• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Osteoarthritis Cartilage. Author manuscript; available in PMC Sep 1, 2012.
Published in final edited form as:
PMCID: PMC3159740

Psychological Health Impact on Two-Year Changes in Pain and Function in Persons with Knee Pain: Data from the Osteoarthritis Initiative

Daniel L. Riddle, PT, PhD, FAPTA, Otto D. Payton Professor,corresponding author Xiangrong Kong, PhD, and G. Kelley Fitzgerald, PT, PhD, FAPTA, Associate Professor



We determined whether baseline depressive symptoms, knee-related confidence and general psychological distress influenced changes in pain and function during two years of follow up.


We included persons in the OAI dataset with baseline pain of 1 or greater on a 0 to 10 scale in at least one knee and no knee or hip surgery during the two-year follow-up (n=3,407). The four outcome variables were repeated chair standing, 20 meter walk and WOMAC Pain and Disability. Linear mixed effects models assessed the association of each mental health variable with the yearly change in each baseline adjusted outcome measure after controlling for covariates.


Depressive symptoms were significantly predictive of worsening in most outcomes. The magnitude of worsening predicted for each year was small. For example, the dichotomized WOMAC Pain model indicated that depressed persons experience more rapid worsening than non-depressed persons at an average rate of 0.59 WOMAC points per year (95%CI 0.176, 1.013, p=0.005). Similar significant but very small effects of depressive symptoms on other outcomes were observed. Knee confidence was not predictive of change. General psychological distress was predictive of change in 20-meter walk and WOMAC Pain.


The most consistent psychological predictor of yearly worsening was baseline depressive symptoms. Although a statistically robust predictor of outcome, given that change was very small and highly dependent on baseline status, our results indicate that a considerable degree of persistent depressive symptoms would be required to have a meaningful effect on future self-reported outcome.

Osteoarthritis and soft tissue disorders of the lower extremities are a major burden for individuals and society. Knee pain is particularly common with a period prevalence of 18.1% of US men and 23.5% of US women aged 60 years and older reporting pain on most days during the previous 6 weeks.1 Knee OA also is common with an estimated prevalence of between 12%2 and 37%3 for persons aged 60 and older. As a result, many efforts have been made to identify and quantify the role of various risk factors in knee symptom or disease progression.412 Most of these studies have been devoted to the identification of demographic, biomarker and disease characteristics, surgical history or other musculoskeletal characteristics that predispose persons to worsening. Only a few studies, to our knowledge, have explored the role of mental health in predicting functional decline or worsening pain in samples of persons with knee pain or knee OA.68,11,13,14 The potential role of mental health in the progression of knee pain and functional decline is important because mental health is modifiable and therefore may be a target for interventions..

Belo and colleagues found that fear of movement independently predicted persistent symptoms one year following baseline measures taken on 480 persons with knee pain.6 Dunlop and colleagues found that depressive symptoms as measured with the CES-D15 independently predicted functional decline (adjusted OR 1.3, 95% CI 1.2, 2.0) in a population based sample of 4,922 persons with self-reported arthrits.7 Depression was not found to independently predict outcome for patients with knee pain or arthritis in studies by Mallen and colleagues, 8 Sharma et al,11 Jinks et al,13 and Holla and colleagues.14

The knee pain/OA literature suggests poor mental health impacts future function but the data are inconsistent. More research is needed to identify and to quantify the extent to which modifiable mental health constructs (e.g. self confidence) and disorders (e.g., depression) potentially influence future pain and function in persons with knee pain. The purposes of this study were to determine whether baseline depressive symptoms, knee related confidence and general psychological distress influenced yearly change in pain, functional status and physical performance during two years of follow up. We hypothesized that depressive symptoms would be the strongest and most consistent predictor of yearly worsening.

