Extending and Enhancing the Osteoarthritis Initiative
Participants Roster

June 2, 2008 (historical)

Michael Nevitt, PhD, UCSF
Charles McCulloch, PhD, UCSF
Gayle Lester, PhD, NIAMS

1. Introduction.

The OAI has had significant success in achieving its initial goals. However, the scientific value of the OAI will be enhanced in several important ways by an extension of clinical and imaging follow-up for another 4 years, to a total of 8 years, with an additional two clinic visits (at 72 and 96 months) and two phone interviews (60 and 84 months). (Details of the measurements proposed at each time-point can be found in the Proposal for Extension of Follow-up in the OAI. The complete protocol for the OAI cohort study is located at www.oai.ucsf.edu.)

The proposed extension will enhance three key aspects of OAI. First, the extended longitudinal nature of the design will allow a better understanding of the natural history and the time course of both early and established clinical disease and their determinants.

Second, and in light of the slowly developing trajectory of knee OA, additional follow-up will permit investigations into the evolution of knee OA pain states and transitions to end-stage symptomatology, functional limitation and total joint replacement. This will strengthen a major objective of OAI, which is to evaluate imaging, biochemical and genetic biomarkers of disease risk, disease activity and progression in relation to key clinical outcomes.

Finally, it will provide increased statistical power for key analyses, including needed gains in power for important subgroups, such as men and African Americans, as well as enhancing power for primary aims in the overall cohort.

2. Progress in achieving the OAI’s goals

The OAI is now in the second half of what has been a highly successful project in terms of enrollment of the cohort, data collection, retention and follow-up and distribution of public use datasets. 4,796 men and women (96% of the goal of 5,000) completed a baseline examination including joint imaging. Participants were assigned to one of three subcohorts, the Progression subcohort of persons with symptomatic knee OA (SxKOA) at enrollment, the larger Incidence subcohort comprised of those without SxKOA but at increased risk for developing it by virtue of having risk factors, and a small group with neither knee OA nor risk factors at enrollment (normal controls). Overall, nearly 21% are ethnic minorities, primarily African Americans, and 42% are men (Table 1). Compared to initial expectations, the overall enrolled cohort had a somewhat greater proportion with SxKOA (32% vs 20%) and women (58% vs 50%), reflecting greater challenges in recruiting men and those without disease. The increased enrollment of those with SxKOA at baseline will actually facilitate investigation of biomarkers of disease progression during the first few years of the study.

Table 1. Baseline characteristics of OAI participants, by major subcohort
Age, mean (SD) 61.4 (9.1) 61.3 (9.2)
BMI, mean (SD) 30.2 (4.9) 28.1 (4.6)
Men 43.0% 41.0%
Nonwhite 29.1% 17.7%


Follow-up visits for joint imaging, biospecimen collection and clinical outcomes occur annually over 4 years (with one six-month interim visit for Progression subcohort participants). The 24-month clinic visit is nearly complete, the 36-month visit is ongoing and the clinics have just begun the 48-month visit. To date, retention is well within projections (Table 2). Despite a lengthy knee MRI exam (˜90 min.) and participant accumulation of MR contraindications, imaging completion rates are high, assuring adequate longitudinal data for imaging biomarker investigations. For example, 81% of all participants enrolled will have baseline and 24-month knee MRIs available for analysis.

Table 2. Follow-up and retention
  12-mo visit 24-mo visit
Had clinic visit (projected) 90% 82%1
Projected clinic visit 90% 84%
% with clinic visit who had knee MRI 97% 94%
% with clinic visit who had biospecimens >99% >99%


1 5% of clinic visits pending as of 5/1/08.

Central assessment of knee MRIs and knee x-rays from the baseline, 12-month and 24-month visits in 600 subjects from the Progression subcohort are in progress, and additional assessments in all subcohorts are planned. These include quantitative and semiquantitative measurement methods applied to both imaging modalities (Table 3).

Table 3. Central image assessments in 600 progression cohort subjects

Knee x-rays (BL, 12-mo, 24-mo)

  • BL K-L grade and IRFs
  • Semiquantitative JSN progression
  • Quantitative JSW progression

MRIs ( index knee: BL, 12-mo, 24-mo)

  • BL whole organ score
  • Semiquantitative cartilage loss
  • Quantitative cartilage loss

In meeting its operational objectives, the OAI is on course to achieve its scientific goals by making available to investigators longitudinal imaging and clinical data through OAI Online www.oai.ucsf.edu. To date, data and images from the baseline, 12-month and 24-month visit have been released for public use (Table 4).

