Prediction of fall risk upon admission to an inpatient stroke
unit
Usual and standard care assessment yields a feasible, useful model.
Cristin McKenna MD, PhD
Kessler Institute for Rehabilitation, Outpatient Musculoskeletal and Neurologic Rehabilitation
Affiliations:
Professor, Physical Medicine and Rehabilitation, Rutgers-New Jersey Medical School
Research Scientist, Kessler Foundation Stroke Lab
Disclosures
Dr McKenna is an employee of Kessler Institute for
Rehabilitation, a Select Medical Corporation
Dr McKenna is on the speakers bureau for Merz,
and Avanir.
No products of those companies will be discussed
in this presentation.
Learning objectives
After the presentation attendees will be able to :
Identify fall prediction instruments used in clinical care and the limitations of these instruments.
Describe barriers to comparing fall risk across populations and institutions.
Make clinical observations regarding fall cause/effect and the limitations of clinical observation.
Identify hidden disabilities in stroke patients that contribute to falls and which impact rehabilitation outcomes
Scope of the problem… even
before they come to rehab
•Falls in outpatients
•Fall morbidity and mortality
•Fall as cause of hospitalization
Risk factors for falls
Clinical Assessment of Fall Risk
History
prior history of falls
Medications
Diuretics, sedatives, anti-hypertensives,
new medications
Physical Exam
gait, tone, cognition
Fall Prediction Instruments
The imperative for improvement
No Gold Standard
Many instruments exist
Need for disease specific instruments?
Definition of fall (poll)
1. lose one's balance and collapse
2. an event which results in a person coming to rest inadvertently on the ground or floor or other lower level
3. An unplanned descent to the floor (or extension of the floor, e.g., trash can or other equipment) with or without injury.
4. Unintentionally coming to the ground or some lower level and other than as a consequence of sustaining a violent blow, loss of consciousness, sudden onset of paralysis as in stroke or an epileptic seizure
5. any unintentional descent to the floor or other horizontal surface
Definition of Fall
seniors most frequently associated ‘‘loss of balance’’ with falling
When defining falls, health care providers mainly talked about the consequences of falling (injury, health, andanatomical landing point)
Our Definition: An unplanned descent with or without injury.
Safety & Quality Aspects
Risk identification
Prevention
Clinical Impact
Financial/Legal impact
How and where are falls and fall
risk data recorded?
Therapy notes
Nurse notes
Physician notes
Coding in ICD-9
Incident Reports
Team meetings
EMR systems
Clinical Fall Prediction Instruments Name Sensitivity (%) Specificity (%)
Conley scale 77 61
Hendrich Risk Model 46 71
Clinical judgment of RN in inpatient rehab setting
88 26
MacAvoy’s Fall Risk Assessment tool
43 70
Clinical Fall Prediction Instruments Name Sensitivity (%) Specificity (%)
Berg Balance Test 77 78
Elderly Fall Screening Test
93 38
Dynamic Gait Index 85 38
Timed Get Up and Go test
87 87
Sensitivity, Specificity, PPV, NPV
Specificity: The capacity of the test to correctly identify diseased individuals in a population “TRUE POSITIVES”. The greater the the sensitivity, the smaller the number of unidentified case “false negatives”.
Specificity: The capacity of the test to correctly exclude individuals who are free of the disease “TRUE NEGATIVES”. The greater the specificity, the fewer “false positives” will be included.
