Running head: HIGH FALL RISK IN ARU 1

Identification of High Fall Risk Patients in Acute Rehab

Vanessa Vonderhaar-Picard DNP, MEd, RN, CNML

An evidence-based doctoral project presented to the Department of Nursing at

Mount St. Joseph University

in partial fulfillment of the Degree:

Doctor of Nursing Practice

Date: May 14, 2019

Susan A. Johnson, PhD, RN - DNP Advisor

Copyright by

Vanessa Vonderhaar-Picard DNP, MEd, RN, CNML

2019 2 HIGH FALL RISK IN ARU

Identification of High Fall Risk Patients in Acute Rehab

Vanessa M. Vonderhaar-Picard

Mount St. Joseph University

3 HIGH FALL RISK IN ARU Table of Contents

Executive Summary………………………………………………………………….……………5

Introduction………………………………………………………………………………………. 6

Problem……………………………………………………………………………………………6

PICOT Question…………………………………………………………………………..…….....9

Literature Review……………………………………………………………………………...…10

Search Description……………………………………………………………………….10

Table 1 Literature Review…………………………………………………………….....11

Summary of Levels of Evidence ………………………………………………………...12

Table 2 Level of Evidence……………………………………………………………….12

Falls………………………………………………………………………………………13

Fall Risk………………………………………………………………………………….13

Acute Rehab Risk Factors for Falls……………………………………………………...15

Psychiatric Risk Factors for Falls ……………………………………………………….16

Morse Fall Scale…………………………………………………………………………17

Table 3 Morse Fall Scale………………………………………………………………...18

Edmonson Psychiatric Fall Assessment Tool.…………..……………………………….18

Table 4 Edmonson Psychiatric Fall Assessment Tool…………………………………..20

Comparison of Edmonson and Morse Fall Scales……………………………………….21

Table 5 Comparison Fall Risk Factors of Edmonson Psychiatric Fall Risk and Morse Fall

Risk………………………………………………………………………………………22

Summary of Literature…………………………………………………………………...22

Evidence-Based Practice Model…………………………………………………………………23

Theoretical Framework……………………………………………………………………..……25 4 HIGH FALL RISK IN ARU Integrated Evidence and Theoretical Framework…………………………………….…27

Table 6 Integrated Evidence-Based Practice Model and Theoretical Framework………28

Project Proposal………………………………………………………………………………….28

Participants………………………………………………………………………..……..29

Setting……………………………………………………………………………………30

Intervention………………………………………………………………………………31

Data Collection…………………………………………………………………………..33

Stakeholders……………………………………………………………………………...33

Driving and Restraining Forces………………………………………………………….34

Budget……………………………………………………………………………………35

Table 7 Project Budget…………………………………………………………………..35

Analysis and Results.………………………………………………………………………….…35

Table 8 Results of Falls Per 1,000 Patient Days Comparing First and Fourth Quarter.....36

Table 9 Results of Estimation of Population Ratio…………………….….……………..37

Significance and Implications……………………………………………………………………38

The Future………………………………………………………………………………………..39

Sustainability……………………………………………………………………………..39

Future Projects………………………………………………………………….………..39

Dissemination………………………………………………………………………...….40

Timeline………………………………………………………………………………………….40

Table 10 Project Timeline………………………………………….……………………41

Conclusion……………………………………………………………………………………….41

Acknowledgments………………………………………………...……………………………...42

References……………………………………………………………………………………….43 5 HIGH FALL RISK IN ARU Appendix A Quarter One Age and MFS Data……………………………………………...……47

Appendix B Quarter Four Age and MFS Data………………………………………………..…47

Appendix C Gender Data……………………………………………………………………...…47

Appendix D Admission Diagnosis Quarter One and Four …………………………………..….48

Appendix E Power Point: ARU Staff Education……………………………………………….49

Appendix F Edmonson Handout…………………………………………………………….…..60

Appendix G Edmonson EPIC Handout ……………………………………………………...….63

Appendix H ARU Reminder Flyer………………………………………………………………64

6 HIGH FALL RISK IN ARU Executive Summary

Patient falls are a prevalent safety issue worldwide, and are the most frequently reported safety event in hospitals. Falls are a costly and complex issue, and despite much research there is not a simple solution to improving fall rates. Patients in the acute rehabilitation unit (ARU) setting are one of the most at risk populations to falls; however, a screening tool for this specific population has not yet been identified. A tool is needed to identify high fall risk patients in the

ARU so appropriate preventative nursing interventions can be implemented.

An exhaustive literature search indicates a fall prediction tool should be utilized in the

ARU; however, a tool has not been validated in this setting. The Edmonson Psychiatric Fall Risk

Assessment Tool (EPFRAT), although indicated for acute psychiatric patients, evaluates age, mental status, elimination, medications, diagnosis, ambulation/balance, nutrition, sleep disturbance, and history of falls, (Edmonson, Robinson, & Hughes, 2011), which are the same risk factors that contribute to Acute Rehab falls. Literature supported evaluating the EPFRAT tool in the ARU to identify high fall risk patients. Accurate identification of high falls risk patients in ARU allows for implementation of correct nursing interventions to prevent falls.

Utilization of the EPFRAT in conjunction with the Morse Fall Scale (MFS) resulted in a decrease of falls per 1,000 patient days, from 9.025 in the first quarter, to 1.947 in the fourth quarter of 2018. In the first quarter, there were 10 falls on the ARU when only the MFS was used. During the fourth quarter when both scales were used, there were only 2 falls. Results indicate that with 95% confidence, using the EPFRAT scale in addition to the MFS decreases in the average number of falls per 1,000 patient days by between 0.44 and 13.9. Further evaluation of the EPFRAT is needed at the site level and should be trialed in another ARU to see if comparable results are reproduced. Falls remain a significant safety issue in the ARU setting, however, use of the EPFRAT assisted in decreasing falls. 7 HIGH FALL RISK IN ARU Identification of High Fall Risk Patients in Acute Rehab

Introduction

The incident of patient falls has become a worldwide, healthcare epidemic. Despite much time, attention, and use of resources, patients are still suffering from the effects of falls. The most frequently reported adverse events in hospitals are falls, and injuries related to falls (Ambutas, Lamb, & Quigley, 2017). According to Clancy (2013) “each year, between

700,000 and 1 million people experience a fall in US hospitals” (p. 195). Christopher, Trotta,

Yoho, Strong, and Dubendorf (2014) predict by 2020, the annual total cost for fall related injuries could increase to 34.4 billion, and Rosario and colleagues predict cost of 54.9 billion

(Rosario, Kaplan, Khonsari, & Patterson, 2014). According to the National Database of Nursing

Quality Indicators (NDNQI) (2017), the fall rate per 1,000 patient days for ARU (units peer group bed size 100-199) is 1.8 – 9.81 with the median of 5.02. Upon literature review, it is clear that falls are a significant issue facing nursing today; however, there is not one answer or solution to solve this challenging and complex issue. Currently in the ARU at The Jewish

Hospital, more than 80% of the patients on the unit are identified as being high fall risk with the current screening tool. The need for a tool to identify high fall risk patients is needed so that the appropriate prevention interventions can be implemented for this population. Fall risk assessment tools are the foundational element of fall prevention programs (Feil & Gardner,

2012). The purpose of this practice change project is to appropriately identify high fall risk patients in the ARU setting.

Problem

Nationally falls are highly prevalent in the ARU setting with estimated rates between

2.92 and 15.9 falls per 1,000 patient days (Frisina, Gullnitz, & Alverzo, 2010). In 2017, The

Jewish Hospital (TJH) ARU experienced a fall rate of 4.67, and in 2018 through February, the 8 HIGH FALL RISK IN ARU fall rate spiked to a high of 13.05, despite a focus on fall reduction. The fall rate in February

2018 was 16.2 per 1,000 patient days, thus remaining above 2017 and exceeds NDNQI median fall rate of 5.02 for similar ARU units (NDNQI, 2017). Many different reasons contribute to why these ARU patients are a higher fall risk compared to other areas.

Patients on acute rehabilitation units embody “one of the most at-risk populations for falls during hospitalization” (Rosario et al., 2014, p. 86). Active promotion of both mobility and independence in the ARU setting contributes to the increased risk of falls (Rabadi, Rabadi, &

Peterson, 2008). ARU patients have more risk factors compared to general hospital patients

(Gilewski, Roberts, Hirata, & Riggs, 2007). During therapy, rehab patients are taught to be as independent as possible but then are limited by safety measures put in place and instruction to not get up without assistance (Cournan, Fusco-Gessick, & Wright, 2018). In addition, patients who fall in the ARU have multiple risk factors, and are often cognitively impaired and impulsive, similar to those in the acute psychiatric setting.

Despite research confirming increased risk of falls in the ARU setting, currently no validated tool to determine fall risk has been identified specific to ARU patients. Often in ARU settings, tools that are utilized, such as the MFS, were intended for the medical surgical hospital setting and identify most ARU patients as high risk, which makes it challenging to determine the true fall risk of patients in ARU (Rosario et al., 2014). When using the MFS in an ARU setting,

75%-90% of patients are identified as high fall risk (Rosario et al., 2014). On average, 80% of patients in TJH ARU are identified as high fall risk patients on the MFS, making it difficult to identify the patients who truly classify as high fall risk. Without an accurate assessment of a patient’s risk for falling, it is challenging to determine appropriate prevention interventions.

High fall risk patients in the ARU setting receive a multitude of interventions, such as: non-skid footwear, bed in low position, hourly rounding or more frequently, medication 9 HIGH FALL RISK IN ARU assessment to educate patient when certain medications could affect safe mobility, use of nurse call prior to getting out of bed, bed and chair alarms, fall arm band, visible indicator outside of patient room, stay with me protocol while toileting or ambulating, fall agreement, and use of gait belt. With 80% of patients receiving all of these interventions, it is hard to differentiate the true high fall risk patients. When the aforementioned interventions are not enough, the use of

AvaSYS cameras or a patient safety companion /sitter can be utilized. The ARU also uses a 3- alarm huddle system for when a patient sets the alarm off more than 3 times in a shift. This process pushes the staff together to huddle, and discuss what interventions are in place and what needed interventions should take place.