Table 4. OAI public data users (4/18/08)
Registered for OAI Online N=890
Researchers who have downloaded clinical datasets N=146
Joint image sets distributed (sets include images from 200 to 5,372 ppt visits) N=87


3. Scientific value of extended follow-up of the OAI cohort

The proposed extension of OAI will enhance the longitudinal nature of the study in several significant ways, enabling researchers to better investigate the natural history and time course of knee OA and its determinants, and to evaluate biomarkers of disease activity and progression in relation to clinically important transitions to end-stage disease.

For most people most of the time, knee OA is a slowly developing and progressing condition (1-3) necessitating relatively long and large studies to investigate its natural history, identify risk factors and assess clinically meaningful patient outcomes. This fact is corroborated by two studies that examined rates of cartilage loss over the first year of follow-up in OAI using images from 160 subjects from the Progression subcohort made available for public use in 2007 (4, 5). Over the first 12 months of follow-up, only modest cartilage loss was observed (2-3%), on average, in knees with symptomatic OA, although the rate of loss showed considerable variability between subjects and locations in the knee (4-7).

Despite recognition of its slowly developing nature, there is a paucity of data on the long-term course of knee OA. Only a handful of mostly small studies have as much as 8 years of follow-up and include comprehensive radiographic or clinical outcomes (1, 3, 8, 9). An extension of imaging and clinical follow-up in OAI will help fill this critical gap. There is a need for such data at both ends of the spectrum of knee OA: on one end the development of early disease and its evolution to established SxKOA, and on the other end progression of established SxKOA to end-stage structural damage and symptomatology, disability and total joint replacement. With the proposed extension, the OAI will be the only long-term study of knee OA that uses state of the art 3 Tesla MR imaging techniques to provide precise and sensitive measurement of cartilage loss and other structural changes at both ends of the disease spectrum and relate these to long-term clinical outcomes.

3a. The value of extended follow-up for investigations of established SxKOA

By virtue of frequent imaging assessments (up to 6 time-points over 4 years), the OAI data address the challenging “chicken or egg” questions with regard to structural degeneration and the factors affecting it. Important questions include: Which tissues are affected first (menisci, synovium, bone, muscle, cartilage)? How do tissue changes unfold spatially and temporally, and in what order? What drives and mediates these structural changes? Various patterns of spatial and temporal evolution of tissue damage can be elucidated over the first several years of the study. Likewise, changes in biochemical markers that accompany these pathoanatomical developments can be identified. Extended follow-up will provide a sufficient number of events (and time) to follow forward from these developments to consequential structural outcomes, such as extensive cartilage loss, and to long-term clinical outcomes.

1 Subjects had clinical/symptomatic tibiofemoral (TF) knee OA in one or both knees. In OAI, this is defined as having at least a definite osteophyte plus frequent knee symptoms — pain, aching or stiffness on most days of a month in the past year — in the same knee.

One of the goals of the OAI is to identify sensitive imaging biomarkers of disease progression in order to reduce the duration and cost of proof of concept studies and clinical trials of new treatments. It is essential that such biomarkers be evaluated in relation to longer-term structural and clinical outcomes. While powerful and precise MRI techniques will detect small amounts of cartilage loss over relatively short periods of time, the clinical significance of these changes remains uncertain. Because structural progression is often episodic, rapid change over any short interval may not be predictive of a long-term poor outcome. Studies of 1-3 years duration looking at the relationship of MRI cartilage loss and clinical outcomes such as knee pain have found weak or modest associations (10, 11). The cumulative trajectory of structural worsening and its relationship to the evolution of clinical outcomes will likely take longer to become evident. This is an area of investigation that will be significantly enhanced with extended follow-up.

The course of pain and physical function limitation in knee OA is highly variable, particularly over the short-term, with a substantial percent of patients reporting improvement over time (2, 8, 12-14). Knee OA clinical trials of a few years duration typically do not observe worsening in knee pain and function in the overall study population, although a subset of patients shows decline and a fraction of these show rapid decline (15, 16). Such variability in clinical trajectory is highly evident in the OAI, with a significant proportion of participants reporting both worsening and improvement (and back again) in knee pain and physical function limitation between evaluations. Investigators will be able to take advantage of the OAI’s repeated measures of structural pathology, symptoms and physical function to help elucidate their interrelationships (17).