Positive Predictive Value: The probability of the disease being present,among those with positive diagnostic test results
Negative Predictive Value: The probability that the disease was absent, among those whose diagnostic test results were negative
Non-diseased
cases
Diseased
cases
Test result value
or
subjective judgement of likelihood that case is diseased
Threshold
Non-diseased
cases
Diseased
cases
Test result value
or
subjective judgement of likelihood that case is diseased
Fall Risk Assessments from
therapist and Nursing
Conley Score
Fall wristband
Berg Balance Test
Clinical judgement
Conley Score
Scoring: Score of 2 points or greater, or a fall during
hospitalization should initiate fall prevention
strategies
The Conley Scale is a 10-point scale that includes the
following:
History: On admission, history of falling in last 3 months (2
points)
Observation: Impaired judgment/lack of safety awareness (3
points)
Agitation (2 points)
Impaired gait, shuffle/wide base, unsteady walk (1 point)
Direct: Do you ever experience dizziness or vertigo? (1 point)
Questions:Do you ever wet or soil yourself on way to the
bathroom? (1 point)
Berg Balance
BERG Patient Name: ____________________________
BALANCE Rater Name: ____________________________
SCALE Date: ____________________________
Balance Item Score (0-4)
1. Sitting unsupported _______
2. Change of position: sitting to standing _______
3. Change of position” standing to sitting _______
4. Transfers _______
5. Standing unsupported _______
6. Standing with eyes closed _______
7. Standing with feet together _______
8. Tandem standing _______
9. Standing on one leg _______
10. Turning trunk (feet fixed) _______
11. Retrieving objects from floor _______
12. Turning 360 degrees _______
13. Stool stepping _______
14. Reaching forward while standing _______
Berg balance Scoring
TOTAL (0–56): _______
Interpretation
0–20, wheelchair bound
21–40, walking with assistance
41–56, independent
Role of Tone in Falls
Spasticity (Ashworth, pectoral muscle
reflex)
Paratonia (Kral’s modified test of
paratonia)
Parkinsonism (Pull Test)
Modified Ashworth Scale for
grading Spasticity Grade
Description
0 No increase in muscle tone
1 Slight increase in muscle tone, manifested by a catch and release, or by minimal resistance at the end of the range of motion when the affected part(s) is moved in flexion or extension
2 Slight increase in muscle tone, manifested by a catch, followed by minimal resistance throughout the remainder (less than half) of the range of movement (ROM)
3 More marked increase in muscle tone through most of ROM, but affected part(s) easily moved
4 Considerable increase in muscle tone, passive movement difficult
5 Affected part(s) rigid in flexion and extension
Cognitive Aspects of Fall Risk
Go/No Go Brain Systems 3-gesture test
Luria’s modified test of echopraxis
Aphasia
spatial neglect Physical exam, bedside pen and paper
screen
Dual task
Ecological validity
Science of Fall Prediction
Current methods:
fall wristband
Standardized instrument
Clinical Judgment
Are there separable items which can be used to predict fall risk?
Standardized Instruments: Why use them?
KEY ELEMENTS FOR ASSESSING RISK OF FALLS
AND MANAGING PATIENTS AT RISK
Step 1: Ideally for all patients but especially those with gait dysfunction
HPI: Inquire about falls in past year
And
History: Stroke, Age ≥ 65 years, Dementia, Vision deficit, Arthritis, arthralgia, Parkinsonism, Depression, Peripheral neuropathy, Polypharmacy, Use of cane or walker, Other condition w/ LE Restricted ADLs, sensorimotor loss
Physical Exam
Physiatric examination, emphasizing:
balance and gait
LE strength, sensation & coordination
mental status
In addition, may consider a
standardized assessment
Hidden Disabilities in stroke
No transfer records
Walk in the room
Which brain hemisphere?
Except when it’s not
How do each of these contribute
to fall risk?
•Aphasia
•Apraxia
•spatial neglect
•Anosagnosia
Rapid bedside assessment
Observe
Ask
Examine
Management
May address:
Underlying disorder
Adjustment of medication
Exercise program
Training in gait and balance
Training in assistive device
Assessment/modification of home environment
Fracture risk: BMD, Vit D
From research to real world
From lab to bedside.
The research presented today is
supported by Select Medical and
Kessler Foundation
Background of Current
Research
Falls in stroke patients represent a significant source of morbidity. Accurate identification of patients at high risk of falling may enable appropriate distribution of fall prevention resources which will in turn enhance care quality and in turn facilitate patient recovery.
Although this problem has been approached by other research groups, there is no single clinically accepted and widely used instrument, thus further work in this area is warranted
We have developed a model that correlates factors measurable within the first 72 hours of admission. We intend to use this set of correlations to develop a predictive model of falls
Objective
To develop a model to predict falls in the inpatient stroke patient population using data stemming from admission documentation and relevant to the rehabilitation population of patients. These factors were chosen because they are routinely collected data on every stroke inpatient.
