-
PDF
- Split View
-
Views
-
Cite
Cite
Jackie Hill, Martyn Lewis, Howard Bird, Do OA patients gain additional benefit from care from a clinical nurse specialist?—a randomized clinical trial, Rheumatology, Volume 48, Issue 6, June 2009, Pages 658–664, https://doi.org/10.1093/rheumatology/kep049
Close -
Share
Abstract
Objectives. To assess whether OA patients attending a clinical nurse specialist (CNS) clinic gain ‘additional benefit’ compared with those attending a traditional junior hospital doctor (JHD) clinic.
Methods. A total of 100 patients with OA attending rheumatology clinics at a UK teaching hospital were randomly allocated to a CNS or JHD clinic and seen at 0, 16, 32 and 48 weeks. The study assessed (i) non-inferiority of the CNS with respect to clinical outcomes (pain, morning stiffness, self-efficacy, physical function and psychological status) and (ii) superiority of the CNS in terms of patient knowledge and satisfaction.
Results. Average pain at follow-up was lower in the CNS group: unadjusted mean difference for the JHD group minus the CNS group was 5.3 (95% CI –4.6, 15.2); adjusted was 1.6 (95% CI –5.7, 8.9). The corresponding effect size estimates were 0.20 (95% CI –0.17, 0.57) and 0.06 (95% CI –0.21, 0.33), respectively. There were similar outcomes in morning stiffness, physical function and self-efficacy. Patient knowledge and satisfaction were statistically significant at the 5% level attaining moderate to large effect sizes in favour of the CNS.
Conclusions. Our findings demonstrate that the clinical outcome of CNS care is not inferior to that of JHD care, and patients attending CNS gain additional benefit in that they are better informed about their disease and significantly more satisfied with care than are their counterparts.
Trial registration. ISRCTN 65487167.
Introduction
The prevalence of rheumatic diseases increases with age [1] and with increasing longevity, and the number of people with a rheumatic disease is expected to keep rising well into the century. The impact of rheumatic disease is pervasive. As well as having a considerable physical, psychological and social impact on the lives of the patient, their family and carers, these diseases impose a significant financial global burden. It has been estimated that at any one time 30% of American adults are affected by joint pain, swelling or movement limitation [2], and in the UK, 7 million adults from a total population of 65 million have health problems associated with arthritis and its related conditions [3]. Surveys carried out in the USA, Canada and Western Europe estimate the prevalence of physical disability associated with these diseases to be 4–5% of the adult population [4].
In the UK, the rise in the number of patients with arthritis, understaffing of medical services and reduced junior hospital doctors’ working hours has meant that while essential medical provision remains intact, it is often at the expense of the psychological, social, rehabilitative and educational needs that are so necessary to enhance patient outcome [5]. This has prompted the UK to reassess how rheumatology patient services are provided and to augment the multidisciplinary team with clinical nurse specialists (CNSs) trained to undertake extended roles. A national survey [6] established that in addition to undertaking the monitoring of drug therapy, CNSs are undertaking activities such as: The role of the CNS continues to develop, and the prescribing of drugs, administering of IA injections [7], hospital admission of patients [8] and setting up of telephone clinics [9] and advice lines [10, 11] are becoming commonplace.
examining the musculoskeletal system;
formulating and carrying out a plan of disease management;
assessing disease status;
managing symptoms;
recommending changes of drug treatment;
making referrals to other health professionals;
addressing physical, psychological and social problems; and
assessing knowledge deficits.
In the UK, this service delivery model was pioneered in the 1980s by nurses from the Academic Unit of Musculoskeletal Disease at the University of Leeds [12,13]. Although CNSs are now widely used in the management of rheumatology patients, there is a paucity of evidence of their effectiveness. To date, five randomized controlled trials have been undertaken on outcomes from nursing clinics, three from the UK [14–16] and two from the Netherlands [17, 18], and all have demonstrated some form of effectiveness. A further study from the Leiden University team concluded that care from the nursing clinics provided equivalent quality of life and utility at a lower cost [19].