It is increasingly recognized that the pain experience of persons with knee and hip OA is heterogeneous, not only in that it can vary greatly over time, but that it also ranges from intermittent pain (often triggered by specific activities that are avoided to control the pain), to pain that is persistent (often punctuated by intense, unpredictable episodes) and difficult to control (18, 19). In addition to an extensive battery of standard questions about knee symptoms, at the 48-month visit the OAI began to collect data using a new OARSI-OMERACT sponsored questionnaire designed to assess the occurrence and intensity of both episodic and persistent knee pain (20). Continued follow-up with this instrument will enhance the OAI’s contribution to understanding the evolution of pain and functional limitation outcome states and how the transition to endstage symptoms affects the risk of developing disability and the utilization of joint replacement surgery (18, 21).

This new pain questionnaire is also a component of the proposed definition of a "theoretical time to the fulfillment of criteria for joint replacement," which is being developed by OARSI/OMERACT to provide a primary endpoint for structure modifying interventions in knee OA (22). Assessment of these criteria for joint replacement during extended follow will provide an additional outcome for evaluation of biomarkers measured earlier in the study.

While knee OA is a common cause of disability in older persons (23, 24), participants in the OAI are relatively young and expected rates of transition to disability early in the study are correspondingly low. To assess disability onset during extended follow-up, at the 48-month visit the OAI has also started to administer the Late Life Disability Instrument (25). Further follow-up with this instrument will add an important dimension to long-term patient outcomes in the study.

3b. The value of extended follow-up for investigation of early knee OA

As previously mentioned, a strength of the OAI is that it encompasses a broad spectrum of disease. The large Incidence subcohort supports a focus on early knee OA, a stage of disease that holds promise for prevention and treatment. Treatment strategies for OA are currently focused on interventions in established clinical disease, designed to slow or delay progression to extensive cartilage loss and other endstage tissue damage, persistent pain, disability and total joint replacement. But there have been no unequivocally successful trials of disease modifying therapies in established SxKOA. Traditional treatment studies also seek to enroll "fast progressors" in order to minimize study size and duration. Yet, it may be proportionately more difficult to slow or reverse the disease process in the most rapidly worsening knees if the joint’s normal biomechanics have already been substantially disrupted (26-28). Thus, it is increasingly recognized that opportunities for intervention earlier in the disease process, before the downward slope has become too steep, should be explored.

For prevention and early intervention to be possible, it is necessary to better define early disease and understand its evolution to clinical OA. Extended follow-up in OAI will greatly enhance such investigations. Many OAI participants had mild symptoms and structural changes when they enrolled and many others are developing such changes during follow-up. Extended follow-up will allow investigators to construct alternative definitions of early disease based on emergent structural changes and symptoms during initial follow-up and to evaluate these definitions in terms of the risk of transition to definite radiographic OA, full-blown clinical disease and beyond.

It will also be possible to search for biomarkers that identify subgroups of knees/subjects with early changes that are most likely to continue on to consequential disease, and which may suggest treatment targets or identify those most likely to benefit from early intervention. For example: In radiographically normal knees that develop symptoms, which MRI changes (e.g. bone marrow abnormalities, meniscal damage, evidence of synovitis from nonenhanced scans, ligamentous lesions, etc.) presage development of definite structural and clinical OA (29-31). In a similar manner, the OAI will generate data on changes in risk factor exposures during the initial years of the study, such as a decline in quadriceps strength, weight changes, new knee injuries and surgeries, the development of meniscal damage, and changes in biochemical markers. Extended follow-up is required to gain sufficient time to assess exposure changes over the first few years in order to evaluate their effect on the subsequent risk of OA development and progression.

Finally, in a traditional cohort study of OA progression, a cross-section of persons with existing disease is enrolled and the resulting sample is heterogeneous in terms of length of time disease has been present and previous rates of progression. Disease of longer duration that has been relatively stable may be overrepresented while recent onset disease that progresses more rapidly may be under-represented, which may bias analyses of progression. Extended follow-up in the OAI will present the unique opportunity to follow knees from onset of disease (uniformly defined in terms of structural and symptom changes) forward to significant structural and clinical outcomes.

