Knee Osteoarthritis Symptom Assessments That Combine Pain and Physical Activity Are Superior to Pain Alone

Grace H. Lo¹, Timothy E. McAlindon², Gillian A. Hawker³, Jeffrey B. Driban⁴, Lori Lyn Price⁵, Jing Song⁶, Charles Eaton⁷, Marc C. Hochberg⁸, Rebecca D. Jackson⁹, C. Kent Kwoh¹⁰, Michael C. Nevitt¹¹ and Dorothy D. Dunlop⁶, ¹Michael E. DeBakey Veterans Affairs Medical Center / Baylor College of Medicine, Houston, TX, ²Division of Rheumatology, Tufts Medical Center, Boston, MA, ³Women's College Research Institute, University of Toronto, Toronto, ON, Canada, ⁴Rheumatology, Tufts Medical Center, Boston, MA, ⁵Biostatistics Research Center, Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, Boston, MA, ⁶Northwestern University Feinberg School of Medicine, Chicago, IL, ⁷Center for Primary Care and Prevention, Memorial Hospital of Rhode Island, Providence, RI, ⁸Department of Medicine, University of Maryland, Baltimore, MD, ⁹Ohio State University, Columbus, OH, ¹⁰School of Medicine, Department of Medicine, Division of Rheumatology and Clinical Immunology, University of Pittsburgh, Pittsburgh, PA, ¹¹Epidemiology & Biostatistics, UCSF (University of California, San Francisco), San Francisco, CA

Meeting: 2012 ACR/ARHP Annual Meeting

Keywords: osteoarthritis and outcome measures

Session Information

Title: Osteoarthritis - Clinical Aspects

Session Type: Abstract Submissions (ACR)

Background/Purpose :

Symptom assessment in knee osteoarthritis (OA) is challenging. Knee pain does not always correlate well with radiographic severity, perhaps because people modify / avoid activities to reduce symptoms. Accelerometers / pedometers can inexpensively and easily quantify physical activity. We hypothesize that symptom assessment accounting for pain intensity in the context of physical activity will improve discrimination across OA radiographic severity levels.

Methods :

We studied Osteoarthritis Initiative (OAI) participants with ≥ 4 days accelerometer monitoring, knee-specific WOMAC pain data, and knee x-rays from the 48-month visit. Accelerometer data included average daily step count and average daily activity counts (i.e., weighted sum of activity frequency and intensity). Four composite knee pain and activity scores (KPAS) were calculated:

KPAS1 = daily step count/(Total WOMAC Pain Score + 1)

KPAS2 = daily activity count/(Total WOMAC Pain Score + 1)

KPAS3 = daily step count/(WOMAC Walking Pain Score + 1)

KPAS4 = daily activity count/(WOMAC Walking Pain Score + 1)

Lower KPAS values reflect greater symptoms, consistent with less activity and/or greater pain. X-rays were read for Kellgren and Lawrence (K-L) grade (0-4). For each participant, only right knee data were evaluated. Total WOMAC pain score, WOMAC walking pain score, accelerometer data, and KPAS scores were tested for normality, and score discrimination by K-L grades using stratified histograms, Kruskal – Wallis testing, and quantile regression analyses (excluding WOMAC walking pain) unadjusted and adjusted for age, sex and BMI.

Results : 1472 participants, mean age 64.9 (± 9.1), mean BMI 28.1 (± 4.8), 43% male, were included. No symptom score was normally distributed, with pain assessments being the most skewed.

Table 1. Kruskal – Wallis testing of symptom assessments across adjacent K-L grades. “X” denotes statistically significant differences.

Symptom Assessments

K-L Grade

Comparisons

WOMAC

WOMAC Walk

Step Count

Activity Count

KPAS1

KPAS2

KPAS3

KPAS4

0 v. 1

1 v. 2

2 v. 3

3 v. 4

*Table 2.* Quantile regression: Differences in median symptom scores across K-L grade groups. Referent group is K-L grade 0. For WOMAC score, higher score = greater symptoms. For all other scores, lower score = greater symptoms.
Outcome	Adjustment Factors	Median Scores	Differences in Median Scores from Referent Group				P for trend
Outcome	Adjustment Factors	KL 0 (Referent)	KL 1	KL 2	KL 3	KL 4	P for trend
WOMAC Total Pain	Unadjusted	0	0	1	1	4	*p=.006*
WOMAC Total Pain	Adjusted^a	0	0	1	1	4	*p<.001*
Step count	Unadjusted	6253	-763*	-774*	-930*	-1957*	*p<.001*
Step count	Adjusted^a	18436	-286*	-242	-222	-299	p=0.20
Activity count	Unadjusted	211358	-13429	-20681*	-30746*	-51700*	*p<.001*
Activity count	Adjusted^a	647624	-1920	732	7038	-3197	p=0.79
KPAS1	Unadjusted	4367	-627*	-1338*	-1674*	-3499*	*p<.001*
KPAS1	Adjusted^a	11709	-387	-1116*	-1245*	-2912*	*p<.001*
KPAS2	Unadjusted	146556	-14023	-40866*	-46057*	-110077*	*p<.001*
KPAS2	Adjusted^a	448042	-6880	-28642*	-29064*	-88068*	*p<.001*
KPAS3	Unadjusted	5560	-500*	-1088*	-1024*	-2842*	*p<.001*
KPAS3	Adjusted^a	16208	-179	-334	-271	-2226*	*p<.001*
KPAS4	Unadjusted	183218	-5603	-25181*	-23770*	-85471*	*p<.001*
KPAS4	Adjusted^a	614784	-892	-7509	-5994	-56155*	*p=.003*
^a Adjusted for age, sex, and BMI * Statistical difference of group median compared to the referent (KL score of 0) median; p<0.05.

Conclusion :

Symptom assessments incorporating pain intensity and physical activity improved discrimination across radiographic OA severity. Pain better discriminates high disease severity while physical activity better distinguishes low severity. Relationships of KPAS measures with x-ray severity were robust to adjustment for traditional OA risk factors. To improve sensitivity, physical activity should be routinely assessed in studies of knee OA symptoms.

Disclosure:

G. H. Lo,
None;

T. E. McAlindon,
None;

G. A. Hawker,
None;

J. B. Driban,
None;

L. L. Price,
None;

J. Song,
None;

C. Eaton,
None;

M. C. Hochberg,
None;

R. D. Jackson,
None;

C. K. Kwoh,
None;

M. C. Nevitt,
None;

D. D. Dunlop,
None.

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/knee-osteoarthritis-symptom-assessments-that-combine-pain-and-physical-activity-are-superior-to-pain-alone/