Fall prevention in older people: past, present and future

Abstract

Over the past 50 years we have transitioned from accepting falls as an inevitable consequence of ageing to something that can and should be prevented. Numerous studies have elucidated the contributors to falls and how to assess a person’s risk of falling. There are many effective approaches to preventing falls in older people including those with cognitive and physical impairments. Exercise is the most tried and tested approach with good evidence that moderate to high intensity balance training is an effective fall prevention strategy. Other successful single modality interventions include enhanced podiatry, home safety interventions, expedited cataract extraction, cardiac pacing for people with carotid sinus hypersensitivity and vitamin D supplementation in people living in care homes. Multiple interventions (everyone receives the same intervention package) and multifactorial interventions (interventions tailored to identified risk factors) are effective particularly in high-risk populations. In more recent years we have seen the emergence of new technologies such as devices and software programs that can offer low-cost interventions which may be more sustainable than our traditional time- and resource-limited approach to prevention. There is still more to be done and a translational focus is needed to ensure that effective interventions are scaled up and delivered to more people while at the same time maximising adherence and maintaining the fidelity of the interventions.

falls, injury, prevention, aged, older people

Key Points

• There is a wealth of evidence to support interventions to prevent falls.

• Sustainable models to prevent falls must reflect the evidence and be adequately resourced.

• New and emerging technologies have a role in fall detection and prevention.

Introduction

‘The liability of old people to tumble and often injure themselves is such a commonplace of experience that it has been tacitly accepted as an inevitable aspect of ageing and thereby deprived of the exercise of curiosity.’ Joseph Sheldon, BMJ 1960 [1].

It was late in the career of British physician and geriatrician Joseph Sheldon that the issue of falls in older people was given prominence. In his landmark BMJ paper in 1960, he decried the dearth of research into what was a common problem for older people and reported on the nature of 500 falls in 202 older people living in the community [1]. In detail he outlines the events leading up to some of the falls including a drop attack in an 86-year-old woman peeling potatoes in her scullery who found herself on the floor after experiencing the ‘unrestrained pull of gravity’. The 500 falls were assigned to categories as outlined in Table 1.

Table 1

Summary of causes of falls in Joseph Sheldon’s BMJ paper, 1960

Accidental falls	171
Drop attacks	125
Trips	53
Vertigo	37
Bad back	20
Posture hypotension	18
Weakness in leg	16
Falling out of bed or chair	10
Uncertain	23
Total	500

Despite Sheldon’s lament, nearly two decades passed before new findings were published on the frequency and causes of falls. In 1977, Norman Exton-Smith (UK) published survey data from 963 people over the age of 65 years and reported incidence rates of falls for men and women in 5-year categories, and documented trips, giddiness, loss of balance and drop attacks as important causes of falls [2]. Since then, there has been a proliferation of research into both the factors predisposing a person to fall as well as strategies to prevent falls in different populations and settings. Many of these studies have been published in Age and Ageing and as we celebrate 50 years of the journal it is important to synthesise the research findings, reflect on lessons learned and consider opportunities for the future.

Risk factor identification

The first phase of research into the causes of falls comprised more systematic ascertainment of risk factors for falls. This initially involved retrospective studies conducted between 1977 and 1988 by a number of eminent geriatricians including Peter Overstall (UK), John Grimley Evans (UK), John Campbell (NZ) John Brocklehurst (UK) and Bernard Isaacs (UK). [3] This was followed by large prospective studies conducted between 1988 and 1995 using regular follow-up for falls and undertaken in more representative samples by researchers including Mary Tinetti (US), John Campbell (NZ), Michael Nevitt and Steven Cummings (US) Jennifer O’Loughlin (Canada), Stephen Lord (Australia) and Heikki Luukinen (Finland) [3]. This and subsequent research allowed the compilation of key fall risk factors in older people as outlined in Table 2.