With 80% of the patients being high fall risk in the ARU, various interventions are put into place. One intervention that is likely overused is bed/chair alarms. Since most patients have bed/chair alarms, any time an alarm sounds, everyone on the unit goes running. With the sensitivity of bed alarms and the amount of bed alarms that go off when a patient is repositioned or moves in his/her bed resulting in false alarms, a major concern has increased around alarm fatigue or desensitization. Because of this, it can be questioned if team members allow the bed alarm to sound without immediate response. This could then lead to high-risk patients possibly being out of bed and on the floor before someone responds. If patients’ risk for falls were more accurately identified, this would allow nursing staff to use the high-risk interventions only on those patients who are truly high risk, thus decreasing the use of bed alarms and alarm fatigue/desensitization of the nursing staff.

The cost of falls in healthcare continues to grow, representing unnecessary burden on the

US healthcare system (Christopher et al., 2014). Additional long-term consequences to the patients who have fallen include but are not limited to anxiety, depression, and increased dependence on others loss of work, and reduced quality of life (Kwan, Kaplan, Hudson- 10 HIGH FALL RISK IN ARU McKinney, Redman-Bently, & Rosario, 2012). In 2000, the annual cost of falls was at $19 billion and is projected to be more than $50 billion in 2020 (Kwan et al., 2012). Health care costs for a person age 72 or above for one fall is $19,440 (Kwan et al., 2012). Centers for

Medicare and Medicaid no longer reimburse hospitals for preventable fall related injuries that occur in the healthcare setting, thus the cost of falls truly impacts hospital budgets (Christopher et al., 2014).

The need for a tool that identifies high fall risk patients in the ARU setting is required so that appropriate nursing prevention interventions can be put into place, leading to more focused fall prevention. Researchers have identified that one tool cannot be used across all settings

(Rosario et al., 2014). “The optimal approach to fall predication in inpatient rehabilitation is not clear” (Vratsistas-Curto, Tiedemann, Treacy, Lord, & Sherrington, 2018, p. 216). No fall prediction tools have been validated in the rehab setting, although the use of a screening tool is suggested (Vratsistas-Curto et al., 2018). The EPFRAT, although indicated for acute psychiatric patients, evaluates age, mental status, elimination, medications, diagnosis, ambulation/balance, nutrition, sleep disturbance and history of falls, (Edmonson, Robinson, & Hughes, 2011), which are the same risk factors that contribute to ARU falls. Understanding the risk factors for the ARU population “is essential to being able to accurately predict patient falls” (Rosario et al., 2014, p.

87). Current use of the MFS reflects mostly all ARU patients are high risk, causing difficulty in identifying truly high-risk patients to appropriately implement additional nursing prevention interventions.

PICOT Question

In patients admitted to acute rehab (P), how does utilization of the Edmonson Psychiatric

Fall Risk Assessment Tool (I) compared to utilization of the Morse Fall Scale (C) affect falls per

1,000 patient days (O) while hospitalized (T)? 11 HIGH FALL RISK IN ARU Literature Review

An exhaustive literature search was completed with focus specific to the MFS and the

EPFRAT to assist in narrowing the focus further, as falls is a broad topic yielding 152,142 articles. Study design was also used to limit data to higher-level evidence. To assure that evidence was not missed, the keyword search, subject-heading search, and title search strategies were all utilized (Melnyk & Fineout-Overholt, 2015), across multiple databases. In addition, collaboration with The Jewish Hospital Librarian and the Librarian at Mount St. Joseph

University occurred to assure a comprehensive review was achieved. The EPFRAT tool is cited in 18 different published articles. The use of Mount St. Joseph University evidence-based care resource was used and this resulted in no additional articles that met criteria of the search.

Search Description

The search was limited to literature published in English, over the last 10 years; however, it was discovered during the search that the original study completed in 1989, by Morse, Morse and Tylko, is a landmark study and continues to add value and was included. The search was not limited to gender or age, but articles specific to the pediatric population were excluded.

The following databases were searched: The Jewish Hospital virtual health science library, PubMed Medline, EBSCO CINAHL Plus with Full-Text, EBSCO MEDLINE Complete with Full-Text, The Cochrane Database of Systematic Reviews, and OVID Joanna Briggs

Evidence Based Practice Database. Using Mount St. Joseph virtual library, EBSCO Psychology and Behavior Science Collection was reviewed. Both Google and Google Scholar were also searched. The following key search terms were used, Fall/Falls, Fall Risk, Fall Risk Assessment,

Fall Risk Assessment Tool, Fall Risk Tool, Fall Scale, Fall Prevention, Fall Prevention Hospital,

MORSE, MORSE Fall, MORSE Fall Risk Assessment, MORSE Fall Risk Assessment Tool,

Edmonson, Edmonson Fall, Edmonson Fall Risk, Edmonson Psychiatric Fall Risk Assessment 12 HIGH FALL RISK IN ARU Tool, Psychiatric Fall, Psychiatric Fall Risk, Psychiatric Fall Risk Assessment, EPFRAT

Validity. Several articles were duplicative across the databases. Over 300 articles were reviewed, first reviewed by titles and those that did not pertain were eliminated. Next, abstracts were reviewed and irrelevant articles were eliminated. Table 1 contains specific databases, search terms, and number of results.

Table 1 Literature Review

Search Term

e

NE

PubMed Medline EBSCO CINAHL EBSCO MEDL OVID Joanna Briggs EBSCO Cochran EBSCO Psych Google Google Scholar Fall/Falls 152142 35532 56094 645 3807 7166 Fall Risk 20560 4218 2572 51 506 580 Fall Risk Assessment 5286 454 371 16 7 57 Fall Risk Assessment 533 205 53 6 3 9 Tool Fall Risk Tool 6 9 6 51 3 15 Fall Scale 67 175 67 1957 2 106 Fall Prevention 2023 8095 2023 49 312 425 Fall Prevention Hospital 0 1 0 279 0 11 MORSE 8672 383 8664 33 328 421 MORSE Fall 70 164 70 669 2 3 MORSE Fall Risk 50 0 42 67 Assessment MORSE Fall Risk 20 0 34 68 Assessment Tool Edmonson 44684 997 44567 1 0 1456 Edmonson Fall 5 0 0 1 0 Edmonson Fall Risk 1 0 1 2 Edmonson Psychiatric 1 0 1 1 0 4 26300 4650 Fall Risk Assessment Tool Psychiatric Fall Risk 1 0 1 830 941 Psychiatric Fall 1 1 1 1282 17 Psychiatric Fall Risk 1 0 803 1200 Assessment EPFRAT validity 49800 4110

13 HIGH FALL RISK IN ARU Articles were eliminated if they did not focus on fall risk tools specific to MFS and

EPFRAT and if data were not specific to the ARU setting. Many articles were eliminated due to focus on prevention strategies and not tools. The remaining 36 suitable articles were reviewed in their entirety resulting in ultimately 11 articles identified that supported the evaluation of the

EPFRAT tool in the ARU setting to distinguish high fall risk patients. These articles were then retained for hierarchy of evidence grading.

Summary of Levels of Evidence

Of the 11 articles retained, 18% (N=2) of the articles supported level I b evidence, and both articles were systematic reviews of nonrandomized trails. Thirty-seven percent (N=4) were single correlational or observational studies, or IV level of evidence. Eighteen percent (N=2) of the studies graded were level V, with one study being comparison and the other an evidence study. Level VI also revealed 18% (N=2) containing one comparative cross-sectional study and one qualitative quality improvement study. Finally, 9% (N=1) was commentary from Agency for Healthcare Research and Quality and contained valuable information related to the role nursing professional’s play in fall reduction. Table 2 contains the hierarchy of evidence graded during this robust literate review.

Table 2 Level of Evidence

Type of Evidence Level of Number of Percentage Evidence Articles of Level with Level (N=11) (N=11) a. Systematic review of randomized control trial I b 2 18% or b. Systematic review of nonrandomized trails a. Single randomized control trial II 0 0 or b. Single nonrandomized trail System review of correlational/observational III 0 0 studies Single correlational /observational study IV 4 37% 14 HIGH FALL RISK IN ARU Systematic review of V 2 18% descriptive/qualitative/physiologic studies Single descriptive/qualitative/physiologic study VI 2 18% Opinion of authorities / Expert committees VII 1 9%

Falls

Falls and falls with injury continue to be a significant issue and burden for both patients and healthcare systems. Falls that occur during rehab stays are complex and often related to the promotion of increased independence (Vratsistas-Curto et al., 2018). A patient fall is defined by

NDNQI as “a sudden, unintentional dissent, with or without injury to the patient, that results in the patient coming to rest on the floor, on or against some other surface, on another person, or on an object” (NDNQI, 2016, p. 2). “Falls and related injuries are the most frequently reported adverse events in the hospital setting” (Ambutas et al., 2017, p. 175).

Falls per 1,000 patient days is currently the national standard of reporting fall rates and helps to normalize data for comparison. Fall rates are the best way to facilitate comparison of falls between hospitals of different sizes (Oliver, Healey, & Haines, 2010). This calculation is simply the number of falls on the unit divided by the number of patient days, multiplied by

1,000. NDNQI and Agency for Healthcare and Research Quality both recommend falls per

1,000 days for consistent comparison and measurement of fall rates nationally and within the organization (NDNQI, 2016).

Fall Risk

According to NDNQI (2016), fall risk is established by individual facilities and based on screening processes or assessment tools. Use of a screening tool is recommended in the literature, though one tool has not been identified as superior to predict falls in the rehab setting.

At TJH, the MFS is currently used hospital wide regardless of the setting. The MFS results in assigning a number between 0 and 125, which indicates one of three fall risk categories. High 15 HIGH FALL RISK IN ARU fall risk is denoted in a score above 45. Moderate fall risk is a score between 25 and 44, and low fall risk is patients scoring 0-24. The aforementioned interventions are assigned based on the fall risk score. Screening tools and prevention strategies are key to preventing falls to at-risk patients, because they can identify the appropriate preventive interventions. Preventing falls in the hospital setting is a high priority at TJH and a quality improvement goal across healthcare organizations (Clancy, 2013). “The increased risk for falls in an acute rehabilitation setting results from the active promotion of fall risk patients allowing patient mobility in independence”

(Rabadi et al., 2008). Fall risk assessment tools are intended to identify high fall risk patients so appropriate interventions can be put in place to decrease the risk of patients falling (Ambutas et al., 2017).

