Strength training in older adults has been shown in many individual trials to be effetive at reducing falls, but what does a systematic review tell us?
Falls are the second leading cause of accidental death worldwide with a large proportion of these occurring in the older population. Understanding how to reduce risk of falling is therefore crucial in helping protect our vulnerable older adults.
Various forms of exercise are used to reduce risk of falls and despite there being hundreds of high quality studies investigating this consensus still has not been reached when it comes to the benefit of strength training on falls risk.
A new systematic review published in the Journal of Medicine aims to assess the effects of strength training (ST), as compared to alternative multimodal or
unimodal exercise programs, on the number of falls in older adults (≥60 years), and help bring us to a point of consensus.
This systematic review was pre-registered on PROSPERO and adhered to the PRISMA guidance both indicative of good methodological rigour. The PICOS used for the search strategy is also fully available further adding to the quality of the review.
Studies published in peer-reviewed journals were eligible with no limitations based on langugage or date. Inclusion criteria was based on PICOS and is as follows:
- P: 60+ years old, community dwelling, residential or nursing or hospital based as well as those living and not living with frailty
- I: Strength training without combination with other exercise
- C: non-exercise controls or multimodal or unimodal exercise interventions
- O: Risk of falls measured as number of falls or falls rate
- S: Supervised RCTs
In total ten databases were searched including CINAHL, Cochrane Library, EBSCO, EMBASE, PEDro, Pubmed, Scielo, Scopus, SPORTDiscus and WoS. The search strategy is available in full and appears logical and robust.
A manual search of references was also performed along with the search criteria and list of studies being sent to seven experts to suggest additional relevant studies.
Three authors performed the searches with disagreements resolved through re-analysis until reaching consensus. Risk of bias was assessed using Cochranes RoB 2.0 tool. The certainty of evidence was assessed using GRADE.
Results & Clinical Implications
In total five RCTs were eligible for review which included 543 participants. All studies included as strength training group with distinct protocols but all followed recommendations for exercise in older adults.
The sessions varied between 1-3x/week with prescription of exercise between 1-3 sets of 6-30 reps with duration between 50-60 minutes. Interventions lasted between 84-365 days with one study having a 3 year follow up.
The risk of bias was mixed across all included studies. One was low risk, two medium and one high risk with the main concerns across all studies being randomisation (difficult to do in these types of studies) and selective reporting. This is because many of the studies did not pre-register a protocol or stats analysis plan. After sensitivity analysis it was found that the poor randomisation did not effect the outcome of anaylsis.
The results of the meta-analysis showed that supervised strength training reduced the risk of falls when compared to unimodal training sessions (balance, agility, stretching, tai chi or self-administered) however the results were not statistically significant.
There was also no significant difference seen in any of the secondary outcomes which included Barthel, BMD, balance, gait speed and SPPB. Interstingly one study investigated cost effectiveness and 1x/wk strength training is -25% cheaper than twice weekly less effective training.
There is a wealth of evidence suggesting strength training is the best strategy to adopt when using exercise to prevent falls however due to the variability in exercise selection and dose and intensity chosen means that definitive advice is unavailable in the form of systematic reviews.