Background/Purpose:  Falls are the leading cause of injury-related morbidity and mortality in older adults. To our knowledge no studies have examined the association between knee OA and fall injuries.

Methods: Using data from the Health ABC Knee Osteoarthritis Substudy, a community-based study of white and black adults ages 70-79 (42% black; 48% men) at baseline (1997-1998), we tested the associations (at the person-level) between knee pain without radiographic OA (ROA), knee ROA without pain, and knee symptomatic ROA (sROA) and incident fall injuries among 962 adults mean (SD) age 74.7 (2.9) years.  We also examined whether these associations differed by sex, obesity status, and fall injury type (fracture vs. non-fracture).  Knee ROA was defined as having a Kellgren-Lawrence grade of ≥2 in at least one knee.  Knee sROA was defined as having both ROA and pain symptoms (during the last 30 days) in the same knee.   Fall injuries were defined using a validated diagnoses code algorithm from linked Medicare claims (99.2% linkage) as any unique event with a fall code (E880-888) and/or non-vertebral fractures (800-804, 807-829).  Cox regression modeling was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).  Covariates associated with the exposure or outcome at p<0.1 were included in multivariate adjusted models. 

Results: The mean (SD) follow-up time was 7.50 (3.02) years. The prevalence of sROA, pain without ROA, ROA without pain, and no ROA or pain was 34.4%, 40.0%, 4.8%, and 20.8%, respectively. Of the 962 participants, 274 (28.5%) had an incident fall injury.  Compared with those without ROA or pain, individuals with pain without ROA (HR= 1.28; 95% CI: 0.89, 1.85), ROA without pain (HR= 1.18; 95% CI: 0.59, 2.39), and sROA (HR=1.24; 95% CI: 0.83, 1.85) did not have a significantly increased risk of fall injuries (Table).  Among men only, and compared with men without ROA or pain, those with pain without ROA (HR= 2.13; 95% CI: 1.01, 4.50, p-value=0.048) and sROA (HR=2.28; 95% CI: 1.00, 5.18, p-value=0.049) had a significantly higher risk of fall injuries (Table).  The association between knee OA and fall injuries did not differ by obesity status or fall injury type.

Conclusion:   Both knee sROA and knee pain without ROA were independently associated with a borderline increased risk of incident fall injuries in men only.    Table. Adjusted risk of fall injuries associated with knee pain without ROA, knee ROA with pain,    and knee sROA           

      Abbreviations: ROA, radiographic osteoarthritis; sROA, Symptomatic ROA; HR, hazard ratio; CI,   confidence intervals; MV, multivariate model   ^aThe reference group comprises participants without ROA or pain in a knee   ^bAdjusted for age, race, sex, site, BMI, smoking, physical activity, health status, hypertension, myocardial infarction,  prior falls, poor vision, NSAISD use, steroid use, calcium use, vitamin D use, and antidepressants   ^*p<0.05