Historically, there are four key components of fall and injury prevention programs, namely implementation of a safe environment, identification of modifiable factors, implementation of interventions targeting those modifiable risk factors to prevent falls, and reduction of risk of injury for those patients who fall (Ambutas et al., 2017). Use of a fall assessment tool allows healthcare professional the ability to measure fall risk factors (Abraham,

2016). It is valuable to measure both the sensitivity and specificity of fall assessment tools to help determine which tool is used for the specialty (Abraham, 2016). Sensitivity testing looks at the ability of the scale to accurately identify who is a fall risk, while specificity is the ability of the scale to identify who is not a fall risk (Abraham, 2016). The sensitivity of a test is its ability to detect the condition when it is present, and the specificity of a test is its ability to exclude the condition in patients without the condition. The overall purpose of a fall risk assessment tool is to measure the appropriate risk factors of the population in order to determine likelihood of fall.

Because of this, it is critical that fall risk tools use the appropriate risk factors for each population. 16 HIGH FALL RISK IN ARU Acute Rehab Risk Factors for Falls

This project focuses on the fall risk of patients on an acute rehab unit. Numerous risk factors lead to patient falls in the ARU setting. Rabadi et al., (2008) identify that the “active promotion of patient mobility and independence” alone lead to increased risk of falls, not even taking into considerations other risk factors (p.104). With 60% of ARU patients having a neurological diagnosis, many risk factors surround stroke patients including, “age, impulsivity, cognitive impairment, severity of neurological impairments, poor balance, urinary incontinence, and sedating and psychotropic drugs” (Rabadi et al., 2008). Rosario et al., (2014) note risk factors in ARU to include “stroke, amputee, cognitive impairment, previous falls, sleep disturbances, medications including tranquilizers, anticonvulsants, and in antihypertensive, advancing age, vertigo, physical impairments, urinary incontinence, and visual and /or hearing impairment” (p.90).

Oliver, Healey, and Haines (2010) evaluated fall data from 1999-2009 and broke the information down in four types of preventions trials, namely design and results of multifactorial fall preventions trials, components of multifactorial fall prevention trials, application of multifactorial fall preventions trials, and singles and dual interventions studies. The study concluded that fall prevention is complex and multiple interventions are necessary to impact outcomes by a multidisciplinary team for best results (Oliver et al., 2010). Rehabilitation units tend to have much higher rates of falls (Oliver et al., 2010). Oliver, Healey, and Haines (2010) note that the MFS is one of the most widely validated tools that has been identified on systematic review; however, organizations using fall prediction tools should test the tool on each population to evaluate the effectiveness of the tool. This study is also one of the most widely cited fall studies being cited 286 times. The EPRFAT assessment was not addressed in this systematic review. 17 HIGH FALL RISK IN ARU Psychiatric Risk Factors for Falls

Because the fall risk assessment tool proposed in this project is primarily used on psychiatric units, risk factors for falls in psychiatric patients will be addressed. Psychiatric patients who fall have a unique set of risk factors that differ from general medical surgical patients, but do have commonalities with many patients in the ARU setting. Edmondson et al.

(2011), found that in the case of the psychiatric patient’s age, specifically younger patients with a mean age of 56.3, fell, which is similar to Rosario et al. (2014), who reports that rehab patients

41-50 years of age are the highest risk. In both the ARU and psychiatric setting, younger patients ranging from 41-53 are more likely to fall than when compared to general medical surgical units, who report those older than 80 are at greatest risk to fall (Edmonson et al., 2011).

Like ARU patients, psychiatric patients are encouraged to be active and participate in activities (Edmonson et al., 2011). Medications and the related side effects play a role and increase the risk of falling for psychiatric patients. These medications include sedatives, antidepressants, and antipsychotic agents, all of which may cause side effects that effect balance, such as dizziness or decreased alertness (Edmonson et al., 2011). Poor nutrition and fluid intake, including loss of appetite, poor eating habits and lack of fluid intake, also contribute to risk of falls for psychiatric patients (Edmonson et al., 2011). In addition, sleep loss is a contributing factor with falls, leading to accidental slips and also impaired awareness of the environment

(Edmonson et al., 2011). Note that many of these fall risk factors in psychiatric patients are the same ARU fall risk factors cited by Rosario and Rabadi (2014). It was verified, that although the

EPFRAT tool is widely utilized, it is not a validated tool, however, superior to the MFS and more sensitive in the psychiatric setting (Edmonson et al., 2011).

18 HIGH FALL RISK IN ARU Morse Fall Scale

The current fall risk assessment tool used on the ARU at TJH is the Morse Fall Scale.

(Chapman, Bachand, & Hyrkäs, 2011). Chapman et al. (2011), completed a descriptive comparison cross sectional study testing four fall risk assessment tools simultaneously in 17 units. Testing evaluated the sensitivity, specificity, and feasibility of four fall risk assessment tools in a clinical setting (Morse, Hendrich II, The New York-Presbyterian Fall and Injury Risk

Assessment Tool, and The Main Medical Center Falls Risk /Interventions) to determine the best tool. Data was collected in May and June 2006 on 17 units (5 surgical, 5 medical, 5 critical care,

1 oncology, 1 birth center), and 1,546 assessments in total were collected. Results indicated sensitivity of MFS 77.2%, and specificity was highest with MFS at 72.8% (Chapman et al.,

2011). Based on the sample, 77% of patients with a medium to high score would be identified with the MFS (Chapman et al., 2011). The use of the MFS is supported with sensitivity of

77.2%, however, the population evaluated was not rehab based.

Morse, Morse, and Tylko (1989) conducted a single correlations case control study. This landmark study continues to be referred to in the literature and is cited 331 times. Morse et al. reported on the development of the MFS to identify patients at risk for falling so prevention strategies could be targeted. This research was conducted in a 1,200 bed urban hospital with a fall rate of 2.5 falls per 1,000 patient days. The study consisted of 100 patients who had fallen and 100 randomly selected patients who had not fallen. Results of this study identified the sensitivity of scale was 78%, positive predictive value 10.3%, specificity was 83%, negative predictive value 99.3%, and interrater reliability score of r = .96. Ultimately, this scale permits identification of the patients at risk for falling so prevention strategies may be targeted; however, it will not predict accidental falls (Morse et al., 1989). While this study supports use of the MFS 19 HIGH FALL RISK IN ARU tool, it was completed in the acute care hospital setting and not specific to the rehab setting.

Table 3 depicts the actual MFS including scores and parameters for the three fall risk categories.

Table 3 Morse Fall Scale©

Risk Factor Scale Score History of Falls Yes 25 (Last 30 days) No 0 Secondary Diagnosis Yes 15 (more than one health issue) No 0 Ambulatory Aide Furniture 30 (uses prior to or during admission) Crutches / Cane / Walker 15 None / Bed Rest/ Nurse assist 0 IV access Yes (intermittent or continuous infusion) 20 No 0 Gait / Transferring Impaired 20 Weak 10 Normal / Bedrest / Immobile(non-ambulatory) 0 Mental Status Forgets Limitations 15 Oriented to Own Ability 0 To obtain the Morse Fall Score, total the scores from the 6 categories above: High Risk = 45 and Higher Moderate Risk = 25-44 Low Risk = 0-24 Morse et al., 1989 Edmonson Psychiatric Fall Risk Assessment Tool (EPFRAT)

Edmonson et al. (2011), designed a qualitative study that began as a quality improvement study in development of a fall risk assessment tool for the inpatient psychiatric population. Nine risk factors were identified and the tool was retrospectively applied to 138 patient records. The

EPFRAT tool was implemented on two inpatient psychiatric units in a 350-bed hospital, located in central Illinois. The MFS and the EPFRAT tools were simultaneously administered to inpatient psychiatric patients. “Sensitivity of the EPFRAT was 0.63, compared with 0.49 for the

MFS; specificity of the EPFRAT was 0.86, compared with 0.85 for the MFS” (Edmonson et al.,

2011, p. 29). Results indicated the EPFRAT tool was more sensitive than the MFS. Edmonson et al. note additional psychometric testing is needed to determine the reliability and validity of 20 HIGH FALL RISK IN ARU the EPFRAT. Sensitivity is the ability of the tool to positively diagnose a condition. Given the results of this study, the EPFRAT was able to identify truly high risk fall patients 63% of the time, compared to the MFS which identified truly high risk fall patients only 49% of the time.

Sensitivity is the ability of the tool to rule out a condition. While the EPFRAT was only slightly higher than the MFS, it was able to rule out high risk fall patients 86% of the time.

Although the EPFRAT tool is not a validated tool, both Abraham (2016) and Slade

(2017) identified that this tool should be used in the psychiatric setting and addressed that tools overall should be specific for and tested in the populations identified. Abraham, in a comparison study including seven different studies, identified the EPFRAT tool as a best practice for the psychiatric setting. Abraham states, “the use of some fall assessment tools may place large percentages of individuals at high risk, limiting the opportunity for increased intervention for actual high-risk patients” (2010, p. 3). Within this statement lies the heart of the problem in which this DNP project aims to address. Slade, in an evidence summary regarding screening tools for mental health settings, also identifies the EPFRAT tool as a best practice and recommends its use in in-patient mental health settings (Slade, 2017).

Although the EPFRAT tool is indicated for acute psychiatric patients, it may also be an appropriate tool for ARU patients because ARU and psychiatric patients have similar risk factors that this tool addresses. The tool evaluates age, mental status, elimination, medications, diagnosis, ambulation/balance, nutrition, sleep disturbance and history of falls, which in fact are also the same risk factors that contribute to ARU falls (Edmonson et al.,

2011). Utilization of the EPFRAT tool in the ARU setting can assist in identifying high fall risk patients, guiding nurses to implement additional fall prevention measures. In the psychiatric setting, when compared to the MFS, the EPFRAT tool was more sensitive (0.63) when to compared to the MFS (0.49) (Edmonson et al., 2011). Table 4 contains the EPFRAT 21 HIGH FALL RISK IN ARU assessment. Those scoring greater than 90 using this tool are considered high fall risk.

Identification of risk factors and understanding specific factors is paramount in being able to accurately predict patient falls in each population (Rosario et al., 2014).

In the ARU setting, age, elimination, nutrition, medication, and sleep, are all things that are discussed, and reviewed daily and are important components in patient care. The aforementioned components of patient care are also risk factors that contribute to falls. One example of how the EPFRAT tool could have assisted the nursing staff in implementing additional interventions is noted from a previous ARU patient who suffered a fall with injury.

