This study evaluated the association between leadership walkrounds (WRs) and caregiver assessments of patient safety climate and risk reduction across 49 hospitals. WRs involve hospital leaders visiting clinical units to openly discuss safety issues with staff. The study found that units where ≥60% of caregivers reported exposure to at least one WR had significantly higher safety climate scores, greater reported risk reduction, and more feedback on actions taken compared to units with <60% exposure. Higher rates of WR participation at the unit level were positively associated with more favorable caregiver assessments of patient safety culture and outcomes.
2. leadership and frontline clinicians about patient safety
concepts and lead to cultural changes such as increased
transparency in discussions of adverse events and an
improved rate of safety-based changes.5
One year following WR implementation in a 1000-bed
tertiary care facility in Canada, there was a 64% to 80%
increase in the number of identified patient safety issues
that had been resolved or had active improvement work
in progress.6 Nine out of 10 clinicians reported that they
felt comfortable when openly and honestly discussing
patient safety issues, and they had enhanced awareness of
patient safety. In another study at a 711-bed tertiary care
473635AJMXXX10.1177/1062860612473635Ameri
can Journal of Medical QualitySchwendimann et al
1University of Basel, Institute of Nursing Science, Basel,
Switzerland
2Patient Safety Training and Research Center, Duke University
Health
System, Durham, NC
3Duke University School of Nursing, Durham, NC
4Duke University Hospital, Durham, NC
5Duke University School of Medicine, Durham, NC
6Pascal Metrics Inc, Washington, DC
7Duke University Department of Psychiatry, Durham, NC
The authors declared no potential conflicts of interest with
respect to
the research, authorship, and/or publication of this article. The
authors
received no financial support for the research, authorship,
and/or
publication of this article.
3. Corresponding Author:
René Schwendimann, PhD, RN, University of Basel, Institute of
Nursing Science, Bernoullistrasse 28, 4056 Basel, Switzerland.
Email: [email protected]
A Closer Look at Associations Between
Hospital Leadership Walkrounds
and Patient Safety Climate and Risk
Reduction: A Cross-Sectional Study
René Schwendimann, PhD, RN,1,2,3 Judy Milne, MSN, RN,2,4
Karen Frush, MD, BSN,2,3,5 Dietmar Ausserhofer, PhD, RN,1
Allan Frankel, MD,6 and J. Bryan Sexton, PhD2,7
Abstract
Leadership walkrounds (WRs) are widely used in health care
organizations to improve patient safety. This retrospective,
cross-sectional study evaluated the association between WRs
and caregiver assessments of patient safety climate and
patient safety risk reduction across 49 hospitals in a nonprofit
health care system. Linear regression analyses using
units’ participation in WRs were conducted. Survey results from
706 hospital units revealed that units with ≥60% of
caregivers reporting exposure to at least 1 WR had a
significantly higher safety climate, greater patient safety risk
reduction, and a higher proportion of feedback on actions taken
as a result of WRs compared with those units with
<60% of caregivers reporting exposure to WRs. WR
participation at the unit level reflects a frequency effect as a
function of units with none/low, medium, and high leadership
WR exposure.