4. Benefits of increased statistical power for key endpoints

Extended follow-up will provide increased statistical power for key endpoints, including needed gains in power for important subgroups, such as men and African Americans, as well as enhancing power for primary aims in the overall cohort. By taking advantage of maturation and aging of the cohort and the likely continued high rates of retention among subjects who have participated in several study examinations, preliminary data suggest that an extension of follow-up will more than double the expected number of total knee replacement and incident radiographic knee OA endpoints.

4a. Total knee replacement

Endstage disease, as represented by the need for and occurrence of total joint replacement, is an important clinical outcome of knee OA, a potential endpoint in clinical trials of disease-modifying treatments(33) and a useful criterion for validating biomarkers of OA progression. The rates of total knee replacement (TKR) observed in OAI during the first 24 months of follow-up are consistent with this being a sample recruited from the community with relatively less severe disease at baseline. When extrapolated over the initial 4 years of follow-up, using conservative assumptions, the observed rates will yield just over 125 TKRs overall (Table 5)2.

Table 5. Expected number of TKRs with 4 and 8 years of follow-up
  4 years 8 years
Women 80 206
Men 46 118
African-Americans 23 60
All participants 126 324


A typical analysis using this data would be a nested case-control study (an efficient design for expensive measurements and uncommon endpoints) of the association of a biomarker, measured in persons (e.g. genetic or biochemical markers), with TKR.

2 Estimated rates of TKR were derived from self-report during the first 24 months of follow-up. Rates were estimated separately by gender and race and projections allow for dropout, aging of the cohort and the fact that not all self-reported TKRs are subsequently confirmed from medical records or study radiograph.

With 4 years of follow-up the OAI will have moderate power for more common exposures (Figure 3). For example, for a baseline marker with a 30% prevalence of abnormal values, the study will have 80% power to detect a hazard ratio of 1.85 or greater for predicting TKR.

Figure 1. Power for biomarker predictor (measured in persons) of TKR: A nested case-control analysis with 4 years of follow-up, ALL subjects eligible

Chart showing power for biomarker predictor after 4 years of follow-up. One axis shows the power and the other axis shows the odds ratio.

As observed in other cohort studies (e.g. SOF, WHI), rates of major disease endpoints are likely to increase over time in OAI as the cohort ages and the "healthy volunteer" effect wanes. Long-term follow-up for endpoints also capitalizes on the fact that the dropout rate declines substantially after the first few years of a study. Conservative projections (Table 5) indicate that the number of TKRs by 8 years will be more than double the number in the first 4 years of follow-up. This will increase power in nested case-control analyses so that for a baseline biochemical marker (or other predictor measured in persons) with a prevalence as low as 15%, we will have 80% power to detect odds ratios of 1.65 or more (Figure 4). In general, power will be greater for predictors like knee MR imaging biomarkers (for a predictor are measured in knees, rather than in persons, correlation between outcomes in a person’s two knees has less of an effect on power) and for measures like quadriceps strength (because it was assessed in the whole cohort and a nested case-control design is not needed).

Figure 2. Power for a biomarker predictor (measured in persons) of TKR: A nested case-control analysis with 8 years of follow-up, ALL subjects eligible

Chart showing power for biomarker predictor after 8 years of follow-up. One axis shows the power and the other axis shows the odds ratio.

Even in the subgroup of men the study will have adequate power for a knee MRI biomarker predicting TKR with 8 years of follow-up. For example in a nested case-control analysis there will be 80% power to detect an odds ratio of 1.86 or greater for a knee MRI biomarker with a 20% prevalence (Figure 3). For a biomarker measured in all men in the cohort (not just in cases and controls) with a 20% prevalence, the minimum detectable hazard ratio (in a time-to-vent analysis) for TKR would be 1.75. For an analysis of a biomarker measured in all African Americans, if an abnormal value is present in 40% the detectable hazard ratio would be just over 2.0. Power for all analyses will be substantially greater in women, with over 200 TKRs expected by 8 years.

Figure 3. Power for a biomarker predictor (measured in knees) of TKR: A nested case-control analysis with 8 years of follow-up, in MEN

Chart showing power for biomarker predictor after 8 years of follow-up in men. One axis shows the power and the other axis shows the odds ratio.

