If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Huo-Luo-Xiao-Ling (HLXL)-Dan, a Traditional Chinese Medicine, for patients with osteoarthritis of the knee: a multi-site, randomized, double-blind, placebo-controlled phase II clinical trial
Address correspondence and reprint requests to: L. Lao, School of Chinese Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, china Tel: 852-25890476; Fax: 852-28725476.
Center for Integrative Medicine, School of Medicine, University of Maryland, Baltimore, USASchool of Chinese Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China
To examine the efficacy and safety of Huo-Luo-Xiao-Ling (HLXL)-Dan, a Traditional Chinese Medicine (TCM), in patients with knee osteoarthritis (OA).
Design
A multi-site, randomized, double-blind, placebo-controlled phase II dose-escalation clinical trial was conducted. Eligible patients who fulfilled American College of Rheumatology criteria were randomized to receive either HLXL or placebo. Clinical assessments included measurement of knee pain and function with the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), patient global assessment (PGA), and knee pain scores every 2 weeks. A Data and Safety Monitoring Board (DSMB) was established to review the data for ensuring the quality of the trial.
Results
In the first stage, 28 participants were randomized to receive either low-dose HLXL-Dan (2400 mg/day) or placebo for 6 weeks. The results showed no statistical difference between the two groups. The study was then re-designed following the recommendation of DSMB. Ninety-two patients were enrolled in the second stage and were randomized to receive either high-dose HLXL-Dan (4000 mg/day for week 1–2, and 5600 mg/day for week 3–8) or placebo for 8 weeks. All outcome assessments showed significant improvements for both groups after 8 weeks but no significant between-group differences. The change (mean ± SD) of WOMAC pain and WOMAC function scores of HLXL and placebo group after 8 weeks were −1.2 ± 1.7 vs −1.4 ± 1.5, and −1.1 ± 1.6 vs −1.3 ± 1.5 respectively. No serious adverse events were reported.
Conclusion
Although safe to use, an 8-week treatment of HLXL-Dan was not superior to placebo for reduction in pain or functional improvement in patients with knee OA.
. It is now recognized that inflammation in joints affected by OA contributes to not only the development of symptoms, including pain and stiffness, but also the progression of structural damage, including cartilage degradation
. Conventional analgesic medications such as acetaminophen, and nonsteroidal anti-inflammatory drugs (NSAIDs) are commonly used for the treatment of OA
. However, their use is limited by concerns regarding potentially serious adverse effects. Nonpharmacologic treatments including patient education, weight loss, exercise, and physical therapy have showed substantial benefit and are recommended to the patients with OA
. Given the limitations of the conventional treatments, complementary and alternative medicine (CAM) are commonly used by patients with OA. It was found that nearly one-third of older adults reported using at least one form of CAM modality for treating OA
. Previous clinical trials and case-series studies suggested that HLXL is effective for various types of arthritis, including gout, rheumatoid arthritis, and OA
. To our knowledge, however, there has been no prior randomized, double-blind, placebo-controlled trial to examine the efficacy of HLXL-Dan for knee OA.
The aims of this phase II dose-escalation, double-blind, randomized clinical trial were to determine whether HLXL improves pain and/or function in patients with OA of the knee who continue to have symptoms despite receiving standard analgesic and/or NSAID treatment, and to identify an optimal dosage of HLXL as well as its safety and tolerability.
Methods
Patient recruitment
The trial was conducted from April 2007 to May 2011 at the University of Maryland Center for Integrative Medicine, Baltimore, Maryland and from April 2010 to May 2011 at the Chinese University of Hong Kong, Hong Kong SAR, China. The trial was approved by the Institutional Review Board (IRB) at both the University of Maryland School of Medicine and Chinese University of Hong Kong. We obtained permission for an Investigational New Drug (IND) from the Food and Drug Administration (FDA), registration number, 70,327. A Data and Safety Monitoring Board (DSMB) was established, which met annually to review the data for ensuring the safety of the trial. The trial was performed in compliance with the Helsinki Declaration and ICH-GCP. We followed the CONSORT recommendations in designing and reporting of controlled trials
Patients meeting the following inclusion criteria were recruited to the trial: age 40 years or older, diagnosis of OA of the knee of at least 6 months duration fulfilling American College of Rheumatology criteria, pain in at least one knee of at least moderate severity (Likert scale [none, mild, moderate, severe, extreme] or visual analog scale (VAS) pain score of at least 40 [0–100]) on most (at least 15) days of the previous 1 month, taking analgesic or NSAIDs for control of pain, documented radiographic changes indicating OA of the knee (Kellgren–Lawrence grade greater than or equal to 2 at the time of screening), stable on arthritis medications for previous 1 month, willingness and ability, with help of a caregiver if necessary, to comply with treatment and follow-up procedures, not pregnant or lactating, use of effective contraception if woman of childbearing potential, and signed consent statement. Exclusion criteria were medical condition that may preclude safe participation in protocol or prevents completion of the study (see Appendix for detailed exclusion criteria).