To date, studies have concentrated on patients with RA and it is not known if a CNS can have a positive impact on patients with other diseases. To assess this deficiency, a study was designed to include patients with OA. The rationale for this was that OA is the most prevalent of the major rheumatic diseases. It was estimated to be the eighth leading non-fatal burden of disease in the world in 1990, and accounts for 2.8% of total years of living with disability [20]. The disease severity varies but is often progressive and at present cannot be prevented or cured. It can affect a number of different joints including the knees, hips, feet, hands and cervical and lumber spine. OA is characterized by joint pain, tenderness, limitation of movement, crepitus, occasional effusion and variable degrees of local inflammation [21]. However, pain, which for many people can be persistent, is the most common reason that patients approach their general practitioner (GP), and >50% of those with OA cite pain as their worse symptom [22]. OA of the weight-bearing joints reduces mobility and this, more than any other rheumatic disease, can impair a person's ability to climb stairs and walk [23]. For those whose hands are affected, simple everyday tasks such as dressing, preparing food and writing become difficult and painful [24]. OA can have a major impact on a patient's quality of life through its effect on work, personal lifestyle and social function.
The comparator group for the study were junior hospital doctors (JHDs), chosen to replicate, as far as possible, traditional practice in secondary care in the UK. It was postulated that whereas clinical outcomes from CNS clinics would be non-inferior to those obtained by JHD clinics, the former may confer additional benefit in relation to non-clinical outcomes such as patient knowledge and satisfaction.
The chosen primary clinical outcome measure was the level of pain. Secondary outcome comparators were function, self-efficacy and psychological health.
Non-clinical outcomes were patient knowledge and satisfaction. The hypotheses were therefore:
H 1: the pain experienced by patients with OA attending a CNS clinic over a 12-month period will be non-inferior to that of patients attending a JHD clinic.
H 2: levels of morning stiffness, self-efficacy, physical function and psychological status of patients with OA attending a CNS clinic will be non-inferior to those of patients attending a JHD clinic.
H 3: levels of knowledge and satisfaction of patients with OA attending a CNS clinic will be superior to those of patients attending a JHD clinic.
Patients and methods
Patients
The patients were a convenience sample who had been referred from primary care and were already attending a consultant rheumatology clinic in a large teaching hospital in the UK. Consecutive patients with a diagnosis of OA in any joint were enrolled and recruitment was based on written, informed consent and fulfilment of the following entry/exclusion criteria.
Inclusion criteria are:
aged ⩾18 years;
positive diagnosis of OA of any joint (defined clinically and radiologically);
able to complete questionnaires unaided; and
have attended the rheumatology clinic on at least two previous occasions thus affording them a basis upon which to assess their satisfaction with care.
Exclusion criteria are:
patients unwilling to receive care from a CNS or JHD;
patients awaiting hospital admission;
patients with concomitant inflammatory arthritis; and
patients who had received care previously from either the JHDs or CNSs involved in patient management.
Study design
The research was a single blind, parallel group study in which patients were randomly allocated to receive care from a CNS or JHD in a clinic setting. The CNS had been working in the specialty for >20 years but had not previously cared for patients with a primary diagnosis of OA. The JHDs (n = 6) had been qualified for a minimum of 2 years and were undertaking rheumatology rotational training. Both the CNS and the JHDs had open access to the consultant rheumatologist (H.B.) for either conferral or referral of patients to him. Patient consultations for both groups were pre-determined; Weeks 0, 16, 32 and 48 to enable results to be directly compared. Although it is accepted that four visits in 12 months does not reflect clinical practice throughout the UK, it was deemed necessary to provide sufficient data within a practicable study timescale. To ensure parity, all 100 patients were allocated a 30-min appointment slot. An independent assessor undertook all study assessments and in order to maintain allocation concealment, patients were requested not to reveal the identity of their practitioner.
