This randomized clinical trial determines the efficacy of a 6-week combination of conservative treatments vs an education comparator for people with thumb base osteoarthritis.
Key Points
Question
Is a combination of conservative treatments including education, splint, hand exercises, and diclofenac sodium, 1%, gel superior to education alone on clinical outcomes in thumb base osteoarthritis?
Findings
In this randomized clinical trial including 204 participants, combined treatments provided small beneficial effects on hand function at 6 weeks, but not pain, compared with an education-only comparator group. At 12 weeks, the between-group differences in hand function and pain were greater, favoring the intervention.
Meaning
In people with thumb base osteoarthritis, combined treatments provided small to medium and potentially clinically beneficial effects on hand function but not pain.
Abstract
Importance
A combination of conservative treatments is commonly used in clinical practice for thumb base osteoarthritis despite limited evidence for this approach.
Objective
To determine the efficacy of a 6-week combination of conservative treatments compared with an education comparator.
Design, Setting, and Participants
Randomized, parallel trial with 1:1 allocation ratio among people aged 40 years and older with symptomatic and radiographic thumb base osteoarthritis in a community setting in Australia.
Interventions
The intervention group (n = 102) received education on self-management and ergonomic principles, a base-of-thumb splint, hand exercises, and diclofenac sodium, 1%, gel. The comparator group (n = 102) received education on self-management and ergonomic principles alone. Intervention use was at participants’ discretion from 6 to 12 weeks.
Main Outcomes and Measures
Hand function (Functional Index for Hand Osteoarthritis; 0-30) and pain (visual analog scale; 0-100 mm) were measured at week 6 (primary time point) and week 12. An α of .027 was used at week 6 to account for co–primary outcomes.
Results
Of the 204 participants randomized, 195 (96%) and 194 (95%) completed follow-ups at 6 and 12 weeks, respectively; the mean (SD) age of the population was 65.6 (8.1) years, and 155 (76.0%) were female. At week 6, hand function improved significantly more in the intervention group than the comparator (between-group difference, −1.7 units; 97.3% CI, −2.9 to −0.5; P = .002). This trend was sustained at 12 weeks (−2.4 units; 95% CI, −3.5 to −1.3; P < .001). Pain scores improved similarly at week 6 (between-group difference, −4.2 mm; 97.3% CI, −11.3 to 3.0; P = .19). At week 12, pain reduction was significantly greater in the intervention group (−8.6 mm; 95% CI, −15.2 to −2.0; P = .01). There were 34 nonserious adverse events, all in the intervention group—mostly skin reactions and exercise-related pain exacerbations.
Conclusions and Relevance
In this randomized clinical trial of people with thumb base osteoarthritis, combined treatments provided small to medium and potentially clinically beneficial effects on hand function but not pain.
Trial Registration
Australian New Zealand Clinical Trials Registry Identifier: ACTRN12616000353493
Introduction
Osteoarthritis is the most common chronic joint disorder, with increasing prevalence particularly owing to increased longevity.1,2 It most commonly affects the hands, among which first carpometacarpal joint (thumb base) osteoarthritis represents a particular clinical challenge because of its significant effect on hand function and limited evidence for efficacious treatments. Thumb base osteoarthritis predominantly affects postmenopausal women. Radiographic signs of osteoarthritis are estimated in approximately one-third of women older than 70 years,3 while approximately 1 in 20 have visited a medical professional because of thumb base pain.4
Recommended conservative treatments include (1) education and training in ergonomic principles; (2) exercises targeting muscle strength, joint mobility, and stability; (3) splinting; and (4) nonsteroidal anti-inflammatory drugs (NSAIDs), with topical formulations preferred over oral NSAIDs because of their superior adverse effect profile.5 Nevertheless, individually, these treatments usually only provide small benefits to patients.5,6 A more effective conservative treatment strategy for thumb base osteoarthritis is lacking.
A combination of safe, efficacious conservative treatments is a potential way to optimize treatment effects and provide meaningful clinical benefits for patients. In this regard, the National Institute for Health and Care Excellence highlighted the investigation of combinations of osteoarthritis treatments as a key research priority.7 The aim of this trial was to determine the effect of a combination of conservative treatments for thumb base osteoarthritis compared with an education comparator.
Methods
Trial Design
The Efficacy of Combined Conservative Therapies on Clinical Outcomes in Patients With Thumb Base Osteoarthritis (COMBO) was a randomized, parallel, 2-arm superiority trial with a 1:1 allocation ratio conducted at the Royal North Shore hospital, a tertiary-care academic hospital in Australia. Participants were randomized into either a combined treatment or a comparator group. The interventions took place from baseline to 6 weeks. Follow-up assessments occurred at 2, 6, and 12 weeks (Figure). Participants were further contacted by telephone or via an online survey at 6 months to assess the primary outcomes and the long-term use of the interventions. Participants in the comparator group were instructed on all treatments received by the intervention group after week 12. These included information about the splint and diclofenac gel, although those were not provided to the comparator group.
Figure. Study Flowchart.
FIHOA indicates Functional Index for Hand Osteoarthritis; KLG, Kellgren-Lawrence grade; VAS, visual analog scale.
A consumer focus group provided advice about treatment selection, outcomes, and duration and frequency of study assessments during the trial genesis. The trial was prospectively registered (protocol in Supplement 1). Approval was obtained from the institutional Human Research Ethics Committee (HREC/15/HAWKE/479). Written informed consent was obtained from all participants. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.
Participants
Participants were recruited from the community in Sydney, Australia, from May 2016 to July 2018 via advertisements on social media, notice boards, medical practice waiting rooms, and our volunteer database. Preliminary screening was conducted by the study physiotherapist via telephone/internet.
Inclusion criteria were the following8:
Age 40 years and older
Thumb base pain at least half of the days in the past month
Average pain rated at 40 or greater on a 0 to 100-mm visual analog scale (VAS), where 0 is no pain and 100 is the worst pain imaginable, over the past 30 days and in the 48 hours prior to screening
Score of 6 or higher on the Functional Index for Hand Osteoarthritis (FIHOA; range, 0-30; higher scores representing worse function)9
Radiographic evidence of osteoarthritis at the first carpometacarpal joint, read by a trained rheumatologist (L.A.D.) (Kellgren-Lawrence grade [KLG] ≥2)10
Main exclusion criteria were presence of concomitant conditions potentially contributing to thumb base pain (eg, inflammatory arthritis) and current use of any of the study treatments. The complete list can be found in Deveza et al.11
Randomization, Allocation Concealment, and Blinding
Participants were assigned to either the intervention or comparator group with a 1:1 allocation via a computer-generated randomization schedule, stratified by osteoarthritis severity (KLG 2 and 3 vs 4) using random blocks of size 2, 4, and 6. Sequentially numbered opaque and sealed envelopes were used to conceal the allocation sequence from researchers enrolling and assessing participants.
