Kolin, M. Melissa DNP, CRNP, RN; Minnier, Tamara MSN, RN, FACHE; Hale, Kathleen M. MSN, RN; Martin, Susan C. MSN, RN; Thompson, Lucy E. MSN, RN
Increased clinical scrutiny by regulatory agencies and healthcare payers, as well as expectations of patients and/or families, can make the delivery of quality patient care challenging. Adverse patient care events precipitated by healthcare workers produce more than 1 million preventable injuries and more than 98,000 deaths at a cost of over $29 billion annually.1 Many of these injuries occur because of the fragmented, frequently interrupted, chaotic work environment experienced by nurses. Daily, newspapers and other media outlets have content related to healthcare and patient injuries. As the national trend to judge quality and award reimbursement based on outcomes increases, every area of nursing practice will be affected. Healthcare reform, managed care, growing price competition, and the tightening of Medicare and Medicaid regulations have steadily increased quality performance expectations for hospitals and healthcare systems. Although the definition of quality in healthcare is a complex, multifaceted, and multidimensional entity, it is often defined by the perception of the "beholder," be it the patient, family, physician, nurse, or organization.2
In 2004, the National Quality Forum published the first set of performance measures that are used to assess nursing's contribution to healthcare quality and patient safety. These nursing-sensitive outcomes, defined as variables responsive to nursing care and interventions, are a major indicator of quality patient care.3 These outcomes-patient falls, urinary tract infections, pneumonia, shock, upper gastrointestinal bleeding, longer hospitalizations, failure to rescue, pressure ulcer development, adverse events such as medication administration errors, 30 day mortality measures, and patient satisfaction measures-are all correlated to nursing care.4
Patient falls, the subsequent treatment, and complications are reported to increase length of stay by more than 3%, with the patient stay extending to an average of 18 days longer, which results in increased resource utilization by more than 32%.5 Falls, second only to medication events, are a large category of reported adverse events in hospitals and the leading cause of nonfatal injuries, which cost the healthcare industry more than $20 billion a year.6 In October 2008, the Centers for Medicare and Medicaid published a list of "reasonably preventable" mistakes and eliminated payment to healthcare organizations for these so-called never events as defined by The National Quality Forum in 2004 as serious complications.3 Patient falls and falls with injury in healthcare organizations are a problem nationwide and are now part of the never-event Centers for Medicare and Medicaid initiative as well as a Joint Commission National Patient Safety Goal.7,8 Patient falls and falls with injury are estimated to happen in up to 20% of hospitalizations, with more than $4,000 in excess charges.9
In February 2008, leaders at the University of Pittsburgh Medical Center (UPMC) formed a system-wide team to prioritize patient falls as one of the most prevalent and important safety threats to our patients. We wanted to investigate internal patient fall assessment and prevention practices in the health system as well as best practices for performance improvement. The team was led by the senior vice president and quality director of the Donald D. Wolff, Jr. Center for Quality Improvement and Innovation at UPMC, with 19 acute care facilities in Western Pennsylvania. In addition, UPMC has a worldwide presence, with facilities in Ireland, Italy, Qatar, and Cypress.
Goals and Expected Outcome
The goal of this team project was to identify best practices from across the country, identify what UPMC does well and not as well, and "prescribe" a best practice package/toolkit for implementation at all UPMC facilities. This project produced a system-wide change in the practice of fall injury assessment and prevention that positively impacted patient care outcomes. Members of the project team include the authors of this article and other expert members from across the health system. Implementation means changing the way nurses think about falls, that is, instead of the traditional method for assessment of patients at risk for falling, accepting that "all" patients are at risk, regardless of age, sex, or diagnosis. The new UPMC process targets those patients at risk for injury as identified through the new assessment tool and processes, with the ultimate goal of decreased injuries reported with falls.
The best practice fall prevention literature review began with CINAHL, Medline, and the Cochrane database, searching for recent studies from the last 5 years, qualified by the terms falls, fall prevention, fall injury prevention, and falls best practice. The literature review did not yield many randomized controlled trials or meta-analyses related to patient fall prevention or interventions. Many of the journal articles found were not research based or were qualitative studies in which insufficient statistical analysis was completed or was not reported, thus making judgment of an effect impossible. Many studies were expert opinion, quality improvement studies and summaries, or descriptive research reports with small sample size. The intervention studies showed that reducing fall risk factors does reduce the likelihood of injury.10,11 Fall prevention programs based on guidelines are commonly referred to as evidence based, but according to Savica, there are actually 3 levels of evidence related to fall prevention and intervention studies.12 Level C interventions are based on meta-analyses of multiple studies, level B interventions are those that have been tested in randomized controlled trials that reduce fall risk factors but not fall, and level A interventions are those that have been tested in randomized controlled trials that do reduce falls. However, few studies, if any, in the literature review were found to be a level A study.