In the USA site, advertisements in local newspapers and local radios were used for patients’ recruitment; in the Hong Kong site, patients in the Rheumatoid clinic were approached for recruitment. After a brief telephone screening, patients were scheduled for an on-site screening examination that included a protocol-driven examination of the knees by a trained physician or nurse practitioner. If eligible, the patient was randomly assigned to either the HLXL treatment group or a placebo control group by a computer-generated process using randomly selected block sizes. We assured the allocation concealment by using a web-based randomization program, which would not reveal the randomization result for each participant until the recruiting coordinator entered the participant ID, following completion of the screening and qualification process. The participants were randomly assigned to HLXL or placebo in a 2:1 ratio at the low dose with a block size ranged 3 to 6. After the DSMB recommended two weeks at the middle dose for safety reasons followed by the high dose, the randomization ratio was set to 1:1 with a block size ranged 2 to 4. All participants and research team members were blinded to group assignment until the end of each dose stage. Only the statistician and data analyst were privy to trial results until after presentation to the DSMB.
Study intervention
HLXL is an ancient prescription that has been used for over 100 years in the treatment of arthritis in China. The original HLXL had only 4 herbs, and there have been many varieties of formula in Traditional Chinese Medicine (TCM) history. The HLXL examined in this study contains 11 individual herbs modified from the original classic formula, and is a popular treatment for inflammatory diseases in China
The herbs were extracted using 70% aqueous acetone. For quality control purposes, high performance liquid chromatography (HPLC) fingerprint was generated (Fig. 1). A single batch of study extracts was used for the entire trial and periodic stability tests were performed every 90 days to ensure the stability of the investigational herbal extracts.
Fig. 1HPLC chromatogram of HLXLa, b. a (1) = Boswellia carterii, (2) = Notopterygium incisum, (3) = Angelica sinensis, (4) = Paeonia lactiflora, (5) = Glycyrrhiza uralensis (6) = Corydalis yanhusuo, (7) = Salvia miltiorrhiza (8) = Ligusticum chuanxiong, (9) = Gentiana macrophylla, (10) = Cinnamomum cassia, (11) = Angelica pubescens. b(1)-1, (1)-2, (1)-3 = Boswellia carterii characteristic peaks 1, 2 and 3 respectively; (2)-1, (2)-2, (2)-3, (2)-4 = Notopterygium incisum characteristic peaks 1, 2, 3 and 4 respectively; etc.
based on the following equation on the selection of a safe starting dose for (non-repeated dosing) initial human trials, as per FDA guidance (134): HED (g/kg) = animal dosage (g/kg) × (Wt of animal ÷ Wt of human)1-b. Whereas HED = human equivalent dose, effective dosage in animal = 0.575 g/kg HLXL, rat weight = 0.25 kg, human weight = 60 kg, 1-b = 0.33, 134 then, human daily dosage = HED × Human weight (60 kg) = about 5 g. HLXL capsules contain 0.370 g/capsule, therefore, daily use in the highest dosage condition will be 14 capsules.
The original dosage escalation design involved three dosage levels: low (6 capsules/day), medium (10 capsules/day), and high (14 capsules/day). Each capsule contained 400 mg HLXL. The subjects received either HLXL or placebo for 6 weeks. The sample sizes were originally estimated at n = 60 per dose level in a treatment to control ratio of 2:1 (40 HLXL, 20 Placebo), and 180 in total (120 HLXL, 60 Placebo).
