The purpose of this study is to evaluate the effect that a new fall prevention training program has on the fall incidence of long-term care facility residents at high-risk of falling.

For Americans age 65 and older, falls are the leading cause of fatal and nonfatal unintentional injury. Although they only comprise about 5% of the elderly population, 20% of all fall-related elderly deaths are of continuous-care facility residents. In the five years after a fall, 75% of fallers in institutional care will die, while only 58% of age-matched non-fallers will die. The increased risk observed in such facilities is a product of a high fall incidence (26-56%) and fall rate (1.4-2.0 falls per person or bed). The incidence of recurrent falling in this population is about 33-40%, with 28% of fallers having five or more falls. Roughly 12-33% of falls result in serious injuries such as fracture, head trauma, soft-tissue injury, or severe lacerations. About 1-4% of falls result in fracture, an injury which comprises the majority (75%) of fall-related admissions from continuous-care facilities to hospitals. Such hospital admissions cost over $31,000 per admission, which contributes to an estimated medical cost of about $1,200 per fall. The proposed research is significant because it addresses the high incidence of fall-related injury and death, in turn reducing notable medical costs.

Based on a rigorous review of more than 1200 participants, supervised exercise has not significantly reduced the rate of falls (rate ratio: 0.74-1.35) or number of fallers (risk ratio: 0.88-1.21) in continuous-care facilities15. The most effective exercises focus on control of upright posture and the response to small gait disturbances (rate ratio: 0.24-0.85, risk ratio: 0.43-1.19). However, the efficacy of these interventions can be improved by safely increasing the magnitude of disturbances (i.e. greater intensity) and focusing on the skills necessary to recover from common fall causes (i.e. greater specificity). Compensatory step training safely induces a fall by delivering large postural disturbances, in turn requiring subjects to arrest a fall through timely, well-placed steps and reduced trunk rotation. Similar demands on step placement and trunk control are evident when recovering from a trip or slip, the most common fall-causes of older adults in the community. Obstacles and slippery surfaces are prevalent causes of falls within care facilities suggesting that trips and slips are also problematic in this environment. For community-dwelling older women, compensatory step training has significantly improved step and trunk kinematics and reduced falls due to trips in the laboratory (odds ratio = 0.13). Furthermore, preliminary evidence suggests that training reduces their fall incidence by 17% in the community. Published and unpublished data have shown that compensatory step training has improved the compensatory stepping response, undisturbed gait, and fall incidence of those with lower extremity amputations. It is likely that such training can benefit the high fall incidence observed in care facilities. The proposed research is innovative because it is a novel application of a state-of-the-art intervention that will likely improve the efficacy of supervised exercise as a means to reduce falls in continuous-care facilities.

A widely implemented prevention strategy in institutional settings is to target residents at high risk of falling. As assessed by subjective measures or fall history, the relative fall risk of high-risk residents is up to 3 times higher than that of low-risk residents. The fall incidence of low-risk residents, as determined by an objective functional test, has not been reported. Recording these statistics would provide perspective on this subpopulation relative to community-dwelling older adults, establishing a benchmark for evaluating the long-term effect of interventions for high-risk residents. Identifying the short-term, functional improvements associated with such interventions may reveal the mechanisms of training-based improvements, informing protocol development and improving its efficacy.

The primary objective is to evaluate the effect that compensatory step training has on the fall incidence of long-term care facility residents at high risk of falling. The investigators hypothesize that compensatory step training will prospectively reduce the fall incidence of high-risk residents to a level not inferior to the fall incidence of low-risk residents.

The study is a non-inferiority trial intended to show that the training of high-risk residents yields a fall incidence not worse than that of low-risk residents. Although such trials are less credible than other designs, the study demands fewer resources.

Functional assessments of fall risk will occur at the care facility. These assessments include the functional reach test, unipedal stance test, tandem stance test, figure-eight walking test, time-up-and-go test, step-up test, Activities-specific Balance Confidence questionnaire, grip strength test, maximum voluntary step length test, and fall history assessment.