The methods adopted in this study follow the guidelines set out by the CONSORT statement for randomized clinical trials [25]. Ethical approval was obtained from the Leeds Research Ethical Committee.
Study entry procedure
All patients who met the entry criteria were referred by a single consultant rheumatologist (H.B.) to the independent assessor who explained the study and provided them with written information. Patients were told that the study was to assess their outcome over a 12-month period and that their care would be managed by either a JHD or a CNS. To avoid bias, emphasis was placed on the assessment of progress rather than who they would consult and the use of the independent assessor was explained by her expertise in clinical assessment. After written consent was obtained, demographic data, current drug therapy and concurrent diseases were noted and patients were given questionnaires to complete. The patients were told that they would receive a further appointment by telephone. Patients were allocated to their group using a computer-generated random code that had been individually placed in sealed envelopes and consecutively numbered. A clinic clerk telephoned the patient with their appointment.
Patient management protocols
Each practitioner was designated a separate room for patient consultations in the rheumatology outpatient area of a large teaching hospital. The patients randomized to the CNS had their medications adjusted, were prescribed adaptation equipment and IA injections, and were provided with psychosocial support as deemed necessary. Patient education, which included all aspects of their disease, its treatment and self-management techniques, was also offered. Patients were also referred to any member of the multidisciplinary team as required. The JHDs managed their patients according to their clinical judgement, changing medications and referring to other members of the team as required. All interventions and consultations made by the practitioners were documented.
Clinical assessments
To test H 1:
the WOMAC [26] was considered for use but was rejected as it is not validated for all the joints affected in the patients in this study. The chosen outcome was pain, a symptom common to all patients, measured using a 100-mm visual analogue scale anchored at each end with 0 (no pain) and 100 (very severe pain).
To test H 2:
duration (in minutes) of morning joint stiffness—capped at 180 min.
Self-efficacy was measured using the ‘pain’ and ‘other symptoms’ subscales from the 11-item version of the Arthritis Self-Efficacy Scale (ASES) [27]. Items are scored on a 10-point numerical scale anchored at 10 (very uncertain) and 100 (very certain). Scores are averaged across items for each subscale with higher scores indicating greater self-efficacy.
AIMS was used to assess physical, social and emotional well-being [28]. It comprises 45 questions grouped in nine Guttman scales measuring mobility, physical activity, dexterity, household activity, social activity, activities of daily living, pain, depression and anxiety. Scales are combined to form three domains, two of which—physical function and psychological status—were used in this study. Scores range from 0 to 10; the low scores equate to high health status.
To test H 3:
patient knowledge was assessed by the validated patient knowledge questionnaire–osteoarthritis (PKQ-OA) questionnaire [29], which consists of 16 questions in four subgroups; disease process, drug therapy, exercise/rest and joint protection/complementary therapy. Scoring is 0–9 for the disease process subgroup and 0–7 for the other subgroups. A composite score, assessing ‘overall knowledge’, was calculated by summing the scores of the subscales (0 = minimum, 30 = maximum; higher scores reflect greater knowledge).
Patient satisfaction was determined by the Leeds Satisfaction Questionnaire (LSQ) [30]. It is a validated 45-question measure of patient satisfaction with care, and comprises six subscales; overall satisfaction, provision of information, empathy, technical competence, attitude to the patient and access and continuity of care. Patients complete the LSQ by 5-point Likert scales ranging from strongly agree to strongly disagree. Subscale scores are derived by averaging scores across subscale items. A composite score, assessing ‘overall satisfaction with care’, was calculated by summing the scores of the subscales (6 = minimum, 30 = maximum; higher scores reflect greater patient satisfaction).
In addition, all consultations, referrals, hospital admissions, treatment changes and non-protocol blood tests and X-rays were noted at each clinic visit.