A 67yo female patient was admitted to the ARU after a large left intraparenchymal hemorrhage. On admission, patient was lethargic, had episodes of disorientation, was not initiating with therapy, and had poor nutrition. The patient was started on neuro stimulants and fell a few days later, suffering a medial orbital wall fracture. Although the MFS was high and the patient had high fall precautions in place like other patients in the ARU, the other triggers of nutrition, sleep disturbance, elimination, and medication could have been evaluated into her overall fall risk. The nurse could have implemented additional interventions with EPFRAT score of 98, as the stimulants peeked, potentially preventing patient fall and injury. In this situation, use of AvaSYS camera or a safety companion could have prevented this fall with the additional triggers noted in the EPFRAT assessment. Table

4 depicts the EPFRAT fall risk assessment tool.

Table 4 Edmonson Psychiatric Fall Risk Assessment ©

Age 8 Less than 50 10 50-79 26 80-over Mental Status -4 Fully Alert/Oriented at all times 12 Agitation/Anxiety 22 HIGH FALL RISK IN ARU 13 Intermittently confused 14 Confused/Disorientation Elimination 8 Independent with control of bowel/bladder 12 Catheter/Ostomy 10 Elimination with Assist 12 Altered elimination (incontinence, nocturia, frequency) 12 Incontinent but Ambulates Independently Medication 10 No Medication 10 Cardiac Medication 8 Psychotropic Medications (including benzodiazepines and antidepressants) OR 12 Increase in these medications and/or PRN (psych, pain medication received in the last 24 hours Diagnosis 10 Bipolar/Schizoaffective Disorder 8 Substance abuse / Alcohol abuse 10 Major Depressive Disorder 12 Dementia/ Delirium Ambulation/Balance 7 Independent/Steady gate / Immobile 8 Proper Use of assistive Devices (cans, walker, w/c) 10 Vertigo/Orthostatic Hypotension/Weakness 8 Unsteady/Requires Assist and Aware of Abilities 15 Unsteady but Forgets Limitations Nutrition 12 Has had very little food or fluids in the past 24 hours 0 No apparent abnormalities with appetite Sleep Disturbance 8 No Sleep Disturbance 12 Report of Sleep Disturbance or patient, family, or staff History of Falls 8 No History of Falls 14 History of Falls in the last 3 months FALL Risk = Score of 90 or greater Edmonson et al., 2011

Comparison of Edmonson and Morse Fall Scales

Table 5 compares both the EPFRAT and MFS identifying which fall risk factors are addressed in each assessment tool. The MFS does not address age, elimination, medications, nutrition, and sleep. These aforementioned factors are critical elements with both psychiatric and ARU patients. The EPFRAT tool does not address IV access; however, this is not a critical element with ARU patients because most patients on the ARU do not have IV access. Once 23 HIGH FALL RISK IN ARU patients are identified by the nurse as high fall risk, then the appropriate interventions can be implemented. The MFS identifies on average 80% of patients as high fall risk in the ARU at

TJH. Because most patients are noted as high fall risk, it’s challenging to identify when extra interventions should be put into place. Utilization of the EPFRAT tool in the ARU may more appropriately identify the truly high fall risk patients since it more specifically assesses fall risk factors than the MFS in this setting, thus leading to appropriate interventions for the truly high fall risk patients, such as the addition of an AvaSYS camera or a safety companion.

Table 5 Comparison Fall Risk Factor of Edmonson Psychiatrics Fall Risk and Morse Fall Risk

Fall Risk Factor Edmonson Morse

Age Yes No Mental Status Yes Yes Elimination Yes No Medication Yes No Diagnosis Yes Yes Ambulation/Balance/Gait Yes Yes Nutrition Yes No Sleep Yes No History of Falls Yes Yes IV Access No Yes (Edmonson et al., 2011; Morse et al., 1989)

Summary of Literature

Due to the complexity of falls and the fact that you cannot do randomized testing to cause someone to experience a fall, most fall research is retrospective and non-randomized. The majority of systematic reviews are specific to fall interventions but the variety of interventions, lack of standardized definition and reporting of falls, multiple tools, risk factors, and variation makes high level research challenging. The literature search revealed variation in levels of evidence supporting the evaluation of piloting the EPFRAT tool to identify high fall risk patients in the ARU. The literature, however, clearly supports evaluation of current practices and 24 HIGH FALL RISK IN ARU opportunity for improved patient safety specific to fall reduction in rehab, as well as supporting the idea that not all fall risk assessment tools are effective for all units.

Evidence-Based Practice Model

When implementing evidence-based practice (EBP), utilization of a model to guide practice is essential and can assist with systematically changing complex and challenging practices (Melnyk & Fineout-Overholt, 2015). Use of an EBP model can help to organize thinking, project approach, and the intended tactics, resulting in strengthened decision making and successful practice change implementation. In addition, use of an EBP provides sequential steps in planning, implementing, and evaluating a project and truly sustaining an EBP. EBP models “facilitate the implementation of research findings into nursing practice” (Schaffer,

Sandau, & Diedrick, 2012, p. 1197).

The model chosen to guide the implementation of this project is the Johns Hopkins

Nursing Evidence-Based Practice (JHNEBP) model. This model is a comprehensive model that addresses the critical aspects of the evidence-based practice process (Schaffer et al., 2012). The goal of EBP “is to promote effective nursing interventions, efficient care, and improve outcomes for patients, and to provide the best available evidence for clinical, administrative, and educational decision making” (Newhouse, Dearholt, Poe, Pugh, & White, 2007). According to

Schaffer et al., “the major focus of the model is a translation of best evidence for nurses at the bedside to using care decisions” (Schaffer et al., 2012, p. 1204). The JHNEBP model contains 3 concepts with 18 total steps: “(1) identification of the practice questions, using a team approach;

(2) collection of the evidence, which involves searching, critiquing, summarizing, determining strength of evidence, and making recommendations; and (3) translation of the evidence for use in practice, which includes determining feasibility of adopting the change in creating an action plan for implementation” (Schaffer et al., 2012, p. 1204). Use of this model allows for nurses at the 25 HIGH FALL RISK IN ARU bedside to translate evidence to clinical, administrative, and educational nursing practice

(Melnyk & Fineout-Overholt, 2015). Engagement of nurses who are closest to the patient at the bedside is a critical component when implementing EBP.

The JHNEBP model and process was chosen because it is practical (Newhouse et al.,

2007) and adaptable for organizations, and most importantly, for bedside nurses to implement

EBP. Utilization of this tool allows nurses at all levels to systematically and successfully complete EBP projects. This rigorous model is the EBP model used by Mercy Health, where this project will occur. In addition to the tools that are provided by the model, other tools have been created in support of utilizing a stair step approach to encourage nurses to participate in evidence-based practice and research. These tools include a proposal outline, contents and committee procedures for project, proposal application, content question development, literature summary reference tool, project management tool, and research committee approval guidelines for Mercy Health. This tool was also chosen based on practicality and applicability, to the aforementioned PICOT question. Furthermore, this model considers internal and external factors, such as culture and quality measures, which are necessary considerations when implementing change.

The first concept in the JHNEBP model is identifying the practice question. The EBP question posed is, how do we identify high fall risk patients in the ARU setting? This step also includes holding an interdisciplinary team meeting consisting of those who contribute to caring for patients in the ARU (Newhouse et al., 2007). The second concept is centered around evidence. This includes evidence collected both at the point of patient contact and within the literature (Newhouse et al., 2007). Currently, the MFS identifies 75-90% of patients as high fall risk, making it difficult to identify those patients who are truly high fall risk (Rosario et al.,

2014). The final concept is translation. This includes reviewing the recommendations identified 26 HIGH FALL RISK IN ARU in the evidence stage and transferring to the identified setting. This step also includes educating and trialing of the EPFRAT tool, evaluating outcomes, and sharing findings.

Theoretical Framework

Use of theoretical framework helps to provide structure and guide projects. “Successful change initiatives use change theory or a planned approach to implement organizational shifts”

(Shirey, 2013, p. 69). Application of theoretical framework helps guide next steps and serves as an operational roadmap for projects.

Utilization of Lewin’s change theory, as a foundation for implementing planned change, allows for a structured approach rather than randomly implementing change. Use of a change theory framework also increases the likelihood of success (Mitchell, 2013). Furthermore, utilizing a practical approach is also necessary when planning change. Lewin’s change theory is simple and easy to understand and ultimately focuses on process and utilization of change frameworks, which can proactively help to eliminate potential problems. Initial utilization of a tool when focusing on planning and strategizing is a critical step in change implementation.

Lewin’s change theory is a three-step approach. The first step is unfreezing when change is needed. This step involves getting ready for the change and is usually initiated when a nurse leader recognizes a problem, such as falls in ARU. Step two is moving or transitioning when change is initiated. This step, focuses on change as a process rather than an event. The third step is refreezing, when equilibrium is established.

Lewin’s framework was chosen because applicability is simple since the theory is easy to apply and helps to simplify the process, therefore helping those affected by the change. In the end, those affected may be more receptive to the change and easily see the desired results.

Strengths of Lewin’s theory include versatility, practicality, simplicity, and ease of understanding (Shirey, 2013). Simplicity of approach can also be viewed as a potential 27 HIGH FALL RISK IN ARU limitation of Lewin’s theory. In addition, the theory is linear and since change is often complex and unpredictable during the refreezing stage, effective evaluation of the change is necessary.

The first step of unfreezing includes recognition of the problem, which in this case is high fall rates in ARU. This step includes creating sense of urgency for change; humanizing each fall in ARU will be one tactic used. Humanizing harms includes speaking about patients by name and telling the story of the patient. Creating a sense of urgency and getting ready for the change to occur. Information was shared with ARU staff regarding national statistics of falls and then unit level statistics. Humanizing each fall requires an in-depth debriefing process after each fall.

Once the debrief is completed with those involved with the incident, the information is then shared with other ARU staff and learning’s are spread across the hospital. With each fall, the patient’s last name and story, including learning’s and also impact of the fall resulting in additional length of stay or resulting in harm, are shared. For example, Ms. Smith is a 72yo, admitted with stroke and left sided weakness; she enjoys gardening and spending time with her

10 grandchildren. Due to her fall, she required three additional days in the hospital and missed her granddaughter’s first birthday party. Discussion with staff will occur on how high fall risk patients are identified in the ARU. Humanizing falls assists in unfreezing by “deliberately creating an emotional stir-up” (Lewin, 1947, p. 211).

