) nurses’ (t [190.94] = 2.21, P b 0.05), and the overall summary
understanding and caring, (4) nurses’ explanation of score (t [144.58] = 2.63, P b 0.05). These results indicate
treatments and tests, (5) and total time spent in the the shorter the LOS, the more likely the patient was to reply
emergency department was calculated at 0.389. The “excellent.”
percentage of patients in the post-bundle intervention
group responding “excellent” to all 5 questions was 0.590.
These calculations yielded an odds ratio of 1.518, indicating Discussion
that patients in the post-bundle implementation group
were 1.5 times more likely to respond “excellent” to all Results from this QI project indicate that patients at our
5 survey questions compared with the patients in the facility were positively influenced by the amount of
pre-intervention arm (z = 2.82, P = 0.004). information shared by our nursing team, time spent in
The relationship between LOS and patient satisfac- the emergency department, and perceptions of caring from
tion was analyzed using separate 2-sample unpooled our nurses. These findings, supported by the literature,
t-tests (Table 5). LOS was measured in hours for the patient indicate that the service nursing bundle of communication,
group that rated time as “poor” to “very good” compared with Hourly Rounding and BSR can have a positive impact on
the patient group that rated time as “excellent.” Examining various attributes associated with patient satisfaction
this relationship between LOS (total time spent) and the metrics. 13,14,19,20
survey questions: quality of care, quality of nursing care, nurses’ Our findings indicate that use of the EBP service
understanding and caring, nurses’ explanation of treatments nursing bundle, coupled with a robust audit process,
and tests, and the overall summary score, statistically significant providing feedback to staff about audit compliance, and
differences were noted among nurses’ explanations of treatment ongoing patient satisfaction scores can improve patient
TABLE 4
Monthly percent of excellent ratings and percentile rankings for 5 key PRC survey questions
PRC Question Before Before After After Before (average) After (average) P value
June July August September Intervention Intervention
June–July August–September
Overall Quality of Care 46% (28.6) 50% (42.5) 59.6% (83.7) 60.0% (85.5) 48% (35.6) 59.8% (84.6) 0.097
Overall Quality of 58% (82.4) 50% (34.4) 66.2% (95.2) 59.6% (81.2) 54% (58.4) 62.9% (88.2) 0.206
Nursing Care
Nurses’ Explanation of 50% (60.9) 56% (81.1) 64.3% (96.4) 63.5% (96) 53% (71) 63.9% (96.2) 0.120
Test/Treatment
Nurses’ Understanding 54% (61.8) 52% (43.9) 58.1% (70.9) 61.5% (87.2) 53% (52.9) 59.8% (79.1) 0.336
and Caring
Total Time spent in 36% (38) 34% (24.7) 50.9% (92.3) 48.1% (84.7) 35% (31.4) 49.5% (88.5) b 0.05 ⁎
ED (LOS)
perceptions of ED quality of care. The “excellent” ratings training. Random audits of the EBP service nursing bundle
and subsequent increase in rankings for overall quality of practices by ED staff are planned. The current overall
care did increase after the bundle implementation across all quality of care score for the emergency department is
5 patient experience factors measured during our QI 54% “excellent” with a percentile ranking of 68.3 (PRC,
project. These results demonstrate the power of Hourly May 2017).
Rounding, bundled with BSR, and creating an environment
in which patients are informed and updated can enhance
their experience. Furthermore, these findings support the Limitations
relationship between LOS and perception of excellent
ratings. Based upon these results, it seems worthwhile to There are several limitations to this QI project, including
devote time and effort to decreasing LOS. the generalizability of the findings to non-rural emergency
Ongoing staff education and hardwiring of the EBP departments outside the Midwest. Our findings may not
service bundle have been critical to our service reflect similar outcomes at other community hospitals or
bundle-implementation success. Two years later, staff academic medical centers. Our pre- and postintervention
members have been re-educated on the service nursing participants were not the same patients; therefore,
bundle or—as we say—“Service 2.0.” We also focus on perception-change measures may have been affected by
educating new hires who did not participate in the original addressing different people. The patient sample size is also a
TABLE 5
t-test for length of stay compared with survey responses
Question Responses Poor–Very Good Responses Excellent Test Statistic P Value
(Mean Time in Hours) (Mean Time in Hours)
Quality of Care 2.46 2.33 0.69 0.494
Quality of Nursing Care 2.59 2.25 1.78 0.076
Nurses’ Explanation of Treatment and Test 2.69 2.18 2.65 0.009 ⁎
Nurses’ Understanding and Caring 2.51 2.31 1.05 0.300
Total Time Spent (LOS) 2.56 2.16 2.21 0.029 ⁎
Overall Score (Summary) 2.55 2.06 2.63 0.010 ⁎
* Indicates a significant difference at the 95th confidence interval.
