Long-term follow-up of exercise interventions aimed at preventing falls in older people living in the community: a systematic review and meta-analysis

Key points

1. 01Pooled across studies, exercise lowered the rate of falls beyond 12 months by 21% (rate ratio 0.79) and the risk of being a faller by 17% (risk ratio 0.83).
2. 02Gait, balance and functional training was the core ingredient, used in 20 of 24 studies, and drove most of the benefit.
3. 03Programmes lasting 6 to 12 months were the most effective, cutting rate and risk of falling by about a third.
4. 04The protective effect appears to last up to roughly two years after the intervention, then declines, but very few trials looked beyond that.
5. 05Single interventions of strength training alone or general walking were not effective for reducing falls.

How it was conducted

Design

Systematic review and meta-analysis of RCTs, cohort studies and secondary analyses with follow-up beyond 12 months

Search

MEDLINE, EMBASE, AMED, CINAHL, psycINFO, PEDro and Cochrane Library from inception to April 2017; 24 studies, 28 comparisons

Participants

7818 community-dwelling adults aged 65 and over (79% women); no specific medical conditions

Intervention

Single or multiple component fall-prevention exercise vs usual care, placebo or non-exercise control

Outcomes

Rate of falls (rate ratio) and number of people falling (risk ratio), random effects model

Analysis

Generic inverse variance random effects in RevMan; I-squared for heterogeneity; Cochrane risk of bias tool

What they found

• Rate of falls beyond 12 months: RaR 0.79 (95% CI 0.71 to 0.88; 23 studies, 6250 participants), a 21% reduction, I-squared 60%.
• Risk of falling beyond 12 months: RR 0.83 (95% CI 0.76 to 0.92; 18 studies, 4690 participants), a 17% reduction, I-squared 55%.
• Multiple component exercise: RaR 0.82 (95% CI 0.74 to 0.91) and RR 0.91 (95% CI 0.85 to 0.96).
• Exercise lasting 6 to 12 months: RaR 0.67 (95% CI 0.56 to 0.80) and RR 0.64 (95% CI 0.48 to 0.85).
• Beyond 24 months: no significant effect on rate (RaR 0.80, 95% CI 0.60 to 1.06; 3 studies) or risk (RR 0.79, 95% CI 0.61 to 1.03; 2 studies).

Limitations

• Moderate to substantial clinical and statistical heterogeneity means results should be read with caution.
• Nearly half the studies had methodological weaknesses, so effects may be overestimated.
• Nine studies used retrospective self-reported falls data, prone to recall bias, rather than prospective diaries.
• Only three studies followed participants beyond two years, so the long-term picture is uncertain.

Why it matters

For patients

Sticking with a balance and strength exercise programme can keep lowering your chance of falling for up to about two years.

For clinicians

Prescribe multi-component, balance-focused programmes of at least six months, and consider ongoing exercise since benefits fade over time.

For readers

Falls prevention exercise has durable benefits, but the evidence for protection past two years is thin.