/clinical/,/clinical/cckm-tools/,/clinical/cckm-tools/content/,/clinical/cckm-tools/content/questionnaires/,/clinical/cckm-tools/content/questionnaires/related/,

/clinical/cckm-tools/content/questionnaires/related/name-97084-en.cckm

201606168

page

100

UWHC,UWMF,

Tools,

Clinical Hub,UW Health Clinical Tool Search,UW Health Clinical Tool Search,Questionnaires,Related

Magnet Children's Hospitals IP Pediatric Fall Prevention Programs

Magnet Children's Hospitals IP Pediatric Fall Prevention Programs - Clinical Hub, UW Health Clinical Tool Search, UW Health Clinical Tool Search, Questionnaires, Related


Magnet Children's Hospitals: Le
Development and Quality Stand
Pediatric Fall Prevention Progra
Elaine Graf PhD, RN, NE-BC

Research & Funding Coordinator, Children's Memorial Hospita
Key words:
Magnet hospitals;
Fall prevention;
Pediatric falls
Magnet hospitals are required to mo
national benchmarks for those indic
most of the pediatric indicators, a ho
taking a leadership role in develo
benchmark for the indicator of choi
hospitals to establish valid pediatric
' 2011 Elsevier Inc. All rights reser
Children's Hospital (designated in 2005), Barbara Bush
Children's Hospital (designated in 2006), and the Children's
ment and led the successful hospital initiative to achieve Magnet Nursing
Recognition in 2001, redesignation in 2005 and 2010. Her research
Journal of Pediatric Nursing (2011) 26, 122–127
Midwest Nursing Research Society.

Corresponding author: Elaine Graf, PhD, RN, NE-BC.
Medical Center Dallas (designated in 2009), have led the
journey to better understand, articulate, prevent, and
establish benchmarks for pediatric inpatient falls.
Up until this time, the only published work related to
fall prevention focused on risk screens and prevention
programs for elderly patients. Further emphasis was placed
on the need for fall prevention programs in 2005 with the
regulatory agency establishment of a National Patient
Safety Goal requiring all acute care hospitals to implement
fall prevention programs to reduce the risk of patient
injury. Without an exemption from this National Patient
interest is the prevention of inpatient pediatric falls. The research has
resulted in the development of the first validated pediatric inpatient fall
risk screen, the General Risk Assessment for Pediatric In-patient Falls
(GRAF PIF
'
) scale. She has presented more than 40 papers at national and
international conferences. Dr. Graf has many publications, including a
chapter on “fluid and electrolytes alterations” in a new pediatric nursing
textbook published in 2008 by Delmar. Dr. Graf is a recipient of the
Children's Memorial Hospital Nurse Research Exemplar Award and is an
active member of American Nurses Association/Illinois Nurses Association,
Sigma Theta Tau International Nursing Research Society, the Society of
Pediatric Nursing, the Illinois Council on Nursing Resources, and the
Hospital Central California (designated in 2004), St. Louis
Dr. Graf came to CMH 14 years ago as the Research & Funding
Coordinator in the Department of Clinical & Organizational Develop-
0
d
cience in nursing/pediatric nu
irginia; and a PhD in Nursing
f Virginia. She has achieved A
r. Graf has been a faculty me
niversities, including Georgetow
tialing Center (ANCC) Magnet recognition are required to
monitor and achieve above the mean or midpoint on national
one of which is hospital falls. Prevention of hospital falls is
an important aspect of the management of patients in acute
care settings. Failure to provide a safe environment can lead
to falls that may result in injury. Such injuries may prolong
hospitalization, may lead to complications, and can decrease
family trust in the health care team. Magnet Children's
hospitals, such as Children's Memorial Hospital (designated
in 2001), Miami Children's Hospital (designated in 2003),
Texas Children's Hospital (designated in 2003), Children's
1
Author Information: Dr. Elaine Graf received a bachelor of science in
nursing from the University of Rochester in Rochester, NY; a master of
s rse practitioner from the University of
V Administration from the Medical College
o NCC certification as a Nurse Executive.
D mber for more than 13 years at several
u n, Old Dominion, and Northern Illinois.
ALL HOSPITALS SEEKING American Nurses Creden-
E-mail address: egraf@childrensmemorial.org.
882-5963/$ – see front matter ' 2011 Elsevier Inc. All rights reserved.
oi:10.1016/j.pedn.2010.12.007
ading Knowledge
ards for Inpatient
ms
1
l, Chicago, IL
nitor nursing-sensitive indicators and be above the mean/median of
ators. When there is no valid national benchmark, as is the case for
spital seeking Magnet designation or redesignation is charged with
ping a mechanism that leads to the establishment of a national
ce. This article will present the efforts taken by Magnet Children's
screening tools and benchmark inpatient pediatric falls.
ved.
benchmarks related to nurse-sensitive outcomes measures,
Safety Goal, all pediatric facilities were challenged to