The study physiotherapist who was unblinded to treatment allocation provided participants with the questionnaires. A blinded assessor performed strength and physical examination measures. Success of blinding was checked at the end of each assessment. The statistician was blinded to group allocation for primary outcome analyses. Participants were unaware of the treatments received by the opposite group to minimize differences regarding treatment expectations.
Interventions
Both groups attended 2 individual, face-to-face treatment sessions with 1 single study physiotherapist (baseline and week 2). The intervention group received a combination of education, splint, hand exercises, and diclofenac sodium, 1%, gel. The comparator group received education alone. The second session was aimed at ensuring the correct performance of exercises by participants in the intervention group and assisting with the progression of the exercises (repetitions and intensity). The comparator group also attended a second session to balance the number of face-to-face visits with the physiotherapist. We emphasized the education component to both groups and clarified any questions at week 2. The duration of the treatment sessions for the intervention and comparator groups was similar (30 and 15 minutes, respectively). Scripts were used by the therapist to standardize the delivery of the intervention.11 We conducted self-monitoring visits to check for protocol fidelity.
Participants were instructed to use the treatments until the 6-week assessment. After this, they were encouraged to continue treatment throughout the follow-up (6 to 12 weeks), but this was at their discretion, to enable evaluation of their treatment choice. At week 12, participants in the comparator group received instructions and recommendations to use the treatments as per the intervention group. All participants were asked not to start other new treatments during the study.
Participants were allowed to use current pain medications provided that the dose had been stable at study entry, per exclusion criteria. Acetaminophen (paracetamol) was permitted as rescue medication (maximum 3000 mg per day).
Education About Osteoarthritis and Ergonomic Principles
Both groups were provided with education about osteoarthritis and ergonomic principles (formerly known as “joint protection”) using a 9-page educational booklet and 2 individual, face-to-face sessions with the study physiotherapist (baseline and week 2).11 The educational booklet did not provide information about exercises or splints.
Splint
The intervention group received a prefabricated neoprene splint (Comfort Cool Thumb CMC Restriction Splint [North Coast Medical]) that incorporated the thumb base and wrist. They were recommended to use the splint during daily activities (minimum 4 hours per day) (eFigure in Supplement 2) and to remove the splint during rest, sleep, exercising, and bathing.
Hand Exercises
The exercise program consisted of 5 exercises to optimize range of motion and improve neuromuscular control of thumb alignment, muscular endurance, and proprioception. These were thumb opposition, paper tearing, line tracing on a ball, using chopsticks to pick up objects, and squeezing a ball. Participants were instructed to perform the exercises at home 3 times per week. The program was adjusted as necessary at week 2. Further details are provided in the protocol (Supplement 1) and the eAppendix in Supplement 2 (all copyrighted material in the eAppendix is reproduced with permission).
Topical NSAID
The intervention group received diclofenac diethylammonium gel (11.6 mg/g) (diclofenac sodium, 1%, gel; Voltaren Emulgel [GlaxoSmithKline]), to apply daily over the thumb base 3 times per day. They received a spatula with a permanent pen mark to standardize the amount to be used (corresponding to approximately 200 mg in an area of 40 cm2).
Outcomes
The primary outcomes were thumb base pain, assessed by VAS (0-100 mm), and hand function, assessed by FIHOA (0-30) at 6 weeks. The FIHOA tool is composed of 10 items scored using a semiquantitative 4-point scale (0-3 units).12
Secondary outcomes were the following11:
Thumb base pain (VAS) and hand function scores (FIHOA) at 2 and 12 weeks, as well as the following outcomes assessed at 2, 6, and 12 weeks:
Grip strength (Jamar hand dynamometer [in kg]) and tip pinch strength (B&L pinch gauge [in kg])
Patient global disease assessment (VAS 0-100 mm, where 0 is very well and 100 is very poor)
Thumb base stiffness (minutes)
Health-related quality of life assessed by the Assessment of Quality of Life–4D instrument (AQoL-4D)13 scored from −0.04 to 1.00, with 1.00 indicating full health14
Use of rescue pain medications
Presence of swelling and tenderness on joint examination (present or absent)
Participant’s global rating of change for pain, function, and overall change, scored using a 5-point Likert scale (much better to much worse)
Percentage of treatment responders at 6 and 12 weeks according to the Outcome Measures in Rheumatology (OMERACT) and Osteoarthritis Research Society International (OARSI) criteria15
Impairments in work and other activities, assessed by the Work Productivity and Activity Impairment Questionnaire–General Health (WPAI-GH)16,17,18
The credibility/expectancy questionnaire was collected at baseline19 (0 to 100, with higher scores demonstrating higher expectation/credibility). Frequency of treatment use as recommended by the physiotherapist and use of rescue medication were monitored using a diary. Adverse events were recorded in the diary.
Statistical Analysis
The sample size estimation accounted for the 2 primary outcomes. Because the minimal clinically important difference (MCID) of the FIHOA is unknown, the calculation was based on detecting a mean difference of 3 points (defined arbitrarily) (range, 0-30) with an SD of 6.2.20 For pain, the calculation was based on detecting an MCID of 20 mm (SD, 20) on a 100-mm VAS.21 As the 2 primary outcomes were correlated (r = 0.49),9 an α of .027 was used as the level of significance for both outcomes at 6 weeks, which preserved an overall 5% level of significance.22 A sample size of 102 participants per group was estimated to achieve at least 80% power accounting for 20% dropouts.
To determine the potential clinical significance of the between-group difference in hand function assessed by the FIHOA questionnaire, we used the anchor-based method to calculate the MCID based on the global ratings of change observed in this trial, as detailed in Terwee et al.23 Briefly, the MCID was estimated based on the difference in the change score of the “slightly better” and “no change” participants.24
Data were analyzed according to the intention-to-treat principle only. The between-group differences in mean change from baseline were compared using independent t test or the Wilcoxon rank-sum test as appropriate. Analyses were further adjusted for baseline score, sex, and KLG using analysis of covariance models fitted separately at 2, 6, and 12 weeks. Standardized mean differences (95% CI) were computed as the adjusted between-group difference in scores divided by the pooled SD of baseline scores.25 Categorical outcomes were analyzed using the χ2 test. The 95% CIs were calculated throughout with the exception of additional 97.3% CIs presented for the co–primary outcomes.
