Abstract
Objective
To quantify the effect of corticosteroids compared to lidocaine‐only injections over 12 weeks among patients with knee osteoarthritis (KOA).
Methods
Participants with KOA were randomized to receive a knee injection of methylprednisolone acetate 1 mL (40 mg) plus 2 mL lidocaine (1%) or 1 mL saline and 2 mL lidocaine. Participants and providers were blinded to treatment allocation using an opacified syringe. The outcome was the average change from baseline of the total Knee Injury and Osteoarthritis Outcome Score (KOOS) (range 0‐100) assessed at 2‐week intervals over 12 weeks. Participants received KOOS questionnaires on their smartphones through a web‐based platform. We used linear mixed‐effects regressions with robust variance estimators to evaluate the association between the intervention and change in KOOS total and subscales (ClinicalTrials.gov identifier NCT03835910; registered 2019‐02‐11).
Results
Of the 33 randomized participants, 31 were included in the final analysis. The predicted mean (SE) change in total KOOS over the 12‐week follow‐up was 9.4 (3.2) in the corticosteroids arm versus −1.3 (1.4) in the control arm (P = 0.003). Of participants, 47% achieved change as large as the minimal clinically important difference (16 units) in the intervention arm compared to 6% of participants in the lidocaine arm. Further, there were greater improvements in the intervention arm for KOOS subscales and for Patient Reported Outcomes Measurement Information System (PROMIS) assessments of pain intensity, behavior, and interference.
Conclusion
Corticosteroid injections demonstrated clinically meaningful improvements in KOA symptoms over 12 weeks of follow‐up. These data support larger studies to better quantify short‐term benefits.
INTRODUCTION
Knee osteoarthritis (KOA) is a high‐priority problem among the aging population, and in particular, among veterans. It is one of the most prevalent and disabling conditions in the US and accounts for high morbidity and costs (1, 2, 3). Few interventions have been shown to demonstrate consistent and convincing clinical benefit for treatment of this common condition.
Intra‐articular corticosteroid injections are one of the most common adjunct approaches to long‐term management of KOA. A 2015 Cochrane review summarized the existing data on the benefits of intra‐articular corticosteroids injections, finding overall low evidence for benefits (4). More recently, a large, randomized trial demonstrated no reductions in pain after corticosteroid injections compared to saline injections when pain was assessed at 3‐month intervals, and a recent meta‐analysis suggested minimal benefits after 3 months (5, 6).
While recent data have suggested a lack of durable responses, prior studies have observed benefits of corticosteroids that peak around 4 to 8 weeks, and management guidelines have largely acknowledged short‐term benefits (7, 8, 9, 10). Current KOA management guidelines recommend a combined approach of nonpharmacologic and pharmacologic interventions, including physical activity and intra‐articular corticosteroids (7). While there is some evidence to support the benefit of this intervention, physicians need a more granular picture of the benefits of corticosteroids to weigh these benefits against the potential risks. To our knowledge, this is among the first trials of its kind in KOA to quantify changes in symptoms using completely remote capture of outcomes at 2‐week intervals.
To elucidate the potential short‐term benefits of intra‐articular corticosteroids and support a larger, more definitive trial, we performed a pilot blinded randomized trial to study the effect of methylprednisolone acetate injections compared to lidocaine‐only injections over a 3‐month period.
PATIENTS AND METHODS
Study setting
This pilot, double‐blind, factorially designed randomized controlled trial enrolled veterans with KOA from the Corporal Michael J. Crescenz Veterans Affairs (VA) Medical Center in Philadelphia (US) (ClinicalTrials.gov identifier NCT03835910; registered 2019‐02‐11). The study was performed in factorial design with an exercise intervention using gamification and social support based on theories from behavioral economics. The factorial design increased the trial efficiency as it allowed for simultaneous assessment of both interventions. In this design, participants were randomized twice. The study was known to be underpowered to evaluate the effect of the exercise intervention in this pilot study, and the results of that analysis have been described elsewhere (11).
Participants between 40 and 80 years old of any race, ethnicity, and sex were included if they met the following criteria: i) had a physician diagnosis of KOA, ii) had at least some radiographic evidence of KOA (a Kellgren‐Lawrence [KL] grade of ≥1), iii) had a clinician who felt that corticosteroid injection(s) were indicated or had previously received palliative injections, iv) expressed interest in increasing their physical activity, and v) reported being able to walk one half mile per day. Exclusion criteria included the following: i) daily use of assistive devices (ie, cane or walker), ii) lack of a smartphone, iii) contraindications to joint injection (eg, intra‐articular hardware), iv) experiencing an acute exacerbation of knee pain, v) or the presence of comorbidities that would limit the patient's ability to exercise. Race and ethnicity were reported using the NIH fixed set of categories.