All questionnaires were given out and collected by the blinded independent assessor. Pain and morning stiffness was assessed at baseline, 16, 32 and 48 weeks; the ASES at 0, 32 and 48 weeks and AIMS, PKQ-OA and LSQ were completed at the beginning and end of the study.
Sample size
To test H 1, the null hypothesis was that the JHD is ‘superior’ to CNS by at least an effect size deemed to be moderate [31]. This means that the effect size for the difference in average pain outcome over 48 weeks follow-up should not exceed 0.5 in favour of JHD clinics. To conclude that CNS was not inferior, the lower boundary of the 95% CI for the difference in mean pain scores for the JHD group minus the CNS group would need to be above the threshold effect size of −0.5, equivalent to a one-sided test with significance level of 2.5%. Given a repeated-measures assessment procedure for the primary outcome measure, 100 patients (50 per clinic group) were needed to test this hypothesis given 80% power, and assuming 15% loss to follow-up and intra-cluster correlation coefficient up to 0.5.
To test H 2 and H 3, the sample size was also powered at ∼95 and 70%, respectively, for assessing statistical significance at each follow-up assessment given the respective effect size margins of 0.8 (large) and 0.5 (moderate).
Statistical analysis
Crude mean values are presented with s.d. Estimates of between-group differences were obtained by statistical modelling of the data. In all cases, mean differences (with 95% CIs) were estimated for the JHD group minus the CNS group. Thus, positive mean differences in pain, stiffness and AIMS subscale scores, and negative mean differences in ASES, knowledge and satisfaction scores between the groups indicate favourable outcomes for the CNS group compared with the JHD group. For the primary outcome measure, statistical estimation was quantified using a random effects linear regression model. A similar modelling procedure was carried out to estimate mean differences in average stiffness between clinic groups during follow-up. Ordinary linear regression was used to compare between-group mean differences in all outcome measures at each follow-up time point. Statistical analyses were carried out both before and after adjustment for the covariates; age, gender, baseline pain score and baseline value of the outcome variable being assessed. Time (weeks) was included as a fixed-effects covariate within unadjusted and adjusted random effects models. Estimates were placed in context of the standardized difference [or effect size, i.e. the ratio of the mean difference in scores (JHD group minus CNS group) to the pooled s.d. of the scores in the two groups], taking into consideration the classification of Cohen [31]; small = 0.2; moderate = 0.5; large = 0.8 [25].
Results
Participants were recruited over a 12-month period and Table 1 summarizes their baseline characteristics. Socio-demographics were well balanced between the study groups but patients in the JHD group had slightly more pain and morning stiffness than those in the CNS group. The flow of participants through the trial is illustrated in Fig. 1. On study completion, there were 41/51 (CNS) and 35/49 (JHD) patients in each cohort. This was due to six CNS and eight JHD patients not attending their first appointment. Two patients from the CNS cohort and five from the JHD subsequently withdrew themselves from the study, and the CNS and JHD withdrew two and one, respectively. Analysis of the age, sex, disease duration, pain and educational level showed no major differences between those patients who remained in the study and those that were lost to follow-up.
Flow of patients through the trial. aPatients did not attend their first clinic appointment (both, after initial contact and after agreeing to subsequent dates). bPatient died or withdrawn from the study due to serious comorbidity.