As discussed above, during the proposed extension of follow-up the OAI will collect data on pain, function and radiographic changes that can be used to identify knees that meet "states of severity defining the theoretical indication for TKR", currently under development by OMERACT/OARSI (22). This new approach will likely increase the number of "total knee replacement" endpoints and reduce the effect of demographic disparities in utilization on endpoint frequency, particularly for African Americans, who have a reduced utilization of joint replacement relative to their need (34). However, likely effects on endpoint numbers cannot be estimated without additional data.

4b. Incident radiographic OA

Although the OAI will have sufficient cases of incident radiographic OA to achieve its primary scientific goals, an extension of follow-up will enhance the study’s power for this endpoint overall and also yield adequate power for analyses in men and African Americans.

Table 6. Expected cases of incident radiographic T-F OA (definite osteophytes) in the Incidence cohort, with 4 and 8 years of follow-up
  4-years 8 years
Women 167 367
Men 100 221
African-Americans 70 158
All participants 267 588


To allow investigations of incident radiographic T-F disease, the OAI will provide a standardized central assessment of this endpoint. Even though radiographic OA is not always accompanied by symptoms, the new development of radiographic changes of knee OA is a signal event in the natural history of the disease. Unlike symptoms, structural OA of the knee as seen on radiograph rarely, if ever, resolves. Its occurrence in knees with symptoms of OA is associated with an increased risk of clinical progression (35, 36) and a high risk of long-term structural damage (1).

The number of incident radiographic T-F knee OA cases expected in the Incidence subcohort with 4 and 8 years of follow-up is shown in Table 63. Figure 4 shows the power for a baseline biochemical marker (or other factor measured in persons) in a nested case-control analysis with incident radiographic OA as the outcome, using data from 4 and 8 years of follow-up. Power for exposures with a prevalence of 20% or greater is excellent with 4 years of follow-up. The study will have 80% power to detect a hazard ratio as small as 1.63 for a 20% exposure and 1.55 for a 40% exposure. With 8 years of follow-up the study will have sufficient power to detect odds ratios as small as 1.54 even for biomarkers with a prevalence of only 10%. Power for parallel analyses to those described above, but using predictors defined from observations made during the first four years of follow-up and incident OA detected during extended follow-up, will be somewhat better than shown in Figure 4a; the number of incident cases is expected to be about 20% greater in the second compared to the first four years.

Fig 4. Power for a biomarker predictor (measured in persons) of incident radiographic T-F OA: A nested case-control analysis in the Incidence subcohort, with 4 and 8 years of follow-up, ALL subjects eligible.

Two charts side by side showing power for biomarker predictor after 4 and 8 years of follow-up. One axis on each chart shows the power and the other axis on each chart shows the odds ratio.

3 Incident radiographic T-F OA was defined as the new development of definite osteophytes in knees without them at baseline. Estimated incidence rates were derived from 30 months of follow-up in the MOST study. MOST also enrolled subjects at risk for knee OA by virtue of risk factors (37) and has performed an assessment of incident T-F OA over 30 months using the same knee radiography protocol as OAI. Rates were estimated separately by gender and race and projections allow for dropout and aging of the cohort.

As mentioned above, power will be higher for imaging biomarkers that are assessed from knee images, due to the reduced impact of inter-knee correlations in these analyses. Figure 5 shows the study’s power for analysis of baseline knee imaging biomarker predictors of incident knee OA with 8 years of follow-up in men and in African Americans. The study will have adequate power in both men and African Americans for exposures with a prevalence of 20% or greater, with 80% power to detect odds ratios as of 1.60 and 1.72, respectively, sufficient for evaluating the performance of knee imaging biomarkers in these important subgroups.

Fig 5. Power for an imaging biomarker predictors (measured in knees) of incident radiographic T-F OA: A nested case-control analysis in the Incidence subcohort, with 8 years of follow-up, Men and African Americans

Two charts side by side showing power for biomarker predictor after 4 and 8 years of follow-up in men and African Americans. One axis on each chart shows the power and the other axis on each chart shows the odds ratio.


