Assessing Impacts of Knee Osteoarthritis Solely With Patient Self-report May Miss Deficits

Older individual sitting on couch holding knee in pain, knee osteoarthritis concept

Objective measures of functional status have been associated with several important treatment outcomes in patients with knee osteoarthritis (OA), including falls and the need for total joint replacement. Yet most investigations assess outcomes by asking patients to use self-report measures, which may not reflect their true capabilities.

In Osteoarthritis and Cartilage Open, researchers at Brigham and Women’s Hospital present additional evidence that both subjective and objective measures are of value in assessing functional status in patients with knee OA. The authors are Faith Selzer, PhD, epidemiologist and administrative director of the Orthopaedic and Arthritis Center for Outcomes Research (OrACORe), Elena Losina, PhD, MSc, co-director of OrACORe and director of the Policy and Innovation eValuation in Orthopaedic Treatments (PIVOT) Center, Jeffrey N. Katz, MD, MS, director of OrACORe, and colleagues.


The researchers analyzed baseline data gathered for the Osteoarthritis Registry of Biomarkers and Imaging Trajectories (ORBIT), a prospective cohort study of adults with knee OA recruited at the Brigham Orthopedic and Arthritis Center.

Patients are eligible if they are at least 40 years old, English-speaking and community-dwelling, have radiographic evidence of OA in one knee, and report at least mild knee pain within the prior four weeks.

Among other components of an in-person visit, interested patients complete two questionnaires about knee pain and function and four questionnaires about pain-related variables:

  • The Knee Injury and Osteoarthritis Outcome Score pain scale (KOOS Pain)
  • The KOOS activities of daily living scale (KOOS ADL)
  • The five-item Mental Health Index (MHI-5, a measure of anxiety and depressive symptoms)
  • The Neuropathic Pain Scale (NPS)
  • The Pain Catastrophizing Scale (PCS)
  • An adapted widespread pain index (WPI; the sum of pain at 23 possible musculoskeletal locations during the past week)

Patients also complete four performance tests:

  • Single-leg balance (SLB)
  • 30-second Sit-to-Stand (STS)
  • Timed-Up-and-Go (TUG)
  • Timed 40-meter fast-paced walk (40-m walk)


This analysis included 101 participants, 63% female, whose average age was 64, and whose average body mass index was 30 kg/m2. The average baseline scores on the KOOS Pain and KOOS ADL were 56 and 67, respectively, on scales of 0–100 (100 best). These scores reflect a moderate degree of knee pain and functional limitation.

Better scores on the KOOS Pain were correlated with the following (r=correlation coefficient; 0=no correlation and 1=perfect correlation):

  • Longer duration on the SLB (r= 0.31)
  • Greater number of repetitions of the STS (r= 0.29)
  • Faster times on the TUG (r= −0.14)
  • Faster times on the 40-m walk (r= −0.34)

Results were similar for scores on the KOOS ADL:

  • Longer duration on the SLB (r= 0.39)
  • Greater number of repetitions of the STS (r= 0.37)
  • Faster times on the TUG (r= −0.24)
  • Faster times on the 40-m walk (r= −0.30)

Adjustment for mental health, neuropathic pain, pain catastrophizing, and widespread pain had little effect on the correlations.

Guidance for the Clinic

These findings indicate that some patient functional capabilities measured with objective performance tests (e.g., walking speed, balance, ability to transfer quickly from sitting to standing) are not captured fully with self-report measures. Objective measures of function should be included in clinical practice and in OA trials.

Leave a Reply