review their approach to screening patients for fall risk and
evaluate the effectiveness of fall prevention programs. The
standard approach for pediatric fall prevention was to treat
all children as at risk of falling, to staff general medical
surgical pediatric units at a ratio of one nurse to every three
to four patients, and to utilize a family-focused model of
care that allows 24-hour parent visitation. Although these
strategies were felt to be effective, no data were available on
pediatric fall rates to validate the impact of this approach.
Research has shown that preventing patient falls begins
with an accurate assessment of a patient's risk of falling
followed by the initiation and ongoing evaluation of a fall
prevention program based on identified risks (Morse, 1997).
Fall prevention programs with the best sustained improve-
ments have included the use of validated fall risk assessment
tools to guide the use of care plan protocols and have used
fall classification data to guide population- or unit-based
interventions (Ignatavicius, 2000; Morse, 1997; Sullivan &
Bandros, 1999). This article will review the preliminary
efforts of Magnet Children's hospitals to understand the
characteristics of inpatient pediatric falls, to establish a
classification system, to develop validated pediatric fall risk
screening tools, and to identify national benchmark thresh-
olds, which can be used to guide practice improvement
The first Children's Hospitals to report descriptive data on
inpatient pediatric falls were Children's Memorial Hospital,
Central California (Cooper & Nolt, 2007; Graf, 2004; Hill-
Rodriquez, Messmer, & Wood, 2007). In a retrospective
review of 100 children who fell, Children's Memorial and
Miami Children's each reported many similarities as shown
in Table 1. Both reported a higher ratio of boys falling than
girls, that falls more frequently occurred when children were
with respiratory or neurological diagnoses, that activity prior
to falling and causes of falls were similar, that falls resulted
in minimal to no injury, that parents were often present
during the fall, and that repeat falls occurred.
Some differences between the samples related to age and
diagnosis. Miami Children's reported that children of all ages
fell; however, the age of child most at risk of falling were
toddlers (percentages were not provided). Although the
Children's Memorial sample concurred that falls occurred in
all age groups, it was adolescents, not toddlers, who fell more
frequently. Children with a psychiatric diagnosis were the
third highest diagnostic group within the Children's
Memorial sample to show a risk of falling. Specific
psychiatric conditions such as attention-deficit/hyperactivity
disorder, impulse control disorders, oppositional defiant
disorders, and disruptive behavior disorders made up this
grouping (Graf, 2005).
Cooper and Nolt (2007) reported similar findings in a
ata
ars)
onths
(seiz
y
123Magnet Children's Hospitals
Miami Children's Hospital, and Children's Hospital of
Table 1 Characteristics of Pediatric Inpatient Falls
Hospital
Miami Children's Hospital 2000 D
(N = 100, ages 6 months to 23 ye
Age Falls reported in all age groups 19–24 m
(most frequent age group to fall)
Gender Male (2:1 ratio)
Time of Day Fell between 8:00 p.m. and 10:00 p.m.
Diagnosis Respiratory/pulmonary/ENT neurological
Description Fell out of bed
Slipped on wet floor
Tripped over equipment
Injury Sustained little to no injury requiring onl
minimal intervention
Supervision Parents frequently present
Repeat Falls N5% fell more than once
Note: Data from Graf (2004).
initiatives in hospitals caring for children.
Characteristics of Inpatient Pediatric Falls
26-month sample of inpatient and outpatient falls from
Children's Hospital of Central California: 63% occurred with
boys; toddlers (25%) and adolescents (23%) were highest
age groups; and activity prior to fall included falling out of
bed, falling while walking, slipping on wet floor, and
tripping over equipment. Fall injuries were minor, and again,
parents were frequently present at the time of the fall. The
study reported the following patient care areas where falls
frequently occurred: emergency department (34%), physical
Children's Memorial Hospital 1998–2003 Data
(N = 100, ages 1 to 18 years)
12–24 months (21%)
3–6 years (26%)
7–10 years (9%)
11–18 years (44%)
Male (2:1 ratio)
Fell between 10:00 a.m. and 12:00 p.m. (24%)
or between 9:00 p.m. and 12:00 a.m. (22%)
ures) Neurological (23%)
Respiratory/pulmonary/ENT (19%)
Psych (13%)
Fell out of bed
Fell walking
Slipped in bathroom
Sustained little to no injury required no
intervention (86%)
Parent present 57%
17% fell more than once while hospitalized