Multiple studies imply that multifactorial interventions seem to have some effectiveness in decreasing fall rates in patient care settings; however, there is inconclusive evidence advocating bed/chair alarms, exercise, or vitamin D supplementation in the prevention of patient fall injuries. Purposeful hourly (ie, scripted or checklist based) nursing rounds seem to be a promising intervention. Although there are few quantitative studies related to purposeful hourly rounding, there is anecdotal evidence that hourly rounding is helpful with both patient satisfaction and fall prevention.13 The key to implementing multifactorial interventions is that they must be individualized to patient need, and most of the reported intervention studies had someone, that is, a champion (a chief nursing office [CNO], physician, unit manager, etc), who was responsible for consistent follow-up on the fall risk issue every day with the staff and verifying that the plan of care was individualized and appropriate to the patient's needs and whether or not the plan was carried out.13
Before adopting a new model, the UPMC practice (per policy and procedure) is to complete a fall risk assessment. The fall risk assessment is completed on admission to both the inpatient and outpatient settings. For patients admitted, the fall assessment is completed once every 24 hours at a minimum, with other events triggering more frequent assessment. The patient assessment is scored as either "low risk" (score of ≤9) or "at risk" (score ≥10) for falling. Based on nursing judgment, appropriate safety interventions are implemented according to policy and the "Catch a Falling Star" protocol, a program carried out to promote the visual identification of a patient at risk for falling, which includes door signage, a falling star documentation/sticker on the Kardex, and a falling star sticker on the patient arm band.
At the redesign team kick-off in February and March 2008, patient fall data were presented to the authors of this project. The data had been statistically analyzed to generate the overall fall rate and fall with injury rate, for patient age, patient sex, commonality of location/activity, and the time of fall. The monthly fall rate and injury rate for each facility in UPMC and corporately were compared with the corporate and national benchmarks. As you can see from the 2008 UPMC Health System data in Figure 1, the fall trend line for falls was increasing over the year.
The UPMC internal fall benchmark for medical-surgical areas was less than 4% per total patient days, rehabilitation was less than 7% per total patient days, and long-term care was less than 7% per total patient days, and the injury index is benchmarked at less than 3% across all patient care areas. This information was presented to the UPMC risk management and safety officers by the clinical leader of the project.
Next, a CNO survey was designed to determine "where we are and where we need to be." The survey was distributed to all business unit CNOs to assess how compliant each facility was following the fall prevention policy and procedure/process. Interestingly, the survey provided the team with valuable information about the variability of compliance across the health system. Whereas 80% of the facilities used the standard patient assessment for fall risk survey, 20% used other tools. Reassessment of fall risk also varied across the health system, with only 66% completed daily. All facilities used a postfall follow-up form, but none were the same. These survey results confirmed an internal audit related to fall prevention that reported notable gaps in practice and compliance across the health system.
National experts from across the country were then contacted about their work with falls. All were interested in our work and collaborated with the team. Multiple fall risk assessment tools were investigated by the team, looking for a validated tool that could be used at the health system during the analysis process; tools included the Hendrich II Fall Risk Model,14,15 the Fall Risk Assessment tool,16 the Morse Fall Scale,17 and the STRATIFY Instrument.18 The National Center for Patient Safety recommends the Morse Fall Scale and the Hendrich Fall Risk Assessment as the preferred risk assessment tool for inpatients with proven reliability and validity. To familiarize the team with the complexity surrounding fall assessment, interventions, communication, and strategies to prevent harm, team members participated in a national educational audio-conference in April 2008.
After reviewing the literature, analyzing data, and planning, a rapid improvement event (RIE) was held for a large interdisciplinary group of clinical experts at UPMC. The ultimate goal of the problem-solving, action-oriented session was to provide an improved fall injury reduction process for our patients; we wanted to change the way interdisciplinary UPMC practitioners think about falls and the traditional assessment of patients related to falls and fall injury prevention. Five work groups were charged to brainstorm, test new ideas, mine and analyze data, and plan the adoption of changes related to fall injury prevention. These 5 work/redesign groups included (1) patient fall/injury risk assessment and reassessment, (2) equipment and interventions to prevent injuries, (3) hospital environment, (4) staff and patient/family education, and (5) postfall follow-up. The groups began redesigning key processes/components of the UPMC falls best practice toolkit/model.