Because the lowest dose was found to have no effect and no adverse effects after the completion of the first 28 subjects, and low recruitment rate was observed, the trial was re-designed. The DSMB recommended merging the medium and high dose stages into one medium to high dose trial and extending the treatment period by 2 weeks. Therefore, in this re-designed medium to high dose trial, the subjects in the HLXL group received the medium dose of HLXL (10 capsules/day or 4000 mg/day) in the first 2 weeks to evaluate safety. If no adverse effects were observed, the dose was increased to 14 capsules per day (5600 mg/day) for the subsequent 6 weeks. Subjects in the placebo group received an equal number of placebo capsule. Our previous phase I trial in seven patients found that the dose of HLXL, 5180 mg daily (7 capsules, 2 times daily), showed no serious adverse effects (data not shown). In addition, the sample size was modified to n = 100 in a 1:1 ratio of HLXL to placebo, 50 subjects in each group. Assumptions for sample size estimation were based on results of the prior pilot study for WOMAC pain and function scales, change from baseline, SD (pain) = 1.75, SD (function) = 1.76, on a scale of 0–10. With power = 0.8, alpha = 0.05% and 4% dropout (actual), we calculated that with 50 patients per group we could detect differences in pain = 0.98 and function = 0.99. Reductions in pain and function scores were greater than these detectable differences; clearly we had adequate power.
The placebo was made from a list of FDA approved ingredients/excipient that matched the color and taste of the real HLXL herbal extract, instead of just white powder used in other placebo trials. Same size and color of capsules were used for the placebo and herbal extract. The equivalent number of capsules used in the treatment group at each dose level were administered to patients assigned to the control group.
Outcome measures
Outcome assessments were conducted at baseline, 2, 4, and 6 weeks after randomization in the low dose stage (low dose trial); after re-designing the study (medium–high dose trial), assessments were performed at 2, 4, 6 and 8 weeks after randomization. The primary outcomes were changes from baseline in the VAS version of Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain and function scores. The primary endpoint was considered the end of the herbal intervention (week 6 for the stage 1 low dose trial and week 8 for the stage 2 medium–high dose trial). The secondary outcomes were change in patient global assessment (PGA) of knee OA severity, change in WOMAC stiffness subscale score, adverse drug reactions, and adherence as measured by pill count and diary.
Statistical analysis
Analyses were carried out under the intention-to-treat principle; that is, all available data for all randomized participants were included in all analyses, with no data imputation. Initial analyses tabulated demographic and baseline characteristics of the study participants by randomization group and dosage level, where the medium and highest dose results were combined and termed high dose.
For outcome analyses, paired t-tests were used to assess changes from baseline by treatment group in primary and (where appropriate) secondary outcomes; independent two-sample Student's t-tests were used to compare treatment groups on change from baseline in continuous outcome measures. Finally, mixed-effect model methods (based on the Generalized Estimating Equation (GEE) principle) were used to compare treatment groups on outcomes with adjustment for covariates identified as potential confounders or variables on which the groups differed at baseline; no variables were found to be significantly associated with outcome or to change the observed effect size when included. Interactions of treatment group with site (Baltimore or Hong Kong) were also investigated and found to be non-significant. Thus, GEE adjusted results were not presented. We did not power the trial to carry out subgroup analyses as part of the primary outcome analyses.
Results
After the completion of 28 subjects in the low dose trial, where no differences were observed between treatment groups and no adverse effects were observed, the sample size and treatment dosage were adjusted following the recommendation of the DSMB (Supplementary Table). In the re-designed medium–high dose trial, we enrolled 92 patients, 53 in the US and 39 in Hong Kong. (Fig. 2). In this report, we present the results from the medium–high dose trial only. At baseline, the mean age was 60 years, 65 (71%) were women and 64 (70%) were of Asian ethnicity with symptomatic OA of the knee (Table I). The mean WOMAC pain and function scores were 4.4 (SD = 1.6) and 4.5 (SD = 1.7), respectively. The dropout rates were not significantly different between the HLXL group (4.3%) and placebo group (6.7%). No serious AEs were identified.
Pain normalized score and function normalized score are on a scale of 0–10, with 10 indicating extreme pain/difficulty related to knee arthritis and 0 indicating no pain/difficultly related to knee arthritis. Overall normalized score is on a scale of 0–30, and is the sum of the pain, stiffness and function normalized scores (each on a scale of 0–10).