therapy (33%), rehabilitation unit (29%), and oncology behavioral issues were nine times more likely to fall while
124 E. Graf
department (26%); however, it did not identify specific
patient diagnoses.
Children's Memorial further compared those who fell
from an anticipated or unanticipated physiological fall
(67 falls) with a control sample of 100 nonfalling patients,
who were matched by age group (b7 years or N7 years) and
unit and month of hospitalization, and found that the
variables of age; number of medications ordered (poly-
pharmacy); presence of a sensory or auditory deficit; drug
group classifications of narcotics, diuretics, chemotherapy,
hypertensives, antipsychotics, or depressants; and the
following patient diagnoses of neurosurgery/brain tumor,
cancer, endocrine/diabetes, psychiatric, cardiac, gastrointes-
tinal/medical, musculature and skin alteration/burns, were
not significant indicators of fall risk. Variables that were
significant in predicting anticipated and unanticipated
physiological falls included male gender (+), length of
stay (LOS) greater than 5 days (+), having more than one
diagnosis (+), communication deficit (+), confusion (+),
developmental delay (+), muscular weakness (+), need for
physical therapy/occupational therapy (+), gait disorder (+),
balance disorder (+), use of assistive devise (+), antiseizure
medication (+), neurological diagnosis/seizure disorder (+),
orthopedic diagnosis (+), general surgery or infectious
disease diagnosis (−), having an IV or heparin lock in
place (−), and presence of parents (−). Direction of the
association is noted as either an increase (+) or a decrease (−)
in likelihood of falling. This review suggests that having
parents present, when their child has intrinsic characteristics
that would put the child at risk of falling, may be a protective
factor against falls. In addition, the fact that the child had an
intravenous or heparin lock in place also served as a
protective factor against falls. Unlike adults with IVs who are
ambulating, children often have an adult or nurse with them
to push the IV pole, thus preventing the child from using the
IV pole as a crutch for balance. A chi-square analysis of
having an IV and parents' presence was highly significant
(p b .00001), suggesting that these two variables may be
linked in some way and may serve as a protector against
these specific fall events (Graf, 2005).
Although many reports note that parents are present at the
time of falls and call for parent diligence to be monitored,
this finding must be reviewed and interpreted cautiously in
light of the fact that all children's hospitals encourage
families to remain an active partner in the care of their child
by welcoming them to stay with their child 24/7. Having a
parent close by helps to decrease a child's anxiety, but long
parental care hours with many disruptions and distractions
and the lack of a restful sleep require nurses to be sensitive to
the demands placed on parents and to reach out to them to
provide for comfort needs and care breaks.
A recent study completed by researchers at Barbara Bush
Children's Hospital in Portland, ME, examined fall likeli-
hood and injury risk within a small sample of 100 patients,
33 of whom fell, and found that children with temperament/
hospitalized than children without a documented behavioral
issue. In addition, they found that children with bleeding
precautions and blood disorders were more than four times
likely to fall while hospitalized as compared with children
without these disorders. Additional predictors included
length of stay, history of fall injury, hyperactivity, bone
fragility, male gender, and age 5–11 years. They did note
that their sample of patients did not include many children
with documented cognitive impairment and/or neurological
disease, which may be why this category was not significant
as compared with other studies (Harvey, Kramlich, Chapman,
Parker, & Blades, 2010).
The Child Health Corporation of America (CHCA), a
corporation owned and operated by 43 Children's hospitals,
25 of which hold Magnet status, has recently commissioned
two studies. The first study was a survey of member hospitals
to identify current practices utilized to reduce the risk of fall
injury in hospitalized children, and the second study was a
large multisite prospective review of pediatric inpatient falls.
The first CHCA-sponsored study surveyed member
hospitals in 2007 to assess fall-related practices. Of 42
hospitals, 29 responded for a return rate of 69%. Findings
showed that there was great variation in fall definitions, fall
classification typologies, and measurement practices for
determining fall and injury rates. The National Quality
Forum definition, “any unplanned descent to the floor,” was
the most common fall definition used (88%). Other
definitions include any unplanned descent where the child
ends up at a lower level from where they started. Under this
second definition, a fall by a child who lost balance while
standing on a bed and fell hitting his or her head on the side
rail would be included but would not be included under the
first definition because the child did not end up on the floor.
Variation in fall definitions will influence fall rate calcula-
tions. With no consensus on how falls should be classified,
many approaches were reported, including classification by
fall type, severity of injury, activity at time of fall, and fall
classification used for adult/geriatric falls. Some reported
classifying pediatric developmental falls, whereas others
excluded all developmental falls. Most of the hospitals
reported calculating fall rates (72%), but only 31% calculated
injury rates. Some included developmental falls with injury
when calculating fall rates, whereas others excluded all
developmental falls from rate calculations. Hospitals
reported using a variety of risk assessment tools (90%),
with only 6 hospitals reporting the use of a screening tool
validated for use in pediatrics. A variety of prevention
strategies were described, but there was no consensus on
what interventions are most effective in preventing falls in
children (CHCA, 2009). The study highlighted the need for
leadership in moving forward to build consensus.
The second study commissioned by CHCA included
26 facilities who shared fall data over six consecutive months
to determine prevalence, fall characteristics, and related
injuries. The final data set included 782 inpatient falls.