Methods
Retrospective chart analysis to
develop model
Prospective chart analysis to test
model
3112 consecutive stroke patients
undergoing inpatient acute
rehabilitation
Study Design
Pertinent data
Functional Independence Measure
(FIM)
Demographic information
Case mix group (CMG), which
incorporates the presence or
absence of comorbidities
Functional Independence Measure
(FIM)
18 Items
• Motor skills (13), Cognitive (5)
• Scale of 1 (total assist) to 7 (no assist)
• Ranges 13-91 Motor, 5-35 Cognitive
• Higher scores = Better function
• Range of total score: 18-126
FIM Subsets
•self-care (nutrition, hygiene, and
dressing)
• sphincter control, mobility (transfer to
bed, chair, and toilet)
• locomotion (walking and climbing stairs)
• communication (comprehension and
expression)
•cognitive function (social relations,
problem solving, and memory)
Results
Variables
Patients
who Fell
Patients who
did not fall Characteristic mean Std. Deviation mean Std. Deviation
age (years) 56.7 12.8 71.8 13.2
Female, n (%) 1409 (50.1) 0.5 136 (45) 0.5
AdmitFIMEating 3.6 1.6 3.7 1.8
AdmitFIMGrooming 2.8 1.3 3.0 1.5
AdmitFIMBathing 2.0 0.9 2.2 1.1
AdmitFIMDressingUpper 2.2 1.0 2.5 1.3
AdmitFIMDressingLower 1.6 0.8 1.9 1.0
AdmitFIMToileting 2.0 1.0 2.3 1.2
AdmitFIMBladderCtrl 2.3 1.5 2.6 1.7
AdmitFIMBowelCtrl 2.7 1.5 3.0 1.7
AdmitFIMBedTransfer 2.0 0.9 2.3 1.0
AdmitFIMToiletTransfer 2.0 0.9 2.2 1.1
AdmitFIMTubTransfer 1.8 1.0 2.1 1.1
AdmitFIMWalkWheelchair 1.2 0.5 1.5 0.9
AdmitFIMStairs 1.0 0.6 1.3 0.9
AdmitFIMComprehension 4.1 1.4 4.2 1.6
AdmitFIMExpression 3.9 1.7 4.0 1.8
AdmitFIMSocialInteraction 4.5 1.7 4.5 1.8
AdmitFIMProblemSolving 3.3 1.4 3.4 1.5
AdmitFIMMemory 3.7 1.5 3.7 1.6
LOS (days) 24.8 14.6 18.5 11.3
CMG 108.5 1.8 107.6 2.4
Tier 0.5 1.1 0.5 1.1
Results
• Gender M>F
• Age Younger>older
• Case mix group (CMG) and Tier assignment
Complexity level
Significant FIM scores
Admission FIM scores which contribute in a statistically
significant manner to the prediction:
• eating
• bed transfer
• stairs
• walking and wheelchair ability
• bowel accident frequency
Effect of FIM on falls
The admission FIM score subsets which contributed in a statistically significant manner to the fall-correlation model were eating, stairs, and toilet transfer.
The high performing to low performing comparison for eating revealed a hazard ratio of 1.15 hazard ratio0.016;
for stairs hazard ratio was 0.691, p-value 0.01;
for toileting hazard ration was 1.3, p-value 0.06
Factors Correlated with Falls
Gender was correlated with falls, with males being more likely to fall. The female vs. male hazard ratio was 0.727, p-value 0.04.
Older patients’ fall risk appeared to be lower than younger patients with patients in the group age 63 or less being more likely to fall than those older than 63. The age hazard ratio was 0.982, p-value 0.003.