Keywords
leadership walkrounds, safety climate, leadership, patient
safety, hospital
4. [email protected] - February 18, 2020 - Read articles at
www.DeepDyve.com
Schwendimann et al 415
hospital, nurses participating in WRs scored significantly
higher in safety climate attitudes (73% positive) com-
pared with controls (53% positive).7 The authors con-
cluded that safety climate scores likely will increase in
clinical areas in which caregivers are engaged in rigor-
ously implemented WRs. However, such implementation
requires significant organizational will in a given hospi-
tal.8 In 2002, the University of Michigan Medical Center
implemented hospital-wide Patient Safety Rounds (PSRs)
as a modified approach of WRs. In a follow-up study,
researchers conducted a retrospective study of more than
900 staff members who had participated in a total of 70
PSRs over the last 4 years. Overall, patient safety was
enhanced as demonstrated by the resolution of a series
of local safety problems (eg, repair of equipment), the
development of safety policies to address safety concerns
at the regional level (eg, intravenous medication policy),
and the initiation of systemwide changes (eg, simulation-
based training on central line catheter insertion). In addi-
tion, clinical staff who participated in PSRs were more
likely to report errors and near misses and were more
likely to perceive that their managers promote patient
safety as compared with those who did not participate in
PSRs. However, it remains unclear to what extent these
differences are attributable to the PSRs.9
Although the concept of WRs is popular and widely
used in North American hospitals,9-12 the broad impact of
5. WRs on specific patient safety attitudes of clinical staff as
well as clinical patient outcomes is less well known. One
of the barriers to rigorous scientific scrutiny of WRs is
that their format varies widely between the institutions
and the leaders facilitating the discussions. For example,
the authors’ experiences with conducting and investigat-
ing WRs have included formats as varied as hallway
conversations, breakroom discussions over snacks, audi-
torium presentations, and “Safe-Tea Time,” during which
general questions were asked over tea. Other variations
range from collecting rigorous minutes, to refusing to let
names or details be written down at all.13,14 To date, little
is known about the frequency of WRs and their associa-
tion with safety culture, although reported WR frequen-
cies range from weekly WRs across the hospital8 to
monthly visits in the same hospital units or clinical areas,
to random visits to random units once a month.6,7
The aim of this study was to evaluate the association
between leadership WRs and caregiver assessments of
patient safety climate, patient safety risk reduction, and
feedback on leadership WR-related actions. The authors
hypothesized that frequency of caregivers’ participation
in WRs (ie, exposure) at the unit level was positively
associated with favorable assessments of patient safety
climate, patient risk reduction, and actions taken as a
result of WRs. More specifically, it was hypothesized
that higher rates of participation in WRs would be associ-
ated with better safety climate, more feedback received
on actions taken as a result of WRs, and patient safety
risk reduction at the unit level.
Methods
Survey data from 49 hospitals of a nonprofit US health
system were used for this cross-sectional study. The
6. study setting consisted of a mix of clinical areas (eg,
medical-surgical units, pharmacy, intensive care units,
radiology) and included caregivers such as nurses, phy-
sicians, technicians, support staff, and others directly or
indirectly involved in patient care. All participating
hospitals had implemented WRs between 2002 and
2005 as part of their quality improvement strategies
related to enhancing patient safety culture, which was
initiated in 2001 and 2003 with leadership WRs train-
ing workshops from the Institute for Healthcare
Improvement15 and annual patient safety, quality, and
risk management conferences, respectively. Thus, all
units within hospitals were included in the overall
health system’s quality improvement strategy that
focused on assessing the individual hospital’s safety
culture, including features such as leadership WR expo-
sure, teamwork climate, and safety climate, in its vari-
ous clinical areas.
Leadership WRs
The WR approach used in this study is based on the
model developed by Allan Frankel3 and consisted of
regular monthly visits by hospital executives (eg, depart-
ment chiefs, chairs, risk managers) to clinical areas to
discuss patient safety issues with caregivers. The execu-
tives were instructed on how to conduct and schedule
their rounds and use rounds to address concerns related
to patient safety. The leaders were given scripted, open-
ended questions that should be asked during each round.
These might include questions about staff members’
specific concerns, such as “What keeps you up at night?”
or “What aspects of the environment are likely to lead to
the next harmful event?” The questions raised in the WR
sessions are intended to initiate group discussion and
stimulate safety problem resolution. It was important for
leaders to explain that their desire was to talk openly and
7. honestly with staff about “what is the next thing that
could harm a patient,” and to reassure staff that any infor-
mation discussed during the WR would be kept confiden-
tial. Safety issues raised during the sessions were
documented and led to actions, which were followed up
by leaders providing feedback to participating staff (eg,
at team meetings).