Participants had to wait at least 4 months after receiving a corticosteroid injection or 6 months after receiving viscosupplementation before they could receive their first study injection. Additionally, participants could receive either unilateral or bilateral injections if they met the criteria mentioned above and depending on clinical need.
The study was approved by the local institutional review board (IRB), and all study participants agreed and signed an informed consent document (IRB# 01795). With a sample size of 32 patients, the study had 80% power to detect a large effect size in the total KOOS score (d = 1.0).
Intervention
Participants with KOA were randomized (1:1) through an online trial platform to receive a double‐blinded intra‐articular knee injection of either methylprednisolone acetate 1 mL (40 mg) plus 2 mL 1% lidocaine or 1 mL saline plus 2 mL 1% lidocaine. Methylprednisolone acetate was chosen because it is the most common agent used at the institution. Only the research pharmacist was aware of treatment allocation, and opacified syringes were used to blind participants and clinical providers. At any point in the intervention period, participants were permitted to receive a single “rescue” corticosteroid injection from their clinical provider if clinically indicated (a requirement from the IRB). At 16 weeks, participants crossed over to receive the alternative injection, though this was delayed by several months for many patients due to the COVID‐19 shutdown. The current analysis focuses only on the initial injection received.
Outcome
The primary outcome for this factorial study was the increase in step counts to assess the effect of the exercise intervention over 10 to 12 months. Participants were provided Fitbit™ activity monitors, and counts were assessed daily and averaged over 2‐week intervals (ignoring days with <1000 steps) through linkage on the participants smartphone. For the current analysis, we focused on the pre‐specified secondary outcome which assessed symptom improvement using the total Knee Injury and Osteoarthritis Outcome Score (KOOS) in 2‐week intervals until week 12. The KOOS is a widely used measure of knee‐related pain and physical functioning that has been validated in both knee injury and KOA populations (12). The KOOS measures symptoms and disability on five subscales: Pain, other Symptoms, Function in Activities of Daily Living (ADL), Function in Sports and Recreation, and Quality of Life (QOL). The KOOS subscores were aggregated and averaged as recommended (12). Individual subscores were also considered. We used 16 for the minimal clinically important difference (MCID) and improvement in KOOS, as previously described (13, 14). Patients were not asked to ascribe responses to a particular knee. Participants completed biweekly surveys on their smartphones through a unique survey link sent out by a web‐based application.
Secondary outcomes and other measures
Secondary outcomes included monthly Patient Reported Outcomes Measurement Information System (PROMIS) questionnaires assessing pain intensity, pain interference, and pain behaviors (15). Pain catastrophizing was assessed with the Pain Catastrophizing Scale (16).
Statistical analysis
Baseline characteristics of those eligible for analyses, and who received corticosteroids or lidocaine only, were evaluated using standard summary statistics, eg, mean and standard deviation (continuous variables) and count and percentage (categorical variables).
We used a modified intention to treat approach to the analysis (excluding two patients after randomization who did not complete any study procedures). The approach to the evaluation for primary and secondary outcomes were specified a priori. Changes in total KOOS and KOOS subscales were stratified by treatment group and were analyzed with a repeated measures approach, and visualized, to assess the average change over the 12‐week period. Linear mixed‐effects regression, with robust variance estimators and unstructured variance‐covariance matrices, evaluated the association between the intervention and KOOS and KOOS subscales over all time points. Covariates in the model included study week (categorical) and baseline KOOS or subscale. This model accounts for the within‐subject correlation due to repeated measures by including subject‐specific random effects. The primary analysis used all available observations except for two observations for one participant in the corticosteroid arm that occurred after a “rescue” injection. Additionally, we explored sensitivity analyses with a last‐observation‐carried‐forward (LOCF) approach for missing data. We also determined the number of participants in each group who achieved a change in KOOS as large as the MCID at any point during the 12‐week follow‐up. Finally, we explored differences in average step counts over all time points in each study arm.