PGA is on a scale of 1–5, with 5 indicating excellent and 1 indicating poor (in response to the question “Considering all the ways that your OA of the knee affects you, how are you feeling today?”
Knee pain is on a scale of 0–100, with 100 indicating worst pain ever and 0 indicating no pain experienced, on average, over the last 48 h.
Right knee
53.6 ± 21.6
54.2 ± 15.1
53.0 ± 26.2
Left knee
49.6 ± 24.2
48.2 ± 23.4
50.9 ± 25.1
Average of both knees
51.6 ± 17.1
51.2 ± 15.9
51.9 ± 18.2
∗ Pain normalized score and function normalized score are on a scale of 0–10, with 10 indicating extreme pain/difficulty related to knee arthritis and 0 indicating no pain/difficultly related to knee arthritis. Overall normalized score is on a scale of 0–30, and is the sum of the pain, stiffness and function normalized scores (each on a scale of 0–10).
† PGA is on a scale of 1–5, with 5 indicating excellent and 1 indicating poor (in response to the question “Considering all the ways that your OA of the knee affects you, how are you feeling today?”
‡ Knee pain is on a scale of 0–100, with 100 indicating worst pain ever and 0 indicating no pain experienced, on average, over the last 48 h.
Although both HLXL and placebo groups showed significant improvement in WOMAC normalized score (primary endpoint), PGA and knee pain scores at week 8, no statistically significant between-group differences were found in any of the efficacy endpoints (Table II). The results were the same when the US site and Hong Kong site were analyzed separately. There were no significant interactions of clinical site and treatment group identified.
Table IITreatment effect as measured at baseline, week 2 and week 8 (high dose only, combined Baltimore and Hong Kong sites)
Pain normalized score and function normalized score are on a scale of 0–10, with 10 indicating extreme pain/difficulty related to knee arthritis and 0 indicating no pain/difficultly related to knee arthritis. Overall normalized score is on a scale of 0–30, and is the sum of the pain, stiffness and function normalized scores (each on a scale of 0–10).
PGA is on a scale of 1–5, with 5 indicating excellent and 1 indicating poor (in response to the question “Considering all the ways that your OA of the knee affects you, how are you feeling today?”).
Knee pain is on a scale of 0–100, with 100 indicating worst pain ever and 0 indicating no pain experienced, on average, over the last 48 h.
Right knee
0.13
0.39
HLXL
47
53.0 ± 26.1
49.8 ± 25.1
−3.2 ± 23.4
0.36
42.0 ± 25.4
−11.5 ± 24.4
<0.01
Placebo
45
54.2 ± 15.1
44.7 ± 19.6
−9.6 ± 15.5
<0.01
39.5 ± 21.7
−15.4 ± 16.1
<0.01
Left knee
0.57
0.35
HLXL
47
50.9 ± 25.1
46.8 ± 29.8
−4.0 ± 23.7
0.25
43.3 ± 29.1
−7.2 ± 23.5
0.04
Placebo
45
48.2 ± 23.4
41.6 ± 21.7
−6.7 ± 20.6
0.04
37.8 ± 24.7
−11.5 ± 18.4
<0.01
Average of both knees
0.22
0.28
HLXL
47
51.9 ± 18.3
48.3 ± 24.4
−3.6 ± 18.5
0.19
42.6 ± 24.6
−9.3 ± 19.0
<0.01
Placebo
45
51.2 ± 15.9
43.1 ± 18.2
−8.1 ± 16.0
<0.01
38.7 ± 21.8
−13.4 ± 15.1
<0.01
∗ P values from paired t-test, comparing change from baseline to week 2 or baseline to week 8, as appropriate.
† P values from independent two-sample Student's t-test, comparing change in HLXL group to change in placebo group.
‡ Pain normalized score and function normalized score are on a scale of 0–10, with 10 indicating extreme pain/difficulty related to knee arthritis and 0 indicating no pain/difficultly related to knee arthritis. Overall normalized score is on a scale of 0–30, and is the sum of the pain, stiffness and function normalized scores (each on a scale of 0–10).
§ PGA is on a scale of 1–5, with 5 indicating excellent and 1 indicating poor (in response to the question “Considering all the ways that your OA of the knee affects you, how are you feeling today?”).
¶ Knee pain is on a scale of 0–100, with 100 indicating worst pain ever and 0 indicating no pain experienced, on average, over the last 48 h.