Although this study did not include a control sample, a
be attributed to physiological causes but are created by
conditions that cannot be predicted before the first fall
run, and pivot. Unlike other fall classifications where the
Progress Toward Validation of Pediatric Fall
Risk Screening Tools
125Magnet Children's Hospitals
occurrence. Examples include an undiagnosed seizure
disorder, undetected lowered blood pressure event resulting
in fainting, or a pathological fracture. When this type of fall
occurs and there is a likelihood that the underlying condition
may recur, nursing interventions should be targeted toward
either preventing a second fall or preventing injury if the
patient falls again. Prevention strategies for this type of fall
include patient education on the specific condition and
targeted prevention of additional falls (Morse et al., 1987).
The initial descriptive study of pediatric in-patient falls
showed that pediatric falls could be classified using these
three categories but needed to include an additional category
for developmental falls, which are falls that are due to a
child's growth and development (Graf, 2004; Harvey et al.,
2010; Razmus, Wilson, Smith & Newman, 2006). Develop-
mental falls are a normal part of how children learn to walk,
regression analysis of falls resulting in injury as compared to
falls without injury will be done that may identify patient
characteristics that increase a child's risk of injury from falls.
Analysis is still underway, and findings have not been
published to date.
Although the descriptive characteristics of inpatient
pediatric falls are showing many similarities that have been
used to guide the development of pediatric fall risk screening
tools, only a few studies have utilized a control group for
comparison (Graf, 2008; Harvey et al., 2010). Great variation
can occur in the types of pediatric patients admitted to
different hospitals. As such, it is very important to review
descriptive quality data carefully and, before making
decisions, compare the findings to those children admitted
who do not fall in order to determine the true significance of
the descriptive findings.
Establishing a Classification System
“The classification of falls is important as methods for
prediction and prevention differ for each type of fall” (Morse,
1997, p. 5). Morse, Tylko, and Dixon (1987), in an analysis
of circumstances resulting in hospitalized adult falls,
revealed three classifications of falls: accidental falls,
anticipated physiological falls, and unanticipated physiolog-
ical falls. Accidental falls are not due to physical factors but
rather environmental hazards or errors of judgment and
therefore are best prevented through environmental strategies
to keep the patient's environment as safe as possible.
Anticipated physiological falls are the only fall group that
can be predicted by using a fall risk screen, as they are due to
physical or physiological factors intrinsic to the patient that
can be identified. Once predicted, they can often be
prevented or the injury minimized by initiating a fall
prevention protocol. Unanticipated physiological falls may
Up until the establishment of a National Patient Safety
Goal to implement and monitor the effectiveness of a fall
prevention program, including the identification of a patient
at risk of falling, pediatric nursing staffs considered all
children at risk of falling and, as such, instituted general
safety principles and fall prevention strategies. With all
children determined to be at risk, little was done to identify if
a subset of children existed who were more at risk of injury
from falls. Until recently, there were no published bench-
marks of pediatric inpatient fall rates or evidence of
successful programs. Validated adult and geriatric fall risk
screens had not been tested in pediatric populations. Razmus
et al. (2006) studied the validity of the Morse Fall Scale and
the Hendrich II Fall Risk Model in predicting pediatric
inpatient falls and found neither to be effective.
Graf (2008) undertook a retrospective case-matched
control study of 100 inpatient pediatric falls and 100 controls
matched by age range (b7 years old or N7 years old) and unit
and month of hospitalization. Charts were reviewed for the
presence of 38 variables identified in the literature as risk
factors. These indicators included all diagnostic and
medication groups, assessment for polypharmacy, gait and
balance alterations, gender, LOS, developmental delay,
communication deficit, need for IV/heparin lock, and
goal is to prevent falls, the goal with developmental falls is
not to prevent but to anticipate falls and keep the child's
work and play environment as safe as possible to prevent
injury. Since developmental falls are normal, it is only
developmental falls that result in injury while the child is
hospitalized that need to be monitored. Developmental falls
that result in injury should be included in fall rate
calculations and classified separately from accidental falls
to provide a clearer picture of the types of falls occurring
on units.
Medical Management Partners, Inc. (MMP), a bench-
marking corporation made up of 20 premier children's
hospitals, including 5 Magnet hospitals, that contract to
share unblinded outcome data gathers and reports on fall
classification percentages on a yearly basis. The most recent
data submitted by 7 participating hospitals demonstrates the
following median values: developmental falls with injury,
9%; accidental falls, 55%; anticipated physiological falls,
27%; and unanticipated physiological falls, 9% (MMP Web
site, www.mmpcorp.com, accessed September 27, 2010).
These data are congruent with other published results
showing that a high percentage of pediatric falls are
accidental or of a type that is not predicable before the
first fall event (Graf, 2008; Kingston, Bryant, & Speer, 2010;
Razmus et al., 2006).