The second step is moving this process and includes education, planning, and implementing changes. In this step, the change is looked at as a process rather than an event

(Shirey, 2013). This step includes transitioning to a new way of thinking regarding falls in the

ARU. Education will occur regarding the EPFRAT tool, including supporting evidence for practice change. A robust plan was shared with the ARU staff on how this new tool will be implemented with time for staff feedback to assure implementation is successful. Planning is necessary to assure project objectives are successfully met (Lewin, 1947). Frequent monitoring 28 HIGH FALL RISK IN ARU of the project with staff feedback and making swift necessary changes will be critical. During this step, a three-month trial will occur using the EPFRAT tool.

The third and final step of refreezing incorporates the new way of thinking about falls in the ARU, hardwiring the process of using the new EPFRAT tool to identify high fall risk, and celebrating success. Outcomes of utilization of the tool and impact to the ARU will be shared.

Celebration of successful implementation contributing to the overall goal of decreasing falls in the ARU will then occur.

Integrated Evidence and Theoretical Framework

Of the chosen EBP model and theoretical framework models, both the JHNEBP and

Lewin’s Theory of Planned Change include a complementary three-step approach. The first stage of unfreezing occurs and practice questions are identified. In this stage, it is identified that the ARU experienced a fall rate of 4.67 falls per 1,000 patient days in 2017, humanization of each fall occurred during team meetings, and identification of the EBP question of how to identify high fall risk patients in ARU occurs. The second step of moving and evidence includes educating, planning implementation, and also the critical step of evaluating the current evidence through an exhaustive literature search. The last stage is refreezing and translation. This step includes implementing a new way of thinking, educating as appropriate, hardwiring the new practice, evaluating the outcomes, sharing findings, and celebrating success of implemented practice change.

Utilization of the JHNEBP model and Lewin’s theoretical framework will assist in guiding the proposed practice change piloting the EPFRAT tool in the ARU. Organized sequential steps will support planning, implementation, and evaluation of this project. It is clear that falls are a substantial issue facing the ARU and admitted patients. Choosing a tool that helps identify high fall risk patients in the ARU can result in decreasing falls in this at- 29 HIGH FALL RISK IN ARU risk population. Table 6 contains an Integrated Evidence-based Practice Model and

Theoretical Framework.

Table 6 Integrated Evidence-based Practice Model and Theoretical Framework

Project Proposal

The purpose of this project is to identify how utilization of the EPFRAT tool compares to utilization of the MFS on falls per 1,000 patient days while patients are hospitalized in the ARU.

The goal of this practice change is to improve patient outcomes by decreasing patient falls in the

ARU through proper identification of high fall risk patients and implementing the correct nursing interventions to prevent falls. The institutional review board (IRB) at Mount St. Joseph

University was consulted. The IRB determined that due to the nature of this practice change,

IRB approval was not necessary.

Utilization of the EPFRAT tool in the ARU setting can assist in identifying high fall risk patients, in turn guiding nurses to implement additional prevention measures, ultimately decreasing falls and falls with injury in this high-risk population. ARU patients are identified as 30 HIGH FALL RISK IN ARU one of the most at-risk populations for high fall risk (Rosario et al., 2014); therefore, it is imperative to identify those high-risk patients in order to proactively enact sufficient fall prevention interventions.

Participants

The unit identified is a 14 bed adult ARU. This unit admits patients 18 years and older. In 2017, the average age of admitted patients was 69.9 and ages ranged from 20-100.

The average length of stay (LOS) is 11.9 days on this unit. More than 60% of the patients admitted to the ARU have a neurological related diagnosis, with the most common admission diagnoses being stroke, cerebrovascular accidents, transient ischemic attacks, brain injuries, brain mass, debility, oncology, amputations, fractures, and other diagnosis. Average daily census on the ARU in 2017 was eleven.

During the first quarter of 2018, there were 106 patients with an average LOS of 10.29 days, and 102 patients in the fourth quarter with an average LOS of 9.89 days. The mean age during the first quarter of 2018 was 72.01, with ages ranging from 37-99 years of age. The mean MFS on admission during the first quarter was 69.01 and 61.32 on discharge (Appendix

A). Of the 106 patients in the first quarter, 90% were high fall risks on admission using the

MFS. In the fourth quarter, the mean MFS on admission was 65.16 and on discharge was

53.38. Any patient with a score above 45 with the MFS is considered a high fall risk. The mean EPFRAT on admission during the fourth quarter was 67.57 and 75.02 on discharge

(Appendix B). Patients with a EPFRAT score above 90 is considered a high fall risk. The

EPFRAT score during the fourth quarter indicated 14% of patients were high fall risk while the MFS indicated 88% of the patients were high fall risk, making it difficult to differentiate the truly high fall risk patients. In the fourth quarter of 2018, the mean age was 66.89 with ages ranging from 21-93. Gender was nearly equal in both quarters. In quarter one, 49.93% 31 HIGH FALL RISK IN ARU were female and 50.07% were male, compared to quarter four 49.95% female and 50.05% male (Appendix C). Neurological related diagnosis comprised 61% of the patent diagnosis in the first quarter and 57.8% in the fourth quarter (Appendix D).

Setting

This practice change will take place at The Jewish Hospital – Mercy Health, in the

ARU. The Jewish Hospital is a 200-bed acute care community hospital, located in Kenwood,

Ohio. The Jewish Hospital was the first Jewish hospital founded in the United States in 1850 to serve the Jewish Community in Cincinnati, welcoming and caring for people of all faiths.

The Jewish Hospital continues to care for people of all faiths, with a legacy of teaching.

The ARU unit at TJH is a 14-bed unit with all spacious private rooms. This unit was designed with welcoming earth tones and each patient room has large windows allowing for natural light. The unit was opened in December of 2015 after being fully renovated, costing

1.8 million dollars. The ARU includes a large physical therapy gym, a full kitchen, and laundry area. Patients admitted to the ARU receive a minimum of 3 hours of therapy daily including physical, occupational, and speech therapy, depending on patient need. All rooms are equipped with an AvaSYS camera for monitoring if necessary, and all beds on the unit have bed alarms with in the bed. Although the unit is spacious and great for promoting activity, and walking the “L” shape of the unit with the nurse’s station located in the middle can be geographically challenging. Based on the unit shape, rooms that are at the end of the hall can be difficult to get to quickly, and it can be problematic to identify which bed alarm is sounding due the spaciousness of the unit. The average nurse to patient ratio on the unit is

1:4, with the maximum nurse to patient ratio 1:6, and assignments are made with consideration of patient acuity.

Intervention 32 HIGH FALL RISK IN ARU The proposed intervention in the ARU is implementation of the EPFRAT tool for all patients to identify high fall risk patients, and to assist in identification of additional appropriate nursing interventions. Meetings will occur with unit level stakeholders to answer questions regarding the proposed practice change and to get buy-in.

Phase one of implementation included developing training tools for nursing staff and therapy support. A PowerPoint presentation was developed for the staff that include the following: an outline of the problem, PICOT question, driving forces supported by the literature to support change, reasoning behind EPFRAT tool selection, specific intervention with ARU staff expectations, comparison of MFS and EPFRAT, a case study, and project evaluation methodology (Appendix E). In addition to the PowerPoint presentation, the pdf available online developed by Edmonson for training with instruction for EPFRAT tool completion will be given to staff and reviewed for questions (Appendix F). The training packet includes step by step instructions on how to complete the assessment in EPIC

(Appendix G). The case study scenario used for training is from an actual ARU patient that fell with injury, and further supports use of EPFRAT tool. Patients, who have an EPFRAT score above 90, will be further assessed for need of AvaSYS cameras or potentially a sitter, as additional interventions of those who meet criteria as a MFS high fall risk.

During phase two of implementation, training (approximately 30 min) was provided to the ARU staff including Nurses, Physical, Occupational, and Speech therapists, during the mandatory unit meeting the month prior to implementation. The PowerPoint, case study, and

EFRAT tool information were reviewed in detail, and questions from attendees were answered. After this training session that 90% of staff attended, an email was sent to all of the ARU staff, including the supporting documents, to assure all staff had the information, including those who were unable to attend. 33 HIGH FALL RISK IN ARU During phase three of implementation, the ARU Director, Manager, Clinical

Coordinators, and Charge Nurses received an additional 30 minutes of training during the monthly Clinical Coordinator meeting. The purpose of this training was for them to be fully comfortable answering simple questions should they arise.

Patients who have an EPFRAT score above 90, will be further evaluated by individual nursing staff for appropriateness of adding an AvaSYS camera or potentially a sitter as additional interventions. If necessary, additional interventions are added based on individual patient circumstances and nursing judgment. This author assured training was complete and was available for any complex questions and to support staff throughout the project. In addition, supplementary information packets were shared with Physical, Occupational, and

Speech Therapy staff, since care on the ARU is multidisciplinary in nature. On several dates prior to and after the go-live, this author rounded on the unit to assure all questions were answered and to assure staff were supported. During the first week of go-live, audits were run in EPIC by this author to assure the tool was completed on each shift with 97% compliance. Daily reminders in huddle and flyers (Appendix H) were placed on the unit.

Implementation of EPFRAT tool practice change will apply to all patients admitted to the ARU at The Jewish Hospital during the fourth quarter of 2018 (October, November and

December). The MFS will continue to be assessed using the current Fall Risk policy. In addition, on admission to ARU and every twelve hours, nursing staff with the AM and PM assessments will complete an EPFRAT assessment. The electronic medical record, EPIC, will be used for documentation purposes. During training, staff was shown how to add this assessment into their workflow so that documentation is electronic, and for ease of use and tracking purposes. Once the project is implemented daily audits will be run and available in

EPIC for other leaders to view, to track compliance in completing an EPFRAT each shift, and 34 HIGH FALL RISK IN ARU implementing additional fall prevention interventions as appropriate.