Baseline characteristics of participants
| . | CNS group (n = 51) . | JHD group (n = 49) . |
|---|---|---|
| Age, mean ± s.d., years | 60.3 ± 9.4 | 61.7 ± 12.2 |
| Females, n (%) | 37 (73) | 37 (76) |
| Males, n (%) | 14 (27) | 12 (24) |
| Married/cohabiting, n (%) | 34 (67) | 30 (61) |
| Divorced/separated, n (%) | 12 (24) | 12 (24) |
| Widowed, n (%) | 5 (10) | 7 (14) |
| Years of full time education, n (%) | ||
| 9 | 5 (10) | 11 (22) |
| 10 | 35 (69) | 26 (53) |
| 11 | 11 (22) | 12 (24) |
| Duration of OA, mean ± s.d., years | 10.2 ± 8.7 | 10.0 ± 8.1 |
| Taking analgesia, n (%) | 38 (75) | 34 (69) |
| Taking NSAIDs, n (%) | 30 (59) | 26 (53) |
| Comorbidity—concomitant disease, n (%) | 39 (76) | 39 (76) |
| . | CNS group (n = 51) . | JHD group (n = 49) . |
|---|---|---|
| Age, mean ± s.d., years | 60.3 ± 9.4 | 61.7 ± 12.2 |
| Females, n (%) | 37 (73) | 37 (76) |
| Males, n (%) | 14 (27) | 12 (24) |
| Married/cohabiting, n (%) | 34 (67) | 30 (61) |
| Divorced/separated, n (%) | 12 (24) | 12 (24) |
| Widowed, n (%) | 5 (10) | 7 (14) |
| Years of full time education, n (%) | ||
| 9 | 5 (10) | 11 (22) |
| 10 | 35 (69) | 26 (53) |
| 11 | 11 (22) | 12 (24) |
| Duration of OA, mean ± s.d., years | 10.2 ± 8.7 | 10.0 ± 8.1 |
| Taking analgesia, n (%) | 38 (75) | 34 (69) |
| Taking NSAIDs, n (%) | 30 (59) | 26 (53) |
| Comorbidity—concomitant disease, n (%) | 39 (76) | 39 (76) |
Baseline characteristics of participants
| . | CNS group (n = 51) . | JHD group (n = 49) . |
|---|---|---|
| Age, mean ± s.d., years | 60.3 ± 9.4 | 61.7 ± 12.2 |
| Females, n (%) | 37 (73) | 37 (76) |
| Males, n (%) | 14 (27) | 12 (24) |
| Married/cohabiting, n (%) | 34 (67) | 30 (61) |
| Divorced/separated, n (%) | 12 (24) | 12 (24) |
| Widowed, n (%) | 5 (10) | 7 (14) |
| Years of full time education, n (%) | ||
| 9 | 5 (10) | 11 (22) |
| 10 | 35 (69) | 26 (53) |
| 11 | 11 (22) | 12 (24) |
| Duration of OA, mean ± s.d., years | 10.2 ± 8.7 | 10.0 ± 8.1 |
| Taking analgesia, n (%) | 38 (75) | 34 (69) |
| Taking NSAIDs, n (%) | 30 (59) | 26 (53) |
| Comorbidity—concomitant disease, n (%) | 39 (76) | 39 (76) |
| . | CNS group (n = 51) . | JHD group (n = 49) . |
|---|---|---|
| Age, mean ± s.d., years | 60.3 ± 9.4 | 61.7 ± 12.2 |
| Females, n (%) | 37 (73) | 37 (76) |
| Males, n (%) | 14 (27) | 12 (24) |
| Married/cohabiting, n (%) | 34 (67) | 30 (61) |
| Divorced/separated, n (%) | 12 (24) | 12 (24) |
| Widowed, n (%) | 5 (10) | 7 (14) |
| Years of full time education, n (%) | ||
| 9 | 5 (10) | 11 (22) |
| 10 | 35 (69) | 26 (53) |
| 11 | 11 (22) | 12 (24) |
| Duration of OA, mean ± s.d., years | 10.2 ± 8.7 | 10.0 ± 8.1 |
| Taking analgesia, n (%) | 38 (75) | 34 (69) |
| Taking NSAIDs, n (%) | 30 (59) | 26 (53) |
| Comorbidity—concomitant disease, n (%) | 39 (76) | 39 (76) |
The location of OA was widespread. Of those patients that attended their Week 1 visit (45 CNS, 41 JHD), the knees were the most affected joints (16 CNS, 22 JHD). This was followed by lumber/cervical spine (11 CNS, 7 JHD), hips (9 CNS, 7 JHD), hands (5 CNS, 4 JHD), ankles/feet (3 CNS, 2 JHD) and shoulders (2 CNS, 2 JHD). The CNS had one patient with affected elbows and one with generalized OA. Four patients in the CNS cohort and three in the JHDs had OA in more than one location, e.g. knees and hips.