Materials and Methods

The Osteoarthritis Initiative

The Osteoarthritis Initiative (OAI) is a publicly and privately funded prospective 4 year longitudinal cohort study designed to identify and follow diverse cohorts of persons aged 45 to 79 years to examine the onset and progression of knee OA. Subjects received no treatment as part of the study. The study was approved by the Institutional Review Board at the University of California, San Francisco.

The sample is community based to the extent that only 6.5% of the sample reported seeing a doctor or other health professional for knee arthritis. No treatment was provided in the OAI but subjects were asked to self-report any treatments received in their communities. Subjects were recruited from: 1) the University of Maryland, Baltimore, Maryland, 2) Ohio State University, Columbus, Ohio, 3) University of Pittsburgh, Pittsburgh, Pennsylvania, and 4) Memorial Hospital of Rhode Island, Pawtucket, Rhode Island.

OAI study sample

Exclusion criteria were the presence of rheumatoid arthritis, bilateral knee arthroplasty or preexisting plans to undergo bilateral (not unilateral) knee arthroplasty in the next 3 years, bilateral OARSI stage 3 (severe) knee OA,16 positive pregnancy test, inability to provide a blood sample, use of ambulatory aids other than a single straight cane for more than 50% of the time, co-morbid conditions that might interfere with 4 year participation, unlikely to reside in clinic area for at least 3 years, current participation in a double-blind randomized controlled trial, and being unwilling to sign informed consent. In addition, because the study measured MRI based changes in a small subsample of subjects, men weighing more than 130 kgs and women weighing more than 114kg were excluded because they were unable to undergo 3.0Tesla MRI(2,295 persons excluded). In total, 27% (N=4,796) of those screened (N=17,457) were admitted to the study. Common reasons for exclusion were drop outs after the initial telephone screen (N=3,321) and admission quotas being met for certain age and gender categories (N=2,954).

Current study sample

The OAI combined incidence and progression sub-cohorts comprise 4,674 persons. Our exclusionary criteria were the following: 1) no pain in either knee at baseline as measured with a 0 to 10 verbal pain rating scale with 0 representing “no pain” and 10 representing “pain as bad as you can imagine”, 2) knee or hip surgery during the follow-up period. We only included persons reporting pain in one or both knees to capture those persons who may seek out healthcare for a knee problem. We excluded persons with knee or hip surgery during the follow-up period to avoid confounding due to potentially substantial changes in their lower extremity related pain or function subsequent to surgery. Thus 1,069 persons were excluded because they reported no baseline pain in either knee and an additional 198 persons were excluded because they underwent either hip or knee surgery during the follow-up period. The total sample for the current study was 3,407 persons.

Outcome variables of interest

We chose two performance-based measures and two self-report measures. The physical performance measures we used were the five chair stands test (standing from a chair 5 times as quickly as possible) measured in stands/second and the 20 meter walk test (at usual pace) measured in meters/second. Both measures are highly reliable with ICCs ranging from 0.93 to 0.98 within and between testers obtained on patients with moderate to severe knee OA.17 These tests were chosen because they represent commonly performed daily activities by persons with knee osteoarthritis and require differing amounts of endurance, strength and balance. Operations manuals for all data collected in the OAI study can be found at http://oai.epi-ucsf.org/datarelease/operationsManuals.asp.

For self-report outcome measures, we used the Pain and Disability sub-scores from the WOMAC scale (Likert 3.1), a commonly used and validated scale for patients with knee osteoarthritis. The WOMAC Pain Scale is scored 0 to 20 with higher scores indicating more severe pain while the WOMAC Disability Scale is scored 0 to 68 with higher scores indicating worse disability.1821 Although OAI investigators collected WOMAC scores for the left and right knees, the WOMAC was originally designed as a person-level self report measure with questions that ask about functional capability associated with activities involving bilateral knee function such as shopping, bending to the floor and standing.18,19 We reasoned that the higher WOMAC score of the two knees reported at each session best reflected the knee-related pain or disability for that person, and therefore used the higher of the left and right sided WOMAC scores for each session.