presence of parents. Twenty variables were found to be
significant. These variables were entered into a principal
cluster analysis to identify variables highly correlated with
each other or collinear. Ten clusters resulted from each
cluster; one predictor variable was chosen based on
univariate analysis results and theoretical soundness. Logis-
tic regression analysis achieved an R
2
of .46 and showed a
subset of five variables that provided the best fit to the data
and correctly predicted 84% of the cases. The variables
selected in the Best Logistic Model are listed in Table 2.
Significant risk factors were length of stay in 5-day
increments, orthopedic diagnosis, physical/occupational
therapy, seizure medication, and being IV/heparin lock
free. This model was used as the foundation of a fall risk
assessment scale named the General Risk Assessment for
fell while hospitalized and 153 children who did not fall
acceptable Cronbach's alpha values ranging from .64 to .77.
A review of the distribution of risk levels and/or tool scores
126 E. Graf
Pediatric In-patient Falls Scale (GRAF PIF
'
). Logistic
regression parameter estimates of the predictor model were
assessed and found to be equal in value, thereby showing no
evidence that any of the risk factors contributed more or less
to the overall risk of falling. As such, each variable was given
a weight of one point. The data were then reanalyzed to
determine score ranges and sensitivity/specificity of various
cutoff points. A cutoff point of 2 resulted in sensitivity/
specificity of 75%/76%, the same level achieved by the
Morse and Hendrich II fall scales (Hendrich, Bender, &
Nyhuis 2003; Morse, 2002). History of fall within the past
month and a fall during hospitalization were added to the
final risk screen as automatic indicators of fall risk.
At the same time, Miami Children's Hospital reviewed
their fall data and developed the Humpty Dumpty Falls
Prevention Program (HDFS) and risk screening tool based on
a retrospective analysis of patients who fell and expert
consensus of fall risk indicators (Hill-Rodriguez, et al.,
2009). The fall risk screening tool differentiates pediatric
patients into either low- or high-risk fall categories based on
identified risk factors of age, gender, diagnosis, cognitive
impairments, environmental factors, length of time post-
surgery/sedation/anesthesia, and medication usage. A score
between 7 and 11 signifies a low risk for falling, whereas a
score between 12 and 23 signifies a high risk of falling
while hospitalized. To validate these cutoff points, a study
of 308 patients was done, which included 153 children who
Table 2 Variables Selected in the Best Logistic Model for
GRAF PIF
'
Variable
Odds
Ratio
Confidence
Limits p
LOS, for each 5 days 1.84 1.30–2.62 .0007
IV/heparin lock free 3.60 1.47–8.33 .0046
Physical therapy/occupational
therapy need
2.88 1.15–7.21 .0241
Seizure medication 4.90 1.99–12.02 .0005
Ortho disorder 4.33 1.23–15.27 .0224
Note: R
2
= .46.
revealed that the GRAF PIF
'
identified 38% of the sample as
scoring at high risk (score of N2). The mean score achieved
within the sample on the HDFS was reported as 14.34 and
resulted in 84% of the sample scoring at high risk. Sixty-
eight percent scored at risk on the CHAMPS tool, 27%
scored at high risk on the Cummings scale, and 20% scored
at high risk on the CNMC tool. Only the GRAF PIF
'
and the HDFS correctly identified the 33 patients who fell
in the sample. This study highlights the importance of
testing screening tools within specific patient populations
to determine best fit prior to adoption.
Development of Benchmark for Inpatient
Pediatric Fall Rates
Two approaches to the establishment of benchmarks for
inpatient pediatric falls have occurred. The first approach
sought to work with existing National Benchmarking
Organizations to either establish a specific indicator for
pediatric falls or to breakout pediatric falls from an
established benchmark data group of participating hospitals
(Graf, 2008; Hill-Rodriguez et al., 2009). The second
approach used was to establish a pediatric falls benchmark-
ing collaborative between a small group of children's
hospitals willing to contract to share inpatient fall data,
establish common goals, and share success stories (Kingston
et al., 2010).
Much effort has gone into encouraging current bench-
marking organizations to monitor pediatric fall prevalence
separate from adult fall prevalence; however, it was not
selected as one of the nursing-sensitive indicators for
who were matched for age, gender, diagnosis, and unit
location. Children with scores of 12 or higher were two
times more likely to fall than children who received scores
in the low-risk category (p = .03). However, the reported
sensitivity was 0.85, and specificity was 0.24. The overall
percentage of patients correctly classified as to their risk of
falling was 59.3%.
Further validation of the GRAF PIF
'
and the HDFS was
achieved in a study undertaken by researchers at the Barbara
Bush Children's Hospital. In an attempt to find the best
pediatric fall risk screening tool for their patient population,
they evaluated five pediatric screening tools, including the
above two tools (Harvey et al., 2010). The other three tools
reviewed were Changes in mental status, History of falls,
Age b3 years old, Mobility problems, Parental involvement,
Safety actions (CHAMPS), Cummings, and a tool developed
at the Children's National Medical Center (CNMC). The
sample included 33 patients who fell and 67 patients who did
not fall during their hospitalization. The study showed that
all tools except the CHAMPS tool achieved reasonably