Table 2

Risk factors for falls

Domain	Risk factor	Association
Psychosocial and	Advanced age	***
demographic factors	Female gender	**
	Race	***
	Living alone	**
	History of falls	***
	Inactivity	**
	Walking aid use	**
	Sleep disturbances	***
	ADL limitations	***
	Alcohol consumption	-
Medical factors	Stroke	***
	Parkinson’s disease	***
	Dementia	***
	White matter lesions	**
	Depression	***
	Incontinence	**
	Acute illness	**
	Vestibular disorders	*
	Arthritis	**
	Foot problems	**
	Dizziness	**
	Orthostatic hypotension	*
Medication factors	Psychoactive medications	***
	Antihypertensives	*
	Antiarrhythmics	*
	Opioids	**
	Anti-inflammatories	-
	Use of 4+ medications	***
Balance and mobility	Impaired stability when standing	**
factors	Impaired stability when leaning and reaching	***
	Inadequate responses to external perturbations	***
	Slow voluntary stepping	***
	Impaired gait and mobility	***
	Impaired ability in standing up	***
	Impaired ability with transfers	***
Sensory and	Poor visual acuity	**
neuromuscular factors	Reduced visual contrast sensitivity	***
	Visual field loss	**
	Poor hearing	*
	Reduced peripheral sensation	***
	Reduced vestibular function	*
	Muscle weakness	***
Psychological factors	Impaired executive functioning	***
	Reduced processing speed	***
	Impaired selective attention	**
	Anxiety	**
	Fear of falling	***
Environmental factors	Poor footwear	*
	Inappropriate spectacles	*
	Home hazards	*
	External hazards	-

^***strong evidence (consistently found in high-quality studies);

^**moderate evidence (usually but not always found);

^*weak evidence (occasionally but not usually found); and—little or no evidence (not found in published studies despite research to examine the issue). Adapted from Lord SR et al. [3].

Interventions to prevent falls

Armed with a raft of potentially modifiable risk factors, the logical next stage in the journey was to determine whether falls could be prevented by addressing one or more risk factors in randomised controlled trials.

The first breakthrough was made in 1994. Tinetti et al. [4] found that by identifying risk factors for falls and implementing tailored intervention strategies, falls could be reduced by 31% in a community dwelling population of people aged 70 years and older. One of the many strengths of this study was the engagement and integral involvement of the general practitioner and home care providers from the outset. This is something that is not the case in many of the intervention studies that have followed and has a significant impact on the generalisability and sustainability of successful intervention programs. Tinetti et al. [5] subsequently implemented their program on a large scale in areas of Connecticut, US, again ensuring that general practitioners and home care services were integral to the intervention. Using neighbouring geographical regions where the program was not rolled out, they demonstrated a beneficial impact of the intervention on rates of serious fall-related injuries and medical service use following a fall.

Soon after Tinetti’s work came the UK-based PROFET trial which targeted people presenting to the emergency department after a fall. Intervention participants underwent comprehensive geriatric assessment in a Day Hospital setting, usually within 2 weeks of the fall, and offered interventions based on identified risk factors and an occupational therapy home assessment and intervention [6]. The intervention was effective in reducing both the risk and rate of falls in this high-risk population and led to the establishment of falls clinics in the UK and beyond; an acknowledgement that falls were a major issue and warranted specialist attention. However, the impact of the approach taken in PROFET has never been adequately replicated and personal reflections as to why PROFET was successful likely relate to the timing (fall prevention was not considered for these people in ED in the 1990s), early assessment (patients were seen in close proximity to their fall and were more engaged with potential prevention strategies) and action focused whereby the interventionists were proactive in ensuring that suggested interventions were implemented. Many other trials that have relied on referral pathways or others to implement interventions within existing services have failed to reduce rate of falls, serious fall injuries and fractures [7, 8].

The alternate approach to fall prevention has been to implement an intervention to address a single risk factor. Exercise has been the intervention modality most studied and there is now a wealth of evidence to tell us that exercise can be effective in preventing falls in older people living in the community. Systematic review evidence indicates exercise must challenge balance and needs to be undertaken in sufficient quantity (dose and duration) to be effective [9]. While few would doubt the efficacy of exercise in preventing falls, it is less clear how to engage older people in effective programs of exercise that extend beyond an arbitrarily defined endpoint determined by the design of time-limited research studies. Exercise needs to be embedded in people’s daily/weekly routines and for most older people does not need to be delivered in a health care setting.

In higher risk populations, caution is needed in interpreting the literature regarding the evidence for exercise as a fall prevention strategy. For example, the evidence to support exercise as a single intervention in residential aged care facilities is not compelling [10], suggesting our approach in this high-risk population needs more consideration. People with significant physical and/or cognitive impairment may need more intensive and supervised exercise—something that is not routinely provided in this setting. Two studies are worthy of note here. Firstly, the FINALEX trial which targeted community dwelling people with Alzheimer’s disease and offered an exercise intervention to maintain physical function [11]. The intervention was delivered by physiotherapists and comprised twice weekly exercise of 60 min duration for 1 year either in the participants home (one on one) or group setting. In the SUNBEAM trial, older people in residential aged care facilities underwent twice weekly exercise of 60 min duration consisting of progressive resistance training and balance training for 25 weeks followed by a maintenance period of a further 6 months [12]. In both studies the rate of falls was lower in the intervention group, but it is less clear if these intensive interventions are generalisable, sustainable or cost-effective in routine settings. Fragmented funding models and a limited grasp of the bigger picture in many countries also creates problems with large-scale investment whereby the potential benefits of an intervention are realised by a different part of the health or social care system.