Results
Participants
We randomized 204 participants (102 to each group) (Figure), with 9 participants (4%) lost to follow-up from baseline to week 6. The mean (SD) age of the population was 65.6 (8.1) years, and 155 (76.0%) were female. Six-month follow-up data were obtained for 91 participants (89%) in the intervention group and 98 (96%) in the comparator group. Baseline characteristics (Table 1) were similar between the groups, except for a greater proportion of women and higher treatment credibility and expectancy in the intervention group. In the intervention group, 63 (66%) and 43 (45%) participants reported using all treatments (exercise, diclofenac sodium gel, and splint) at 6 and 12 weeks, respectively. Accidental unblinding of the blinded assessor (assessing grip and pinch strength and joint examination) occurred in 25 of 204 participants (12.3%), mostly due to participants revealing their treatment allocation.
Table 1. Demographic and Clinical Characteristics of Participants by Group and in Total at Baseline.
| Characteristic | No. (%) | ||
|---|---|---|---|
| Comparator (n = 102) | Intervention (n = 102) | Total (n = 204) | |
| Age, mean (SD), y | 65.2 (8.5) | 66.0 (7.8) | 65.6 (8.1) |
| BMI, mean (SD)a | 28.4 (6.6) | 28.9 (6.4) | 28.7 (6.5) |
| Female sex | 68 (66.7) | 87 (85.3) | 155 (76.0) |
| Marital status | |||
| Never married | 6 (5.9) | 6 (5.9) | 12 (5.9) |
| Widowed | 6 (5.9) | 10 (9.8) | 16 (7.8) |
| Divorced/separated | 19 (18.6) | 24 (23.5) | 43 (21.1) |
| Married | 71 (69.6) | 62 (60.8) | 133 (65.2) |
| Household income per year, $ | |||
| Negative or nil income | 5 (4.9) | 4 (3.9) | 9 (4.4) |
| 1-33 799 | 19 (18.6) | 25 (24.5) | 44 (21.6) |
| 33 800-88 399 | 45 (44.1) | 37 (36.3) | 82 (40.2) |
| 88 400-207 999 | 22 (21.6) | 28 (27.5) | 50 (24.5) |
| >208 000 | 8 (7.8) | 4 (3.9) | 12 (5.9) |
| Highest education level | |||
| Less than secondary | 3 (2.9) | 5 (4.9) | 8 (3.9) |
| Completed secondary | 16 (15.7) | 12 (11.8) | 28 (13.7) |
| Diploma or certificate | 33 (32.4) | 32 (31.4) | 65 (31.9) |
| Tertiary undergraduate | 21 (20.6) | 21 (20.6) | 42 (20.6) |
| Tertiary postgraduate | 29 (28.4) | 32 (31.4) | 61 (29.9) |
| Cultural and ethnic group | |||
| Australian | 45 (44.1) | 52 (51.0) | 97 (47.5) |
| British | 17 (16.7) | 20 (19.6) | 37 (18.1) |
| Irish | 6 (5.9) | 8 (7.8) | 14 (6.9) |
| Other | 34 (33.3) | 22 (21.6) | 56 (27.5) |
| Index thumb, right | 60 (58.8) | 54 (52.9) | 114 (55.9) |
| Dominant hand, right | 88 (86.3) | 91 (89.2) | 179 (87.7) |
| Bilateral hand (any joint, self-reported) | 49 (48.0) | 55 (53.9) | 104 (51.0) |
| Comorbid interphalangeal joint painb | 29 (28.4) | 28 (27.5) | 57 (27.9) |
| Erosive osteoarthritisc | 21 (20.6) | 22 (21.6) | 43 (21.1) |
| Other joints with osteoarthritis (any other joint) | 84 (82.4) | 88 (86.3) | 172 (84.3) |
| Knee | 53 (63.1) | 60 (68.2) | 113 (65.7) |
| Foot | 31 (36.9) | 34 (38.6) | 65 (37.8) |
| Neck | 28 (33.3) | 36 (40.9) | 64 (37.2) |
| Spine | 26 (30.9) | 25 (28.4) | 51 (29.6) |
| Hip | 21 (25.0) | 24 (27.3) | 45 (26.2) |
| Shoulder | 19 (22.6) | 20 (22.7) | 39 (22.7) |
| Ankle | 16 (19.0) | 18 (20.4) | 34 (19.8) |
| Wrist | 15 (17.9) | 18 (20.4) | 33 (19.2) |
| Elbow | 7 (8.3) | 7 (7.9) | 14 (8.1) |
| Stiffness, mean (SD), min | 10.3 (21.9) | 7.6 (14.4) | 9.0 (18.6) |
| Symptoms duration, y | |||
| <1 | 11 (10.8) | 11 (10.8) | 22 (10.8) |
| 1-5 | 48 (47.0) | 52 (50.9) | 100 (49.0) |
| >5 | 43 (42.2) | 39 (38.2) | 82 (40.2) |
| Kellgren-Lawrence grade | |||
| 2 | 48 (47.1) | 40 (39.2) | 88 (43.1) |
| 3 | 36 (35.3) | 44 (43.1) | 80 (39.2) |
| 4 | 18 (17.6) | 18 (17.6) | 36 (17.6) |
| Subluxation ratio, mean (SD), mmd | 0.36 (0.08) | 0.37 (0.08) | 0.37 (0.08) |
| Medications | |||
| None | 37 (36.3) | 38 (37.3) | 75 (36.8) |
| Paracetamol | 45 (44.1) | 43 (42.2) | 88 (43.1) |
| NSAIDs | 40 (39.2) | 32 (31.4) | 72 (35.3) |
| Opioids | 3 (2.9) | 7 (6.9) | 10 (4.9) |
| Reported general good health | 96 (94.1) | 93 (91.2) | 189 (92.6) |
| SCQ [0-36], median (IQR) | 5 (3-7) | 5 (3-7) | 5 (3-7) |
| VAS thumb pain [0-100], mean (SD) | 58.4 (14.1) | 57.3 (13.1) | 57.9 (13.6) |
| FIHOA [0-30], mean (SD) | 10.5 (4.1) | 10.8 (4.0) | 10.7 (4.0) |
| Treatment credibility, mean (SD)e | 74.2 (18.2) | 80.2 (13.1) | NA |
| Treatment expectancy, mean (SD)e | 54.6 (20.6) | 60.7 (16.8) | NA |
Abbreviations: BMI, body mass index; FIHOA, Functional Index for Hand Osteoarthritis; IQR, interquartile range; NA, not applicable; NSAIDs, nonsteroidal anti-inflammatory drugs; SCQ, self-administered comorbidity questionnaire; VAS, visual analog scale.
Calculated as weight in kilograms divided by height in meters squared.
Self-reported pain in 1 or more interphalangeal (IP) joints (among all proximal and distal IPs and thumb IP) in the index hand.
Presence of erosive osteoarthritis on radiograph in either thumb base or IP joints.
