Key messages from scientific research that's ready to be acted on
Got It, Hide thisde Souto Barreto P, Rolland Y, Vellas B, et al. Association of Long-term Exercise Training With Risk of Falls, Fractures, Hospitalizations, and Mortality in Older Adults: A Systematic Review and Meta-analysis. JAMA Intern Med. 2018 Dec 28. [Epub ahead of print]
In older people, does exercise training for at least 1 year reduce risk for falls, falls that cause injuries, fractures, hospitalization, or death?
People 65 years of age and older are more likely to experience falls, injuries from falls, broken bones, hospitalization, and even death. It is unclear whether exercise can help reduce the risk for these events.
The researchers did a systematic review based on studies available up to March 2018. They found 46 randomized controlled trials with 22,709 people.
The key features of the RCTs were:
· average age was 73 years, 66% were women, and most lived in the community;
· most trials included multiple types of exercise, such as aerobic exercise plus strength training plus balance training;
· exercise was usually about 3 times per week for 50 minutes at a moderate intensity;
· exercise was often in a group-based supervised format or a mix of group-based with home-based unsupervised exercises;
· exercise continued for an average of a year and half; and
· studies usually compared exercise with other treatments.
Compared with control groups, exercise:
In older people, long-term exercise training reduces risk for falls and falls that cause injuries but not fractures, hospitalization, or death.
Outcomes | Number of trials (and people) | Effect of exercise |
Falls | 20 trials (4420 people) | About 5 fewer people out of 100 would fall. |
Falls that cause injuries | 9 trials (4481 people) | About 4 fewer people out of 100 would fall. |
Fractures | 19 trials (8410 people) | Groups did not differ for fractures. |
Hospitalization | 12 trials (5939 people) | Groups did not differ for hospitalizations. |
Death | 29 trials (11,441 people) | Groups did not differ for deaths. |