CMG which incorporates the presence or absence of co morbidities was also correlated with falls. Increasing burden of CMG had a hazard ratio of 1.3, p-value of 0.0001
Select Medical Innovative Fall
Predictor for Stroke Inpatients
SeMIFall
Sensitivity (60%)
Specificity (58%)
Positive predictive value (13.8%)
Negative predictive value (93.9%)
Patients identified to be high fall risk
who fell
(true positives)
75
Patients identified to be high fall risk
who didn’t fall (false positives)
467
Falls that happened in
patients who were not identified as
high fall risk (false negatives)
50
Patients who did not fall and who were correctly
identified to be not at risk for falling (true negatives)
657
What is SeMIFall?
The SeMIFall is a novel fall prediction
instrument based on subsets of FIM,
demographics, and medical stratification
systems such as case mix groups and
tiers. Further work in this area will seek
additional factors to improve the
sensitivity and specificity of this test.
Implications for Practice
Our results suggest that stroke patients
who fall during their inpatient stay can be
identified within the first 72 hours of
admission to acute rehabilitation.
This instrument uses no additional time
beyond routine clinical assessments
Prediction of fall risk upon admission to an
inpatient stroke unit: usual and standard care
assessment yields a feasible, useful model.
Cristin McKenna MD, PhD1, Joan Alverzo PhD, CRRN1, Jeffrey Yangang Zhang PhD2, Peii Chen PhD2, Pasquale G. Frisina, PhD1, Annie Kutlik1, Mooyeon Oh-Park MD2, Mahesh Lellella MBBS2, A.M. Barrett, MD2
1.Select Medical Corporation - Kessler Institute for Rehabilitation, 1199 Pleasant Valley Way, West Orange NJ 07052
2.Kessler Foundation, 1199 Pleasant Valley Way, West Orange NJ 07052
Types of falls (poll)
accidental falls
anticipated physiologic falls
unanticipated physiologic falls
extrinsic
Hospital-Acquired Conditions Inpatient Prospective Payment System (IPPS) Fiscal Year (FY) 2009 Final
Rule, CMS included 10 categories of conditions that were selected for the
HAC payment provision.
The 10 categories of HACs include:
1. Foreign Object Retained After Surgery
2. Air Embolism
3. Blood Incompatibility
4. Stage III and IV Pressure Ulcers
5. Falls and Trauma 1. Fractures
2. Dislocations
3. Intracranial Injuries
4. Crushing Injuries
5. Burns
6. Electric Shock
6. Manifestations of Poor Glycemic Control
7. Catheter-Associated Urinary Tract Infection (UTI)
8. Vascular Catheter-Associated Infection
9. Surgical Site Infection Following: CABG, Bariatric Surgery, Orthopedic Procedures
10. Deep Vein Thrombosis (DVT)/Pulmonary Embolism (PE):
Total Knee Replacement, Hip Replacement
https://www.cms.gov/hospitalacqcond/06_hospital-acquired_conditions.asp
Affected Hospitals
The Present on Admission (POA) Indicator requirement and Hospital-Acquired Conditions (HAC) payment provision only apply to Inpatient Prospective Payment Systems (IPPS) Hospitals.
At this time, the following hospitals are exempt from the POA Indicator and HAC:
Critical Access Hospitals (CAHs)
Long-term Care Hospitals (LTCHs)
Maryland Waiver Hospitals
Cancer Hospitals
Children's Inpatient Facilities
Rural Health Clinics
Federally Qualified Health Centers
Religious Non-Medical Health Care Institutions
Inpatient Psychiatric Hospitals
Inpatient Rehabilitation Facilities
Veterans Administration/Depart of Defense Hospitals
Which types of falls are we held
responsible for?
Community fall risk
CMS perspective on falls
Unintended consequences
Fall Risk Perspectives
Within disease
Within institution
Individual
Summary
Clinical: How to identify and manage
fall risk in your patients
Research: Impact of prediction of fall
risk on function
Standardized Scales
Hidden disabilities
Alarm Fatigue
Alarm Fatigue
Relationship to falls
Avoiding adverse event
Improving outcomes
Personalized therapy plan
Helpful resources
http://www.rehabmeasures.org
http://www.aan.com/go/practice/guideli
nes
http://www.strokecenter.org/profession
als/
http://www.cdc.gov/HomeandRecreatio
nalSafety/Falls/index.html
REFERENCES
Aizen, E., I. Shugaev, et al. (2007). "Risk factors and characteristics of falls during inpatient rehabilitation of elderly patients."