Radiographic assessment of the degree of radial subluxation of the first metacarpal base off the trapezium, divided by the width of the first metacarpal base, as detailed in Deveza et al11 and Riordan et al.26
The credibility/expectancy questionnaire was collected at baseline19 (0 to 100, with higher scores demonstrating higher expectation/credibility).
There were 34 adverse events, all in the intervention group, with mild to moderate severity in 28 participants. These mostly consisted of skin reactions owing to splint use (n = 12) or diclofenac gel (n = 3). One participant reported gastrointestinal upset. There were 17 episodes of transient increased thumb pain related to the exercises. One participant reported a thumb injury unrelated to the study. All adverse events resolved with no sequelae. There were no serious adverse events.
Primary Outcomes
There was no between-group difference in pain from baseline to week 6 (mean difference, −4.2; 97.3% CI, −11.3 to 3.0; P = .19) (Table 2). Pain scores improved in both groups. Change in hand function was significantly different between the groups at week 6 (between-group difference, −1.7; 97.3% CI, −2.9 to −0.5; P = .002), favoring the intervention (Table 2). Adjusted analyses did not substantially change these results. The standardized mean difference (Cohen d) indicated the difference in function was small (−0.40; 97.3% CI, −0.68 to −0.11).25 The observed differences at week 6 and week 12 in the FIHOA scores were superior to the calculated MCID of −1.3.
Table 2. Primary Outcomes.
| Time point | Summary of variables, mean (SD) [No.] | Change from baseline, mean (95% CI) | Between-group difference in change [intervention − comparator], difference (95% CI) | |||||
|---|---|---|---|---|---|---|---|---|
| Comparator | Intervention | Comparator | Intervention | Unadjusted | P value | Adjusted | P value | |
| VAS pain | ||||||||
| Baseline | 58.4 (14.1) [102] | 57.3 (13.1) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 45.2 (21.4) [99] | 41.6 (20.6) [96] | −13.2 (−17.0 to −9.3)a | −16.6 (−21.0 to −12.2)a | −3.4 (−9.2 to 2.4) | .24 | −3.4 (−9.1 to 2.3) | .23 |
| 6-wk | 40.4 (22.5) [99] | 36.2 (22.6) [96] | −17.3 (−21.7 to −13.0)a | −21.5 (−26.1 to −16.9)a | −4.2 (−10.5 to 2.1) | .19 | −3.4 (−9.6 to 2.7) | .27 |
| −4.2 (−11.3 to 3.0)b | ||||||||
| 12-wk | 43.9 (23.5) [98] | 35.5 (22.1) [96] | −13.6 (−17.9 to −9.2)a | −22.2 (−27.2 to −17.2)a | −8.6 (−15.2 to −2.0) | .01 | −8.5 (−15.0 to −2.1) | .009 |
| FIHOA | ||||||||
| Baseline | 10.5 (4.1) [102] | 10.8 (4.0) | NA | NA | NA | NA | NA | NA |
| 2-wk | 9.5 (4.4) [99] | 9.5 (3.9) [97] | −0.8 (−1.4 to −0.2)a | −1.4 (−2.1 to −0.6)a | −0.5 (−1.5 to 0.4) | .28 | −0.4 (−1.3 to 0.4) | .35 |
| 6-wk | 9.4 (4.7) [99] | 8.3 (3.9) [96] | −0.9 (−1.7 to −0.2)c | −2.6 (−3.4 to −1.9)a | −1.7 (−2.8 to −0.7) | .002 | −1.5 (−2.5 to −0.5) | .002 |
| −1.7 (−2.9 to −0.5)b | ||||||||
| 12-wk | 9.5 (4.4) [98] | 7.6 (4.4) [96] | −0.9 (−1.6 to −0.2)c | −3.3 (−4.1 to −2.4)a | −2.4 (−3.5 to −1.3) | <.001 | −2.2 (−3.3 to −1.1) | <.001 |
Abbreviations: FIHOA, Functional Index for Hand Osteoarthritis; VAS, visual analog scale.
P < .01.
97.3% CI (primary outcomes).
P < .02.
Secondary Outcomes
At week 12, there was greater improvement in pain and hand function in the intervention group vs the comparator group (between-group difference, −8.6; 95% CI, −15.2 to −2.0; P = .01 for pain; −2.4; 95% CI, −3.5 to −1.3; P < .001 for function) (Table 2). Standardized mean differences were greater (medium) for both outcomes (function: −0.56; 95% CI, −0.83 to −0.29; pain: −0.63; 95% CI, −1.10 to −0.16).
At week 6, the intervention group had greater improvements compared with the comparator group in the mental health domain of the AQoL-4D (between-group difference, 5.6; 95% CI, 2.2 to 8.9; P = .001) and grip strength (1.6; 95% CI, 0.2 to 3.1; P = .03) (Table 3). The between-group difference in tip-pinch strength became significant in the adjusted analysis, favoring the intervention (0.4; 95% CI, 0.1 to 0.7; P = .01).
Table 3. Secondary Outcomesa.