limitation, as it was related to the volume allotted by our nursing bundle has the potential to positively impact the
PRC contract, which specified a certain amount for patient experience and can position your emergency
monthly phone calls for the emergency department. The department to be ready for ED CAHPS.
QI project team assumed the nurses used the service nursing Since the conclusion of this QI project, SOMC
bundle when unobserved. The QI project team leader was volunteered as one of 18 client hospitals who participated
an employee, thereby potentially influencing the staff in the PRC 2017 ED CAHPS pilot study. PRC used the
participation. The Hawthorne effect may have been a “Discharged to Community” version of the CMS ED
factor in staff audits. The amount of time available for data Patient Experience of Care (EDPEC) survey currently in
collection was limited to 4 months (June to September development. 21 PRC selected 247 patients who were seen
2015), which may have yielded very different results from a in the SOMC ED during January to February 2017. PRC
project during the peak flu and pneumonia season. was able to complete 42 phone interviews, which
Discharge calls were occurring during this period but corresponds to a response rate of 18%. Doctor/nurse
not included in this original study because a process change communication composite scores in the pilot group ranged
occurred during this time frame. The impact of discharge from 66 to 88, with SOMC scoring 77. This reaffirmed to
calls on our patient satisfaction ratings is unknown. Future our team the ongoing need to use EBP service nursing
studies could evaluate all 4 bundle elements (communica- bundle strategies to improve the patient experience.
tion, rounding, BSR, and discharge phone calls) and their Note: AIDET®, Bedside Shift Report SM, and Hourly
impact on patient satisfaction. Rounding® are registered trademarks of The Studer Group,
LLC. All rights reserved.
11. Wolf JA, Niederhauser V, Marshburn D, LaVela SL. Defining patient 18. Applebaum SH, Habashy S, Malo JL, Shafiq H. Back to the future: revisiting
experience. Pat Exp J. 2014;1(1):7-19. Kotter's 1996 change model. http://www.ressearchgate.net/profile/Steven_
12. Beryl Institute. Defining patient experience. 2016. http://www.theberylinstitute. Applebaum/publications/242339672_Back_to_the_future_revisiting_Kotter's_
org/?page=DefiningPatientExperience. Accessed December 1, 2015. 1996_change_model/links/0f31752f14502e718d000000.pdf. Accessed May 15,
2015.
13. Weigand L. Customer Service: the nursing bundle. J Emerg Nurs.
2013;39(5):454-455. 19. Boudreaux ED, O'Hea EL. Patient satisfaction in the emergency
14. Studer Q. Hardwiring Excellence. Gulf Breeze, FL: Fire Starter Publishing; 2003. department: a review of the literature and implications for practice. J Emerg
Med. 2004;26(1):13-26.
15. Gregory S, Edwardson N, Tan D, Gamm L, Tilrico M. Bedside shift
reports: what does the evidence say? J Nurs Adm, 2014;44(10):541-545. 20. Handel DA, French LK, Nichol J, Momberger J, Fu R. Associations
between patient and emergency department operational characteristics
16. Mitchell MD, Trotta RL, Lavenberg JG, Umscheid CA. Hourly rounding and patient satisfaction scores in an adult population. Ann Emerg Med.
to improve nurse responsiveness. J Nurs Adm. 2014;44(9):462-472. 2014;64(6):604-608.
17. R Core Team. R: A language and environment of statistical computing. 21. Professional Research Consultants. ED CAHPS Pilot. 2017. http://
R Foundation for statistical Computing, Vienna, Austria. 2015. http:// www.prccustomresearch.com/services/patient-experience/. Accessed
www.R-project.org/. April 27, 2017.