For study purposes, 2 questions from a version of the
Safety Attitudes Questionnaire (SAQ) were used; these
questions were modified to include specific WR-related
items: (a) “Does your clinical area use Patient Safety
[email protected] - February 18, 2020 - Read articles at
www.DeepDyve.com
416 American Journal of Medical Quality 28(5)
Leadership WalkRounds to discuss with senior leaders
any issues that could harm patients or undermine the safe
delivery of care? (Yes; No; Not Sure)” and (b) “How
often did you participate? (0, 1, 2, 3-4, 5-7, 8 or more, Not
Sure).” The other WR items on the survey were the fol-
lowing: “Did you receive feedback about actions taken as
a result of the WalkRounds? (Yes, No, Not Sure)” and
“Do you think that 1 or more patient safety risks were
reduced as a result of the WalkRounds? (Yes, No,
Not Sure).”
Patient Safety Climate
Patient safety climate, defined as “the perceived level of
commitment to and focus on patient safety within a given
unit,” is a consensus assessment about patient safety-
related norms and behaviors. Thus, patient safety atti-
8. tudes of caregivers were measured with the 7-item
patient safety climate subscale from the SAQ. Each item
is rated on a 5-point Likert-type scale ranging from dis-
agree strongly to agree strongly. Examples of questions
included statements such as “I would feel safe being
treated here as a patient” or “In this clinical area, it is
difficult to discuss mistakes.” There is growing evidence
that the SAQ elicits attitudes that are responsive to
interventions associated with caregivers’ behaviors, such
as communication and teamwork to improve patient
safety.14,16
Patient Safety Risk Reduction
The WR item “Do you think that 1 or more patient safety
risks were reduced as a result of the WalkRounds? (Yes,
No, Not Sure)” was the self-reported caregiver assess-
ment of the extent to which patient safety issues were
elicited and acted on to reduce risk. Anecdotal examples
include double-checking medication before administer-
ing, supported ambulation from bed to wheelchairs, and
unit staff briefings before morning shifts, among others.
Feedback on WR-Related Actions
The nature of leadership WRs is to discuss patient safety
issues and concerns at the unit level and include appro-
priate follow-up actions using the leader’s position to
make things happen at hierarchical levels above indi-
vidual units. Accordingly, providing feedback to the
involved unit caregivers about actions taken as a result of
these WRs is essential to keep momentum and build trust
in management’s ability to solve patient safety problems.
For instance, unit managers and safety officers track
planned measures at the unit or departmental level fol-
lowing WRs for updates and communicate this follow-up
information to caregivers and senior leadership with the
9. aim of supplying accurate feedback to ensure completion
of improvement tasks.3
Therefore, the modified SAQ included the item “Did
you receive feedback about actions taken as a result of the
WalkRounds? (Yes/No/Not Sure)” to capture the care-
giver’s experience.
Data Collection
After the patient safety survey was initiated by the health
system, in each hospital, the questionnaire was adminis-
tered (paper and pencil) in all clinical patient care areas
to caregivers who had a working commitment of at least
50% in a specific care area and who had been working
full-time or part-time for at least 4 weeks. Participation
was voluntary. Completing the questionnaire implied
consent and took approximately 10 minutes. Data collec-
tion took place between May and September 2006 as a
partial evaluation of the hospital’s strategies to enhance
patient safety. Data from returned questionnaires were
scanned, exported to statistical software, and checked for
consistency.
Statistical Analysis
Descriptive analyses such as frequencies, percentages,
means (±standard deviation [SD]), and graphs were used
to describe demographics and the 3 variables: safety cli-
mate, exposure to leadership WRs, and patient safety risk
reduction.
Units were included in the analyses if at least 60% of
the respective caregivers completed the questionnaire
and there were at least 5 respondents per unit.
Leadership WR exposure was aggregated at the unit
level by calculating the percentage of respondents within
10. a unit who reported participating in leadership WRs at
least once. Safety climate was calculated for individual
respondents by taking the average of the scaled items and
then calculating the percentage of respondents within a
unit who reported positive safety climate (ie, proportion
of those who agreed slightly or strongly).17,18
Patient safety risk reduction and received feedback
about actions taken as a result of WRs were both calcu-
lated as the proportion of respondents within a unit who
answered “Yes,” that 1 or more patient risks were reduced
as a result of WRs or that they received feedback about
actions taken as a result of WRs, respectively (ie, respon-
dents who answered “No” or “Not Sure” to these items
were excluded).
