The experience of pain and emergent osteoarthritis of the knee

Osteoarthritis Cartilage. 2001 Aug;9(6):527-32. doi: 10.1053/joca.2000.0429.


Discrepancies exist between radiographic osteoarthritis of the knee (OAK) and report of knee joint pain. Little is known about how these two definitions of osteoarthritis (OA) and their correlates differ between African American (AA) and Caucasian (CA) women.

Objective: We compared the prevalence of radiographic OAK and knee joint pain in AA and CA women, and the congruency of these outcomes according to age, body size, and knee injury.

Design: A cross-sectional study of African American and Caucasian women aged 40-53 years (N=829) in Southeast Michigan used the Kellgren and Lawrence Atlas of Standard Radiographs of Arthritis to characterize radiographs of both knee joints (weight bearing) and self-report of knee pain.

Results: Current pain was a significantly more sensitive predictor of radiographic OAK among AA women (Se=0.51) compared to CA women (Se=0.35). Specificity was similar between AA women (Sp=0.77) and CA women (Sp=0.82). Positive predictive value was significantly greater for AA compared with CA women (PV+=0.40 and PV+=0.15, respectively). The odds of having radiographic OAK increased with BMI >32 kg/m(2) in both groups. Knee pain was related to BMI in CA women, but not AA women. Previous knee injury was associated with knee pain in both AA and CA women (OR=3.0 and OR=2.4).

Conclusions: Joint pain in AA women was more likely to be associated with radiographic OAK as compared with CA women. This suggests differences in these two groups in both how pain is experienced in the OAK process and in the prevalence of non-OAK related pain in knee joints.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Adult
  • Age Factors
  • Analysis of Variance
  • Black People*
  • Body Constitution
  • Body Mass Index
  • Chi-Square Distribution
  • Confidence Intervals
  • Cross-Sectional Studies
  • Female
  • Humans
  • Knee Injuries / complications
  • Least-Squares Analysis
  • Logistic Models
  • Michigan / epidemiology
  • Middle Aged
  • Odds Ratio
  • Osteoarthritis, Knee / diagnostic imaging
  • Osteoarthritis, Knee / ethnology*
  • Osteoarthritis, Knee / etiology
  • Pain / diagnostic imaging
  • Pain / ethnology*
  • Pain / etiology
  • Predictive Value of Tests
  • ROC Curve
  • Radiography
  • Sensitivity and Specificity
  • White People*