  1. Wolfe F, Lane NE. The longterm outcome of osteoarthritis: rates and predictors of joint space narrowing in symptomatic patients with knee osteoarthritis. J Rheumatol 2002;29(1):139-46.
  2. van Dijk GM, Dekker J, Veenhof C, van den Ende CH. Course of functional status and pain in osteoarthritis of the hip or knee: a systematic review of the literature. Arthritis Rheum 2006;55(5):779-85.
  3. Schouten JSAG, van den Ouweland FA, Valkenburg HA. A 12 year follow up study in the general population on prognostic factors of cartilage loss in osteoarthritis of the knee. Ann Rheum Dis 1992;51:932-937.
  4. Hunter DJ, Niu J, Zhang Y, Totterman S, Tamez J, Dabrowski C, et al. Change in cartilage morphometry: a sample of the progression cohort of the Osteoarthritis Initiative. Ann Rheum Dis 2008.
  5. Eckstein F, Maschek S, Wirth W, Wyman B, Hudelmaier M, Nevitt M, et al. Change in Femorotibial Cartilage Volume and Subregional Cartilage Thickness Over 1 Year - Data from the Osteoarthritis Initiative Progression Subcohort. Arthritis Rheum 2007;56(9S):S283.
  6. Maschek S, Wirth W, Helllio Le Graverand M, Wyman B, Hudelmaier M, Hitzl W, et al. Cartilage Thickness Changes in the Knee Over 1 Year and Associated Risk Factors - Data from the Osteoarthritis Initiative Progression Subcohort. Osteoarthritis Cartilage 2007;15(S3).
  7. Wirth W, Maschek S, Helllio Le Graverand M, Wyman B, Hudelmaier M, Nevitt M, et al. Sensitivity to Change of Cartilage Thickness for Different Subregions of the Knee - Data from the Osteoarthritis Initiative Progression Subcohort. Osteoarthritis Cartilage 2007;15(S3).
  8. Dieppe P, Cushnaghan J, Tucker M, Browning S, Shepstone L. The Bristol 'OA500 study': progression and impact of the disease after 8 years. Osteoarthritis Cartilage 2000;8(2):63-8.
  9. Felson DT, Zhang Y, Hannan MT, Naimark A, Weissman BN, Aliabadi P, et al. The incidence and natural history of knee osteoarthritis in the elderly. The Framingham Osteoarthritis Study. Arthritis Rheum 1995;38(10):1500-5.
  10. Raynauld JP, Martel-Pelletier J, Berthiaume MJ, Beaudoin G, Choquette D, Haraoui B, et al. Long term evaluation of disease progression through the quantitative magnetic resonance imaging of symptomatic knee osteoarthritis patients: correlation with clinical symptoms and radiographic changes. Arthritis Res Ther 2006;8(1):R21.
  11. Amin S, Niu JB, LaValley M, Guermazi A, Grigoryan M, Hunter DJ, et al. Cartilage loss of the knee on MRI is not related to progression of knee pain or disability. Arthritis Rheum 2004;50(9S):141.
  12. Felson DT. The course of osteoarthritis and factors that affect it. Rheumatic Disease Clinics of North America 1993;19:607.
  13. 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(512):205-11.
  14. Paradowski PT, Englund M, Lohmander LS, Roos EM. The effect of patient characteristics on variability in pain and function over two years in early knee osteoarthritis. Health Qual Life Outcomes 2005;3:59.
  15. Brandt KD, Mazzuca SA. Lessons learned from nine clinical trials of disease-modifying osteoarthritis drugs. Arthritis Rheum 2005;52(11):3349-59.
  16. Bingham CO, 3rd, Buckland-Wright JC, Garnero P, Cohen SB, Dougados M, Adami S, et al. Risedronate decreases biochemical markers of cartilage degradation but does not decrease symptoms or slow radiographic progression in patients with medial compartment osteoarthritis of the knee: results of the two-year multinational knee osteoarthritis structural arthritis study. Arthritis Rheum 2006;54(11):3494-507.
  17. Zhang YQ, Nevitt MC, Niu J, al. E. Reversible MRI Features and Knee Pain Fluctuation: The MOST Study. . Osteoarthritis Cartilage 2007;15(Suppl C):17.