Predictor variables

Key predictor variables of interest

The key predictor variables were the following: a) the commonly endorsed SF-12 Mental Component Summary (MCS) 22. This 12-item self-report instrument is norm referenced (mean of 50 for general US population) with scores ranging from 0 to 68 with higher scores indicating higher mental function.23 b) The 20-item Center for Epidemiologic Studies Depression Scale (CES-D 20).. The CES-D is commonly used and ranges from 0 to 60 and a score of 16 or higher is generally accepted as the cut-score for indicating probable clinical depression.15,2426 c) The single item knee confidence scale from the KOOS Quality of Life Scale. The item asks “How much are you troubled with lack of confidence in your knee(s)?” The item is scored on a Likert scale from 0 (not at all) to 4 (Extremely). We chose this item because, in our experience, patients with knee pain sometimes report a lack of confidence in their knees and we suspected that lower confidence may impact negatively on future pain or function. We found no reliability evidence for this single KOOS item but similar Likert-type single-item scales of complex health constructs demonstrate reliability in the 0.7 to 0.8 range.27

Potential confounding variables

The approach to selecting covariate predictors was to identify variables found to be associated with knee pain or functional loss from large sample longitudinal studies of persons with either knee pain or knee OA.411 Predictor variables examined represent the full depth and breadth of data encompassing demographic and socioeconomic characteristics, variables measuring general health, arthritis specific health and physical examination results.

Age, gender, race, annual income and social support were the demographic and socioeconomic variables. The degree of social support indicated the number of persons living in the household. The validated Charlson comorbidity index, scored 0 to 31,28 BMI and Physical Activity Scale for the Elderly (PASE), 29 a validated self-report measure of daily function were included.

Subjects completed a variety of arthritis-related health measures including symptom duration, history of traumatic knee injury, whether the subject had generalized OA and whether the person had bilateral knee OA as determined by weight-bearing fixed flexion radiographs.30

Quadriceps and hamstring muscle torque tests (Newton Cm/bodyweight in kgs) each coded as the weakest score were included because of the association between thigh muscle strength and osteoarthritis.31 Varus and valgus alignment, measured in degrees was coded as the knee with the most severe alignment.11 The extent of knee flexion contracture, measured in degrees, was coded as the knee with the worst knee flexion contracture.

Data analysis

To assess the influence of each baseline psychological factor on the change of each outcome variable measured on a continuous scale, linear mixed effects model was used to model the outcome (measured at baseline and year 1 and year 2) as a function of time (t=0, 1,2). Linear mixed effects model is a regression model that is commonly used to assess effects of independent variables on the outcome variable. Specifically, it is adopted here to take into account the correlations between repeated measurements on the same individual. Also the intercept and slope parameter that measures the change per year are assumed to be normally distributed random effects. The mean values of the random effects allow estimation of population average trend across all individuals, whereas the variances of the random effects allow different variations from the population average for different individuals. The assumption of a linear trend over the 2-year follow-up was examined by graphical inspection and by testing the statistical significance of the term for second order of time. Both showed that the linear assumption was acceptable.

Baseline pain or functional status is the key determinant for a person’s pain or function measures at follow-up visits, and individuals with different levels of baseline pain or functional status were expected to have different levels of change during the two years of follow-up. 8,9 Baseline pain or function is an important confounder when estimating effects of psychological variables and was adjusted for in all analyses. To simplify interpretation, individuals were first categorized using quartiles of the baseline outcome measurements. This grouping variable and its interaction with time were included in the model as fixed effects predictor variables in order to account for the variation in trends determined by baseline pain or functional status. The psychological factor and its interaction with time were also treated as fixed effects to estimate the population average change of the outcome variable associated with the psychological factor, after controlling for the baseline pain or function level. All psychological factors were measured on continuous scales. Depression is a disorder that occurs along a continuum,25,32 but because the CES-D depression measure has a widely used and accepted cutpoint of ≥16 for probable clinical depression, we also modeled depression on this dichotomous scale.15

The effect of each psychological factor on pain or functional status change was estimated in multiple models that controlled for potential confounding variables. Specifically, a linear mixed effect model (as described above) for each confounding variable was fit, and if the confounding variable had significant influence on change in outcome measure with p-value<0.15, the variable was included in the multiple models.

