WASHINGTON — Postmenopausal women with a prior fall or those 80 years of age or older have a significantly greater risk of a subsequent fall, according to data presented at an international symposium sponsored by the National Osteoporosis Foundation.
Specifically, the investigators found that women with a prior fall had an odds ratio of 2.7 and those 80 years or older had a odds ratio of 1.5 for a future fall, based on a analysis of potential risk factors for falls among 66,134 women enrolled in the National Osteoporosis Risk Assessment (NORA) study, said Dr. Elizabeth Barrett-Connor, chief of epidemiology in the department of family and preventive medicine at the University of California at San Diego.
The NORA study enrolled over 200,000 community-dwelling, postmenopausal women between 1997 and 1999. Women had to be at least 50 years old without a diagnosis of osteoporosis. They could not have had a bone mineral density measurement in the previous year or be taking an osteoporosis-specific medication. At baseline, BMD was measured at the heel, forearm, or finger. The women were followed up at 1, 3, and 6 years with surveys asking if they'd had a fracture in the previous year.
The sample of women in this study responded to all of the surveys. At baseline, the average age was 63 years. Most (91%) were white. The average T score was −0.78. In all, 38% reported at least one fall in the past year.
The researchers included a long list of potential risk factors. These included age, body mass index, a self-rating of health as being poor/fair, functional limitations, smoking and alcohol use, early menopause, height loss, peripheral T score, personal history of fracture after age 45, maternal history of fracture and/or osteoporosis, first-degree relatives with a history of fracture, history of estrogen therapy, calcium supplementation, use of certain medications (oral corticosteroids, thyroid medication, an osteoporosis-specific drug), history of depression, osteoporosis self-knowledge, and self-report of a fall within the previous 12 months at the year 1 survey. They also included arthritis, coronary artery disease, hypertension, diabetes, kidney/liver disease, breast cancer, other cancers, memory problems, stroke, hyperthyroidism, hypothyroidism, epilepsy, poor vision, and poor hearing.
In addition, history of depression and of stroke increased the risk of falling by over 40%. An additional nine factors were identified that significantly increased fall risk by 9%–23%. The number of baseline risk factors was linearly associated with a risk of falling.
The NORA study has several limitations. First, participants were volunteers and may not be a representative sample. Second, falls were self-reported and limited to a 12-month recall period. The gaps between surveys likely mean falls were underreported. Longitudinal attrition resulted in a slightly younger and healthier analytic sample. No data were collected on factors known to be associated with falls, such as prescription medications, environment, gait, balance, and muscle strength. Lastly, the cause of falls was not known.
Dr. Barrett-Connor disclosed research support from several pharmaceutical companies and is a consultant for Merck & Co. Two of her collaborators are employees of Merck.