Several single non-exercise interventions targeted to people with particular risk factors have also been found to prevent falls in trials [3]. There is evidence to support the following: a multifaceted podiatry intervention in people with disabling foot pain; insertion of a cardiac pacemaker in people with cardioinhibitory carotid sinus hypersensitivity; cataract removal in those with operable cataracts; gradual reduction of centrally acting medications; GP-based medication reviews, vitamin D supplementation in those who are vitamin D deficient, replacement of multifocal glasses with single lens glasses for those who regularly walk outdoors and occupational therapy fall prevention interventions in high-risk people.

Optimal approaches to the prevention of falls in hospitals and residential care facilities are less clear. In hospitals, multifaceted programs have been found to prevent falls in longer stay settings but risk factor screening interventions have not been found to work in large trials in acute hospital settings [13]. The most promising interventions focus on communication between patients and staff about the risk of falls and safe mobility on the ward particularly when using the bathroom. There is evidence that vitamin D can prevent falls in residential care, probably because levels are low among residents [10]. A well-supported multifaceted program from Germany that included exercise can prevent falls [14] but has not been replicated in other settings when implemented with fewer resources. Two recent trials involving the provision of protein and nutrition supplements [15] and multifactorial interventions involving extensive staff training have added to the evidence base for effective strategies for preventing falls in residential care [16].

Advances in past 10 years and future directions

The last decade has seen new technologies emerge as potential strategies for fall identification and prevention [17]. Several studies have demonstrated it is feasible to monitor activity in older people living at home with wearable sensors. Applications developed for smart phones can accurately perform long-term activity monitoring [18], offering scope for automated fall detection and incorporation of remote fall assessments into clinical care. However, wearable fall detection devices are still not sufficiently reliable, primarily due to their inability to distinguish falls from other activities. The research into remote fall risk assessments has likely reported over-optimistic results due to small samples, overfitting and lack of external validation. The derived models are therefore unlikely to work well in everyday use or provide useful prognostic tools [19]. Future prospective studies are needed with designs that permit external validation.

Interactive, exercise-based videogames (exergames) may comprise an effective means for delivering multimodal exercise. Exergames combine player movement, enjoyment and performance feedback; factors that may promote adherence to exercise in older people. A systematic review conducted in 2014 identified 37 trials that investigated a variety of exergames (Wii balance boards, dance mats and virtual reality systems) in clinical trials in older people. Most were of a pilot nature, but most reported exergame training improved physical (e.g. balance and strength) and cognitive (e.g. attention, executive function) measures [20]. More recently, with the Standing Tall program we have seen the benefits of using iPAD technology to remotely implement an exercise program that considers participant preferences, progresses exercise and provides inbuilt monitoring of performance over time [21].

While much has been achieved in preventing falls and fall-related injury over the last 50 years, there is still much more to do. More work needs to be undertaken to maximise the uptake of interventions that have been shown to be effective. Many of the trials have been time and funding limited and there has been insufficient investment in implementation and scale-up of interventions. Research now needs to establish the most successful approaches to implementation. For example, only a small proportion of community living older people regularly participate in the types of exercise found to prevent falls. Strategies to address this important evidence to implementation gap may include health professionals encouraging their clients to be more active, systems to support health professionals to refer patients to suitable community exercise opportunities and support of community organisations to deliver suitable exercise opportunities.

Where scale-up has occurred there have been challenges with fidelity and a tendency to modify interventions based on resource availability rather than science. That said—generalisability and sustainability necessitate better alignment with our existing health and social care models as well as a good understanding of the barriers and enablers for participant uptake and adherence. A siloed approach to preventing falls is unlikely to be the way forward and embracing the concepts of preventing and reversing frailty and preserving function and independence is likely to yield better dividends both for older people and the health system.

Finally, it should not go unnoticed that the majority of risk factor and intervention studies have been undertaken in high-income countries with well-developed health care systems. At a global level this is not where the greatest challenge lies going forward and there is an urgent need to implement fall prevention interventions aligned with frailty and preserving function in low- and middle-income settings. While it is likely that the main principles behind fall prevention strategies such as exercise will apply across countries, the design and delivery of programs will need refinement to ensure the programs are appropriate and affordable in different settings.

Acknowledgements

To all the researchers and clinicians who have contributed to fall prevention activities over the decades.

Declaration of Conflicts of Interest

None.

Declaration of Sources of Funding

None.