Data Collection

Pre implementation. Pre implementation data collection will include recording the following for each admission in the ARU during the first quarter of 2018 (January 1, 2018 thru March 31, 2018) in an EXCEL spreadsheet: medical record number, sex, age, admission diagnosis, admission date, discharge date, length of stay, fall during stay, date of fall, more than one fall, time of fall, fall witnessed or un witnessed, reason for fall, use of sitter/camera at time of fall, MFS score on admission, MFS at discharge, and MFS at time of fall if occurred. Pre implementation quarter one data was collected and placed in the data collection tool. Data was collected utilizing a report of all ARU patients admitted during the first quarter and EPIC was used to obtain patient specific data.

Post implementation. Post implementation data collected during the fourth quarter of 2018 (October 1, 2018 thru December 31, 2018) included, previously mentioned data points and additional data points including EPFRAT score on admission, EPFRAT at discharge, EPFRAT at time of fall if occurred, and additional interventions due to EPFRAT.

Fourth quarter post implementation data was collected using the same report of patients discharged from the ARU and EPIC for patient specific data.

Stakeholders

According to Reavy, stakeholders are formal or informal members of the project team that have a vested interest in the outcome of the practice change (Reavy, 2016). For this project, formal stakeholders include a multi-disciplinary approach. The DNP student is responsible for designing, planning, and implementing this project. A variety of The Jewish

Hospital – Mercy Health Administrators have a vested interest to decrease falls in the ARU in order to assure patient safety and decrease patient harms. Decreasing falls impacts overall 35 HIGH FALL RISK IN ARU organizational cost, length of stay, patient injuries and risk of litigation, in addition to being patient centered. Additional stakeholders include the ARU team, leadership, nursing, support staff, therapies, and physician. The ARU team desires to provide high patient centered quality care to all patients who are cared for on the unit. ARU patients are directly affected and impacted by results of implemented practice change, and are also considered stakeholders benefiting from the practice change.

Driving and Restraining Forces

A change is needed in the ARU to accurately identify high fall risk patients. The current tool (MFS) identifies about 80% of ARU patients as high fall risk making it difficult to identify who the true high fall risk patients are. In 2017, the fall rate in ARU was 4.67 falls per 1,000 patient days, and 2018 falls spiked alarmingly high in January and February alarmingly high at 13.05 falls per 1,000 patient days. The 2018 fall rate per 1,000 patient days was 4.56, with no significant change from 2017. Decreasing the fall rate will decrease overall hospital cost, and improve length of stay associated with patient falls. According to

Butcher (2013), a patient who experiences a fall has an average length of stay increase of 6.27 days. In addition, decreasing patient falls will increase patient confidence and improve overall patient satisfaction.

Potential barriers to implementing the EPFRAT tool in the ARU include engagement from a multidisciplinary team, nursing compliance to completing an additional tool, and nurse buy-in. The ARU culture is also an opportunity for improvement due to the nature of the workflow with multidisciplinary teamwork and the application of teamwork to patient centered care. Staff engagement in the ARU has been minimal in the past, however, has recently improved. Another contributing factor to the high fall rate in ARU includes the physical layout of the unit, as the unit is set up in a large “L” shape, and the nurse station is 36 HIGH FALL RISK IN ARU not in a central location to all patients, which creates a lag in nurse response time.

Budget

The cost for implementing this practice change is minimal, as it supported by the organization. All staff training time will occur during mandatory monthly unit meetings, in which minimal training supplies will be provided. The cost of the statistician will also be paid for by the organization. Approximately 20 Nurses, 6 Patient Care Assistants, and 20

Therapy staff will require training.

Table 7 Project Budget

Training Item Budget Statistician – Joe Nolan Consultation Free Food for meeting to engage ARU unit level $125 stakeholders in practice change Training supplies: laminated cards with $150 EPFRAT tool for badges, reminders posters to complete tool in staff lounge and for computers Food for initial training for ARU staff $300 RN & Therapy (training will occur during mandatory units meeting and breakfast will be provided) Snacks for Super user training $50 Statistician – Joe Nolan ($50hr) to help run $150 statistics Graeter’s Ice Cream Social to share findings $200 with ARU staff and thank staff for participation Total $975

Timeline

Utilization of a milestone timeline identifies a clear sequence of events and helps to track progress along the way. A milestone timeline supports successful implementation, ensures key deliverables are achieved, and the project is completed in a timely fashion. Table

8 is a timeline for implementing the practice change to identify high fall risk patients in the

ARU utilizing the EPFRAT. 37 HIGH FALL RISK IN ARU Table 10 Project Timeline

Practice Change Milestone Start Date Completion Date Project approval January 25, 2018 April 5, 2018 Meet with Statistician and April 8, 2018 April 30, 2018 discuss development of data collection tool, and data points to be collected Complete and submit IRB May 1, 2018 June 15, 2018 application for MSJ Present project and get buy in July 1, 2018 August 1, 2018 from unit level stakeholders Design and Complete September 1, 2018 September 30, 2018 training and education with ARU staff  9/18 ARU staff training completed  9/24 ARU Leadership training completed

Complete Data collection September 1, 2018 September 30, 2018 tool Implement practice change October 1, 2018 December 31, 2018 Gather data January 1, 2019 January 31, 2019 Evaluate outcomes and February 1, 2019 March 31, 2019 analyze results Share findings and April 11, 2019 Ongoing disseminate results of practice change

Analysis and Results

The primary goal of this practice change was to decrease falls in the ARU. Primary data was assessed as fall rate per 1,000 patient days. Falls per 1,000 patient days is currently the national standard of reporting and helps to normalize data for comparison. Use of fall rates can help to identify if the implemented plan is improving outcomes of falls. Falls per 1,000 patient days, falls with injury per 1,000 patient days, and raw number of falls were recorded using the data collection spreadsheet. Although the measure of fall rates per 1,000 days is the best way of facilitating an assessment between hospitals of different sizes, it can also disguise the problem at hand (Oliver et al., 2010). 38 HIGH FALL RISK IN ARU Utilization of the EPFRAT in conjunction with the MFS resulted in a decrease of falls per

1,000 patient days, from 9.025 in the first quarter of 2018, to 1.947 in the fourth quarter of 2018.

In the first quarter of 2018, there were 10 falls on the ARU when only the MFS was used.

During the fourth quarter when both scales were used, there were only 2 falls. During the fourth quarter of 2018 when the EPFRAT was used, falls decreased 80%. Eight less patients experienced a fall with use of the EPFRAT tool, thus supporting use of the EPFRAT tool in the

ARU setting.

Table 8 Results of Falls Per 1,000 Patient Days Comparing First and Fourth Quarter

ARU FALLS 10 8 6 4

2 Falls per 1,000 Pt Days 0 Q1 Q4 Series1 9.025 1.947

Of the ten falls in the first quarter when only the MFS was used, three falls were witnessed by staff and safely lowered to the ground without injury. Seven falls were unwitnessed, with four patients sliding out of the bed or wheelchair, all of which were potentially preventable. Half of the falls occurred on day shift between 7a-7p and the other half on nights between 7p- 7a. The average LOS for the ten fall patients during the first quarter was 16 days.

Half of the patients who fell were male and half were female who fell, with one female patient falling twice during her admission. One injury occurred in the first quarter resulting in a large right frontal scalp hematoma, when only using the MFS. 39 HIGH FALL RISK IN ARU During the fourth quarter while the EPFRAT was used, no injuries occurred. Of the two falls in the fourth quarter, both were not preventable. One fall was witnessed and lowered to the ground. On day 10 of admission at 1510, the patient was working with her husband practicing transfer training for pending discharge. A physical therapist was present and she was safely lowered to the ground when she became weak. The second patient who was considered a fall was repositioning in bed at 0640 on day three of admission. His alarm sounded and when nurses responded, he was found with his left shoulder touching the ground and his feet were still in the bed, which meets the NDNQI fall definition. Average LOS for the two fall patients was 14 days.

Thus it is of significance to note that not only did the number of falls decrease with using the

EPFRAT tool, the occurrence/severity of injury also was much less.

Ratio estimation was used to estimate the average number of falls per 1,000 patient days for both quarter one pre implementation when only the MFS was used and quarter four post implementation when both the MFS and the EPFRAT were used. Resulting averages and standard errors were then used to estimate the difference in ratios. Results indicate that with

95% confidence, using the EPFRAT scale in addition to the MFS is associated in decreasing the number of average number of falls per 1,000 patient days by between 0.44 and 13.9.

Table 9 Results of Estimation of Population Ratio

MORSE Scale EPFRAT Scale Difference Number of observations 106 102 Ratio Estimate (falls / 1000 pt 9.16 1.98 7.18 days) Standard Error 2.97 1.59 3.37 95% Confidence Interval (3.23,15.10) (0,5.18) (0.45, 13.92)

Significance and Implications 40 HIGH FALL RISK IN ARU The goal of decreasing patient falls in the ARU with one of the most at risk populations was achieved. Using the right risk assessment tool for each specific patient population is imperative. There is not one tool that can be used across all populations. “Proper identification and precise assessment of individuals at risk are important components of fall prevention programs” (Abraham, 2016, p. 1). Assurance of using the proper tool assists to identify truly high fall risk patients allowing for appropriate nursing interventions is crucial. Using the wrong tool in a clinical area may exacerbate fall risk and injury (Abraham, 2016). The results of this study suggest that the EPFRAT is more effective in identifying high risk fall patients. The

EPFRAT score during the fourth quarter indicated 14% of patients were high fall risk while the

MFS indicated 88% of the patients were high fall risk. This more exclusive delineation of who is truly a high risk for falling allows better decision-making for needed interventions.

Falls continue to be one of the most frequently reported safety events in hospitals.

Fall risk assessment tools are the foundational element of fall prevention programs (Feil &

Gardner, 2012). In addition to the impact to our patients, the increasing regulatory and reimbursement pressures hospitals are facing, shows the urgent need to prevent falls (Feil &

Gardner, 2012). Assuring we are using an appropriate tool to identify high fall risk patients based on the setting and population is a crucial step in keeping our patients safe. Falls impact many things including quality of life, human suffering, cost, patient experience, and LOS, to mention a few. Use of the right fall risk assessment tool, “offers the advantage of process standardization the key to high reliability” (Feil & Gardner, 2012, p. 79). The risk assessment tool alone will not prevent falls but pared with effective fall prevention interventions, having the right tool like the EPFRAT in the ARU will make a significant impact for patients.

