Randomized Controlled Trial of Prevention of Fall-related Fractures and Other Injuries by Physical Exercise Among Community Dwelling Elderly Women

Background : Osteoporotic fractures are a major cause of morbidity and disability in elderly women. Most of these fractures result from a fall. Falls prevention measures largely applicable in the population may contribute to significantly reduce the number of fractures in the population. Recent randomized controlled trials have shown that exercise programs of relatively low intensity, but centered on balance retraining and targeted to subjects at moderate risk, can reduce the risk of falls by 30% in community-dwelling elderly people (Province MA et al 1995 ; Gillepsie LD et al 2003 ; Chang JT et al 2004). But, the effectiveness of exercise in preventing injuries caused by falls, in particular the most serious ones such as those accompanied by fractures, has not been demonstrated. Most previous trials are underpowered to examine the effect of fall prevention exercise programmes on severe injurious falls. Furthermore, there is no consensus about the outcomes of fall related injuries that should be evaluated in controlled trials, and the injurious falls definitions vary substantially across studies. Besides falls leading to fractures and other serious injuries, falls leading to medical care use ought to be considered as well, since their cost for society is high and their burden on the health care system heavy. Falls causing minor injuries are also important, as they may also have serious consequences, such as depression, fear of falling, and activity restriction.

Objectives : To evaluate the effectiveness of exercise in preventing falls resulting in fractures or other physical traumas. Also, to better understand the mechanisms by which physical exercise prevents injurious falls (e.g., improvement in physical functional abilities, global physical activity level, general physical and mental health, and self-confidence), and to determine the individual factors associated with a good long-term adherence to the program.

Method : Randomized controlled trial. Participants will be women aged 75 to 85 years old, living at home, but with impaired physical functional capacity (as assessed by simple functional tests). Women will be recruited using general population lists, such as voter-registration or supplemental health-insurance membership rolls. They will be randomized in 2 groups : one that will receive the intervention, and a control group without it. Each group will include 1000 women followed for 2 years, which will allow to demonstrate a 30% reduction in the frequency of serious falls in the intervention group (if the incidence of serious falls is around 8% per year). Falls and injuries occurrence will be monitored by asking women to return monthly calendar cards where they can mark the date of any falls. A telephone interview will then be realized in case a fall is reported to record the circumstances and consequences of the fall.

Intervention : The intervention will last 2 years and will be implemented in partnership with the association SIEL (Sport, Initiatives, et Loisir) Bleu in 20 study centers located in large and medium-sized cities. It will include weekly exercise sessions in small groups, led by a qualified and specially trained instructor, and supplemented by simple exercises to do at home. The exercise program will focus on progressive balance retraining but will also include strength/resistance, coordination and flexibility training exercises.

Evaluation criteria : The main outcome measure will be the incidence of serious falls defined as " an unexpected event in which the participant comes to rest on the ground, floor, or lower level " and accompanied by a fracture or another serious injury (including head injuries requiring hospitalization, joint dislocations, severe sprains, other non-specified serious joint injuries, and lacerations requiring suturing) (Buchner DM et al 1993). The effectiveness of the intervention on all injurious falls including those leading to more moderate injuries (such as bruises, cuts, abrasions or reduction in physical function for at least 3 days, or if the participant sought medical help (Campbell AJ et al 1997) will also be assessed. Falls associated with an intrinsic major event or an overwhelming hazard (e.g., traffic accident) will be excluded. Intermediate outcome measures, which may help to explain the effect (or absence of effect) of the intervention, will be changes in physical functional capacity (e.g., balance, gait), global physical activity level, feeling of self-efficacy (fear of falling), and general physical and psychological well-being (quality of life). These changes will be evaluated on a sample of participants (160 per group) that will be re-examined at 1 year and 2 years.