The analyses were based on observed data likelihood and missing follow-up data were treated as missing at random (MAR), following Rubin's definition of missing data mechanisms.33,34 It was observed that subjects with higher levels of pain and poorer function at baseline were more likely to drop out. Given the determinative effect of baseline pain and function status on follow-up measures, the missingness at follow-up may depend on baseline measurements but will likely not depend on the unobserved follow-up values. Therefore the MAR assumption was accepted after adjusting for baseline pain or function status. All analyses were conducted in SAS 9.2 using two-sided tests and significance level 0.05. The proportion of explained variance for each model was calculated. 35


Baseline characteristics for the 3,407 participants are reported in Table 1. Summary scores over the two year period for each outcome measure, reported in quartiles, are summarized on Table 2. For the WOMAC Pain and Function scores, the first quartile of scores represents the best scores (least pain and highly functioning) while the fourth quartile represents subjects who had the worst scores (highest pain and lowest functioning). For the repeated chair stand and 20 meter walk test, the first quartile represents subjects with the worst scores while the fourth quartile indicates the best performance.

Table 1
Characteristics of 3407 persons in the study
Table 2
Descriptive Statistics for the Longitudinal data for key outcome variables

The loss to follow-up rate can be derived from Table 2. For example, 3,405 of the 3,407 (99.9%) subjects included in this study completed the baseline WOMAC Pain score, 3,146 (92.3%) completed the 1-year follow-up and 3,007 (88.3%) completed the 2-year follow-up. Subjects lost to follow-up reported more pain and functional loss and had more impaired performance at baseline than subjects with complete data (see Table 3).

Table 3
Comparison of Baseline Outcome Scores of Subjects with and without missing data at either the 1- or 2-year follow-up

Yearly changes in WOMAC Pain and Disability

Results for univariate models adjusted for baseline WOMAC Pain or Disability Scores and multiple models adjusting for all potential confounders appear in Table 4. After adjusting for multiple covariates, the Time X CES-D depression interaction was highly significant (p = <0.001) for predicting one and two-year changes in WOMAC Pain and Function. For example, each point increase in depressive symptoms at baseline resulted in a .02 point average increase in WOMAC Pain scores each year. Using the dichotomized CES-D scores, WOMAC Pain scores of depressed persons increased by 0.59 points each year and WOMAC Disability scores increased by 2 points each year compared to non-depressed persons.

Table 4
Summary of key findings from modified univariate and multivariate models adjusting from time. Each model describes the association between the psychological predictor variable and yearly changes in each key outcome measure over two years.

Baseline confidence, measured by question #3 from the KOOS Quality of Life Scale was not predictive of change in WOMAC Pain (p = 0.148) or Disability (p = 0.12). The multiple model using the baseline SF-12 MCS score significantly predicted one and two-year changes in WOMAC pain (p = 0.025). The SF-12 MCS score (p = 0.106) was not significantly related to one and two-year WOMAC Disability change scores. The proportion of explained variance in all models ranged from 23.3% to 28.5%.

Yearly changes in 20-meter walking pace and repeated chair stand pace

The univariate and multiple models for predicting changes in one- and two-year 20-meter walking pace and repeated chair stand pace scores appear in Table 4. For example, the Time X CES-D depression score interaction was significant (p < 0.001) indicating that for every point increase in depressive symptoms, an average reduction of 0.0009 meters per second in 20-meter walking pace occurs each year. The KOOS confidence question was not a significant predictor of future 20-meter walking pace (p = 0.068) or repeated chair stand pace (p = 0.218). The SF-12 MCS score X Time interaction was significant (p = 0.029) indicating that for every point increase in MCS score, 20-meter walking pace increased by .0004 meters per second, on average, per year. Tables of the complete multiple models for predicting yearly WOMAC Pain and 20 meter walk changes appear in the supplementary file.