  18. Hawker GA, Stewart L, French MR, Cibere J, Jordan JM, March L, et al. Understanding the pain experience in hip and knee osteoarthritis - an OARSI/OMERACT initiative. Osteoarthritis Cartilage 2008;16(4):415-422.
  19. Duncan R, Peat G, Thomas E, Hay E, McCall I, Croft P. Symptoms and radiographic osteoarthritis: not as discordant as they are made out to be? Ann Rheum Dis 2007;66(1):86-91.
  20. Hawker GA, Davis AM, French MR, Cibere J, Jordan JM, March L, et al. Development and preliminary psychometric testing of a new OA pain measure - an OARSI/OMERACT initiative. Osteoarthritis Cartilage 2008;16(4):409-414.
  21. Peat G, Thomas E, Croft P. Staging joint pain and disability: a brief method using persistence and global severity. Arthritis Rheum 2006;55(3):411-9.
  22. Gossec L, Hawker G, Davis AM, Maillefert JF, Lohmander LS, Altman R, et al. OMERACT/OARSI initiative to define states of severity and indication for joint replacement in hip and knee osteoarthritis. J Rheumatol 2007;34(6):1432-5.
  23. Peat G, McCarney R, Croft P. Knee pain and osteoarthritis in older adults: a review of community burden and current use of primary health care. Annals of Rheumatic Diseases 2001;60:91-97.
  24. Guccione AA, Felson DT, Anderson JJ, Anthony JM, Zhang Y, Wilson PWF, et al. The effects of specific medical conditions on the functional limitations of elders in the Framingham Study. Am J Public Health 1994;84:351-358.
  25. Sayers SP, Jette AM, Haley SM, Heeren TC, Guralnik JM, Fielding RA. Validation of the Late-Life Function and Disability Instrument. J Am Geriatr Soc 2004;52(9):1554-9.
  26. Brandt KD, Radin EL, Dieppe PA, van de Putte L. Yet more evidence that osteoarthritis is not a cartilage disease. Ann Rheum Dis 2006;65(10):1261-4.
  27. Felson DT, Kim YJ. The futility of current approaches to chondroprotection. Arthritis Rheum 2007;56(5):1378-83.
  28. Sharma L. The role of varus and valgus alignment in knee osteoarthritis. Arthritis Rheum 2007;56(4):1044-7.
  29. Javaid MK, Lynch JA, Tolstykh I, Guermazi A, Roemer F, Aliabadi P, et al. Preradiographic knee MRI findings are associated with the onset of knee pain: the MOST study. Arthritis Rheum 2007;56(9S):124.
  30. Karachalios T, Zibis A, Papanagiotou P, Karantanas AH, Malizos KN, Roidis N. MR imaging findings in early osteoarthritis of the knee. Eur J Radiol 2004;50(3):225-30.
  31. Cibere J, Hongbin J, Thorne A, Garnero P, Poole R, Lobanok T, et al. Urine CTX-II and C2C are differentially associated with radiographic and preradiographic knee OA: Results of a population-based study using MRI. Arthritis Rheum 2006;54(9S):160.
  32. Birrell F, Croft P, Cooper C, Hosie G, Macfarlane G, Silman A. Health Impact of pain in the hip region with and without radiographic evidence of osteoarthritis: a study of new attenders to primary care. Ann Rheum Dis 2000;59:857-63.
  33. Altman RD, Abadie E, Avouac B, Bouvenot G, Branco J, Bruyere O, et al. Total joint replacement of hip or knee as an outcome measure for structure modifying trials in osteoarthritis. Osteoarthritis Cartilage 2005;13(1):13-9.
  34. Kane RL, Wilt T, Suarez-Almazor ME, Fu SS. Disparities in total knee replacements: a review. Arthritis Rheum 2007;57(4):562-7.
  35. Hochberg MC, Lawrence RC, Everett DF, Cornoni-Huntley J. Epidemiologic associations of pain in osteoarthritis of the knee: data from the National Health and Nutrition Examination Survey and the National Health and Nutrition Examination-I Epidemiologic Follow-up Survey. Semin Arthritis Rheum 1989;18(4 Suppl 2):4-9.
  36. Davis MA, Ettinger WH, Neuhaus JM, Mallon KP. Knee osteoarthritis and physical functioning: evidence from the NHANES I epidemiologic followup study. J Rheumatol 1991;18:591-598.
  37. Felson DT, Nevitt MC. Epidemiologic studies for osteoarthritis: new versus conventional study design approaches. Rheum Dis Clin North Am 2004;30(4):783-97, vii.