The results of the descriptive summary of clinical outcomes between the CNS and JHD groups can be seen in Table 2. The difference in the primary outcome, ‘average’ pain at follow-up (Table 3), was 5.30 in the unadjusted analysis and 1.61 in the adjusted analysis relating to effect sizes of 0.20 and 0.06, respectively. These were small mean differences relative to the s.d. of the data and the direction favoured the CNS group (Fig. 2). Thus the null hypothesis of ‘superiority’ with respect to JHD clinics can be rejected as the CNS group was ‘non-inferior’ to the JHD group. Mean differences (effect sizes), after adjustment, were −1.38 (−0.05) at 16 weeks, 10.4 (0.41) at 32 weeks and −2.54 (−0.09) at 48 weeks; therefore, the CNS group was more likely to have a better outcome than the JHD group at 32 weeks compared with 16- and 48-week follow-up. Similar patterns of results were observed across the other outcome measures with all 95% CIs for the effect size estimates lying within the ranges corresponding to non-inferiority of the CNS group (equivalence and/or CNS superiority); differences being greatest at 32-week follow-up.
Comparison of self-reported ‘average’ pain (primary outcome) during follow-up. ‘Superiority’ and ‘equivalence’ margins are defined in accordance with the a priori definition of a significant clinical effect, i.e. a standardized difference in excess of a ‘moderate’ effect size [33].
Descriptive summary of outcomes, stratified by study group
| Outcome measure . | Follow-up . | JHD group, mean ± s.d. . | CNS group, mean ± s.d. . |
|---|---|---|---|
| Primary outcome | |||
| VAS-pain | Average | 52.6 ± 26.6 | 47.2 ± 26.6 |
| Secondary outcomes | |||
| VAS-pain | Week 0 | 59.1 ± 24.7 | 51.9 ± 24.6 |
| Week 16 | 58.1 ± 24.8 | 56.9 ± 26.1 | |
| Week 32 | 60.5 ± 25.5 | 47.0 ± 25.6 | |
| Week 48 | 54.3 ± 30.2 | 55.3 ± 27.6 | |
| M-S | Average | 51.9 ± 60.2 | 37.9 ± 58.6 |
| Week 0 | 55.5 ± 54.2 | 50.6 ± 52.0 | |
| Week 16 | 58.9 ± 59.3 | 53.1 ± 54.9 | |
| Week 32 | 69.7 ± 62.5 | 44.8 ± 54.5 | |
| Week 48 | 64.5 ± 59.3 | 61.4 ± 65.7 | |
| ASES-Pain | Week 0 | 45.1 ± 20.1 | 47.5 ± 18.8 |
| Week 32 | 49.0 ± 20.1 | 56.4 ± 17.6 | |
| Week 48 | 51.5 ± 21.7 | 56.0 ± 15.3 | |
| ASES-Other | Week 0 | 52.0 ± 19.2 | 55.8 ± 21.6 |