Our findings indicate that of the three psychological measures examined in the OAI study, depressive symptoms as measured with the CES-D is the most robust independent predictor of yearly changes in self-reported pain, disability, walking pace and chair stand performance. The magnitude of the effect of depressive symptoms on each outcome is small, particularly for the performance-based measures to the extent that yearly reductions in performance-based measures, while statistically robust, are not clinically meaningful.

We use two examples from Table 4 to illustrate the prognostic impact of depression on future WOMAC Pain and Function. For example 1, the multiple Beta coefficient for the dichotomized WOMAC Pain model from Table 4 indicates that, relative to non-depressed persons, those with depression (CES-D ≥16) can expect to experience more rapid worsening at the average rate of 0.6 WOMAC points per year. For example 2, the multiple Beta coefficient for the dichotomized WOMAC Disability model from Table 4 indicates that, relative to non-depressed persons, those with depression can expect to have more rapid worsening at an average rate of approximately 2 points per year. Both examples illustrate that the patient’s depression would need to persist for multiple years to produce an appreciable impact on future pain and disability.

Our study has several strengths and these include a large sample size, an extensive and well defined list of covariates and an 88% follow-up over the 2-year period. We also reported the yearly impact of each mental health measure in the measurement units of interest (e.g. WOMAC Pain points) to allow for clear interpretation and clinical application. However, our study also had several limitations. The magnitude of yearly worsening predicted by the CES-D and SF-12 MCS scores was very small, primarily because changes over the two years were also small and were in the direction of group-level improvements. We suspect that much of these group level improvements, particularly for the worst scoring quartiles are due to selective loss to follow-up particularly for persons with higher initial disability or pain (Table 3). Nevertheless, since follow-up pain or performance highly depended on baseline status, we adjusted for baseline function or pain status in all analyses. Therefore, the missing at random assumption on loss-to-follow-up was acceptable and parameter estimates using the adjusted linear mixed effect models are unbiased. Some improvement also may be attributed to health monitoring by interested researchers and participation in a large federally funded study.

Improvement over time in large cohorts with knee pain or OA is not unusual. Belo and colleagues noted dramatic improvement in patients with knee pain over a 1 year period.6 Holla and colleagues reported an approximate 1 point mean improvement on a 100 point scale with large variation (sd=9.84) in WOMAC Disability scores for 832 persons with knee pain over a 2-year period.14 Given that our sample showed improvements over the two-year study period, depression was still found to be a statistically robust and consistent predictor of worsening. particularly for more depressed patients. This requires an assumption that similar changes would occur beyond the time of our follow-up and this may not be the case. In addition, our models using the continuous depression measure may be potentially limited because the relationship between depression score and changes in pain and functional status may not be linear.

Evidence collected on a variety of different types of patients has indicated that pain and depression have reciprocal effects on one another such that current depression or changes in depression predicts subsequent pain and vice versa.32,3638 The effects we found were robust and in the direction of current depression influencing future pain and functional status. We acknowledge that there are likely to be reciprocal and confounding effects of pain and potentially functional status on depression and that we only assessed and quantified the effects in the direction of depression influencing future pain and function. A recently published trial suggests that optimal depression and pain treatment in depressed arthritic patients reduces pain and enhances function.39 It is likely that treatments directed toward both pain and depression will have superior effects to treatment directed only to pain or to depression in patients with both musculoskeletal pain and depression.

We were unable to test for the potential impact of important personality traits such as neuroticism or perceived helplessness40 or mental health constructs such as pain catastrophizing or self efficacy.41,42 It is possible that these or other traits and constructs may impact future pain and function in patients with knee pain. Finally, only 27% of persons screened were actually entered into the study and this selection bias limits generalizability particularly to persons above MRI weight limits.