Several prespecified hypothesized factors were tested as potential predictors of greater improvements in KOOS, independent of treatment. These included whether the participant believed the injection would help on a Likert scale (ie, agree or strongly agree), receipt of the exercise incentive, the presence of pain catastrophizing, and KL grade.
While the study was underpowered to evaluate the primary outcome of changes in step counts, we evaluated the changes in daily step counts between groups using similar mixed‐effects models and adjusting for baseline step counts.
Analyses were performed using STATA 14 software (StataCorp, LP, College Station, TX).
RESULTS
Of 36 participants who were consented for participation, 33 were randomized and received the intervention. Two participants received an injection but were withdrawn because they were not adherent to study procedures (ie, did not complete any surveys), resulting in 31 included in the final analysis (Supplementary Figure 1). Characteristics of the participants are shown in Table 1. There were no statistically significant differences between treatment arms in age, sex, body mass index (BMI), comorbid conditions, KL grade, and baseline KOOS. Only one participant randomized to the lidocaine arm had a KL grade of less than 2. The exercise incentive was received by 53% of those who received corticosteroids plus lidocaine and 50% among those who received lidocaine only. The majority of participants (20/31) also received bilateral injections (Table 1). Thirteen percent of responses for the total KOOS were incomplete throughout the follow‐up period.
Table 1.
Basic characteristics of the study groups
| Corticosteroid + lidocaine | Lidocaine only | |
|---|---|---|
| N | 15 | 16 |
| Age (y) | 55.5 (16.9) | 63.5 (1.9) |
| Female, N (%) | 2 (13%) | 1 (6%) |
| BMI (kg/m2) | 35.5 (6.1) | 31.9 (5.2) |
| Current smoker, N (%) | 3 (20%) | 2 (13%) |
| Highest KL grade, N (%) | ||
| 1 | 0 (0%) | 1 (6%) |
| 2 | 5 (33%) | 2 (12%) |
| 3 | 5 (33%) | 3 (19%) |
| 4 | 5 (33%) | 10 (63%) |
| Baseline total KOOS | 38.6 (16.3) | 41.9 (3.9) |
| KOOS Pain | 40.4 (4.6) | 49.3 (4.2) |
| KOOS Symptoms | 56.9 (4.6) | 48.5 (4.1) |
| KOOS Activities | 45.2 (5.2) | 51.6 (4.1) |
| KOOS Sports/Recreation | 45.2 (5.2) | 51.6 (4.1) |
| KOOS Quality of Life | 21.3 (5.1) | 28.9 (4.7) |
| Exercise incentive, N (%) | 8 (53%) | 8 (50%) |
| Prior injections, N (%) | 15 (100%) | 16 (100%) |
| Knee injected, N (%) | ||
| Bilateral | 12 (80%) | 10 (63%) |
| Left only | 4 (27%) | 3 (19%) |
| Right only | 0 | 2 (13%) |
| Believe injections will help (agree or strongly agree), N(%) | 11 (73%) | 12 (75%) |
| Comorbidities, N (%) | ||
| Diabetes | 6 (40%) | 7 (44%) |
| Hypertension | 5 (33%) | 8 (50%) |
| Hyperlipidemia | 4 (27%) | 7 (44%) |
| Coronary artery disease | 1 (7%) | 2 (13%) |
| Gout | 3 (20%) | 6 (38%) |
| Sleep apnea | 8 (53%) | 6 (38%) |
| Depression | 4 (26%) | 5 (31%) |
| Anxiety | 6 (40%) | 5 (31%) |
| Spine disease | 3 (20%) | 8 (50%) |
| Any cancer | 0 (0%) | 2 (13%) |
Continuous measures presented as mean (SD) or median (IQR).
Abbreviations: BMI, body mass index; IQR, interquartile range; KL, Kellgren‐Lawrence; KOOS, Knee Osteoarthritis Outcome Score; SD, standard deviation.