| Time point | Summary of variables, mean (SD) [No.] | Change from baseline, mean (95% CI) | Between-group difference in change [intervention − comparator], difference (95% CI) | |||||
|---|---|---|---|---|---|---|---|---|
| Comparator | Intervention | Comparator | Intervention | Unadjusted | P value | Adjusted | P value | |
| PGA of disease activity | ||||||||
| Baseline | 41.8 (21.5) [102] | 37.1 (21.8) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 36.7 (22.0) [99] | 32.2 (21.3) [96] | −4.9 (−9.6 to −0.2)b | −5.0 (−9.1 to −0.9)b | −0.2 (−6.4 to 6.1) | .95 | −2.7 (−8.3 to 2.7) | .32 |
| 6-wk | 37.6 (22.7) [99] | 30.9 (21.3) [96] | −3.9 (−8.9 to 1.1) | −6.4 (−11.1 to −1.6)b | −2.4 (−9.3 to 4.4) | .48 | −5.5 (−11.4 to 0.4) | .06 |
| 12-wk | 37.5 (24.0) [98] | 28.6 (22.0) [96] | −3.8 (−8.1 to 0.6) | −8.7 (−13.2 to −4.1)b | −4.9 (−11.1 to 1.4) | .12 | −8.0 (−13.7 to −2.3) | .005 |
| AQoL-4D independent living | ||||||||
| Baseline | 89.9 (12.2) [102] | 88.9 (12.7) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 89.4 (12.4) [99] | 89.8 (10.7) [97] | −0.6 (−2.7 to 1.5) | 0.5 (−1.8 to 2.7) | 1.0 (−2.0 to 4.1) | .50 | 1.1 (−1.6 to 3.8) | .42 |
| 6-wk | 89.6 (13.0) [99] | 91.6 (10.3) [96] | −0.3 (−2.6 to 1.9) | 2.2 (0.3 to 4.1)b | 2.5 (−0.4 to 5.5) | .08 | 2.4 (−0.2 to 5.1) | .07 |
| 12-wk | 90.5 (12.7) [98] | 92.5 (8.8) [96] | 0.3 (−2.1 to 2.8) | 3.1 (1.0 to 5.2)b | 2.8 (−0.4 to 6.0) | .08 | 2.4 (−0.3 to 5.1) | .08 |
| AQoL-4D mental health | ||||||||
| Baseline | 70.4 (15.3) [102] | 68.1 (16.6) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 71.1 (15.5) [99] | 70.9 (16.1) [97] | 0.4 (−1.7 to 2.6) | 2.5 (−0.1 to 5.2) | 2.1 (−1.3 to 5.5) | .22 | 0.8 (−2.4 to 3.9) | .63 |
| 6-wk | 70.9 (16.2) [99] | 74.3 (16.3) [96] | 0.3 (−1.8 to 2.5) | 5.9 (3.3 to 8.5)b | 5.6 (2.2 to 8.9) | .001 | 5.1 (1.8 to 8.3) | .002 |
| 12-wk | 70.9 (17.1) [98] | 73.1 (17.3) [96] | 0.3 (−1.9 to 2.6) | 4.7 (2.4 to 7.1)b | 4.4 (1.1 to 7.7) | .009 | 3.9 (0.6 to 7.2) | .02 |
| AQoL-4D relationship | ||||||||
| Baseline | 87.1 (16.1) [102] | 90.4 (12.1) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 89.1 (13.7) [99] | 90.3 (12.4) [97] | 1.6 (−0.6 to 3.7) | −0.6 (−2.1 to 1.0) | −2.1 (−4.8 to 0.5) | .10 | −1.2 (−3.6 to 1.3) | .34 |
| 6-wk | 88.4 (14.3) [99] | 90.9 (11.6) [96] | 1.1 (−1.0 to 3.3) | 0.1 (−1.6 to 1.8) | −1.0 (−3.7 to 1.7) | .46 | −0.1 (−2.6 to 2.5) | .95 |
| 12-wk | 87.1 (14.3) [98] | 91.1 (11.1) [96] | −0.6 (−2.9 to 1.8) | 0.3 (−1.3 to 2.0) | 0.9 (−2.0 to 3.8) | .53 | 1.5 (−1.1 to 4.1) | .26 |
| AQoL-4D senses | ||||||||
| Baseline | 88.1 (11.3) [102] | 91.3 (9.5) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 89.7 (10.3) [99] | 92.6 (8.6) [97] | 1.2 (−0.6 to 3.0) | 0.6 (−0.9 to 2.1) | −0.7 (−3.0 to 1.7) | .57 | 0.5 (−1.6 to 2.7) | .62 |
| 6-wk | 90.0 (10.2) [99] | 91.9 (9.9) [96] | 2.0 (0.2 to 3.8)b | 0 (−1.8 to 1.8) | −2.0 (−4.5 to 0.5) | .11 | −0.7 (−3.0 to 1.6) | .54 |
| 12-wk | 88.7 (10.7) [98] | 93.2 (8.9) [96] | 0.8 (−1.2 to 2.8) | 1.3 (−0.5 to 3.0) | 0.5 (−2.1 to 3.1) | .71 | 1.9 (−0.5 to 4.3) | .11 |
| AQoL-4D total | ||||||||
| Baseline | 83.9 (9.6) [102] | 84.7 (9.3) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 84.8 (8.7) [99] | 85.9 (8.5) [97] | 0.7 (−0.6 to 1.9) | 0.7 (−0.6 to 2.1) | 0.1 (−1.7 to 1.9) | .93 | 0.3 (−1.4 to 1.9) | .76 |
| 6-wk | 84.7 (9.5) [99] | 87.2 (8.7) [96] | 0.8 (−0.6 to 2.2) | 2.1 (0.9 to 3.2)b | 1.3 (−0.5 to 3.1) | .16 | 1.5 (−0.2 to 3.2) | .08 |
| 12-wk | 84.3 (10.0) [98] | 87.5 (8.2) [96] | 0.2 (−1.3 to 1.7) | 2.4 (1.2 to 3.5)b | 2.1 (0.3 to 4.0) | .02 | 2.2 (0.4 to 4.0) | .01 |
| Grip strength (index hand) | ||||||||
| Baseline | 22.1 (10.3) [102] | 20.4 (8.3) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 23.3 (11.0) [92] | 22.4 (8.3) [91] | 1.0 (0 to 2.1) | 2.1 (1.2 to 3.0)b | 1.0 (−0.4 to 2.4) | .15 | 1.2 (0 to 2.6) | .06 |
| 6-wk | 23.7 (10.6) [93] | 23.7 (8.7) [93] | 1.4 (0.4 to 2.4)b | 3.0 (2.0 to 4.1)b | 1.6 (0.2 to 3.1) | .02 | 2.2 (0.6 to 3.8) | .005 |
| 12-wk | 24.7 (11.3) [92] | 24.5 (9.2) [93] | 2.4 (1.2 to 3.6)b | 4.0 (2.8 to 5.1)b | 1.6 (−0.1 to 3.2) | .06 | 2.0 (0.4 to 3.6) | .01 |
| Tip-pinch strength (index hand) | ||||||||
| Baseline | 3.4 (1.5) [102] | 3.1 (1.1) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 3.5 (1.6) [92] | 3.4 (1.1) [91] | 0.2 (0 to 0.4) | 0.2 (0 to 0.4)b | 0.1 (−0.2 to 0.3) | .67 | 0.1 (−0.1 to 0.3) | .36 |
| 6-wk | 3.5 (1.5) [93] | 3.6 (1.1) [93] | 0.1 (−0.1 to 0.4) | 0.4 (0.2 to 0.6)b | 0.2 (−0.1 to 0.5) | .11 | 0.3 (0 to 0.6) | .01 |
| 12-wk | 3.7 (1.5) [92] | 3.8 (1.2) [93] | 0.3 (0.1 to 0.5)b | 0.6 (0.4 to 0.8)b | 0.3 (0 to 0.6) | .07 | 0.3 (0 to 0.6) | .01 |
| Stiffness | ||||||||
| Baseline | 10.3 (21.9) [102] | 7.6 (14.4) [102] | NA | NA | NA | NA | NA | NA |
| 2-wk | 13.4 (39.6) [98] | 5.4 (14.8) [96] | 3.6 (−2.8 to 10.1) | −1.8 (−4.3 to 0.8) | −5.4 (−12.3 to 1.6) | .12 | −5.7 (−12.9 to 1.4) | .11 |
| 6-wk | 12.7 (42.2) [99] | 7.2 (17.1) [96] | 2.3 (−5.7 to 10.3) | −0.2 (−3.9 to 3.5) | −2.6 (−11.4 to 6.3) | .56 | −1.8 (−10.6 to 6.9) | .68 |
| 12-wk | 10.1 (24.7) [98] | 8.0 (34.8) [96] | −0.4 (−5.5 to 4.7) | 0.5 (−7.0 to 8.1) | 0.9 (−8.1 to 9.9) | .84 | −0.7 (−9.3 to 7.9) | .87 |
Abbreviations: AQoL-4D, Assessment of Quality of Life–4D instrument; PGA, patient global assessment.
