Part 1 of 3 — A physiotherapist’s perspective for residential aged care management.
What proportion of your falls burden is attributable to unmet need, rather than physical frailty alone?
As a physiotherapist working across residential aged care, I review a falling resident on referral almost every week. The environment audit is clear. The care plan is current. The mobility aid is correctly prescribed and in use. And still the incidents accumulate. When I review the clinical notes, a consistent pattern emerges: a resident living with dementia, repeatedly attempting to mobilise unassisted, at a consistent time of day — documented by staff as “wandering,” “restlessness,” or “non-compliant with the care plan.”
I want to put a direct clinical question to you as operational and clinical leaders: what proportion of your falls burden is attributable to unmet need, rather than physical frailty alone?
The incidence data you’re already carrying
Falls remain the leading cause of hospitalised injury and injury death among older Australians. People aged 65 and over are hospitalised for falls at a rate of 3,275.3 per 100,000 population — roughly 12 times the rate recorded for adults aged 25–44 (268.1 per 100,000). One in five hospitalised falls among older Australians occurs within a residential aged care facility.
A retrospective cohort study across 25 Australian residential aged care facilities in Sydney, encompassing 6,163 residents and 3.9 million resident-days, recorded a crude incidence rate of 7.14 falls per 1,000 resident-days. At the admission level, 43.4% of residents experienced at least one fall, 32.0% experienced an injurious fall, and 16.6% required hospitalisation as a direct result. Nearly two-thirds of these falls occurred in residents’ own rooms.
Residents are also entering care older and with higher acuity than in previous years — the mean age at admission is now 85 for women and 83 for men — and a growing proportion carry a dementia diagnosis. This is where the falls data becomes disproportionate.
Why dementia changes the risk calculus
The relationship between cognitive impairment and falls risk is not marginal. A meta-analysis of 4,654 older adults with Alzheimer’s disease found a pooled annual fall prevalence of 44.27%, against a pooled global prevalence of approximately 26.5% for older adults generally. The same analysis found an average of 1.30 falls per person per year in the Alzheimer’s cohort, compared with approximately 0.3 falls per person per year in cognitively healthy older adults — and 42.08% of fallers with Alzheimer’s disease experienced recurrent falls within the study period. A separate meta-analysis reported that individuals with Alzheimer’s disease had approximately three times the odds of falling compared with cognitively healthy peers.
A large Italian nursing home cohort study (32 facilities, residents aged 65 and over) reinforces this from a different angle: cognitive impairment carried a hazard ratio of 1.30 (95% CI 1.10–1.53) for time to first fall, independent of age, sex and comorbidity.
The clinical model that explains the pattern
There is a well-established framework in the dementia care literature that accounts for this elevated risk beyond biomechanics alone: the Unmet Needs Model, developed by Cohen-Mansfield and colleagues. It proposes that as dementia progresses, residents lose the capacity to communicate ordinary needs — hunger, thirst, pain, boredom, loneliness, the need for meaningful activity — and simultaneously lose the capacity to meet those needs independently.
The need itself doesn’t disappear. Only the ability to name it does.
In the original study underpinning this model, 89 residents across six nursing homes were formally assessed for unmet needs; an average of three unmet needs per resident was identified, with boredom or sensory deprivation, loneliness or the need for social interaction, and the need for meaningful activity rated as the most prevalent. Each of these is a plausible falls trigger — a resident who gets up unassisted to find company, to relieve unaddressed pain, to locate a toilet, or simply to move because no structured activity has been offered — hiding behind a behaviour label that provides no clinical information about its cause.
Why generic falls risk assessment misses this cohort
Most facilities I work with operate a genuinely compliant falls prevention program: admission screening, environmental audit, medication review, correctly fitted mobility aids. These remain necessary, and none of what follows suggests removing them.
However, standard multifactorial falls risk assessment was developed for residents who can reliably self-report — “I feel dizzy,” “I’m afraid of falling,” “my hip is sore.” In a resident with moderate-to-advanced dementia, that self-report channel is substantially compromised. The consequence is a recurring blind spot: the risk assessment is completed, the care plan is updated, the environment is modified — and the resident continues to fall, because the driver was never purely physical. It was boredom in the pre-dinner lull. It was unaddressed pain presenting only as restlessness. It was the need to mobilise, because no toileting round had been offered in three hours.
Current national guidance now reflects this. The Australian Commission on Safety and Quality in Health Care’s 2025 Falls Guidelines for Residential Aged Care Services list neurocognitive status — including dementia, delirium, depression and behavioural symptoms — as a core domain of a comprehensive falls risk assessment, not a supplementary consideration. This represents a substantive shift from the 2009 edition, reflecting a resident population with materially higher care complexity than 15 years ago.
The clinical question for your next incident review
Before your next falls review defaults to increased supervision or environmental modification alone, I’d recommend a preceding clinical question be documented as standard: what need might this resident have been attempting to meet?
That question doesn’t replace clinical rigour — it closes a gap most falls programs currently carry. In the next article in this series, I examine why documenting a behaviour label — “wandering,” “agitation,” “resisting care” — is not equivalent to a clinical assessment, and won’t move your facility’s falls rate.
Sources
Sources referenced include the Australian Commission on Safety and Quality in Health Care’s 2025 Falls Guidelines for Australian Residential Aged Care Services, the Australian Institute of Health and Welfare’s national falls and aged care data, a retrospective cohort study of falls epidemiology across 25 Australian residential aged care facilities (Bail et al.), meta-analyses of fall risk in Alzheimer’s disease, an Italian multi-facility nursing home cohort study of the Tinetti POMA as a fall predictor, and the original Unmet Needs Model studies (Cohen-Mansfield and colleagues).
This article is general information for operational planning purposes and does not replace individual clinical assessment or facility-specific policy review.