Over all time points over 12 weeks, there was a greater improvement in total KOOS in the corticosteroid group compared to the lidocaine‐only group (β: +10.8 [3.7‐17.8]; P = 0.003). The mean (SE) change over the entire 12‐week follow‐up (predicted from the regression model) for the corticosteroid arm was 9.4 (3.2) versus −1.3 (1.4) in the lidocaine‐only arm. The total KOOS scores over the 12 weeks of follow‐up are shown in Figure 1 and Supplementary Table 1. There were significant differences in the change from baseline in total KOOS score at 4, 6, 10, and 12 weeks. In sensitivity analyses, there was a similar effect of corticosteroid injections on total KOOS score when using a LOCF approach for missing data (β: +10.2 [3.5‐16.9]; P = 0.003). Results from this analysis over all time points are shown in Table 2. Results were also similar after adjusting for age, BMI, the exercise incentive intervention, and baseline KL grade (full models not shown). Among the intervention arm, 7 of 15 (47%) achieved a mean change reaching or exceeding the MCID at some point during follow‐up whereas only 1 of 16 (6%) achieved a mean change reaching or exceeding the MCID among the lidocaine‐only arm. The risk difference was 41% (47%‐6%), with a number needed to treat (NNT) of approximately 3.
Figure 1.
Change from baseline in total KOOS score over time in the corticosteroid and lidocaine‐only arms. KOOS, Knee Osteoarthritis Outcome Score.
Table 2.
The changes in KOOS score [Mean (SE)] in the corticosteroid and lidocaine‐only arms adjusting for the baseline value using a last‐observation‐carried‐forward approach
| Week 0 | Week 2 | Week 4 | Week 6 | Week 8 | Week 10 | Week 12 | |
|---|---|---|---|---|---|---|---|
| Corticosteroid | 0 (N = 15) | 7.77 (3.6) (N = 15) | 12.7 (3.8) (N = 15) | 11.4 (4.4) (N = 15) | 11.0 (4.8) (N = 15) | 10.1 (4.5) (N = 15) | 8.5 (4.1) (N = 15) |
| Lidocaine | 0 (N = 16) | 1.44 (1.6) (N = 16) | 1.25 (1.9) (N = 16) | −0.97 (1.7) (N = 16) | −3.1 (2.5) (N = 16) | −3.5 (2.0) (N = 16) | −1.4 (2.7) (N = 16) |
| P value | — | 0.11 | 0.01 | 0.01 | 0.01 | 0.009 | 0.05 |
Abbreviation: KOOS, Knee Osteoarthritis Outcome Score.
The changes in individual KOOS subscales over all follow‐up time points are shown in Figure 2 and Supplementary Table 2. In analyses incorporating all time points, there were significantly greater improvements in the corticosteroid arm for KOOS pain (β: +11.3 [2.2‐20.4]; P = 0.01), KOOS ADL (β: +10.3 [2.7‐17.8]; P = 0.008), KOOS sports/recreation (β: +13.2 [1.5‐24.8]; P = 0.03), and KOOS QOL (β: +15.3 [6.5‐24.0]; P = 0.001). A significant difference was not observed for KOOS other symptoms (β: +1.4 [−6.0 to 8.9]; P = 0.71).
Figure 2.
Mean change (SE) in KOOS subscales over time in corticosteroid and lidocaine‐only arms. ADL, Activities of Daily Living; KOOS, Knee Osteoarthritis Outcome Score; QOL, Quality of Life.
Compared to the lidocaine‐only arm, the corticosteroid arm demonstrated significant improvements in PROMIS assessments of pain intensity (β: −5.1 [−8.7 to −1.6]; P = 0.004), behavior (β: −2.2 [−3.7 to −0.7]; P = 0.004), and interference (β: −4.4 [−6.7 to −2.1]; P < 0.001). The changes over time in PROMIS pain measures are shown in Figure 3.
Figure 3.
Change in PROMIS pain measures over 12 weeks in the corticosteroid and lidocaine‐only arms. PROMIS, Patient Reported Outcomes Measurement Information System.
There were no associations between the total change in KOOS and baseline factors such as receipt of the exercise incentive, pain catastrophizing, expected benefits from the injection, the presence of a joint effusion or KL grade (not shown). There was not a significant association between the baseline KOOS and the change in KOOS (β: 0.14 [−0.06 to 0.34]; P = 0.18).
Over the 12 weeks of follow‐up, average weekly step counts were significantly lower among those who received corticosteroid injections (β: −1874 [−3455 to −293]; P = 0.02).
Adverse events and rescue injections
A total of eight participants experienced serious adverse events during the trial, and none were considered related to study participation by the study team and IRB (Supplementary Table 3). One participant in the corticosteroid arm received a “rescue” injection around week 10.