Based on the results of the functional reach test and unipedal stance test, subjects will be identified as low-risk (functional reach > 8" AND unipedal stance > 5 s) or high-risk (functional reach ≤ 8" OR unipedal stance ≤ 5 s). Those at low risk of falling and those at high risk of falling who do not qualify for training will have their falls recorded for one year. Those at high risk of falling who qualify for training will undergo compensatory step training, undergo a second set of functional assessments, and then have their falls recorded for one year.

For high-risk subjects, health screenings will be assessed by a medical physician or nurse practitioner. If the subject does not meet the exclusion criteria, then no training will take place and the subject will have their falls monitored for one year. For high-risk subjects who otherwise pass all exclusion criteria, hip bone mineral density will be assessed using DXA in the Mayo Clinic Charlton Clinical Research Unit. If the subject does not meet the bone mineral density exclusion criterion (femoral neck or total hip bone mineral density t-score less than -2.5), then no training will take place and the subject will have their falls monitored for one year. If a DXA scan is of poor quality (e.g. there is artifact from subject movements), it will be repeated. At most, two DXA scans will be taken for each subject.

Subjects will participate in compensatory step training for up to 6 sessions in 7 to 30 days. Each session is comprised of anterior and posterior step training on a microprocessor-controlled treadmill (ActiveStep®, Simbex, Lebanon, NH). Training consists of a progression of anterior or posterior treadmill belt movements applied as the subject is standing or walking. The training intensity (magnitude of disturbance delivered) is progressive and dependent on subject performance. Subjects will be instructed to respond with single or multiple steps in order to prevent a fall. All subjects will be outfitted with a safety harness to prevent injury. This training specifically focuses on aspects important to trip and slip recovery. Up to 72 disturbances will be delivered each session.

Falls will be recorded for one year as per facility operations and with fall-tracking questionnaires.

Adults 50 years or older living in continuous-care retirement facilities will be recruited for this study. The investigators anticipate that 150 subjects will participate in the study, with 30 of those subjects participating in the compensatory step training.

Inclusion Criteria:

• Subjects must be 50 years or older, living in continuous-care retirement facilities.

Exclusion Criteria:

1. For participation in study:

   • Three or more errors on the Six-Item Screener for Cognitive Impairment
   • Individuals who cannot provide consent for themselves

2. For participation in compensatory step training:

   • Physician or nurse practitioner approval to participate in the study with the guidelines that subjects should not participate if any of the following is applicable:
   • Unable to walk one block without stopping or using a walking aid (self-reported)
   • Acute illness at the time of functional assessment or training
   • Body mass greater than 114 kg
   • Body size too large for wearing a safety harness
   • Lower extremity joint replacement within a year prior to participation
   • Femoral neck or total hip bone mineral density t-score less than -2.5
   • Based on a dual-energy X-ray absorptiometry (DXA) scan in the Mayo Clinic Charlton Clinical Research Unit (CRU), with no medication changes that may affect bone mineral density at the hip (e.g. starting prednisone). During the visit to the CRU, the subject's height and weight will be measured.
   • If the subject has had a hip DXA scan within 6 months of participation, the exclusion criterion will be based on the previous scan.
   • If the subject has had bilateral hip replacement, then the subject will not qualify for compensatory step training.
   • Dementia
   • Parkinson's disease
   • A history of stroke
   • A history of back surgery
   • Bulging vertebral discs
   • Spine, hip, or lower extremity fracture within a year prior to participation
   • Open lesions on the lower extremity
   • In case of lower extremity amputation, stump volume fluctuations within a year of training
   • Use of a pacemaker
   • Use of an ostomy pouch
   • Pregnancy
   • Doctor recommendation to avoid moderate physical activity or exercise
   • Any neural, muscular, or skeletal condition or injury that precludes safe participation in the training, at the discretion of the clinician or study staff.