During this practice change, buy in and engagement from the nursing staff was a factor that this author believes contributed to the successful results. This buy in will be necessary to 41 HIGH FALL RISK IN ARU sustain the results. During informal interviews, the staff agreed that the EPFRAT tool was more specific to the patients in the ARU, offering a more accurate assessment resulting in more appropriate interventions. As consistent with the JHNEBP, the final step of translation of evidence to practice with the engagement of those closest to the practice is critical when implementing EBP.

The Future

Sustainability

The need for a more accurate tool in the ARU is clear and, identification of high-risk patients is challenging. The need for a tool to identify high fall risk patients remains desperately needed so that the appropriate prevention interventions can be implemented for this critical population. During this practice change, there was a significant reduction in ARU fall rates. It is difficult to assess fall prevention interventions and consistency of implementation. Throughout this practice change, other initiatives may have impacted results (new nurse manager, Hawthorne effect, staff turnover, education on fall prevention interventions done during the second and third quarter, and increased awareness due to poor results during the first quarter); however based on results of this practice change, the ARU at TJH will continue to use the EPFRAT in conjunction with the MFS to identify high-risk patients. “Staff involvement is a key factor in successful fall prevention interventions” (Zhao et al., 2019, p.86).

Future Projects

Moving forward continued use of the EPFRAT in the ARU is necessary for further assessment and larger sample size. Of importance, the need to conduct further testing of the

EPFRAT tool in other ARU would establish accuracy of use and impact, assuring rigorous assessment of this population. This author suggests trialing the EPFRAT in another Mercy 42 HIGH FALL RISK IN ARU Health Cincinnati ARU to compare results and then spreading to other ARU in the Bon Secours

Mercy Health enterprise if successful.

The suggestion was made by ARU staff to use the EPFRAT tool independently of the

MFS so that nurses do not need to complete two scales. Once the tool has been used in another

ARU to validate results, this could be considered and evaluated. Policy changes would be necessary and submission to the IRB to evaluate this proposed practice change would be needed based on the decision to use both tools or just one tool.

Dissemination

Dissemination of EBP to other health care professionals is necessary so practices can be replicated or applied in other settings (Forsyth, Wright, Scherb, & Gaspar, 2010).

This author wishes to present this ARU practice change as a poster presentation at the REACH conference in 2020. The REACH conference is the national annual education conference for rehabilitation nurses presented by the Association for Rehabilitation Nurses. “Poster presentations are a very effective method of communicating research findings and provide the opportunity to meet other researchers” (Halligan, 2008, p. 41).

The key to growth and development of the nursing professions is the dissemination of clinical innovations and findings (Halligan, 2008). In addition to this national conference, this data will be presented to the Bon Secours Mercy Health Chief Nurse Executives in an upcoming meeting in the summer of 2019. Furthermore, this author would like to submit this practice change to the Rehabilitation Nursing Journal, once additional data is gained post implementation in another ARU.

Conclusion

Active promotion of both mobility and independence in the ARU setting contributes to the increased risk of falls (Rabadi et al., 2008). In 2017, the fall rate in the ARU was 4.67 falls 43 HIGH FALL RISK IN ARU per 1,000 patient days, with little change in 2018 4.56 falls per 1,000 patient days, despite much effort. Falls spiked alarmingly high in January and February of 2018 at 13.05 falls per 1,000 patient days, indicating a need for change. Patients who fall in the ARU have multiple risk factors, and are often cognitively impaired and impulsive, similar to those in the acute psychiatric setting. The EPFRAT, although indicated for acute psychiatric patients, evaluates age, mental status, elimination, medications, diagnosis, ambulation/balance, nutrition, sleep disturbance, and history of falls, (Edmonson, Robinson, & Hughes, 2011), which are the same risk factors that contribute to ARU falls. The answer and solution to reduction in falls is complex, but it is clear that something has to change. The purpose of this practice change was to appropriately identify high fall risk patients in the ARU setting, resulting in the ability to identify high fall risk patients, allowing for appropriate nursing interventions, and ultimately resulting in decreasing ARU fall rates. The need for a tool to identify high fall risk patients is desperately needed so that the appropriate prevention interventions can be implemented for this critical population.

Although fall prevention programs are complex and multifaceted, this project was successful in decreasing patient falls by 80% in the ARU. During the fourth quarter when the

EPFRAT was utilized falls were at a unit low of 1.947 falls per 1,000 patient days. Continued assessment of the EPFRAT tool in the ARU setting is necessary, however data suggest a significant decrease in falls with use during this practice change. Consequences of falls include serious injury, increased cost, increase length of stay, loss of independence, and decreased quality of life. Use of the EPFRAT to accurately identify high fall risk patients in the ARU and implement appropriate nursing interventions was highly successful.

44 HIGH FALL RISK IN ARU

45 HIGH FALL RISK IN ARU Acknowledgments

Statistical analysis and interpretation work was completed with the assistance of the Northern

Kentucky University Burkardt Consulting Center (Highland Heights, KY).

46 HIGH FALL RISK IN ARU References

Abraham, S. (2016). Looking for a psychiatric fall risk assessment tool. Annals of Psychiatry

and Mental Health, 4 (2), 1061.

Ambutas, S., Lamb, K.V., & Quigley, P. (2017). Fall reduction and injury prevention toolkit:

Implementation on two medical-surgical units. MEDSURG Nursing, 26 (3), 175-197.

Butcher, L. (2013). The no-fall zone. Hospitals & Health Networks, 87(6), 26-30.

Chapman, J., Bachand, D., & Hyrkäs, K. (2011). Testing the sensitivity, specificity and

feasibility of four falls risk assessment tools in a clinical setting. Journal of Nursing

Management, 19, 133-142. doi:101111/j.1365-2834.2010.01218.x

Christopher, D.A., Trotta, R.L., Yoho, M.A., Strong, J., & Dubendorf, P. (2014). Using process

improvement methodology to address the complex issue of falls in the inpatient setting.

Journal of Nursing Care Quality, 29 (3), 204-214.

Clancy, C.M. (2013). Evidence-based toolkit helps organizations reduce patient falls. Journal of

Nursing Care Quality, 28 (3), 195-197. doi: 10.1002/rnj.308

Cournan, M., Fusco-Gessick, B., & Wright, L. (2018). Improving patient safety through video

monitoring. Rehabilitation Nursing, 43(2), 111-115.

Edmonson, D., Robinson, S., & Hughes, L. (2011). Development of the Edmonson psychiatric

fall risk assessment tool. Journal of Psychological Nursing, 49 (2), 29-36.

Feil, M., & Gardner, L.A. (2012). Falls risk assessment: A foundational element of falls

prevention programs. Pennsylvania Patient Safety Advisory, 9 (3), 73-81.

Forsyth, D.M., Wright, T.L., Scherb, C.A., & Gaspar, P.M. (2010). Disseminating evidence-

based practice projects: Poster design and evaluation. Clinical Scholars Review, (3) 1,

14-21. doi: 10.1891/1939-2095.3.1.14 47 HIGH FALL RISK IN ARU Frisina, P.G., Gullnitz, R., & Alverzo, J. (2010). A time series analysis of falls and injury in the

inpatient rehabilitation setting. Rehabilitation Nursing, 35(4), 141-146.

Gilewski, M.J., Roberts, P., Hirata J., & Riggs, R. (2007). Discriminating high fall risk on an

inpatient rehabilitation unit. Rehabilitation Nursing, 32(6), 234-240.

Halligan, P. (2008). Poster presentations: Valuing all forms of evidence. Nurse Education and

Practice, (8), 41-45. doi: 10.1016/j.nepr.2007.02.005

Kwan, F., Kaplan, S., Hudson-McKinney, M., Redman-Bently, D., & Rosario, E. (2012).

Comparison of fallers and nonfallers at an inpatient rehabilitation facility: A

retrospective review. Rehabilitation Nursing, 37(1), 30-36.

Lewin, K. (1947). Group decisions and social change. In T. M. Newcomb & E.

L. Hartley (Eds.), Readings in social psychology (pp. 330–344). New York, NY: Henry

Holt.

Melnyk, B. & Fineout-Overholt, E. (2015). Evidence-based practice in nursing and healthcare:

A guide to best practice (2nd ed.). Philadelphia, PA: Wolters Kluwer.

Mitchell, G. (2013) Selecting the best theory to implement planned change. Nursing

Management, 20(1), 32-37.

Morse, J.M., Morse, R.M. & Tylko, S.J. (1989). Development of a scale to identify the fall prone

patient. Canadian Journal of Aging, 8, 366-377.

National Database of Nursing Quality Indicators. (2017). Nurse management clinical unit detail:

Acute Rehabilitation. Retrieved from

https://members.nursingquality.org/MicroStrategy/asp/Main.aspx

National Database of Nursing Quality Indicators. (2016). Guidelines for data collection and

submission on patient falls indicator. Retrieved from

https://www.pressganey.com/solutions/clinical-excellence/nursing-quality 48 HIGH FALL RISK IN ARU Newhouse, R.P., Dearholt, S.L., Poe, S.S., Pugh, L.C., & White, K.M. (2007). Johns Hopkins

nursing evidence-based practice: Model & guidelines. Indianapolis, IN: Sigma Theta

Tau.

Oliver, D., Healey, F., & Haines, T.P. (2010). Preventing falls and fall- related injuries in

hospitals. Clinics in Geriatric Medicine, 26, (4), 645-692. doi:10.1016/j.cger.2010.06.005

Polit, D. F., & Beck, C. T. (2014). Essentials of nursing research appraising evidence for

nursing practice (8th ed.). Philadelphia, PA: Lippincott, Williams & Wilkins.

Rabadi,M.H., Rabadi, F.M., & Peterson, M. (2008). An analysis of falls occurring in patients

with stroke on an acute rehabilitation unit. Rehabilitation Nursing, 33, (3), 104-109.

Reavy, K. (2016). Inquiry and leadership: A resource for the DNP project. Philadelphia, PA:

FA Davis Company.

Rosario, E.R., Kaplan, S.E., Khonsari, S., & Patterson, D. (2014). Predicting and assessing fall

risk in acute inpatient rehabilitation facility. Rehabilitation Nursing, 39, 86-93.

doi10.1002/rjn.114

Schaffer, M. A., Sandau, K. E., & Diedrick, L. (2012). Evidence-based practice models for

organizational change: Overview and practical applications. Journal of Advanced

Nursing, (69)5, 1197-1209. doi: 10.1111/j.1365-2648.2012.06122.x

Shirey, M. R. (2013). Lewin’s theory of planned change as a strategic resource. Journal of

Nursing Administration, (43)2, 69-72.