DISCUSSION
This study observed a significant reduction in pain, and improvement in function and quality of life, among participants with KOA who received a corticosteroid injection compared to those who received lidocaine only over a 12‐week period. While most KOOS subscales improved, there was no improvement in mechanical symptoms, an observation that perhaps makes sense given the mechanism of action of the intervention. Overall, the effect of corticosteroids was large and supports further study to confirm these benefits with greater precision. This study suggests that to achieve a meaningful response for a single patient at any point in the 12‐week follow‐up, the number needed to treat is 3. Due to the small sample size, this number may be subject to imprecision; however, it supports the potential value and clinical importance of the intervention over the short term.
Prior studies have found inconsistent results regarding the benefit of corticosteroid injections. This variability likely stems from differences in study design, resulting in substantial heterogeneity between studies (4). Differences in the timing of outcome evaluation are likely to have played a role in heterogeneous results. A recent study, measuring outcomes after 3 months from the receipt of the injections found no benefit (5). Our study, which assessed outcomes at 2‐week intervals over a 12‐week period, identified meaningful changes over the observation period, particularly in the first 4 to 10 weeks.
Our study demonstrated placebo effects that were lower than those observed for other trials. The use of an opacified syringe and the blinding of allocation to the investigator and coordinators ensured no unblinding of the participants. Placebo effects may have been minimized because outcomes were assessed in participants’ homes and in real time (17). In other words, participants may have had more freedom to report their symptoms honestly without hoping to please interviewers. In addition, since all participants had previously received injections, the novelty of an interventional procedure may have had limited impact on their perception of pain.
As previously described, the exercise intervention had a moderate effect on step counts over 10 months, walking an average of 1119 (95% CI: −562 to 2799) more steps per day compared to the control arm (P = 0.19) (11). The current pilot study was underpowered to adequately evaluate the interaction between the two interventions. However, the larger, multi‐site study is currently underway (ClinicalTrials.gov identifier NCT05035810; registered 2021‐09‐05). Surprisingly, we found lower physical activity levels over 12 weeks among those receiving corticosteroids compared to lidocaine only. These results are preliminary but suggest a complex and counterintuitive relationship between pain and exercise, at least in the context of behavioral incentives to promote changes in this behavior.
The primary limitation of the current study as compared to some other trials is the small sample size, which can result in low precision and residual confounding (by chance). In addition, despite not being an exclusion criterion, all participants reported previous receipt of corticosteroid injection(s). Our study sample may have been more likely to respond, overall, than patients who were injection naïve. Since most participants had previously received injections, the novelty of an interventional procedure may have had limited impact on their perception of pain. Nonetheless, this study suggests that corticosteroids can provide clinically relevant benefits in this population. Further study is needed to determine whether the effect of corticosteroid injections varies in an injection‐naïve population. This study did not assess long‐term safety of corticosteroid injections, and further study in this area is also needed. It is worth noting that we used lidocaine as the comparator to mimic a patient's typical experience in a clinical visit; however, the use of lidocaine might enhance the placebo effect. Despite this, we did not observe a strong placebo effect. The strengths of this study include the double‐blind design; that study procedures were performed within usual care; and the real‐world, at‐home assessment of outcomes with low participant burden, which may have limited participation bias.
In conclusion, intra‐articular injections with corticosteroid plus lidocaine compared to lidocaine only demonstrated clinically meaningful improvements over a 12‐week follow‐up in this study. These data support the clinical utility of corticosteroids combined with lidocaine, for at least short‐term benefits, among patients who have previously received such injections. Larger studies to better quantify short‐term benefits for corticosteroid injections for a broader range of patients with KOA are warranted.
AUTHOR CONTRIBUTIONS
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Drs. Arkema and Sjöwall had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design
Baker, Ogdie, Neogi, White, England, Wysham, Scanzello
Acquisition of data
Baker, Olave, Leach, Doherty, Gillcrist, Quinones
Analysis and interpretation of data
Baker Olave, Xiao, Scanzello
Supporting information
Disclosure Form:
Appendix S1: Supporting Information
ACKNOWLEDGMENTS
Dr. Baker would like to acknowledge funding through a Veterans Affairs Clinical Science Research & Development Career Merit Award (I01 CX001703) a Rehabilitation Research & Development Merit Award (I01 CX003644) and SPiRE Award (I21 RX003157). The contents of this work do not represent the views of the Department of the Veterans Affairs or the United States Government.
The contents of this article do not represent the official views of the Department of Veterans Affairs (VA) or the US Government.
Supported by a VA Rehabilitation Research & Development SPiRE Award (I21 RX003157) and Merit Award (I01 RX003644).
