JAGS 50:671–678, 2002 © 2002 by the American Geriatrics Society 0002-8614/02/$15.00 Musculoskeletal Pain and Risk for Falls in Older Disabled Women Living in the Community Suzanne G. Leveille, PhD,* Jonathan Bean, MD, MS,* † Karen Bandeen-Roche, PhD, ‡ Rich Jones, PhD,* Marc Hochberg, MD, MPH, § and Jack M. Guralnik, MD, PhD  OBJECTIVES: To determine whether musculoskeletal pain increased risk for falls in older women with disabilities. DESIGN: Prospective population-based cohort study. SETTING: The city and county of the eastern area of Bal- timore. PARTICIPANTS: One thousand two women aged 65 and older, participants in the Women’s Health and Aging Study, representing the one-third of older women who were living at home with disabilities, followed semiannu- ally for 3 years beginning in 1991. MEASUREMENTS: Pain was categorized into four groups according to severity and location. Widespread pain was defined as pain in the upper and lower extremities and in the axial skeletal region, with moderate to severe pain in at least one region (  4 on a 10-point numeric rating scale, 10 = excruciating pain). Moderate to severe lower extremity pain that did not meet criteria for widespread pain was the next category. The reference category was no pain or mild pain in one site. The additional category of “other pain” was pain that did not fit into the other three groups. The occurrence of falls and fall-related injuries were assessed at each interview. RESULTS: Of the 940 women who participated in at least one follow-up examination, 39% fell in first year; of the survivors, 36% fell in Year 2, and 39% in Year 3. Af- ter adjusting for several major risk factors for falls, women with widespread pain had an increased likelihood of fall- ing during follow-up (adjusted odds ratio (AOR) = 1.66, 95% confidence interval (CI) = 1.25–2.21) compared with those with no or mild pain in only one musculoskele- tal site. Women who had other musculoskeletal pain but not widespread pain or lower extremity pain also had an increased risk of falls (AOR = 1.36, 95% CI = 1.02– 1.82). Among women with musculoskeletal pain, risk for falls was lower in those who used daily analgesic medica- tion. Risk for recurrent falls and self-reported fractures due to falls was also elevated in women with musculoskel- etal pain, most consistently in women with widespread pain. CONCLUSIONS: Musculoskeletal pain, particularly wide- spread pain, is a substantial risk factor for falls in older women with disabilities. These findings add an important dimension to our understanding of the multifactorial pro- cesses leading to falls in older persons. J Am Geriatr Soc 50:671–678, 2002. Key words: pain; falls; osteoarthritis; fibromyalgia; aging; musculoskeletal; women alls are a major cause of disability, healthcare utiliza- tion, and mortality in the older population. Fall-re- lated deaths in older persons have been on the increase in the past 15 years, and rates are highest in those aged 85 and older, 1 the fastest-growing segment of the older popu- lation. An estimated 87% of fractures in adults aged 65 and older are due to falls. 2 In 1994, falls and their sequelae in persons aged 65 and older in the United States cost an estimated $20.2 billion. 3 Although many risk factors for falls have been examined in several longitudinal studies, 4–6 musculoskeletal pain, a major contributor to disability in older persons, 7–9 has rarely been evaluated as a fall hazard. Osteoarthritis (OA), assessed by self-report, has been implicated as a cause of falls in older persons, 5,6 possibly related to postural instability and reduced muscle strength in the lower extremities. 10,11 However, radiographic hip OA in the Study of Osteoporotic Fractures was found to be protective for falls. 12 The concordance between self- reported arthritis and radiographic disease is poor, but the concordance between self-reported arthritis and knee pain is much higher, 13 lending credence to the notion that self- reported arthritis is more an indication of pain symptoms than confirmed disease. The relationship between pain and falls is unknown and requires study. In general, the From the *Hebrew Rehabilitation Center for Aged, Boston, Massachusetts; † Department of Physical Medicine and Rehabilitation, Harvard Medical School, Spaulding Rehabilitation Hospital, Boston, Massachusetts; ‡ The Johns Hopkins University School of Public Health, Baltimore, Maryland; § Departments of Medicine, and Epidemiology and Preventive Medicine, University of Maryland, Baltimore, Maryland; and  Epidemiology, Demog- raphy, and Biometry Program, National Institute on Aging, Bethesda, Mary- land. Address correspondence to Suzanne Leveille, PhD, Hebrew Rehabilitation Center for Aged Research and Training Institute, 1200 Centre Street, Boston, MA 02131. E-mail: leveille@mail.hrca.harvard.edu F