With respect to adverse reactions, our previous Phase I pilot study had identified skin rash, GI events and possibly small increases in transaminases as potential AE's for the high dose HLXL (unpublished data). In this study, no additional toxicities of the HLXL compound were identified. Specifically, there were no ALT (alanine aminotransferase) or AST (aspartate aminotransferase) elevations above the upper limit of normal. Nausea and mild bloating were reported by a few subjects, probably related to the investigational drug, but did not lead to dropout of subjects. Two individuals developed rash. One subject discontinued HLXL treatment after developing an allergic reaction; another developed transient urticaria but completed the study. One individual developed worsening of hypertension and hypokalemia. In a comparison with the placebo control, the frequency and intensity of reported adverse events in the HLXL group were not significantly different from those in the control group (Table III).
Table IIISummary of adverse events by treatment groups
In this randomized, double-blinded, placebo-controlled phase II dose-escalation clinical trial evaluating the effect and safety of HLXL-Dan, we found that, although safe to use, an 8-week treatment of this classic Chinese herbal formula in the proposed capsule format was not superior to placebo control for pain management or functional improvement in patients with knee OA. Despite the wide use of HLXL with different ways of disposure in treating arthritis in the Chinese population, based on the results of this study, we cannot recommend HLXL as an adjunctive therapy for treating knee OA.
There were several reasons for examining HLXL. First, this formula is widely used to treat joint pain, known as “Bi” syndrome in TCM
. Second, prior to our clinical trial, we tested the efficacy and mechanisms of action of this formula in an inflammatory pain animal model. In those studies, HLXL-Dan significantly suppressed complete Freund's adjuvant (CFA)-induced inflammatory pain
. Therefore, HLXL could be a potential effective treatment for knee OA and is warranted to be examined by a rigorously designed clinical trial to examine its efficacy.
In contrast to most clinical trials of Chinese herbal medicine that were often criticized for lack of adequate methodological design and implementation
, this clinical trial was rigorously designed with a carefully calculated sample size, adequate randomization and allocation concealment, a valid placebo control, standard outcome measures, and adequate follow-up time. It was monitored by an external DSMB. A botanical core consisting of a team of experts in botanical preparation ensured the quality and stability of the herbal product. A single batch of high-quality extract was employed throughout the trial, and the product was monitored and found to be stable throughout the trial. We are confident that our data are reliable and accurate.
The study was planned to recruit 60 participants in the stage 1 low dose trial, and 60 participants each in the medium dose and high dose trials. However, since we did not find any significant difference after completion of the first 28 participants during the stage 1 low dose trial, we followed the recommendation of the DSMB to re-design the study. The timely adjustment of the protocol ensured that research resources and participants' efforts would not be wasted.
The present phase II clinical trial is designed based on our previous animal studies
, in which an optimal dose of HLXL with maximal efficacy and minimal adverse events was determined. The optimal dose was found to be safe and without any observable adverse events. Therefore, it was converted into human equivalent dose and examined in this study. Our results indicated that the optimal dose of HLXL (5600 mg/day) is also safe and well-tolerated in human, without causing any significant adverse events and serious adverse event.
Despite its wide usage and its anti-inflammatory properties, which were confirmed by our pre-clinical studies, HLXL-Dan in the specific form of standardization and formulation showed no benefits beyond placebo in any of the dosages tested in this phase II clinical trial. Thus, based on the results of this trial, we cannot recommend its use in the treatment of knee OA. However, several factors might explain the inconsistencies between these findings and those of our basic science research: First, this formula might not be suitable for knee OA, which is not characterized by a systemic inflammatory component. HLXL-Dan might be more suitable for systemic inflammatory arthritis such as rheumatoid arthritis. Second, the formula might not be potent enough to provide additional symptom relief for patients already on background analgesics and/or NSAIDs. The classic formula was used in ancient times when no other analgesics were available. Third, TCM, as is well known, is an individualized modality in which a single disease may be treated with various formulae chosen to fit patients’ symptoms and constitutions. Standard placebo-controlled trial methods might not be the most effective way to examine the efficacy of TCM formulae or assess the effectiveness of TCM practice. A carefully designed trial that employs individualized treatment and medications might be a better way to evaluate the usefulness of Chinese herbal formulas. Fourth, in the present study, the herbs were extracted using 70% aqueous acetone, which has been found to be an efficient extraction method for Chinese herbal medicine
. In tradition, however, the herbs were usually by water extraction. Acetone is more efficient than water in extracting ingredients from herbal medicine, but it extracts both non-polar and polar-components. Therefore, it is possible that some water insoluble ingredients were also extracted by acetone, which may affect the absorption in human, leading to differences in treatment efficacy.
