Article Text

THU0638 Leg Length Inequality as A Risk Factor for Disability in A Community-Based Longitudinal Study
  1. Y.M. Golightly1,
  2. C. Alvarez1,
  3. J. Cantrell1,
  4. J.B. Renner1,
  5. C.G. Helmick2,
  6. J.M. Jordan1
  1. 1University of North Carolina, Thurston Arthritis Research Center, Chapel Hill, NC
  2. 2Centers for Disease Control and Prevention, Atlanta, GA, United States


Background Leg length inequality (LLI), a condition in which one limb is longer than the other, is associated with osteoarthritis (OA) and pain of the knee and hip. OA is a leading cause of disability, but it is not known whether LLI is a risk factor for disability when accounting for OA.

Objectives This longitudinal analysis examined the hazard of incident and progressive disability by LLI in a large, bi-racial (African American and Caucasian) community-based study in the United States of men and women 45+ years old with and without OA.

Methods Participants were enrolled from 1991–1997 (original cohort) or 2003–2004 (enrichment cohort); 2369 participants had complete baseline LLI and baseline and follow-up (1999–2003 or 2006–2010) disability data available for analyses. With the participant supine, right and left lower extremity lengths were measured by trained examiners using a tape measure from anterior superior iliac spine to distal medial malleolus. LLI was defined as a difference between limbs ≥2 cm at baseline. Self-reported disability was assessed with the Stanford Health Assessment Questionnaire (HAQ) Disability Index, which assesses 20 functions covering 8 domains (dressing, arising, eating, walking, reaching, gripping, chores, hygiene) scored on a scale of 0–3 (no difficulty to unable). HAQ scores were calculated by averaging the scores of the 8 domains. Incident Disability was present if baseline HAQ =0 and follow-up HAQ >0, and Progression of Disability was present if baseline HAQ < follow-up HAQ. Separate parametric Weibull time-to-event models estimated hazard ratios of the association of LLI and disability outcomes, adjusting for time of LLI assessment (1991–1997 or 1999–2004), age, sex, race, body mass index (BMI), radiographic knee or hip OA, knee or hip problems (injury, surgery, fracture, or congenital anomaly), and knee or hip symptoms (pain, aching, stiffness on most days). Interactions between LLI and covariates were examined (p<0.05 considered statistically significant).

Results Characteristics of the total sample were: mean ± standard deviation (SD) baseline age 60 ± 9.7 years; 66% women; 31% African American; mean ± SD baseline BMI 30 ± 6.2 kg/m2; 23% knee OA; 26% hip OA; 21% knee problems; 8% hip problems; 44% knee symptoms; 36% hip symptoms; median follow up time 7.6 years, range 3.6–17.8 years. 154 participants (6.5%) had a LLI ≥2 cm. Compared to those without LLI, the adjusted hazard of Incident Disability was 39% higher among participants with LLI (adjusted hazard ratio [aHR] 1.39, 95% confidence interval [CI] 1.002, 1.93). For the model with Progression of Disability, a statistically significant interaction was noted between baseline LLI and time of baseline LLI assessment; thus, results were stratified by time of assessment. Among participants with baseline LLI collected during 1999–2004, the adjusted hazard for Progression of Disability was 60% higher among participant with LLI vs. no LLI (aHR 1.60, 95% CI 1.12, 2.28).

Conclusions LLI appears to be a modifiable risk factor for incident and progressive disability outcomes when accounting for OA. Treatment for LLI, such as shoe lift therapy, should be examined as an approach to prevent disability progression.

Disclosure of Interest None declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.