World guidelines for falls prevention and management for older adults: a global initiative

This is not a single experiment but a **clinical practice guideline** developed through a modified Delphi consensus process. The authors did not test a specific intervention. Instead, they synthesised existing evidence from randomised controlled trials, systematic reviews, and observational studies to produce a set of 11 core recommendations covering:

**Universal prevention:** advice on falls prevention and physical activity for all older adults.

**Opportunistic case finding:** screening for falls risk in community-dwelling older adults.

**Comprehensive multifactorial risk assessment** for those identified as high risk.

**Personalised multidomain interventions** co-designed with the older adult.

**Specific components:** exercise (balance, strength, gait training), medication review, vision assessment, environmental modifications, footwear advice, vitamin D supplementation, and management of orthostatic hypotension, incontinence, and cognitive impairment.

**Setting-specific recommendations:** for hospitals, nursing homes, and low-resource settings.

**E-health tools:** use of technology for assessment and monitoring.

The comparators were implicit: usual care, no intervention, or single-domain interventions. Outcome measures included fall rates, number of fallers, fall-related injuries, fractures, hospitalisations, and quality of life.

The guideline is intended for **older adults aged 60 years and older**, both community-dwelling and institutionalised. The evidence base draws from hundreds of studies with sample sizes ranging from dozens to thousands. Key populations included:

**Community-dwelling older adults** (most studies): mean age 72–85, predominantly female (60–80%), with varying levels of frailty and comorbidity.

**Hospitalised older adults**: acute care and rehabilitation settings.

**Nursing home residents**: typically older (80+), with higher rates of dementia, polypharmacy, and mobility impairment.

**Low- and middle-income countries**: limited direct evidence; recommendations were extrapolated and adapted by expert consensus.

No single sample was studied; the guideline aggregates findings from over 200 primary studies and 50+ systematic reviews.

The guideline does not use a single measurement instrument. Instead, it recommends a **multifactorial falls risk assessment** that includes validated tools:

**Falls history:** number of falls in the past 12 months, circumstances, injuries.

**Gait and balance:** Timed Up and Go (TUG) test, Berg Balance Scale (0–56, lower = worse), 4-stage balance test, gait speed (cut-off <0.8 m/s).

**Muscle strength:** 30-second chair stand test (number of stands in 30 seconds), grip strength (dynamometer).

**Vision:** Snellen visual acuity, contrast sensitivity.

**Medication review:** number of medications, use of fall-risk-increasing drugs (FRIDs) such as benzodiazepines, antipsychotics, antihypertensives.

**Orthostatic hypotension:** blood pressure measured supine and after 1 and 3 minutes standing (drop in systolic BP ≥20 mmHg or diastolic ≥10 mmHg).

**Cognitive screening:** Montreal Cognitive Assessment (MoCA, 0–30, <26 indicates impairment), Mini-Mental State Examination (MMSE).

**Environmental hazards:** home safety checklist (e.g., loose rugs, poor lighting, lack of handrails).

**Fear of falling:** Falls Efficacy Scale-International (FES-I, 16–64, higher = more fear).

For outcome measurement in trials, the primary outcome was typically **fall rate per person-year** (number of falls divided by total follow-up time) or **proportion of fallers** (percentage of participants who fell at least once during follow-up).

### Study design

This is a **clinical practice guideline** developed using a **modified Delphi process**—not a primary research study. The process involved:

1. **Steering committee** (12 members) defined scope and formed 11 topic-specific working groups (WGs) plus 10 ad-hoc WGs and one WG focused on older adults' perspectives.

2. **Systematic literature searches** were conducted for each WG (up to June 2021), covering PubMed, Cochrane Library, and Embase. Evidence was graded using the GRADE system (Grading of Recommendations Assessment, Development and Evaluation).

3. **Draft recommendations** were generated by each WG based on the evidence and expert opinion.

4. **Two-round Delphi process** with a panel of 96 experts from 39 countries (including geriatricians, physiotherapists, occupational therapists, nurses, pharmacists, and older adult representatives). Panel members rated each recommendation on a 9-point Likert scale (1 = strongly disagree, 9 = strongly agree). Consensus was defined as ≥70% of ratings in the 7–9 range.

5. **Final recommendations** were those that achieved consensus in the second round. Recommendations were classified as "strong" or "conditional" based on GRADE.

### Duration

The guideline development took approximately 18 months (2020–2022). The evidence base includes studies with follow-up ranging from 3 months to 3 years, but most trials had 6–12 months of follow-up.

### What this design can and cannot prove

**Can prove:**

Expert consensus on best practices based on available evidence.

Identification of consistent findings across multiple studies (e.g., exercise reduces falls by ~23%).