monitoring by the Pediatric Data Quality Systems Collab-
orative in 2007 (Hill-Rodriguez et al., 2009). As a
participating member of MMP, a benchmarking corporation
made up of 20 premier children's hospitals, including 5
Magnet hospitals, Children's Memorial Hospital requested
began reporting comparative data in 2005. Four hospitals
have consistently participated and have demonstrated fall
Children's Hospitals have been instrumental in guiding
research, practice, and quality improvement processes
both within their organizations and by reaching out to the
professional community to share best practices that keep
children safe. Through these efforts, national benchmark
data are more available, and classification characteristics
fall-prone patient. The Gerontologist, 27, 516−522.
Razmus, I., Wilson, D., Smith, R., & Newman, E. (2006). Falls in
hospitalized children. Pediatric Nursing, 32, 568−572.
127Magnet Children's Hospitals
rates per 1,000 patient days ranging from 1.1 to 0.37, with a
median rate of 0.77.
Kingston et al. (2010) reported on a partnership
established between three Magnet children's hospitals to
benchmark falls. They standardized definitions and calcu-
lation processes, set annual goals for improvement, and
shared fall risk reduction strategies and best practices. Falls
were classified, and developmental falls were excluded from
fall rate calculations as they were felt to be normal even if
the fall resulted in an injury. At the start of the project, the
reported fall rates per 1,000 patient days were 0.81, 1.02,
and 1.20. The collaborative goal for 2006 was to lower the
fall rate to 0.62. The average fall rate 1 year later was 0.87
per 1,000 patient days, with only one hospital meeting the
threshold goal. The collaborative also provided an oppor-
tunity to review fall events for injury, noting that only 36%
of falls resulted in an injury. Of these injuries, only 2% of
the falls resulted in a moderated injury requiring suturing,
casting, traction, or a need for neurological consultation
(Kingston et al., 2010).
From these benchmarking outcomes, there is growing
evidence that pediatric inpatient fall rates are very different
from adult/geriatric reported rates, and injuries are not as
severe. Setting a fall rate threshold at less than 1/1,000
patient days appears to be justified and reasonable. Further
discussion and guidelines need to focus on establishing a
threshold for pediatric fall injury rates. Monitoring of injury
severity is critical and may be the most valuable measure of
the quality of the fall prevention program in use.
Summary
Knowledge and quality monitors related to pediatric
inpatient falls have grown steadily since 2005. Magnet
for a fall event and fall classifications. Discussions also
centered on what data were important to gather and report to
member hospitals. Member hospitals requested comparative