Linear regression analyses were used to test the
assumption that the rate of caregiver participation in WRs
at the unit level was positively associated with safety
climate, patient risk reduction, and feedback about WR
actions. The predictor variable, unit exposure to WRs,
[email protected] - February 18, 2020 - Read articles at
www.DeepDyve.com
Schwendimann et al 417
was categorized as follows: (a) 0% to 19% of caregivers
reported participating in WRs = no or low exposure; (b)
20% to 59% = medium exposure; and (c) ≥60% = high
exposure. The thresholds for this categorization came
from earlier studies using the SAQ.8,14,19 Because the data
aggregated at the unit level were hierarchically structured
(units within hospitals), the authors considered multilevel
11. modeling to be appropriate and included hospitals as a
random effect.
All analyses were performed using IBM SPSS Statistics,
version 19.0.1 (IBM Inc, Armonk, NY) and STATA
11.2 (StataCorp LP, College Station, TX).
Results
A total of 19,053 SAQ surveys were received from
49 hospitals and included 706 clinical and nonclinical
units (4-38 per hospital); the units included medical and
surgical patient units, critical care units, hospital pharma-
cies, and med tech laboratories (overall response rate =
80.2%; range = 64% to 100% per hospital). Mean and
median number of respondents per unit were 27 (SD 19)
and 22, respectively. Respondent characteristics are
shown in Table 1.
Leadership WRs
Based on the item “Does your clinical area use Patient
Safety Leadership WalkRounds to discuss with senior
leaders any issues that could harm patients or undermine
the safe delivery of care,” the percentage of people
within units who answered “Yes” ranged from 0% (49
units) to ≥80% (16 units). More specifically, using the
item “How often did you participate?” 100 units (14.2%)
had 0 respondents who self-reported participating at
least once in WRs, whereas the proportion of respon-
dents who reported that they themselves had participated
at least once ranged from 4% to 100% in the remaining
606 units (85.8%).
Only 52 (7.4%) of the 706 hospital units in this study
reported WR exposure above the 60% threshold (60% of
respondents participating in WRs at least once).
12. Safety Climate
In 419 (59.3%) of the total 706 units, ≥60% of the respon-
dents reported a positive safety climate. Overall, the propor-
tion of respondents per unit reporting a “good safety climate”
ranged from 0% (1 unit) to 100% (35 units; Figure 1).
Patient Safety Risk Reduction
Of the total 706 units, the percentages of respondents per
unit reporting safety risk reduction following WR expo-
sure ranged from 0% (62 units) to 100% (1 unit; Figure 2).
More specifically, ≥60% of the respondents reported
patient safety risk reduction as a result of WR exposure
in only 41 (5.8%) of the units.
Table 1. Respondent Characteristics (N = 19 053).
Female sex, n (%)a 15 337 (82.8)
Professional categories, n (%)a
Nurses (registered nurse, nurse manager,
LVN/LPN)
8463 (46.3)
Support staff (eg, unit assistant, clerk,
secretary)
2960 (16.2)
Technicians (eg, EKG, laboratory, radiology) 2253 (12.3)
Physicians (attending/staff/resident
physician)
1267 (7.0)
13. Therapists (eg, respiratory, PT, OT, speech,
dietician)
1280 (7.0)
Pharmacist 331 (1.8)
Nurse practitioner/physician assistant 105 (0.6)
Medical administrator 30 (0.2)
Others (eg, maintenance personnel) 1594 (8.7)
Years in organization, n (%)a
<1 year 2324 (12.8)
1-2 years 2640 (14.6)
3-4 years 5432 (28.5)
5-10 years 2301 (12.1)
11-20 years 2798 (14.7)
>20 years 2633 (13.8)
Abbreviations: LVN, licensed vocational nurse; LPN, licensed
practical nurse; EKG, electrocardiogram; PT, physical therapist;
OT,
occupational therapist.
aMissing data: sex = 541; professional categories = 770; years
in
organization = 925.
Figure 1. Safety climate, percentage of respondents reporting
“good safety climate” (n = 706 units).