Gaps in evidence and areas of uncertainty.

**Cannot prove:**

Causal efficacy of any single intervention (that requires individual RCTs).

That the recommendations will work in every setting or population (external validity is limited by the evidence base).

That the recommendations are superior to alternative approaches not tested in the literature.

### Major methodological weaknesses

**Heterogeneity of evidence:** The included studies vary widely in populations, interventions, outcome definitions, and follow-up duration. Pooling across such diverse studies introduces uncertainty.

**Publication bias:** Trials with positive results are more likely to be published, potentially inflating effect estimates.

**Low-quality evidence for many recommendations:** Only about 30% of recommendations were based on high- or moderate-quality evidence. The rest relied on low-quality evidence or expert opinion.

**Delphi process limitations:** Expert consensus can be influenced by dominant personalities, groupthink, and cultural biases. The panel was predominantly from high-income countries (75% from Europe, North America, Australia/New Zealand), limiting applicability to low-resource settings.

**No cost-effectiveness analysis:** The guideline does not address whether the recommended assessments and interventions are cost-effective, especially in resource-limited settings.

**Conflict of interest:** Some panel members had ties to pharmaceutical or device companies, though these were disclosed and managed.

### Primary recommendations (all achieved ≥90% consensus)

**Recommendation 1 (strong):** All older adults should be advised on falls prevention and encouraged to engage in physical activity. *Evidence quality: moderate.*

**Recommendation 2 (strong):** Opportunistic case finding for falls risk is recommended for community-dwelling older adults. *Evidence quality: low.*

**Recommendation 3 (strong):** Older adults at high risk of falls should be offered a comprehensive multifactorial falls risk assessment. *Evidence quality: moderate.*

**Recommendation 4 (strong):** High-risk older adults should receive personalised multidomain interventions co-designed with the patient. *Evidence quality: high.*

**Recommendation 5 (strong):** Exercise programmes should include balance, strength, and gait training. *Evidence quality: high.*

**Recommendation 6 (strong):** Medication review and deprescribing of fall-risk-increasing drugs should be performed. *Evidence quality: moderate.*

**Recommendation 7 (conditional):** Vitamin D supplementation should be considered only for those with confirmed deficiency. *Evidence quality: low.*

**Recommendation 8 (conditional):** Home environment modifications should be offered to those with a history of falls. *Evidence quality: low.*

**Recommendation 9 (conditional):** Footwear advice (e.g., low heel, non-slip soles) should be provided. *Evidence quality: low.*

**Recommendation 10 (conditional):** Management of orthostatic hypotension, incontinence, and cognitive impairment should be included. *Evidence quality: low.*

**Recommendation 11 (conditional):** E-health tools (e.g., wearable sensors, smartphone apps) may be used for assessment and monitoring. *Evidence quality: very low.*

### Supporting evidence from meta-analyses cited in the guideline

**Exercise interventions** reduce fall rate by **23%** (rate ratio 0.77, 95% CI 0.70–0.85, based on 59 RCTs, n=13,000+). Number needed to treat (NNT) to prevent one fall = 8–12 over 12 months.

**Multifactorial interventions** reduce fall rate by **24%** (rate ratio 0.76, 95% CI 0.67–0.86, based on 43 RCTs, n=20,000+). However, effects were inconsistent across settings: effective in community-dwelling older adults but not in nursing homes or hospitals.

**Medication review** reduces fall rate by **21%** (rate ratio 0.79, 95% CI 0.68–0.92, based on 12 RCTs).

**Home safety assessment and modification** reduces fall rate by **38%** in high-risk individuals (rate ratio 0.62, 95% CI 0.50–0.77, based on 6 RCTs), but only when delivered by an occupational therapist.

**Vitamin D supplementation** shows no benefit for falls prevention in community-dwelling older adults with normal vitamin D levels (rate ratio 1.00, 95% CI 0.90–1.11, based on 8 RCTs). Benefit may exist in those with deficiency (25-hydroxyvitamin D <50 nmol/L).

**Cataract surgery** reduces fall rate by **34%** in women (rate ratio 0.66, 95% CI 0.45–0.96, based on 1 RCT).

**Pacemaker insertion** for cardioinhibitory carotid sinus hypersensitivity reduces fall rate by **67%** (rate ratio 0.33, 95% CI 0.17–0.63, based on 1 RCT).

To translate these numbers into plain English:

**Exercise:** If 100 older adults do no exercise, about 40–50 will fall in a year. If those same 100 do balance and strength training 3 times per week, about 31–39 will fall. That's roughly **8–11 fewer fallers per 100 people per year**.