Unadjusted analysis.
P < .01.
At week 12, the intervention group had greater improvements compared with the comparator group in the mental health domain of the AQoL-4D (4.4; 95% CI, 1.1 to 7.7; P = .009) and total AQoL-4D score (2.1; 95% CI, 0.3 to 4.0; P = .03) (Table 3). There were no between-group differences in strength, patient global assessment, and joint stiffness duration, although differences in strength and patient global assessment became significant in the adjusted analysis, favoring the intervention.
Among participants with joint swelling at baseline (n = 82, 40%), the percentage with absence of joint swelling at 6 and 12 weeks was similar between groups (eTable 1 in Supplement 2). In contrast, a greater percentage of the comparator group reported improvement in joint tenderness at 6 and 12 weeks (eTable 1 in Supplement 2). There was no between-group difference at any time point for WPAI-GH (eTable 2 in Supplement 2).
Fewer participants in the intervention group reported using rescue pain medication between 2 and 6 weeks. Although this trend remained from week 6 to 12, the difference was not significant.
The intervention group had a greater rate of responders according to OMERACT-OARSI criteria at all time points (Table 4), but this was only significant at 12 weeks. The number of participants reporting themselves as much better or slightly better in the intervention group was greater compared with the comparator for pain, function, and overall change at 6 and 12 weeks (Table 4).
Table 4. Rates of Responders According to OMERACT-OARSI Criteria and Rates of Perceived Improvement in Pain, Function, and Overall Change Stratified by Group and Time Point.
| Outcome and time point | No. | Comparator, No. (%) | No. | Intervention, No. (%) | P value |
|---|---|---|---|---|---|
| OMERACT-OARSI criteria | |||||
| 2-wk | 99 | 42 (42.4) | 97 | 50 (51.5) | .20 |
| 6-wk | 99 | 45 (45.5) | 96 | 56 (58.3) | .07 |
| 12-wk | 98 | 44 (44.9) | 96 | 60 (62.5) | .01 |
| Meaningful change in pain | |||||
| 2-wk | 99 | 37 (37.4) | 97 | 42 (43.3) | .39 |
| 6-wk | 99 | 46 (46.5) | 96 | 64 (66.7) | .004 |
| 12-wk | 98 | 39 (39.8) | 95 | 67 (70.5) | <.001 |
| Meaningful change in function | |||||
| 2-wk | 99 | 34 (34.3) | 96 | 41 (42.7) | .23 |
| 6-wk | 99 | 40 (40.4) | 96 | 62 (64.6) | .001 |
| 12-wk | 98 | 38 (38.8) | 95 | 67 (70.5) | <.001 |
| Meaningful change overall | |||||
| 2-wk | 99 | 37 (37.4) | 97 | 44 (45.4) | .25 |
| 6-wk | 99 | 47 (47.5) | 96 | 69 (71.9) | .001 |
| 12-wk | 98 | 41 (41.8) | 95 | 71 (74.7) | <.001 |
Abbreviations: OARSI, Osteoarthritis Research Society International; OMERACT, Outcome Measures in Rheumatology.
Primary Outcomes and Use of Intervention at 6 Months
In the intervention group, mean (SD) pain scores increased slightly compared with week 12 (39.2 [23.1]), whereas mean (SD) function scores continued to improve (7.2 [4.5]) (eTable 3 in Supplement 2). In the comparator group, mean (SD) scores for both pain (38.0 [23.7]) and hand function (7.8 [5.1]) improved compared with week 12.
Most participants in both groups (>70%) reported continuing to use ergonomic measures at 6 months (eTable 3 in Supplement 2), whereas 41% of the intervention group and 64% of the comparator group reported undertaking hand exercises. In both groups, less than 25% of participants reported using the splint and diclofenac sodium gel. The main reasons for nonuse are reported in eTable 4 in Supplement 2.
Discussion
This trial showed that a combination of education on osteoarthritis and ergonomic principles, diclofenac sodium, 1%, gel, hand exercises, and splint provided small improvements in hand function but no greater improvement in pain at week 6 compared with education alone. At week 12, the intervention provided greater improvements in both outcomes with a larger effect size. The effect of the intervention was potentially clinically meaningful for function but not for pain. The intervention was superior to the comparator in several secondary outcomes. The primary outcomes remained relatively stable at 6 months in the intervention group.
It is of note that participants in the comparator group also received an intervention (education on osteoarthritis and ergonomic principles) that has been recommended to all patients in current hand osteoarthritis recommendations.5 A previous study showed that self-management strategies resulted in a greater proportion of responders at 6 months according to the OMERACT-OARSI criteria compared with no intervention, although only one-third of participants were responders and there were no differences at 3 and 12 months in that study.27
The individual treatments were selected based on evidence of their efficacy, good safety profile, and availability for clinical use. Diclofenac sodium gel provides small pain benefits in hand osteoarthritis28 and is recommended over systemic NSAIDs owing to its better safety profile.5 Similarly, a recent meta-analysis including 7 studies showed that hand exercises provided small improvements in pain and function of uncertain clinical benefits in hand osteoarthritis.29 Splinting provides moderate to large effects on pain in the medium to long term (>3 months) compared with no intervention, based on a recent meta-analysis of 4 trials.30 The effects on function are generally smaller. In contrast, when compared with a placebo splint, splinting did not provide additional clinical benefits for individuals with thumb base osteoarthritis participating in a self-management program.31
Previous studies on combined treatments in thumb base osteoarthritis have provided conflicting results. Stukstette et al32 found no difference between education on self-management and ergonomic principles, home exercises and splinting, as appropriate, and written educational material at 3 months. Another study33 compared education on ergonomic principles and hand exercises vs education only and found greater improvements in grip strength and hand function in the combined treatment group. Boustedt et al34 showed greater improvements in pain, function, grip strength, and stiffness in a group receiving a 5-week program that included education on ergonomic principles, splinting, hot packs, and home exercises compared with education alone at 1 week and 1 year postintervention. To our knowledge, the present study is the first to investigate a combination of nonpharmacologic and pharmacologic treatments in thumb base osteoarthritis.