[email protected] - February 18, 2020 - Read articles at
www.DeepDyve.com
418 American Journal of Medical Quality 28(5)
14. Actions Taken as a Result of WRs
Percentages of respondents per unit reporting feedback
about actions taken as a result of WRs ranged from 0%
(79 units) to 100% (1 unit; Figure 3). More specifically,
≥60% of the respondents reported feedback about actions
taken as a result of WR in only 34 (4.8%) of the 706 units.
Association Between Leadership WR
Exposure, Safety Climate, Patient Safety
Risk Reduction, and Feedback About WR
Actions Taken
Units with ≥60% of caregivers reporting at least 1 WR
exposure had a significantly higher safety climate com-
pared with those units in either of the other categories
(Figure 4). In the multilevel regression analysis testing
for associations between unit WR exposure and unit
safety climate, significant variability between hospitals
was observed (variance = 55.02; 95% confidence interval
[CI] = 29.14-103.91). In addition, units in which ≥60% of
caregivers reported at least 1 WR exposure reported sig-
nificantly higher patient safety risk reduction compared
with units in either of the other categories (Figure 4). The
regression analysis for unit WR and unit patient safety
risk reduction, considering hospital level as random
effects, revealed significant variability between hospitals
(variance = 38.58; 95% CI = 21.36-69.67). Furthermore,
units in which ≥60% of caregivers reported at least 1 WR
exposure reported significantly higher feedback about
WR actions compared with those units in either of the
other categories (Figure 4). The regression analysis for
unit WR and received feedback about actions taken as
a result of WR, considering hospital level as random
effects, revealed significant variability between hospitals
(variance = 24.26; 95% CI = 12.21-48.18).
15. Discussion
This retrospective multicenter study on the relationship
between WRs and patient safety climate, patient safety
risk reduction, and WR feedback included more than
19 000 caregivers from 706 clinical areas in 49 hospitals
of a large nonprofit health system in the United States. A
strong association was observed between exposure of
caregivers to WRs and all 3 outcome variables at the
hospital unit level. These associations may reflect a
frequency effect as a function of units with none/low,
medium, and high leadership WR exposure, whereby
these effects varied significantly between participating
hospitals.
Differences in safety climate scores of caregivers in
units with a higher WR exposure compared with units
with a relatively low WR exposure might indicate an
effect of this widely applied intervention on team perfor-
mance and perceptions. It appears that WRs elicited
patient safety issues and potential solutions within these
caregiver teams, allowing them not only to express their
concerns to WR leaders but also to openly discuss identi-
fied aspects of what may harm patients and how the care
teams may eliminate those risks.
Figure 2. Patient safety risk reduction, percentage of
respondents reporting patient safety risk reduction following
walkround (n = 706).
Figure 3. Actions taken as a result of the walkround (WR),
percentage of respondents reporting received feedback about
actions taken as a result of the WR (n = 706).
[email protected] - February 18, 2020 - Read articles at
www.DeepDyve.com
16. Schwendimann et al 419
Caregivers’ notions of patient safety risk reduction
and feedback about actions taken as a result of the WRs
also revealed significant differences between units with
relatively high WR exposure compared with those units
with relatively low WR exposure. This indicates that
WRs likely impact caregiver behavior toward patient
safety. Although the characteristics of patient safety risk
reduction and actions taken as a result of the WRs were
not explored in detail, findings indicate that greater num-
bers of caregivers being exposed to WRs was associated
with greater risk reduction and more feedback on
WR-related actions at the unit level. From anecdotal evi-
dence, caregiver teams used a variety of teamwork
behaviors such as structured information delivery, criti-
cal language, and debriefing20 to resolve patient safety
defects and provide safe care.
Units with relatively high WR exposure showed favor-
able results in terms of safety climate scores when com-
pared with units having a relatively low WR exposure.
One could argue that WRs lead to proactively identifying
safety risks, facilitating open discussion of solutions, and
providing feedback on actions taken as a result of the dis-
cussion. These activities likely positively influence safety
climate, because units with more exposure to these expe-
riences reported a significantly higher safety climate.