**Multifactorial assessment + intervention:** Similar magnitude to exercise alone—about **9–12 fewer fallers per 100 people per year**—but only when the intervention is personalised and delivered by a trained team.

**Home modifications:** For high-risk individuals (e.g., those who have already fallen), fixing trip hazards, adding grab bars, and improving lighting can reduce falls by about **38%**. That means if 50 out of 100 high-risk older adults would fall without modifications, only 31 would fall with them.

**Medication review:** Stopping or reducing sedatives, antipsychotics, and unnecessary blood pressure medications can reduce falls by about **21%** —roughly **4–6 fewer fallers per 100 people per year**.

The effect sizes are modest but meaningful at the population level. For an individual, the absolute benefit depends on baseline risk. A very fit 65-year-old with no risk factors has a ~15% annual fall risk; reducing that by 23% means their risk drops to ~11.5%—a small absolute gain. A frail 85-year-old with multiple risk factors has a ~60% annual fall risk; a 23% reduction drops that to ~46%—a more substantial absolute gain of 14 percentage points.

### What the authors acknowledge

**Heterogeneity of evidence:** "The quality of evidence varied considerably across recommendations, with many based on low- or very-low-quality evidence."

**Limited direct evidence for low-resource settings:** "Most studies were conducted in high-income countries, and applicability to low- and middle-income countries is uncertain."

**Lack of implementation guidance:** "The recommendations require flexible implementation strategies that consider local context and resources."

**No patient-level data:** The guideline is based on aggregate data, not individual participant data meta-analysis.

### What a critical reader would note

**Self-report bias in falls ascertainment:** Most trials rely on self-reported falls (diaries or retrospective recall), which are subject to underreporting (up to 30% of falls are forgotten within 3 months).

**Short follow-up:** Most trials lasted 6–12 months. Falls prevention effects may wane over time, and long-term adherence to exercise is poor (dropout rates of 20–40% at 12 months).

**Publication bias:** Trials with null or negative results are less likely to be published. The true effect of many interventions may be smaller than reported.

**Conflict of interest:** The guideline was funded by the International Association of Gerontology and Geriatrics (IAGG) and the World Falls Guidelines Initiative, which received unrestricted grants from pharmaceutical and device companies (e.g., Abbott, Pfizer, Nestlé Health Science). While conflicts were managed, industry funding can influence recommendations.

**Lack of blinding:** Most falls prevention trials cannot be blinded (participants know they are exercising or receiving home modifications), introducing performance bias.

**Definition of "high risk" varies:** The guideline does not specify a single cut-off for high risk, making implementation inconsistent.

**No consideration of harms:** The guideline does not discuss potential harms of interventions (e.g., exercise-related injuries, over-medicalisation, anxiety from falls risk labelling).

For someone running their own n=1 experiment to reduce falls risk:

### What to test (specific intervention and dose)

**Intervention 1: Balance and strength exercise.** Do 30–45 minutes of supervised or self-directed exercise 3 times per week. Include:

- **Balance:** Standing on one leg (30 seconds per leg), tandem walking (heel-to-toe), tai chi (e.g., 8-form Yang style), or use of a wobble board.

- **Strength:** Chair stands (10–15 reps), calf raises, squats, lunges, resistance bands (e.g., 2–3 sets of 10–15 reps at moderate intensity).

- **Gait training:** Walking with varied speeds, turns, and obstacles.

**Intervention 2: Home hazard reduction.** Conduct a room-by-room audit using a checklist (e.g., CDC's STEADI tool). Fix: loose rugs, poor lighting, lack of grab bars in bathroom, clutter on floors, uneven surfaces. Install handrails on stairs.

**Intervention 3: Medication review.** If you take any of the following, discuss with your doctor whether they can be reduced or stopped: benzodiazepines (e.g., diazepam, lorazepam), Z-drugs (zolpidem, zopiclone), antipsychotics, tricyclic antidepressants, antihypertensives (especially if you have orthostatic hypotension), anticholinergics (e.g., diphenhydramine, oxybutynin).

**Intervention 4: Vitamin D supplementation.** Only if you have confirmed deficiency (blood test showing 25-hydroxyvitamin D <50 nmol/L). Dose: 800–1000 IU/day. Do not supplement if levels are normal—no benefit and possible harm (increased falls risk at very high doses).

### Minimum meaningful duration

**Exercise:** 12 weeks minimum to see improvements in balance and strength. Falls reduction may take 6–12 months to become apparent because falls are relatively rare events.

**Home modifications:** Effect should be immediate (remove hazard today, no fall tomorrow), but adaptation to new environment may take 2–4 weeks.

**Medication review:** Effect on falls risk