Association between Depressive Symptom Subtypes and Disease Severity in Knee Osteoarthritis

Alan Rathbun¹, Megan Schuler², Elizabeth Stuart³, Michelle Shardell⁴, Michelle S. Yau⁵, Joseph Gallo⁶ and Marc C. Hochberg⁷, ¹Epidemiology and Public Health, University of Maryland School of Medicine, Baltimore, MD, ²Rand Coportation, Boston, MA, ³Mental Health, Biostatistics, and Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, ⁴Translational Gerontology Branch, National Institute on Aging, Baltimore, MD, ⁵Institute for Aging Research, Hebrew SeniorLife, Harvard Medical School, Boston, MA, ⁶Mental Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, ⁷University of Maryland School of Medicine, Baltimore, MD

Meeting: 2018 ACR/ARHP Annual Meeting

Keywords: depression, functional status, Knee, osteoarthritis and pain

Session Information

Date: Monday, October 22, 2018

Title: Epidemiology and Public Health Poster II – ARHP

Session Type: ACR Poster Session B

Session Time: 9:00AM-11:00AM

Background/Purpose: Latent and modifiable factors, such as depressive symptoms, may affect the course of knee OA. Depression is clinically heterogeneous, and effects on pain and disability may manifest differently by depressive symptom subtypes. This study evaluated trajectories of pain and disability by depressive symptom subtypes among individuals who have or are at risk for knee OA. Methods: Participants (n=4486) were enrolled in the Osteoarthritis Initiative. Latent class analysis was applied to the 20-Item Center for Epidemiological Studies Depression Scale measured at baseline and used to assign participants to one of four depressive symptom subtypes identified previously (“No Symptoms”, “Catatonic”, “Anhedonic”, and “Melancholic”). OA disease severity was assessed annually over four years using the pain and disability subscales (rescaled range = 0-100) of the WOMAC. Analyses were stratified by those with (n=1626) and without (n=2860) symptomatic knee OA at baseline, defined as pain on most days of a month in the past 12 months. Propensity score weights were used to balance the four depressive symptom subtypes on baseline confounders: age, sex, race, education, smoking status, alcohol consumption, health insurance, employment status, BMI, Kellgren-Lawrence grade, history of knee injury, analgesic use, and total WOMAC score. Non-response weights were computed with logistic regression and were used to account for missing data. Weighted estimating equations were used to estimate pain and disability trajectories and to evaluate between-group differences in disease severity by baseline depressive symptom subtype during the follow-up period. Results: Among participants with symptomatic knee OA, the “Melancholic” depressive symptom subtype had more severe pain (β=3.61; 95% CI: -0.15, 7.39; P=0.060) and significantly greater disability (β=5.36; 95% CI: 1.41, 9.32; P=0.007) than the “No Symptoms” subtype (Table 1). In individuals without symptomatic knee OA, the “Anhedonic” subtype had significantly worse pain (β=1.33; 95% CI: 0.20, 2.47; P=0.020) and disability (β=1.34; 95% CI: 0.27, 2.42; P=0.014) compared to the “No Symptoms” subtype (Table 1). Differences in pain and disability between the “Catatonic” and “No Symptoms” subtypes were small and not statistically significant in both participants with and without symptomatic knee OA. Conclusion: Findings indicate that the effects of depressive symptoms on pain and disability are largest in persons with symptomatic knee OA who exhibit “Melancholic” symptomology. “Melancholic” depressive symptoms are characterized by the inability to feel pleasure, decreased energy and movement, and somatic complaints; and thus, may be a modifiable risk factor for worsening disease severity and potential target for intervention in patients with symptomatic knee OA.

Table 1. Differences in pain and disability by baseline depressive symptom subtype among Osteoarthritis Initiative participants with (n=1626) and without (n=2860) symptomatic knee OA over four years of follow-up.
Depressive Symptom Subtype	Pain			Disability
Symptomatic OA	β	95% CI	P Value	β	95% CI	P Value
No Symptoms	REF	REF	REF	REF	REF	REF
Catatonic	-0.09	-3.10, 2.91	0.950	0.10	-2.87, 3.07	0.945
Anhedonic	0.19	-2.11, 2.51	0.867	0.42	-1.79, 2.65	0.705
Melancholic	3.61	-0.15, 7.39	0.060	5.36	1.41, 9.32	0.007
Non-symptomatic OA	β	95% CI	P Value	β	95% CI	P Value
No Symptoms	REF	REF	REF	REF	REF	REF
Catatonic	0.76	-0.81, 2.34	0.340	0.11	-1.13, 1.35	0.862
Anhedonic	1.33	0.20, 2.47	0.020	1.34	0.27, 2.42	0.014
Melancholic	1.30	-1.08 3.69	0.284	1.83	-0.48, 4.15	0.120

Disclosure: A. Rathbun, None; M. Schuler, None; E. Stuart, None; M. Shardell, None; M. S. Yau, None; J. Gallo, None; M. C. Hochberg, Bioberica, 5,EMD Serono, 5,Novartis Pharma AG, 5,Plexxikon, 5,Pfizer, Inc., 5,Proximagen, 5,Regeneron, 5,Samumed, LLC, 5,Theralogix LLC, 5.

To cite this abstract in AMA style:

Rathbun A, Schuler M, Stuart E, Shardell M, Yau MS, Gallo J, Hochberg MC. Association between Depressive Symptom Subtypes and Disease Severity in Knee Osteoarthritis [abstract]. Arthritis Rheumatol. 2018; 70 (suppl 9). https://acrabstracts.org/abstract/association-between-depressive-symptom-subtypes-and-disease-severity-in-knee-osteoarthritis/. Accessed .

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/association-between-depressive-symptom-subtypes-and-disease-severity-in-knee-osteoarthritis/