These findings are in line with another study in which
WRs were associated with increased caregivers safety
climate perceptions.8 That study relied on a pre–post
design comparing 21 clinical areas in 2 hospitals over
time, whereas the current study findings are based on a
17. 1-time observation of a multicentre sample of more than
700 hospital units showing strong associations between
safety climate and the rate of WR exposure. Although
examples of patient safety risk reduction activities were
not explored in detail and rely solely on caregiver self-
report, these results are clearly in line with another study
showing that risk reduction related to specific actions
resulting from WRs is a common strategy to provide safe
care.9 WR data from the current study suggest that WRs
encouraged teams to speak up about errors and potential
safety risks, as well as to adopt practices shared through
“lessons learned.” In another study, improvements in
safety culture and facilitated reporting led to substantial
improvements in safety practices such as equipment
replacement, process changes, and effective communica-
tion.10 Such safety practices also were employed in many
of the clinical units in the current study, according to
anecdotal evidence. Each may have influenced caregivers’
perceptions of patient safety climate. However, the fact
remains that more than 3 times the number of units with
>60% participation in WRs had positive patient safety
climate scores compared with units with participation
below the 60% threshold. Therefore, the authors attribute
the positive safety climate to WR exposure, especially
in those units with high WR exposure rates compared
with those with relatively low exposure.
This study reinforces the argument that “leaders must
see patient safety problems as problems of their system
(eg, different units within hospitals) not of their employ-
ees,”1 supported by the use of WRs that allow leaders to
tap into questions such as “How did this incident occur?”
or “What in our treatment chain could harm a patient?”
rather than “Who did it?” or “Who was at fault?” In this
study, hospital units that participated in WRs with senior
19. 90
100
0%-19% exposed to WR
(n=259 units)
20%-59% exposed to WR
(n=395 units)
≥60% exposed to WR
(n=52 units)
(p=0.149)
(p=0.000)
(p=0.000)
(p=0.000)
(p=0.000)
(p=0.000)
Pa�ent safety risk reduc�on
Feedback about WR ac�ons
Figure 4. Association between low, medium, and high
walkround (WR) exposure, safety climate, risk reduction, and
feedback
about WR actions taken (n = 706 units).
[email protected] - February 18, 2020 - Read articles at
www.DeepDyve.com
20. 420 American Journal of Medical Quality 28(5)
leaders demonstrated enhanced patient safety perfor-
mance, as evidenced by improved safety culture scores
and increased risk reduction efforts. Indeed, units with
high WR exposure reported some of the highest safety
climate scores and most robust risk reduction strategies.
Study Limitations
Several study limitations must be recognized. First, the
reliance on cross-sectional surveys allows for observa-
tions and associations to be made, whereas causal rela-
tionships between WR frequency and improvements in
safety climate could not be established. Second, patient
safety “risk reduction” was noted on the questionnaire as
“Do you think one or more patient safety risks were
reduced as a result of WalkRounds? (Yes, No, Not
Sure),” such that exploring the specific patient safety risk
reduction strategies or the format and content of WRs
was beyond the scope of the current study. Nevertheless,
the results suggest that higher WR frequency was associ-
ated with larger proportions of respondents reporting that
they thought one or more patient safety risks were
reduced as a result. Third, the science of assessing and
improving safety culture is not yet mature, and the use of
a 60% positive threshold is based on previous experience
and a priori thresholds used in previous safety culture
studies. Fourth, other activities and safety practices (eg,
risk reduction practices such as error reporting and dis-
cussion or safety debriefing) that have been used in the
study units in addition to the WRs may have contributed
to the positive patient safety climate (above the 60%
threshold) in many units. Nevertheless, the 60% thresh-
21. old is quite conservative relative to other standards,21 the
response rates for this study were relatively high, and the
sample was large. Moreover, the focus on rates of WR
exposure rather than format or content, while a limita-
tion, is also a methodological option that others may
choose to explore, as it allows for systematic compari-
sons between units, over time, regardless of the type of
unit, background of the senior leader, length of the WR,
location of the WR, and dynamics between the caregivers
and the senior leaders.
Conclusion
These findings from a large retrospective study indicate
that WR participation is associated with a positive safety
climate, safety risk reduction, and feedback about WR
actions. This study substantiates preliminary findings on …