It is of note that intervention use (all components) at 6 months was low (4%). At 6 months, most participants self-reported using ergonomic strategies (71%), while less than half continued to use each of the other treatments. However, the primary outcomes remained relatively stable compared with the week 6 visit, which may suggest that participants were able to self-manage their condition, choosing to continue the treatments that most suited them.
This study has several strengths. We followed the 2015 OARSI recommendations for hand osteoarthritis trials to ensure optimal methodological quality.8 Second, although participants were aware of the intervention received, which is nearly universal in trials including exercises, they were unaware of the treatment offered to the other group. Despite this, participants in the intervention group reported higher treatment credibility and expectancy, although the clinical significance of the differences between groups is unclear. As both measures have been shown to correlate with outcomes,19 it is possible that this difference may have contributed to greater improvements in the intervention group.
Limitations
Limitations of this trial include the relatively short follow-up duration, which we attempted to address with a 6-month contact with participants to assess treatment continuation and primary outcomes. The FIHOA questionnaire was a recommended outcome when the trial was designed8,35; however, it contains questions that may be outdated and do not appropriately assess the impact of the disease on function when the nondominant hand is the index hand. Thus, other tools with greater sensitivity to change may be recommended for future trials. It is also of note that approximately one-third of participants in each group were receiving NSAIDs at baseline, more often for pain elsewhere. They were asked to continue using the same medications and not to start any new treatment during the study. Furthermore, we asked all participants not to start new treatments other than the allocated intervention during the study, but we did not monitor cointervention use. The trial’s results are not generalizable to patients with less severe symptoms (ie, VAS pain <40). Lastly, no cost-effectiveness analysis was carried out.
Conclusions
In conclusion, the trial showed that a combination of conservative treatments for thumb base osteoarthritis conferred small to medium benefits on hand function that were potentially clinically meaningful. Effects on pain were smaller and clinically no better than education alone. These results provide clinicians with evidence for efficacy of a relatively safe conservative intervention for patients with thumb base osteoarthritis.
Trial Protocol
eFigure. Comfort Cool Thumb CMC Restriction Splint.
eAppendix. COMBO thumb base OA exercises.
eTable 1. Absence of swelling and tenderness by timepoint and stratified by baseline swelling/tenderness.
eTable 2. Work Productivity and Activity Impairment Questionnaire–General Health (WPAI-GH) stratified by group and time-point.
eTable 3. Primary outcomes and intervention use at 6 months.
eTable 4. Reason for not using the intervention as recommended by the study therapist at 6 months.
Data Sharing Statement
References
- 1.Bitton R. The economic burden of osteoarthritis. Am J Manag Care. 2009;15(8)(suppl):S230-S235. [PubMed] [Google Scholar]
- 2.Lawrence RC, Felson DT, Helmick CG, et al. ; National Arthritis Data Workgroup . Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. part II. Arthritis Rheum. 2008;58(1):26-35. doi: 10.1002/art.23176 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Wilder FV, Barrett JP, Farina EJ. Joint-specific prevalence of osteoarthritis of the hand. Osteoarthritis Cartilage. 2006;14(9):953-957. doi: 10.1016/j.joca.2006.04.013 [DOI] [PubMed] [Google Scholar]
- 4.Moriatis Wolf J, Turkiewicz A, Atroshi I, Englund M. Prevalence of doctor-diagnosed thumb carpometacarpal joint osteoarthritis: an analysis of Swedish health care. Arthritis Care Res (Hoboken). 2014;66(6):961-965. doi: 10.1002/acr.22250 [DOI] [PubMed] [Google Scholar]
- 5.Kloppenburg M, Kroon FP, Blanco FJ, et al. 2018 update of the EULAR recommendations for the management of hand osteoarthritis. Ann Rheum Dis. 2019;78(1):16-24. doi: 10.1136/annrheumdis-2018-213826 [DOI] [PubMed] [Google Scholar]
- 6.Zhang W, Doherty M, Leeb BF, et al. EULAR evidence based recommendations for the management of hand osteoarthritis: report of a Task Force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT). Ann Rheum Dis. 2007;66(3):377-388. doi: 10.1136/ard.2006.062091 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.National Institute for Health and Clinical Excellence. Osteoarthritis: Care and Management in Adults. National Institute for Health and Clinical Excellence; 2014. [PubMed] [Google Scholar]
- 8.Kloppenburg M, Maheu E, Kraus VB, et al. ; OARSI Hand Clinical Trial Recommendations Work Group . OARSI clinical trials recommendations: design and conduct of clinical trials for hand osteoarthritis. Osteoarthritis Cartilage. 2015;23(5):772-786. doi: 10.1016/j.joca.2015.03.007 [DOI] [PubMed] [Google Scholar]
- 9.Dreiser RL, Maheu E, Guillou GB, Caspard H, Grouin JM. Validation of an algofunctional index for osteoarthritis of the hand. Rev Rhum Engl Ed. 1995;62(6)(suppl 1):43S-53S. [PubMed] [Google Scholar]
- 10.Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957;16(4):494-502. doi: 10.1136/ard.16.4.494 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Deveza LA, Hunter DJ, Wajon A, et al. Efficacy of combined conservative therapies on clinical outcomes in patients with thumb base osteoarthritis: protocol for a randomised, controlled trial (COMBO). BMJ Open. 2017;7(1):e014498. doi: 10.1136/bmjopen-2016-014498 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Visser AW, Bøyesen P, Haugen IK, et al. Instruments measuring pain, physical function, or patient’s global assessment in hand osteoarthritis: a systematic literature search. J Rheumatol. 2015;42(11):2118-2134. doi: 10.3899/jrheum.141228 [DOI] [PubMed] [Google Scholar]
- 13.Busija L, Pausenberger E, Haines TP, Haymes S, Buchbinder R, Osborne RH. Adult measures of general health and health-related quality of life: Medical Outcomes Study Short Form 36-Item (SF-36) and Short Form 12-Item (SF-12) Health Surveys, Nottingham Health Profile (NHP), Sickness Impact Profile (SIP), Medical Outcomes Study Short Form 6D (SF-6D), Health Utilities Index Mark 3 (HUI3), Quality of Well-Being Scale (QWB), and Assessment of Quality of Life (AQoL). Arthritis Care Res (Hoboken). 2011;63(suppl 11):S383-S412. doi: 10.1002/acr.20541 [DOI] [PubMed] [Google Scholar]
