Preventing Fall Injury

Falls are the leading cause of injury, premature institutionalization, and long-term disability in elderly adults worldwide, with a fall-related fatality in the United States every 19 minutes.¹ According to the Centers for Disease Control and Prevention,² 3 million people over 65 years of age receive emergency room treatment for fall injuries at an average cost of $30 000. The annual cost of fall injuries was more than $50 billion in 2015.^1,2 Community-based interventions effective in preventing falls include exercise, medication, and nutritional management as well as improving safety of the local environment.³ Evidence supporting interventions designed to reduce hospital inpatient falls is less clear despite considerable research aimed at reducing this growing problem. Those injured due to falling during a hospital stay incur higher costs, including a 6-day longer hospital stay, than non-fallers.⁴ Programs have worked to prevent other “never events,” such as wound infections or pressure ulcers, but mixed results have been reported for preventing falls or fall-related injuries in hospitals. This month’s Evidence Corner reviews a randomized controlled trial (RCT)⁵ and a prospective observational study⁶ that offer important clues on how to prevent hospital inpatient falls. 

RELATED CONTENT

Reduction of Hospital-acquired Pressure Injuries Using a Multidisciplinary Team Approach: A Descriptive Study

Hydrocellular Foam Versus Hydrocolloid Plate in the Prevention of Pressure Injuries

Development of a Manual for the Prevention and Treatment of Skin Tears

Reference: Haines TP, Hill AM, Hill KD, McPhail S, Oliver D, Brauer S, Hoffmann T, Beer C. Patient education to prevent falls among older hospital inpatients: a randomized controlled trial [published online November 22, 2010]. Arch Intern Med. 2011;171(6):516–524.

Rationale: Intensive fall-prevention education of cognitively intact elderly patients has been reported to reduce patient falls in acute and subacute care settings, but cost effectiveness of such interventions merits further study.

Objective: Determine the cost effectiveness of an in-hospital fall-prevention program, while measuring the economic burden of falls and the cost of a person being a faller. 

Methods: A cost-effectiveness analysis applied a decision tree model of incremental costs and effects on fall rates and proportion of inpatients falling in acute and subacute settings in Australia. Consenting patients at least 60 years of age were randomly assigned to receive usual care plus either (A) the intervention consisting of standardized multimedia educational materials with trained health care professional follow-up (N = 401, of whom 310 were cognitively intact) or (B) multidisciplinary usual care (N = 381, of whom 280 were cognitively intact) consisting of routine fall-risk screening and alert items such as patient arm bands. A research assistant blinded to the treatment group recorded patient falls and related radiological, medical, nursing, and clinical investigations and treatments, including length of stay (LOS). Site-specific per diem costs were evaluated in A$ as of 2008. Acute care admissions resulting from rehabilitation setting falls also were included. Standardized acute hospital and rehabilitation costs unrelated to falls and those directly related to falls per specific incident reports or medical records were calculated for each patient and corroborated by weekly and predischarge interviews. Time spent completing and documenting nursing and medical assessments was determined empirically and recorded as 15 minutes for each uncomplicated fall or 30 minutes for a fall with injury, with cost based on corresponding local wage rates, seniority, and setting. Iterative cohort regression analyses estimated the burden of falls based on increased LOS for “fallers” versus “non-fallers” after adjusting for other factors contributing to a longer stay in the setting. A decision tree model provided the framework for calculating incremental cost effectiveness as the following: (Cost of Providing Intervention to 100 Patients + Cost of Fallers Among the 100 Patients) – Cost of Fallers Among 100 Patients Receiving Usual Care. The difference was divided by the number of falls prevented or by the number of fallers prevented to derive cost per fall or faller prevented. Sensitivity analyses were conducted for (1) the proportion of patients with intact cognition on each ward, (2) cost to a health service of a patient being a “faller,” and (3) effectiveness of the intervention. The 95% confidence interval indicated statistical significance.

Results: Among cognitively intact patients, there were 4% fewer fallers and fewer falls per 100 patients in the intervention group as compared with patients receiving usual care. The intervention had an incremental cost of A$526 to prevent one faller or A$294 to prevent one fall among 100 patients. Overall cost per fall was A$12 469 and the cost per faller was A$24 927, with outliers included. This difference was driven mainly by rehabilitation rather than acute care costs and by a few outliers with very long stays. Rehabilitation costs nearly doubled for those with cognitive impairment (P < .001). 

Authors’ Conclusions: Conservative modeling suggests that if at least 4% of cognitively intact patients fall on a ward under usual care conditions, providing this educational intervention with professional follow up can prevent falls and reduce health service costs. 

Reference: Cuttler SJ, Barr-Walker J, Cuttler L. Reducing medical-surgical inpatient falls and injuries with videos, icons and alarms. BMJ Open Qual. 2017;6(2):e000119. doi: 10.1136/bmjoq-2017-000119.

Rationale: Hospital falls did not change substantially from 2010 to 2014 and efforts to prevent them have proven minimally cost effective without significant injury reduction. 