| Week 32 | 54.6 ± 20.1 | 62.3 ± 17.4 | |
| Week 48 | 57.1 ± 21.8 | 62.7 ± 14.7 | |
| AIMS-PF | Week 0 | 3.1 ± 1.7 | 2.9 ± 1.6 |
| Week 48 | 3.4 ± 1.7 | 2.7 ± 1.6 | |
| AIMS-Psych | Week 0 | 3.9 ± 2.1 | 3.8 ± 1.8 |
| Week 48 | 3.9 ± 2.0 | 3.7 ± 1.9 | |
| Knowledge | |||
| Disease process | Week 0 | 6.1 ± 1.3 | 6.3 ± 1.4 |
| Week 48 | 6.3 ± 1.4 | 7.0 ± 1.4 | |
| Drug therapy | Week 0 | 3.9 ± 1.5 | 3.5 ± 1.7 |
| Week 48 | 3.6 ± 1.5 | 4.3 ± 1.6 | |
| Exercise and rest | Week 0 | 5.3 ± 1.4 | 5.1 ± 1.9 |
| Week 48 | 5.4 ± 1.6 | 6.3 ± 0.9 | |
| Joint protection and complementary therapy | Week 0 | 3.6 ± 1.3 | 3.4 ± 1.5 |
| Week 48 | 3.7 ± 1.5 | 4.3 ± 1.5 | |
| Overall score | Week 0 | 18.9 ± 3.9 | 19.2 ± 5.1 |
| Week 48 | 19.0 ± 4.9 | 21.9 ± 4.3 | |
| Satisfaction | |||
| Overall satisfaction | Week 0 | 3.2 ± 0.7 | 3.4 ± 0.9 |
| Week 48 | 3.6 ± 0.8 | 4.0 ± 0.8 | |
| Provision of information | Week 0 | 3.1 ± 0.7 | 3.3 ± 0.6 |
| Week 48 | 3.4 ± 0.7 | 4.1 ± 0.6 | |
| Empathy | Week 0 | 3.2 ± 0.7 | 3.3 ± 0.7 |
| Week 48 | 3.4 ± 0.6 | 3.9 ± 0.6 | |
| Technical competence | Week 0 | 3.8 ± 0.5 | 3.9 ± 0.6 |
| Week 48 | 4.0 ± 0.6 | 4.3 ± 0.5 | |
| Attitude to the patient | Week 0 | 3.4 ± 0.6 | 3.7 ± 0.7 |
| Week 48 | 3.7 ± 0.6 | 4.1 ± 0.6 | |
| Access and continuity with care | Week 0 | 3.1 ± 0.6 | 3.2 ± 0.7 |
| Week 48 | 3.2 ± 0.7 | 3.8 ± 0.6 | |
| Overall score | Week 0 | 19.6 ± 3.3 | 20.7 ± 3.7 |
| Week 48 | 21.3 ± 3.5 | 24.2 ± 3.4 |
| Outcome measure . | Follow-up . | JHD group, mean ± s.d. . | CNS group, mean ± s.d. . |
|---|---|---|---|
| Primary outcome | |||
| VAS-pain | Average | 52.6 ± 26.6 | 47.2 ± 26.6 |
| Secondary outcomes | |||
| VAS-pain | Week 0 | 59.1 ± 24.7 | 51.9 ± 24.6 |
| Week 16 | 58.1 ± 24.8 | 56.9 ± 26.1 | |
| Week 32 | 60.5 ± 25.5 | 47.0 ± 25.6 | |
| Week 48 | 54.3 ± 30.2 | 55.3 ± 27.6 | |
| M-S | Average | 51.9 ± 60.2 | 37.9 ± 58.6 |
| Week 0 | 55.5 ± 54.2 | 50.6 ± 52.0 | |
| Week 16 | 58.9 ± 59.3 | 53.1 ± 54.9 | |
| Week 32 | 69.7 ± 62.5 | 44.8 ± 54.5 | |
| Week 48 | 64.5 ± 59.3 | 61.4 ± 65.7 | |
| ASES-Pain | Week 0 | 45.1 ± 20.1 | 47.5 ± 18.8 |
| Week 32 | 49.0 ± 20.1 | 56.4 ± 17.6 | |
| Week 48 | 51.5 ± 21.7 | 56.0 ± 15.3 | |
| ASES-Other | Week 0 | 52.0 ± 19.2 | 55.8 ± 21.6 |
| Week 32 | 54.6 ± 20.1 | 62.3 ± 17.4 | |