- 14.Hawthorne G, Richardson J, Osborne R. The Assessment of Quality of Life (AQoL) instrument: a psychometric measure of health-related quality of life. Qual Life Res. 1999;8(3):209-224. doi: 10.1023/A:1008815005736 [DOI] [PubMed] [Google Scholar]
- 15.Pham T, van der Heijde D, Altman RD, et al. OMERACT-OARSI initiative: Osteoarthritis Research Society International set of responder criteria for osteoarthritis clinical trials revisited. Osteoarthritis Cartilage. 2004;12(5):389-399. doi: 10.1016/j.joca.2004.02.001 [DOI] [PubMed] [Google Scholar]
- 16.Reilly MC, Zbrozek AS, Dukes EM. The validity and reproducibility of a work productivity and activity impairment instrument. Pharmacoeconomics. 1993;4(5):353-365. doi: 10.2165/00019053-199304050-00006 [DOI] [PubMed] [Google Scholar]
- 17.Reilly MC, Gooch KL, Wong RL, Kupper H, van der Heijde D. Validity, reliability and responsiveness of the Work Productivity and Activity Impairment Questionnaire in ankylosing spondylitis. Rheumatology (Oxford). 2010;49(4):812-819. doi: 10.1093/rheumatology/kep457 [DOI] [PubMed] [Google Scholar]
- 18.Zhang W, Bansback N, Boonen A, Young A, Singh A, Anis AH. Validity of the work productivity and activity impairment questionnaire—general health version in patients with rheumatoid arthritis. Arthritis Res Ther. 2010;12(5):R177. doi: 10.1186/ar3141 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Devilly GJ, Borkovec TD. Psychometric properties of the credibility/expectancy questionnaire. J Behav Ther Exp Psychiatry. 2000;31(2):73-86. doi: 10.1016/S0005-7916(00)00012-4 [DOI] [PubMed] [Google Scholar]
- 20.Moe RH, Garratt A, Slatkowsky-Christensen B, et al. Concurrent evaluation of data quality, reliability and validity of the Australian/Canadian Osteoarthritis Hand Index and the Functional Index for Hand Osteoarthritis. Rheumatology (Oxford). 2010;49(12):2327-2336. doi: 10.1093/rheumatology/keq219 [DOI] [PubMed] [Google Scholar]
- 21.Villafañe JH, Cleland JA, Fernández-de-Las-Peñas C. The effectiveness of a manual therapy and exercise protocol in patients with thumb carpometacarpal osteoarthritis: a randomized controlled trial. J Orthop Sports Phys Ther. 2013;43(4):204-213. doi: 10.2519/jospt.2013.4524 [DOI] [PubMed] [Google Scholar]
- 22.Gupta, SS. Probability integrals of multivariate normal and multivariate t1. Ann Math Stat. 1963;34(3):792-828. http://www.jstor.org/stable/2238463 [Google Scholar]
- 23.Terwee CB, Roorda LD, Dekker J, et al. Mind the MIC: large variation among populations and methods. J Clin Epidemiol. 2010;63(5):524-534. doi: 10.1016/j.jclinepi.2009.08.010 [DOI] [PubMed] [Google Scholar]
- 24.Copay AG, Subach BR, Glassman SD, Polly DW Jr, Schuler TC. Understanding the minimum clinically important difference: a review of concepts and methods. Spine J. 2007;7(5):541-546. doi: 10.1016/j.spinee.2007.01.008 [DOI] [PubMed] [Google Scholar]
- 25.Cohen J. Statistical Power Analysis for the Behavioural Sciences. Academic; 1977. [Google Scholar]
- 26.Riordan E, Robbins S, Deveza L, et al. Radial subluxation in relation to hand strength and radiographic severity in trapeziometacarpal osteoarthritis. Osteoarthritis Cartilage. 2018;26(11):1506-1510. doi: 10.1016/j.joca.2018.06.014 [DOI] [PubMed] [Google Scholar]
- 27.Dziedzic K, Nicholls E, Hill S, et al. Self-management approaches for osteoarthritis in the hand: a 2×2 factorial randomised trial. Ann Rheum Dis. 2015;74(1):108-118. doi: 10.1136/annrheumdis-2013-203938 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Altman RD, Dreiser RL, Fisher CL, Chase WF, Dreher DS, Zacher J. Diclofenac sodium gel in patients with primary hand osteoarthritis: a randomized, double-blind, placebo-controlled trial. J Rheumatol. 2009;36(9):1991-1999. doi: 10.3899/jrheum.081316 [DOI] [PubMed] [Google Scholar]
- 29.Østerås N, Kjeken I, Smedslund G, et al. Exercise for hand osteoarthritis. Cochrane Database Syst Rev. 2017;1:CD010388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Buhler M, Chapple CM, Stebbings S, Sangelaji B, Baxter GD. Effectiveness of splinting for pain and function in people with thumb carpometacarpal osteoarthritis: a systematic review with meta-analysis. Osteoarthritis Cartilage. 2019;27(4):547-559. doi: 10.1016/j.joca.2018.09.012 [DOI] [PubMed] [Google Scholar]
- 31.Adams J, Barratt P, Arden NK, et al. O03: the OA thumb therapy trial: a three arm multi-centre randomised controlled trial of the clinical effectiveness and efficacy of splints for symptomatic thumb base OA. Rheumatology. 2020;59(suppl 2). [Google Scholar]
- 32.Stukstette MJ, Dekker J, den Broeder AA, Westeneng JM, Bijlsma JW, van den Ende CH. No evidence for the effectiveness of a multidisciplinary group based treatment program in patients with osteoarthritis of hands on the short term; results of a randomized controlled trial. Osteoarthritis Cartilage. 2013;21(7):901-910. doi: 10.1016/j.joca.2013.03.016 [DOI] [PubMed] [Google Scholar]
- 33.Stamm TA, Machold KP, Smolen JS, et al. Joint protection and home hand exercises improve hand function in patients with hand osteoarthritis: a randomized controlled trial. Arthritis Rheum. 2002;47(1):44-49. doi: 10.1002/art1.10246 [DOI] [PubMed] [Google Scholar]
- 34.Boustedt C, Nordenskiöld U, Lundgren Nilsson A. Effects of a hand-joint protection programme with an addition of splinting and exercise: one year follow-up. Clin Rheumatol. 2009;28(7):793-799. doi: 10.1007/s10067-009-1150-y [DOI] [PubMed] [Google Scholar]
- 35.Kloppenburg M, Bøyesen P, Smeets W, et al. Report from the OMERACT Hand Osteoarthritis Special Interest Group: advances and future research priorities. J Rheumatol. 2014;41(4):810-818. doi: 10.3899/jrheum.131253 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial Protocol
eFigure. Comfort Cool Thumb CMC Restriction Splint.
eAppendix. COMBO thumb base OA exercises.
eTable 1. Absence of swelling and tenderness by timepoint and stratified by baseline swelling/tenderness.
eTable 2. Work Productivity and Activity Impairment Questionnaire–General Health (WPAI-GH) stratified by group and time-point.
eTable 3. Primary outcomes and intervention use at 6 months.
eTable 4. Reason for not using the intervention as recommended by the study therapist at 6 months.
Data Sharing Statement