Archives of Gerontology and Geriatrics 44(1): 1-12.
Ashburn, A., D. Hyndman, et al. (2008). "Predicting people with stroke at risk of falls." Age and Ageing 37(3): 270-276.
Buella, C. J., E. Martin, et al. (2008). "Validation of an adapted Falls Efficacy Scale in older rehabilitation patients." Archives of
Physical Medicine and Rehabilitation 89(2): 291-296.
Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Web–based Injury Statistics
Query and Reporting System (WISQARS). November 30, 2010.
Divani, A. A., G. Vazquez, et al. (2009). "Risk Factors Associated With Injury Attributable to Falling Among Elderly Population
With History of Stroke." Stroke 40(10): 3286-3292.
Englander, F., T. J. Hodson, et al. (1996). "Economic dimensions of slip and fall injuries." Journal of Forensic Sciences 41(5):
733-746.
Forster, A. and J. Young (1995). "Incidence and consequences of falls due to stroke – a systematic Inquiry." British Medical
Journal 311(6997): 83-86.
Gilewski, M. J., P. Roberts, et al. (2007). "Discriminating high fall risk on an inpatient rehabilitation unit." Rehabilitation
Nursing 32(6): 234-240.
Gillen, R., H. Tennen, et al. (2005). "Unilateral spatial neglect: Relation to rehabilitation outcomes in patients with right
hemisphere stroke." Archives of Physical Medicine and Rehabilitation 86(4): 763-767.
Granek, E., S. P. Baker, et al. (1987). " Medications and diagnoses in relation to falls in a long-term care." Journal of the
American Geriatrics Society 35(6): 503-511.
Jorgensen, L., T. Engstad, et al. (2002). "Higher incidence of falls in long-term stroke survivors than in population controls -
Depressive symptoms predict falls after stroke." Stroke 33(2): 542-547.
Kalra, L., I. Perez, et al. (1997). "The influence of visual neglect on stroke rehabilitation." Stroke 28(7): 1386-1391.
Kim, E. A. N., S. Z. Mordiffi, et al. (2007). "Evaluation of three fall-risk assessment tools in an acute care setting." Journal of
Advanced Nursing 60(4): 427-435.
Lee, J. E. and D. S. Stokic (2008). "Risk factors for falls during inpatient rehabilitation." American Journal of Physical Medicine
& Rehabilitation 87(5): 341-350.
Lovallo, C., S. Rolandi, et al. (2010). "Accidental falls in hospital inpatients: evaluation of sensitivity and specificity of two risk
assessment tools." Journal of Advanced Nursing 66(3): 690-696.
Mackintosh, S. F., K. D. Hill, et al. (2006). "Balance score and a history of falls in hospital predict recurrent falls in the 6
months following stroke rehabilitation." Archives of Physical Medicine and Rehabilitation 87(12): 1583-1589.
Maeda, N., J. Kato, et al. (2009). "Predicting the Probability for Fall Incidence in Stroke Patients Using the Berg Balance
Scale." Journal of International Medical Research 37(3): 697-704.
Nyberg, L. and Y. Gustafson (1995). "Patient falls in stroke rehabilitation – a challenge to rehabilitation strategies." Stroke
26(5): 838-842.
Nyberg, L. and Y. Gustafson (1997). "Fall prediction index for patients in stroke rehabilitation." Stroke 28(4): 716-721.
Paolucci, S., G. Antonucci, et al. (2001). "The role of unilateral spatial neglect in rehabilitation of right brain-damaged ischemic stroke
patients: A matched comparison." Archives of Physical Medicine and Rehabilitation 82(6): 743-749.
Paolucci, S., G. Antonucci, et al. (1996). "Predicting stroke inpatient rehabilitation outcome: The prominent role of neuropsychological
disorders." European Neurology 36(6): 385-390.