Prevalence of depression in persons with chronic arthritic or knee pain ranges from 15 to 20%.32 Prevalence of depression of 11.8% in our study is slightly lower than these estimates but also support the clinically important link between co-occurring pain and depression. Cross sectional correlations among baseline depression and WOMAC Pain (Pearson r = 0.26) and Disability (r= 0.29) in our study were low. These data suggest that subsequent changes in pain and disability were not simply attributable to high baseline correlations among pain, disability and depression.

Two studies examined similar heterogeneous mixes of persons with and without confirmed knee OA to determine if baseline depression was an independent predictor of outcome.6,8,14 Mallen and colleagues8 and Holla et al14 found that depressive symptoms were not predictive of outcome. We suspect that the reason for differences between our study and those reported by others is that we had a much larger sample and we treated the prognostic variables as continuous variables in most analyses.

In conclusion, our study has implications for clinicians in that persistent depressive symptoms may lead to self reported pain and function worsening for patients with knee pain. Actual physical performance does not appear to be impacted in a clinically meaningful way by depression or general psychological distress. Future studies are needed that examine the costs and effectiveness of interventions designed to treat depression and pain in persons with knee pain and co-occurring depression. Studies of potential mediating effects of changes in depression and pain on outcome in groups of patients with higher levels of depression are also needed.

Supplementary Material


Funding Acknowledgement: The OAI is a public-private partnership comprised of five contracts (N01-AR-2-2258; N01-AR-2-2259; N01-AR-2-2260; N01-AR-2-2261; N01-AR-2-2262) funded by the National Institutes of Health, a branch of the Department of Health and Human Services, and conducted by the OAI Study Investigators. Private funding partners include Merck Research Laboratories; Novartis Pharmaceuticals Corporation, GlaxoSmithKline; and Pfizer, Inc. Private sector funding for the OAI is managed by the Foundation for the National Institutes of Health. This manuscript was prepared using an OAI public use data set and does not necessarily reflect the opinions or views of the OAI investigators, the NIH, or the private funding partners.” The investigators on this paper are not part of the OAI investigative team. 3. The funding source played no role in the design of the study or the drafting of this manuscript.


Author contributions: DLR contributed to (1) the conception and design of the study, acquisition of data, or analysis and interpretation of data, (2) drafting the article and revising it critically for important intellectual content, (3) final approval of the submitted version. XK and GKF contributed to (1) the conception and design, analysis and interpretation of data, (2) drafting the article and revising it critically for important intellectual content, and (3) final approval of the submitted version.

Conflict of Interest. No conficts or interest were identified by the authors.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Daniel L. Riddle, Departments of Physical Therapy and Orthopaedic Surgery, Virginia Commonwealth University, Richmond, Virginia 23298-0224, Phone: 804-828-0234, Fax: 804-828-8111, ude.ucv@elddirld.

Xiangrong Kong, Department of Family Population and Reproductive Health and Department of Biostatistics, The Johns Hopkins University, Baltimore, MD, USA.