data on fall rates for both inpatient and outpatient falls and
comparative data on fall classification percentages. MMP
Sullivan, R., & Badros, K. K. (1999). Recognizing risk factors to prevent
patient falls. Nursing Management, 30,37−41.
have shown that pediatric falls differ greatly from adult
and geriatric patient populations and as such need to be
addressed with different approaches and strategies.
References
Child Health Corporation of America. (2009). Pediatric falls: State of the
science. Pediatric Nursing, 35, 227−231.
Cooper, C. l., & Nolt, J. D. (2007). Development of an evidence-based
pediatric fall prevention program. Journal of Nursing Care Quality, 22,
107−112.
Graf, E. (2004). Identifying predictor variables associated with pediatric in-
patient fall risk assessments. Proceedings from the 5th National
Conference Evidence-based Fall Prevention Conference, Clearwater, FL.
Graf, E. (2005). Pediatric hospital falls: Development of a predictor model
to guide pediatric clinical practice. A paper presented at the Sigma
Theta Tau International 38th Biennial Convention, Indianapolis, IN.
Graf, E. (2008). Pediatric fall risk assessment & classification: Two
hallmarks for a successful inpatient fall prevention program. SPN News,
17,3−5.
Harvey, K., Kramlich, D., Chapman, J., Parker, J., & Blades, E. (2010).
Exploring and evaluating five paediatric falls assessment instruments
and injury risk indicators: An ambispective study in a tertiary care
setting. Journal of Nursing Management, 18, 531−541.
Hendrich, A. L., Bender, P. S., & Nyhuis, A. (2003). Validation of the
Hendrich II fall risk model: A large concurrent case/control study of
hospitalized patients. Applied Nursing Research, 16,9−21.
Hill-Rodriguez, D., Messmer, P., Williams, P., Zeller, R. A., Williams, P.,
Wood, M., & Henry, M. (2009). The Humpty Dumpty Falls Scale:
A case control study. Journal for Specialists in Pediatric Nursing, 14,
22−32.
Hill-Rodriguez, D., Messmer, P. R., & Wood, M. L. (2007). Humpty
dumpty sat on a wall: Developing a pediatric falls prevention program
and scale. A paper presented at the 17th Annual Society of Pediatric
Nurses Convention, Milwaukee, WI.
Ignatavicius, D. (2000). How do you help your staff rise to the fall-
prevention challenge? Nursing Management, 31,27−30.
Kingston, F., Bryant, T., & Speer, K. (2010). Pediatric falls benchmarking
collaborative. Journal of Nursing Administration, 40, 287−292.
Morse, J. (1997). Preventing patient falls. Thousand Oaks, CA: Sage
Publications, Inc.
Morse, J. (2002). Enhancing the safety of hospitalization by reducing patient
falls. American Journal of Infection Control, 30, 376−380.
Morse, J. M., Tylko, S. J., & Dixon, H. A. (1987). Characteristics of the
that pediatric fall data be added as a benchmarked data
element in 2004. Work began to establish clear definitions