After the RIE, the assessment/reassessment work group planned and conducted tests of change using a convenience sample of patients within UPMC in a comparison study with the current UPMC assessment tool and the "ABCS" tool, a tool that was originally created and tested at the VA Health System Tampa facility to identify populations at increased risk for harm/injury due to a fall based on age older than 85 years, bone pathophysiology, coagulation status, and surgical wound sutures. This test provided valuable information to the team regarding the sensitivity of the ABCS tool. The ABCS tool was tested randomly on 250 patients across the health system, and, although having valuable attributes, it resulted in low sensitivity, with findings that lacked specificity to meet the needs of the UPMC population (Figure 2). This tool's low sensitivity was attributed to the population differences between the VA Health System and UPMC's patient population. Another adapted tool-a hybrid tool-was piloted on 154 patients, yielding the same results of low sensitivity. A simple yet comprehensive assessment tool was then developed by the redesign team using expert opinion, from working clinicians as temporal validation. With only 3 questions and some checkbox items (Figure 3), this tool testing proved to be specific enough to predict injury risk and reduced the time that a nurse needed to complete the assessment by more than 85%.
The new Fall Injury Assessment and Interventions tool was an improvement in identifying patients at risk for injury; it also guides the nursing staff in the selection of appropriate, targeted interventions to be used based on the assessed risk. Standard safety interventions were designated for the 3 risk groups. The first group of interventions are applied to all patients and includes basic safety interventions such as bed locked and in low position and call bell in reach. Level 1 interventions are applied to patients who screen positively to 1 question on the tool and include, in addition to the mandatory universal interventions, a yellow fall injury risk arm band to identify patients at a glance, focused lightning rounding, and bed alarm use based on nursing clinical judgment. The level 2 interventions are applied to patients who screen positively to 2 or more questions and include, in addition to the universal and level 1 interventions, mandatory bed alarms, low bed level, and appropriate room location, as examples.
In multiple studies, purposeful hourly nursing rounds have been implicated in the improvement of patient satisfaction and patient safety. The UPMC implemented "Lightning Rounds" using a checklist to promote consistency in rounding, with education provided to nursing staff at all business units (Figure 4). Comprehensive nursing education was conducted before the implementation of the new Fall Injury Risk Assessment and Interventions Tool (Figure 5).
As a method of standardization for the health system, the refinement of the postfall follow-up documentation form was completed; similar to a mini root cause analysis, it gathers pertinent patient information after a fall. The form has the manager/supervisor note, the unit name, and the date and time of the fall; discusses the circumstances of the fall; and identifies the root cause (Figure 6 shows the areas of evaluation) and recommendations/corrective action. This information is then incorporated into the UPMC Health System event reporting system (Figure 6). The UPMC information technology department and the Center for Quality and Innovation are working on an interface to connect the system electronic medical record with the event reporting system, thus enabling pertinent patient medical record information to load directly into the system. This will ensure that data collection related to falls with injury will be consistent, resulting in reliable data analysis and the ability to benchmark data across the health system and nationally.
During the RIE, the environmental work group recommendations were grouped into 2 categories: new construction and existing construction. Recommendations included installation of grab rails, placement of glow strips/signage along handrails to illuminate the path to the bathroom, improved bathroom lightings including automatic motion sensor lights, use of lever doorknobs, equipment cord bundling to keep out of the patient's pathway, as well as an environmental checklist for staff to use.
Staff and patient communication and education updates were completed with 2 new patient safety brochures developed, including a summary of "Falls Fast Facts" for staff and an educational brochure for patients and families for hospital safety and safety at home. "Fear of Falling: It's a Balancing Act" is a DVD available for patients and families to improve patient safety and prevention of falls.
The implementation of the new fall injury assessment and intervention toolkit was rolled out across the health system in June 2009 for paper documentation sites and September 2009 for electronic healthcare record sites. Postimplementation data collection, data analysis, and benchmarking of falls with injuries occurring at 3, 6, and 12 months are carried out with indicators that include total number of falls, total number of falls with injury, days/intervals between falls, patient demographics, compliance with nursing documentation, compliance with recommended intervention implementation, and compliance with postfall documentation using random medical record audits across the health system.