G. Kelley Fitzgerald, Department of Physical Therapy, University of Pittsburgh, Pittsburgh, Pennsylvania.

Reference List

1. Andersen RE, Crespo CJ, Ling SM, Bathon JM, Bartlett SJ. Prevalence of significant knee pain among older Americans: results from the Third National Health and Nutrition Examination Survey. J Am Geriatr Soc. 1999;47:1435–1438. [PubMed]
2. Quintana JM, Arostegui I, Escobar A, Azkarate J, Goenaga JI, Lafuente I. Prevalence of knee and hip osteoarthritis and the appropriateness of joint replacement in an older population. Arch Intern Med. 2008;168:1576–1584. [PubMed]
3. Lawrence RC, Felson DT, Helmick CG, et al. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part II. Arthritis Rheum. 2008;58:26–35. [PMC free article] [PubMed]
4. Amin S, Baker K, Niu J, et al. Quadriceps strength and the risk of cartilage loss and symptom progression in knee osteoarthritis. Arthritis Rheum. 2009;60:189–198. [PMC free article] [PubMed]
5. Belo JN, Berger MY, Reijman M, Koes BW, Bierma-Zeinstra SM. Prognostic factors of progression of osteoarthritis of the knee: a systematic review of observational studies. Arthritis Rheum. 2007;57:13–26. [PubMed]
6. Belo JN, Berger MY, Koes BW, Bierma-Zeinstra SM. Prognostic factors in adults with knee pain in general practice. Arthritis Rheum. 2009;61:143–151. [PubMed]
7. Dunlop DD, Semanik P, Song J, Manheim LM, Shih V, Chang RW. Risk factors for functional decline in older adults with arthritis. Arthritis Rheum. 2005;52:1274–1282. [PMC free article] [PubMed]
8. Mallen CD, Peat G, Thomas E, Lacey R, Croft P. Predicting poor functional outcome in community-dwelling older adults with knee pain: prognostic value of generic indicators. Ann Rheum Dis. 2007;66:1456–1461. [PMC free article] [PubMed]
9. Peters TJ, Sanders C, Dieppe P, Donovan J. Factors associated with change in pain and disability over time: a community-based prospective observational study of hip and knee osteoarthritis. Br J Gen Pract. 2005;55:205–211. [PMC free article] [PubMed]
10. Roos EM, Bremander AB, Englund M, Lohmander LS. Change in self-reported outcomes and objective physical function over 7 years in middle-aged subjects with or at high risk of knee osteoarthritis. Ann Rheum Dis. 2008;67:505–510. [PubMed]
11. Sharma L, Cahue S, Song J, Hayes K, Pai YC, Dunlop D. Physical functioning over three years in knee osteoarthritis: role of psychosocial, local mechanical, and neuromuscular factors. Arthritis Rheum. 2003;48:3359–3370. [PubMed]
12. Slemenda C, Heilman DK, Brandt KD, et al. Reduced quadriceps strength relative to body weight: a risk factor for knee osteoarthritis in women? Arthritis Rheum. 1998;41:1951–1959. [PubMed]
13. Jinks C, Jordan KP, Blagojevic M, Croft P. Predictors of onset and progression of knee pain in adults living in the community. A prospective study. Rheumatology (Oxford) 2008;47:368–374. [PubMed]
14. Holla JF, Steultjens MP, Roorda LD, Heymans MW, Ten WS, Dekker J. Prognostic factors for the two-year course of activity limitations in early osteoarthritis of the hip and/or knee: Results from the CHECK cohort. Arthritis Care Res (Hoboken ) 2010 [PubMed]
15. Radloff LS. The CES-D scale: a self report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401.
16. Altman RD, Hochberg M, Murphy WA, Jr, Wolfe F, Lequesne M. Atlas of individual radiographic features in osteoarthritis. Osteoarthritis Cartilage. 1995;3 (Suppl A):3–70. [PubMed]
17. Gill S, McBurney H. Reliability of performance-based measures in people awaiting joint replacement surgery of the hip or knee. Physiother Res Int. 2008;13:141–152. [PubMed]
18. Bellamy N, Buchanan WW. A preliminary evaluation of the dimensionality and clinical importance of pain and disability in osteoarthritis of the hip and knee. Clin Rheumatol. 1986;5:231–241. [PubMed]
19. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol. 1988;15:1833–1840. [PubMed]