Petitpierre, N. J., A. Trombetti, et al. (2010). "The FIM (R) instrument to identify patients at risk of falling in geriatric wards: a 10-year
retrospective study." Age and Ageing 39(3): 326-331.
Ramnemark, A., L. Nyberg, et al. (1998). "Fractures after stroke." Osteoporosis International 8(1): 92-95.
Rapport, L. J., J. S. Webster, et al. (1993). "Predictors of falls among right-hemisphere stroke patients in the rehabilitation setting."
Archives of Physical Medicine and Rehabilitation 74(6): 621-626.
Rizzo, J. A., R. Friedkin, et al. (1998). "Health care utilization and costs in a Medicare population by fall status." Medical Care 36(8):
1174-1188.
Robertson, M. C., N. Devlin, et al. (2001). "Economic evaluation of a community based exercise programme to prevent falls." Journal
of Epidemiology and Community Health 55(8): 600-606.
Ruchinskas, R. (2003). "Clinical prediction of falls in the elderly." American Journal of Physical Medicine & Rehabilitation 82(4): 273-
278.
Saverino A., E. Benevolo, et al. (2006). “Falls in a rehabilitation setting: functional independence and fall risk.” Eura Medicophys
42(3): 179-184.
Stevens, J. A., P. S. Corso, et al. (2006). "The costs of fatal and non-fatal falls among older adults." Injury Prevention 12(5): 290-295.
Stone, S. P., P. W. Halligan, et al. (1993). "The incidence of neglect phenomena and related disorders in patients with an acute right
or left-hemisphere stroke." Age and Ageing 22(1): 46-52.
Suzuki, T., S. Sonoda, et al. (2005). "Incidence and consequence of falls in inpatient rehabilitation of stroke patients." Experimental
Aging Research 31(4): 457-469.
Teasell, R., M. McRae, et al. (2002). "The incidence and consequences of falls in stroke patients during inpatient rehabilitation:
Factors associated with high risk." Archives of Physical Medicine and Rehabilitation 83(3): 329-333.
Tinetti, M. E. and M. Speechley (1989). "Prevention of falls among the elderly." New England Journal of Medicine 320(16): 1055-1059.
Tsur, A. and Z. Segal (2010). "Falls in Stroke Patients: Risk Factors and Risk Management." Israel Medical Association Journal 12(4):
216-219.
Tutuarima, J. A., R. J. Dehaan, et al. (1993). "Number of nursing staff and falls – a case-control study on falls by stroke patients in
acute-care settings." Journal of Advanced Nursing 18(7): 1101-1105.
Tutuarima, J. A., J. H. P. vanderMeulen, et al. (1997). "Risk factors for falls of hospitalized stroke patients." Stroke 28(2): 297-301.
Ugur, C., D. Gucuyener, et al. (2000). "Characteristics of falling in patients with stroke." Journal of Neurology Neurosurgery and
Psychiatry 69(5): 649-651.
Vassallo, M., R. Stockdale, et al. (2005). "A comparative study of the use of four fall risk assessment tools on acute medical wards."
Journal of the American Geriatrics Society 53(6): 1034-1038.
Vlahov, D., A. H. Myers, et al. (1990). "Epidemiology of falls among patients in a rehabilitation hospital." Archives of Physical Medicine
and Rehabilitation 71(1): 8-12.
Wagner, L.M., V.L. Phillips, et al. (2009). “Falls among community-residing stroke survivors following inpatient rehabilitation: a
descriptive analysis of longitudinal data.” BMC Geriatrics Oct 14; 9: 46.
Webster, J. S., L. A. Roades, et al. (1995). " Rightward orienting bias, wheelchair maneuvering and fall risk." Archives of Physical
Medicine and Rehabilitation 76(10): 924-928.
Cristin McKenna MD, PhD [email protected]
Kessler Institute for Rehabilitation, Outpatient
Musculoskeletal and Neurologic Rehabilitation
Professor, Physical Medicine and Rehabilitation
Rutgers-New Jersey Medical School
Research Scientist, Kessler Foundation Stroke Lab
What to do ?!?!
Open floor for
discussion