One of the rural business units and paper documentation sites in the Medical Center, UPMC Horizon implemented the new Fall Injury Risk Assessment and Interventions process on June 1, 2009. In fiscal year 2008, Horizon had 174 falls and 3 falls with injury, with an injury rate of 1.7. Data for the first 3 months after implementation (Figure 7) show an average fall rate per 1,000 patient days of 2% and a zero injury rate per 10,000 patient days, certainly promising results. In addition, UPMC Health System data show promising results with the complete implementation of the new program, notably a decreasing trend line as seen in Figure 8. The fall team meets regularly, with in-depth analysis of fall rates and fall injury rates occurring at regular intervals as we strive to improve patient care and patient safety at every level.
As we move into the next era of healthcare, which includes some form of healthcare reform, managed care, pay for performance, never events, and a more educated healthcare consumer, the emphasis on the provision of quality patient care and positive patient care outcomes will only increase. Clearly, there will be the need for research into the relationship between optimal management and quality outcomes including nursing-sensitive adverse clinical outcomes related to the lack of efficiency or inadequate processes in the current healthcare delivery system. More accurate and consistent measures of acuity and quality are needed to explain the complex relationship between nurse processes and quality of care and patient safety. No longer can healthcare leaders continue to manage complex organizations as they have in the past with flawed processes and systems. Future emphasis must look at new management methods, best research available, and new best practice methods in the provision of safe, quality patient care. This project has developed a fall injury prevention toolkit that is used across the 19-business unit healthcare system targeting the reduction of falls with injury and has produced a system-wide change that has positively affected patient care outcomes at the UPMC.
The authors thank Suzanne Rita, Sentera Health System, Viriginia, and Patricia Quigley, VA Health System, Tampa, Florida, for their guidance and support.
1. Greiner A, Knebel B. Health Professions Education: A Bridge to Quality
. Washington, DC: National Academies Press; 2001.
2. Currie V, Harvey G, West E, McKenna H, Keeney S. Relationship between quality of care, staffing levels, skill mix and nurse autonomy. J Adv Nurs
3. Agency for Healthcare Research and Quality (AHRQ). Hospital nurse staffing and quality of care. Res Action
4. Needleman J, Buerhaus P, Mattke S, Stewart M, Zelevinsky K. Nurse staffing and quality of care in hospitals in the United States. N Engl J Med
5. Clark PA, Leddy K, Drain M, Kaldenberg D. State nursing shortages and patient satisfaction. J Nurs Care Qual
6. Agency for Healthcare Research and Quality (AHRQ). Patient safety and quality, patient safety events make hospital stays longer and more costly. Available at www.ahrq.gov/research
. Accessed March 2, 2010.
7. Joint Commission on Accreditation of Healthcare Organizations. 2008 Joint Commission National Patient Safety Goals: practical strategies and helpful solutions for meeting these goals. Available at http://www.jcrinc.com/subscribers/patientsafety.asp
. Accessed March 2, 2010.
9. Inouye SK, Brown CJ, Tinetti ME. Medicare nonpayment, hospital falls, and unintended consequences. N Engl J Med
10. Coussement J, Paepe LD, Schwendimann R, Denhaerynyck Dejaeger E, Milisen K. Interventions for preventing falls in acute- and chronic-care hospitals, a systematic review and meta-analysis. J Am Geriatr Soc
11. Cuming RG, Sherrington C, Lord SR, et al. Cluster randomized trial of a targeted multifactorial intervention to prevent falls among older people in hospital. Br J Med
12. Stevens A. What does an evidence-based fall prevention program mean? Defining levels of evidence. J Am Geriatr Soc
13. Meade CM, Bursell AL, Ketelsen L. Effects of nursing rounds on patients' call light use, satisfaction and safety. Am J Nurs
14. Hendrich A, Nyhuis A, Kippenbrock T, Soja ME. Hospital falls: development of a predictive model for clinical practice. Appl Nurs Res
15. Hendrich A, Bender PS, Nyhuis A. Validation of the Hendrich II Fall Risk Model: a large concurrent case/control study of hospitalized patients. Appl Nurs Res
16. Parsons NS, Cryer C, Underwood M, et al. Development and preliminary examination of the predictive validity of the Falls Risk Assessment Tool (FRAT) for use in primary care. J Public Health
17. Quigley PA, Hahm B, Gibson W, et al. Reducing Serious Injury from Falls in Tow Veteran's Hospital Medical-Surgical Units. J Nurs Care Qual
18. Milisen K, Staelens N, Schwendimann R, et al. Fall prediction in inpatients by bedside nurses using the St. Thomas's Risk Assessment Tool in Falling Elderly Inpatients (STRATIFY) Instrument. J Am Geriatr Soc
© 2010 Lippincott Williams & Wilkins, Inc.