Slade, S. (2017). Falls in inpatient mental health settings: Screening and risk assessment

tools. The Joanna Briggs Institute of EBP Database, JBI12176.

Vratsistas-Curto,A., Tiedemann, A., Treacy, D., Lord, S.R., & Sherrington, C. (2018).

External validation of approaches to prediction of falls during hospital rehabilitation

stays in development of a new simpler tool. Journal of Rehabilitation Medicine, 50, 49 HIGH FALL RISK IN ARU 216-222. doi: 10.2340/16501977-2290

Zhao, Y., Bott, M., He, J., Kim, H., Park, S. H. & Dunton, N. (2019). Evidence on fall and

injurious fall prevention interventions in acute care hospitals. JONA: The Journal of

Nursing Administration, 49(2), 86–92. doi: 10.1097/NNA.0000000000000715

50 HIGH FALL RISK IN ARU Appendix A

Quarter One Age and MFS Data

Q1 Variable Mean StDev Minimum Q1 Median Q3 Maximum AGE 72.01 12.47 37.00 61.75 72.00 81.00 99.00 MFS on Admission 69.01 18.60 25.00 60.00 65.00 85.00 110.00 MFS on DC 61.32 17.89 25.00 50.00 65.00 75.00 100.00

51 HIGH FALL RISK IN ARU Appendix B

Quarter Four Age and MFS Data

Q4 Variable Mean StDev Minimum Q1 Median Q3 Maximum AGE 66.89 14.34 21.00 59.75 69.50 78.00 93.00 MFS on Admission 65.15 20.49 0.00 50.00 60.00 85.00 100.00 MFS on DC 53.38 19.08 0.00 40.00 60.00 65.00 100.00

52 HIGH FALL RISK IN ARU Appendix C

Gender Data

Gender Count Q1 Percent Q1 Count Q4 Percent Q4 F 52 49.93% 50 49.95% M 54 50.07% 52 50.05% Total 106 100.00% 102 100.00%

53 HIGH FALL RISK IN ARU Appendix D

Admission Diagnosis Quarter One and Four

Count Count Diagnosis Q1 Diagnosis Q4

*Other Neurologic Condition 22 *CVA 17 *CVA 18 Other Ortho 16 Debility non - Cardiac/Pulm 9 Debility 11 Cardiac 7 *Other non-traum SCI 10 *Other BI 7 *Neuromuscular Disorder 9 *Other NT Spinal Cord Cond 6 *Other BI 8 *TBI 6 *Traumatic BI 5 Other Orthopaedic 5 Cardiac 4 Other Traumatic SC Condition 4 *Neurological 4 Bilateral Hip Fx 3 *Other Neurologic Disorder 3 Unilateral Hip Fx 3 *Non Traumatic BI 2 Debility 2 Unilat Hip Repl 2 *Neurologic 2 Unilat Knee Repl 2 *Neuromuscular Disorder 2 Bilat Knee Repl 1 *NTBI 2 Fx Femur 1 AKA 1 Hip Fx 1 BKA 1 Maj Mult Fx 1 Maj Mult Trauma 1 MS 1 Major Mult Fx 1 *Non-Traum BI 1 MS 1 Unilat AKA 1 Other Medically Complex Cond 1 Unilat BKA 1 Other Otho 1 Unilat Hip Fx 1 Unilateral Hip Repl 1 Total 102 Total 106 *Indicates Neurologic Diagnosis

54 HIGH FALL RISK IN ARU

Appendix E

Power Point: ARU Staff Education

ARU Staff Education

Identification of High Fall Risk Patients in ARU Vanessa Vonderhaar-Picard Mount St. Joseph University

55 HIGH FALL RISK IN ARU

Problem

● Yearly 700,000 - 1 million people fall in US hospitals (Clancy, 2013, p. 195)

● By 2020 annual total cost for fall related injuries could increase to 34.4 billion (Christopher, Trotta, Yoho, Strong, & Dubendorf, 2014)

● ARU patients are “one of the most at risk populations for falls during hospitalization” (Rosario, Kaplan, Khonsari, & Patterson, 2014, p. 86)

● Understanding the risk factors for ARU “is essential to being able to accurately predict patient falls” (Rosario et al., 2014, p. 87)

● Current ARU fall tools are intended for the medical surgical hospital setting and identify 75-90% of ARU patients as high risk (Rosario et al., 2014).

56 HIGH FALL RISK IN ARU

Problem Continued… ● National ARU fall rate per 1,000 patient days nationally is 1.8 – 9.81 with the median 5.02. (National Data Base of Nursing Quality Indicators, 2017)

● The Jewish Hospital ARU fall rate 4.67 per 1,000 patient days in 2017, and 5.72 YTD 2018 with 1st Q at 9.18.

● Falls are a serious safety issue

● Falls are the leading cause of injury in adults 65 and older (Christopher et al., 2014)

● CMS stopped reimbursement of cost associated with preventable injury from falls in 2008 (Clancy, 2013)

● Higher fall rates occur among older patients with neurological conditions in the rehabilitation setting (Ambutas, Lamb, Quigley, 2017)

57 HIGH FALL RISK IN ARU

PICOT Question

In patients needing acute rehab (P), how does utilization of the Edmonson Psychiatric Fall Risk Assessment Tool (I) compared to utilization of the Morse Fall Scale (C) affect falls per 1,000 patient days (O) while hospitalized(T)?

58 HIGH FALL RISK IN ARU Driving Forces for Change Supported by the Literature ● Current tool identifies 80% of patients as high fall risk

● One tool can not be used across all settings

● No current validated tool to identify high fall risk in ARU making it difficult to identify when additional Nursing interventions are needed

● Decrease fall rates in ARU current rate 5.7 2018 YTD / Q1 9.18

● Decrease hospital cost

● Improve length of stay associated with patient falls

● Increase patient confidence

● Improve patient experience

59 HIGH FALL RISK IN ARU

Why the EPFRAT tool

● The EPFRAT although indicated for acute psychiatric patients, evaluates age, mental status, elimination, medications, diagnosis, ambulation/balance, nutrition, sleep disturbance and history of falls (Edmonson, Robinson, & Hughes, 2011)

● These same risk factors contribute to ARU falls and falls with injury.

● Morse Fall Scale (MFS) identifies 75-90% of patients as high fall risk making it difficult to assure appropriate interventions are in place (Rasario et al., 2014)

● The EPFRAT tool was more sensitive (0.63) when compared to the MFS (0.49) (Edmonson et al., 2011)

60 HIGH FALL RISK IN ARU

Intervention

• Utilization of the Edmonson Psychiatric Fall Risk Assessment Tool in the Acute Rehab Unit • EPFRAT Assessment to be completed with MFS (starting Oct 1, 2018) on admission to unit and every 12 hours • All EPFRAT Scores of 90 or above evaluate for Camera /Sitter / Additional High Fall Risk Interventions

61 HIGH FALL RISK IN ARU Edmonson Psychiatric Fall Risk Assessment ● Age

● Mental Status

● Elimination

● Medications

● Diagnosis

● Ambulation/Balance

● Nutrition

● Sleep Disturbance

● History of Falls

62 HIGH FALL RISK IN ARU

MFS vs EPFRAT

63 HIGH FALL RISK IN ARU

ARU Case Study

● 67yo Female admitted to ARU post stroke (large left intraparenchymal hemorrhage)

8/11-8/14 Admission to ARU – poor initiation and participation with therapy - lethargy

8/15 Neuro stimulant

8/21 Fall

8/22 Fall w injury Medial Orbital Wall Fracture

64 HIGH FALL RISK IN ARU

Evaluation of Practice Change

● Identify high fall risk patients in the ARU setting using the Edmonson Psychiatric Fall Risk Assessment Tool (EPFRAT) tool.

● Utilize appropriate nursing interventions, evaluate patients for EPFRAT score above 90 for AvaSYS camera or sitter. Goal

● Decrease falls per 1,000 patient days in ARU.

65 HIGH FALL RISK IN ARU Appendix F

Edmonson Handout

Edmonson Psychiatric Fall Risk Assessment ©

Date & Initials Complete Daily & upon admission *More than one item may be circled in each category if appropriate for the patient. Age 8 Less than 50 10 50-79 26 80-over Mental Status -4 Fully Alert/Oriented at all times 12 Agitation/Anxiety 13 Intermittently confused 14 Confusion/Disorientation Elimination 8 Independent with control of bowel/bladder 12 Catheter/Ostomy 10 Elimination with Assist 12 Altered elimination (incontinence, nocturia, frequency) 12 Incontinent but Ambulates Independently Medications 10 No Medications 10 Cardiac Medications 8 Psychotropic Medications (Including benzodiazepines and antidepressants) O R 12 Increase in these medications and/or PRN (psych, pain) medication received in the last 24 hours Diagnosis 10 Bipolar/ Schizoaffective Disorder 8 Substance abuse/Alcohol abuse 10 Major Depressive Disorder 12 Dementia/ Delirium Ambulation/Balance 7 Independent/Steady gait/Immobile 8 Proper Use of Assistive Devices (cane, walker, w/c) 10 Vertigo/Orthostatic Hypotension/Weakness 8 Unsteady/Requires Assist and Aware of Abilities 15 Unsteady but Forgets Limitations Nutrition 12 Has had very little food or fluids in the past 24 hours 0 No apparent abnormalities with appetite Sleep Disturbance 8 No Sleep Disturbance 12 Report of Sleep Disturbance by patient, family or staff History of Falls 8 No History of Falls 14 History of Falls in the last 3 months TOTAL Add all nine columns

** FALL RISK = SCORE OF 90 OR GREATER **

Edmonson Psychiatric Fall Risk Assessment © Page 1 of 3

66 HIGH FALL RISK IN ARU

67 HIGH FALL RISK IN ARU

68 HIGH FALL RISK IN ARU Appendix G

Edmonson EPIC Handout

69 HIGH FALL RISK IN ARU Appendix H

ARU Reminder Flyer

Starting October 1, 2018 Complete the EPFRAT Assessment in EPIC with the Morse Fall Scale.

EPFRAT Score above 90 consider AvaSYS Camera or Sitter 70 HIGH FALL RISK IN ARU