Objective: Conduct a prospective observational study evaluating effects on falls and fall-related injuries in one hospital’s medical-surgical unit of a volunteer-administered program to educate patients and staff about fall prevention using video education, patient fall-risk-alert icons, and bed-exit alarms.

Methods: After RCT evidence in PubMed and CINAHL databases revealed that timed toileting, patient education, and icons alerting staff to patient fall risk reduced likelihood of falls or related injury, the falls task force of a US urban safety net teaching hospital initiated a falls-improvement project and monitored its effects on patients in 2 mainly surgical units and 2 acute care units caring for elderly patients during baseline care from January 2009 to September 2010 (75 293 patient days) or after implementing a fall-prevention program (January to December 2015; 42 580 patient days). Key elements of the program were: (A) specialized beds with 3-mode bed-exit alarms were introduced into the hospital; (B) a multilingual 4-minute video for bedside computer viewing to inform patients and staff how to reduce patient risks of falling; (C) researcher-trained volunteer staff, fluent in the patient’s language(s), who encouraged patients to wait for staff, accept staff presence for toileting, and activate call lights if a patient exits the bed; and (D) a set of icons to be displayed at the head of each patient’s bed alerting staff to elevated patient fall risk. Outcomes were incidence of all falls per 1000 patient days and proportion of falls in each class of injury severity. Falls were defined as unintentional descent to a lower elevation with or without injury. All falls were classified for injury severity and entered as deidentified data into a voluntary nursing measures unit-based database as a fall without injury, with minor injury (bruise, abrasion, or pain lasting 24 hours), moderate injury (laceration requiring closure), or major injury (fracture, subdural hematoma, blood products required, or death). Each patient’s fall-risk was measured using the Schmid Fall Risk Assessment Tool⁷ and cross checked against the patient’s Braden Scale, with a score less than or equal to 18 indicating a risk of a pressure ulcer. During the follow-up test period, January to December 2015, the specialized beds, video, and icons were in use in all of the studied medical-surgical units. Fall incidence rate ratios (IRR) during baseline and test periods were compared. P < .05 indicated statistical significance. 

Results: Of the 2915 falls that occurred during the study period, 20% caused patient injury. The IRR for all falls during 2015 of 0.80 reflected a 20% reduction in falls compared with baseline (P = .01). Falls resulting in serious injury decreased from 21% during baseline to 16% during 2015 (IRR = 0.60; P = .02). During baseline, there were 5 moderate and 7 major fall-related injuries compared with 1 moderate fall-related injury in 2015. The IRR for combined moderate and major fall-related injuries in 2015 was 0.15, reflecting an 85% improvement compared with baseline (P = .03). The proportion of falls observed by staff members increased from 13% during baseline to 27% during 2015, reflecting increased staff vigilance (P = .01). Though Braden risk of pressure ulcers on admission and restraint use remained unchanged throughout the study, hospital-acquired pressure ulcers decreased from 3.24 during baseline to 0.42 during 2015 (P < .001), while the proportion of patients under close nurse assistant supervision paradoxically decreased from 17.87% to 11.83% (P < .001). The icons were not widely used, but patients appreciated the fall-prevention video in their own languages and visits by volunteer assistants who spoke their language. This level of patient engagement in fall prevention plus the monitored bed-exit systems were practiced universally on all study medical-surgical units during 2015. 

Authors’ Conclusions: A brief fall-prevention educational video and use of a 3-mode bed-exit alarm effectively prevented falls, independently of recognized fall-risk factors such as patient age, gender, or morbidity.

Falls occur when patient, environment, and caregiver characteristics increase their likelihood. An unattended patient on pain medication in dire need of a toilet and without access to walking assistance is a fall waiting to happen. Groggy patients may not realize how important it is to wait for assistance. Family, friends, or trained volunteers or well-attended bed-exit alarms can make the difference between a serious fall injury and a normal recovery. These 2 studies^5,6 suggest patient engagement is a key aspect of fall prevention in hospitals. In the prospective observational study,⁶ patient engagement was fostered by volunteer assistants speaking the patient’s language while engaging them in a fall-prevention video in their language. It is not possible to gauge what portion of the effect is due to the bed-exit alarm and what portion is due to the educational video or the multilingual volunteers who administered it and encouraged patient cooperation. In the RCT,⁵ cognitive impairment obscured the effects of an otherwise effective fall-prevention program. Other patient conditions that impede patient capacity to understand educational materials or carry out measures to prevent hospital falls include taking 2 or more medications that increase fall risk, such as cardiovascular agents, central nervous system drugs, analgesics, or endocrine drugs,⁸ or having severe Parkinson’s disease.⁹ Hospital settings offer special challenges for falls research. Shared accountability allows a single breach of vigilance in the fall-prevention chain to undo rigorous care by other patient care team members. Multiple facets of care that prevent falls spread responsibility across specialties, making cost-effectiveness studies difficult and rarely focusing efforts where they are most needed on high-risk patients or activities. Short hospital stays make it necessary to use statistics that require large numbers of patients to reveal statistically significant efficacy or cost effectiveness of fall-prevention programs. Despite these challenges, the 2 studies summarized here highlight aspects of fall-prevention approaches that can work for hospitalized patients.