20. Bellamy N, Kirwan J, Boers M, et al. Recommendations for a core set of outcome measures for future phase III clinical trials in knee, hip, and hand osteoarthritis. Consensus development at OMERACT III. J Rheumatol. 1997;24:799–802. [PubMed]
21. Bellamy N. The WOMAC Knee and Hip Osteoarthritis Indices: development, validation, globalization and influence on the development of the AUSCAN Hand Osteoarthritis Indices. Clin Exp Rheumatol. 2005;23:S148–S153. [PubMed]
22. Ware J, Jr, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34:220–233. [PubMed]
23. Ware JE, Kosinski M, Turner-Bowker DM, Gandek B. How to Score Verson 2 of the SF-12 Health Survey. Lincoln, Rhode Island: QualityMetric Incororporated; 2002.
24. Blalock SJ, DeVellis RF, Brown GK, Wallston KA. Validity of the Center for Epidemiological Studies Depression Scale in arthritis populations. Arthritis Rheum. 1989;32:991–997. [PubMed]
25. Hybels CF, Pieper CF, Blazer DG. The complex relationship between depressive symptoms and functional limitations in community-dwelling older adults: the impact of subthreshold depression. Psychol Med. 2009;39:1677–1688. [PMC free article] [PubMed]
26. Lewinsohn PM, Seeley JR, Roberts RE, Allen NB. Center for Epidemiologic Studies Depression Scale (CES-D) as a screening instrument for depression among community-residing older adults. Psychol Aging. 1997;12:277–287. [PubMed]
27. McDowell I. Measures of self-perceived well-being. J Psychosom Res. 2010;69:69–79. [PubMed]
28. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–383. [PubMed]
29. Martin KA, Rejeski WJ, Miller ME, James MK, Ettinger WH, Jr, Messier SP. Validation of the PASE in older adults with knee pain and physical disability. Med Sci Sports Exerc. 1999;31:627–633. [PubMed]
30. Brandt KD, Mazzuca SA, Conrozier T, et al. Which is the best radiographic protocol for a clinical trial of a structure modifying drug in patients with knee osteoarthritis? J Rheumatol. 2002;29:1308–1320. [PubMed]
31. Wood L, Peat G, Thomas E, Hay EM, Sim J. Associations between physical examination and self-reported physical function in older community-dwelling adults with knee pain. Phys Ther. 2008;88:33–42. [PubMed]
32. Bair MJ, Robinson RL, Katon W, Kroenke K. Depression and pain comorbidity: a literature review. Arch Intern Med. 2003;163:2433–2445. [PubMed]
33. Hogan JW, Roy J, Korkontzelou C. Handling drop-out in longitudinal studies. Stat Med. 2004;23:1455–1497. [PubMed]
34. Little R, Rubin D. Statistical Analysis With Missing Data. John Wiley and Sons; 2002.
35. Xu R. Measuring explained variation in linear mixed effects models. Stat Med. 2003;22:3527–3541. [PubMed]
36. Geerlings SW, Twisk JW, Beekman AT, Deeg DJ, van TW. Longitudinal relationship between pain and depression in older adults: sex, age and physical disability. Soc Psychiatry Psychiatr Epidemiol. 2002;37:23–30. [PubMed]
37. Hurwitz EL, Morgenstern H, Yu F. Cross-sectional and longitudinal associations of low-back pain and related disability with psychological distress among patients enrolled in the UCLA Low-Back Pain Study. J Clin Epidemiol. 2003;56:463–471. [PubMed]
38. Williams LS, Jones WJ, Shen J, Robinson RL, Kroenke K. Outcomes of newly referred neurology outpatients with depression and pain. Neurology. 2004;63:674–677. [PubMed]
39. Kroenke K, Bair MJ, Damush TM, et al. Optimized antidepressant therapy and pain self-management in primary care patients with depression and musculoskeletal pain: a randomized controlled trial. JAMA. 2009;301:2099–2110. [PMC free article] [PubMed]
40. Costa PT, Jr, McCrae RR, Zonderman AB. Environmental and dispositional influences on well-being: longitudinal follow-up of an American national sample. Br J Psychol. 1987;78 ( Pt 3):299–306. [PubMed]
41. Sullivan MJL, Bishop S, Pivik J. The Pain Catastrophizing Scale: Development and Validation. Psych Assessment. 1995;7:524–532.
42. Lorig K, Holman H. Arthritis Self-Efficacy Scales measure self-efficacy. Arthritis Care Res. 1998;11:155–157. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...