An important lesson that we learnt from this study is that, prior to a large scale phase III trial, a smaller scale phase II trial is necessary, because this approach allows us to evaluate the efficacy and safety of an herbal formula in a relatively small scaled, cost-effective trial. It has been reported that a number of large scale studies on herbal medicine failed to confirm their benefits seen in earlier preliminary studies
, the results of the present phase II trial suggested that HLXL in 5600 mg/day is not warranted to be further examined in a large scale phase III trial.
In conclusion, our study found that an 8-week treatment of HLXL-Dan capsules was safe and well-tolerated, but not superior to placebo control for pain management or functional improvement in patients with knee OA. Future studies employing individualized treatment and medications might be more appropriate to assess the efficacy of Chinese medicine treatment.
Author contributions
Conception and design: LL, MH, DYWL, AMKG, PL and BB; Analysis and interpretation of the data: LL, MH, PL, KC and BB; Drafting of the article: LL, MH, PL, KC and BB; Critical revision of the article for important intellectual content: LL, MH, DYWL, AMKG, HHSF, PL, EKL, LST and BB; Final approval of the article: all authors; Provision of study materials or patients: DYWL, HHSF, EKL and LST; Statistical expertise: AMKG, PL and KC; Obtaining of funding: LL, MH, DYWL, AMKG, HHSF and BB; Administrative, technical or logistic support: DYWL, HHSF, KC and LST. LL and BB took responsibility for the integrity of the work as a whole, from inception to finished article.
Role of funding source
This study was supported by NIH/NCCAM grant P01 AT002605-01A1 from the National Center for Complementary and Alternative Medicine (NCCAM) at the National Institutes of Health. Its contents are solely responsibility of the authors and do not necessarily represent the official views of NCCAM. NCCAM had no involvement in the study design, collection, analysis and interpretation of data; in the writing of the manuscript; and in the decision to submit the manuscript for publication.
Competing interest statement
All authors declared no conflicts of interest.
Acknowledgements
We thank Dr Lyn Lowry for her English edit, Dr Wing-Fai Yeung, Ms Wing-Lok Lam and Ms Mei-Kuen Wong for their editorial assistance, Dr Hongjie Zhang (UIC) for his phytochemical technical assistance, Dr Stephen W. Hoag for his assistance in the formulation of placebo herbal capsule and Dr Tena Li for her assistance in subject recruitment.
Appendix. Exclusion criteria
Uncontrolled angina and/or congestive heart failure, severe chronic obstructive pulmonary disease, active treatment for cancer, major psychiatric disease, other severe systemic disease, or significant abnormalities on screening physical examination and laboratory tests that reveal clinically important abnormalities of hematological, cardiac, pulmonary, metabolic, renal, hepatic, gastrointestinal or other systems, intra-articular (IA) corticosteroid injection of either knee within a 3 month interval immediately prior to baseline screening, IA hyaluronates in either knee within the past 6 months, Tidal lavage or arthroscopy of either knee within the past 12 months, current use of topical capsaicin cream on either knee, blood pressure over 180 mm Hg systolic or 100 mm Hg diastolic (at either the second or third of triplicate measurements), use of any constituent herb in HLXL within the past 3 months, current use of Chinese herbs for arthritis, use of oral prednisone in the past 30 days, history or clinical indications of bleeding diathesis, including current use of anti-coagulants, use of any investigational drug within the past 30 days, inflammatory arthritis (e.g., rheumatoid or psoriatic arthritis), currently participating in another intervention research study, unwilling to be randomized, plan to move residence away from the immediate area within the next 2 months, drug or alcohol abuse sufficient to hinder compliance with treatment or follow-up procedures.
Appendix A. Supplementary data
The following is the supplementary data related to this article:
01215-7&id=gr1.jpg){kind=link}
01215-7&id=gr2.jpg){kind=link}