Author disclosures are available at https://onlinelibrary.wiley.com/doi/10.1002/acr2.11596.
REFERENCES
- 1. Golightly YM, Allen KD, Jordan JM. Defining the burden of osteoarthritis in population‐based surveys. Arthritis Care Res (Hoboken) 2016;68:571–3. [DOI] [PubMed] [Google Scholar]
- 2. Rivera JC, Amuan ME, Morris RM, et al. Arthritis, comorbidities, and care utilization in veterans of operations enduring and Iraqi freedom. J Orthop Res 2017;35:682–7. [DOI] [PubMed] [Google Scholar]
- 3. Deshpande BR, Katz JN, Solomon DH, et al. Number of persons with symptomatic knee osteoarthritis in the US: impact of race and ethnicity, age, sex, and obesity. Arthritis Care Res (Hoboken) 2016;68:1743–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Jüni P, Hari R, Rutjes AW, et al. Intra‐articular corticosteroid for knee osteoarthritis. Cochrane Database Syst Rev 2015;2015:CD005328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. McAlindon TE, LaValley MP, Harvey WF, et al. Effect of intra‐articular triamcinolone vs saline on knee cartilage volume and pain in patients with knee osteoarthritis: a randomized clinical trial. JAMA 2017;317:1967–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Donovan RL, Edwards TA, Judge A, et al. Effects of recurrent intra‐articular corticosteroid injections for osteoarthritis at 3 months and beyond: a systematic review and meta‐analysis in comparison to other injectables. Osteoarthritis Cartilage 2022;30:1658–69. [DOI] [PubMed] [Google Scholar]
- 7. Kolasinski SL, Neogi T, Hochberg MC, et al. 2019 American College of Rheumatology/Arthritis Foundation guideline for the management of osteoarthritis of the hand, hip, and knee. Arthritis Care Res (Hoboken) 2020;72:149–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Yavuz U, Sökücü S, Albayrak A, et al. Efficacy comparisons of the intraarticular steroidal agents in the patients with knee osteoarthritis. Rheumatol Int 2012;32:3391–6. [DOI] [PubMed] [Google Scholar]
- 9. Beyaz SG. Comparison of efficacy of intra‐articular morphine and steroid in patients with knee osteoarthritis. J Anaesthesiol Clin Pharmacol 2012;28:496–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Conaghan PG, Hunter DJ, Cohen SB, et al. Effects of a single intra‐articular injection of a microsphere formulation of triamcinolone acetonide on knee osteoarthritis pain: a double‐blinded, randomized, placebo‐controlled, multinational study. J Bone Joint Surg Am 2018;100:666–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Leach W, Olave M, Doherty C, et al. A remote behaviorally‐designed intervention to promote physical activity in patients with knee osteoarthritis: results of a pilot randomized clinical trial [abstract]. Arthritis Rheumatol 2022;74 Suppl 9. [Google Scholar]
- 12. Roos EM, Roos HP, Lohmander LS, et al. Knee Injury and Osteoarthritis Outcome Score (KOOS)—development of a self‐administered outcome measure. J Orthop Sports Phys Ther 1998;28:88–96. [DOI] [PubMed] [Google Scholar]
- 13. Singh JA, Luo R, Landon GC, et al. Reliability and clinically important improvement thresholds for osteoarthritis pain and function scales: a multicenter study. J Rheumatol 2014;41:509–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Angst F, Aeschlimann A, Michel BA, et al. Minimal clinically important rehabilitation effects in patients with osteoarthritis of the lower extremities. J Rheumatol 2002;29:131–8. [PubMed] [Google Scholar]
- 15. Stone AA, Broderick JE, Junghaenel DU, et al. PROMIS fatigue, pain intensity, pain interference, pain behavior, physical function, depression, anxiety, and anger scales demonstrate ecological validity. J Clin Epidemiol 2016;74:194–206. [DOI] [PubMed] [Google Scholar]
- 16. Osman A, Barrios FX, Kopper BA, et al. Factor structure, reliability, and validity of the Pain Catastrophizing Scale. J Behav Med 1997;20:589–605. [DOI] [PubMed] [Google Scholar]
- 17. Zimbroff DL. Patient and rater education of expectations in clinical trials (PREECT). J Clin Psychopharmacol 2001;21:251–2. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Disclosure Form:
Appendix S1: Supporting Information