| Week 48 | 57.1 ± 21.8 | 62.7 ± 14.7 | |
| AIMS-PF | Week 0 | 3.1 ± 1.7 | 2.9 ± 1.6 |
| Week 48 | 3.4 ± 1.7 | 2.7 ± 1.6 | |
| AIMS-Psych | Week 0 | 3.9 ± 2.1 | 3.8 ± 1.8 |
| Week 48 | 3.9 ± 2.0 | 3.7 ± 1.9 | |
| Knowledge | |||
| Disease process | Week 0 | 6.1 ± 1.3 | 6.3 ± 1.4 |
| Week 48 | 6.3 ± 1.4 | 7.0 ± 1.4 | |
| Drug therapy | Week 0 | 3.9 ± 1.5 | 3.5 ± 1.7 |
| Week 48 | 3.6 ± 1.5 | 4.3 ± 1.6 | |
| Exercise and rest | Week 0 | 5.3 ± 1.4 | 5.1 ± 1.9 |
| Week 48 | 5.4 ± 1.6 | 6.3 ± 0.9 | |
| Joint protection and complementary therapy | Week 0 | 3.6 ± 1.3 | 3.4 ± 1.5 |
| Week 48 | 3.7 ± 1.5 | 4.3 ± 1.5 | |
| Overall score | Week 0 | 18.9 ± 3.9 | 19.2 ± 5.1 |
| Week 48 | 19.0 ± 4.9 | 21.9 ± 4.3 | |
| Satisfaction | |||
| Overall satisfaction | Week 0 | 3.2 ± 0.7 | 3.4 ± 0.9 |
| Week 48 | 3.6 ± 0.8 | 4.0 ± 0.8 | |
| Provision of information | Week 0 | 3.1 ± 0.7 | 3.3 ± 0.6 |
| Week 48 | 3.4 ± 0.7 | 4.1 ± 0.6 | |
| Empathy | Week 0 | 3.2 ± 0.7 | 3.3 ± 0.7 |
| Week 48 | 3.4 ± 0.6 | 3.9 ± 0.6 | |
| Technical competence | Week 0 | 3.8 ± 0.5 | 3.9 ± 0.6 |
| Week 48 | 4.0 ± 0.6 | 4.3 ± 0.5 | |
| Attitude to the patient | Week 0 | 3.4 ± 0.6 | 3.7 ± 0.7 |
| Week 48 | 3.7 ± 0.6 | 4.1 ± 0.6 | |
| Access and continuity with care | Week 0 | 3.1 ± 0.6 | 3.2 ± 0.7 |
| Week 48 | 3.2 ± 0.7 | 3.8 ± 0.6 | |
| Overall score | Week 0 | 19.6 ± 3.3 | 20.7 ± 3.7 |
| Week 48 | 21.3 ± 3.5 | 24.2 ± 3.4 |
VAS-pain: pain measured by visual analogue scale (0–100 scale; 0 = no pain, 100 = maximum pain); M-S: duration of morning stiffness (0–180 scale; measured in minutes); ASES-Pain: Arthritis Self-Efficacy Subscale—‘Pain’; ASES-Other: Arthritis Self-Efficacy Subscale—‘Other symptoms’ (10–100 scale; 10 = minimum self-efficacy, 100 = maximum self-efficacy); AIMS-PF: AIMS physical functioning; AIMS-Psych: AIMS psychological status (0–10 scale; 0 = highest health state, 10 = lowest health state).
Descriptive summary of outcomes, stratified by study group
| Outcome measure . | Follow-up . | JHD group, mean ± s.d. . | CNS group, mean ± s.d. . |
|---|---|---|---|
| Primary outcome | |||
| VAS-pain | Average | 52.6 ± 26.6 | 47.2 ± 26.6 |
| Secondary outcomes | |||
| VAS-pain | Week 0 | 59.1 ± 24.7 | 51.9 ± 24.6 |
| Week 16 | 58.1 ± 24.8 | 56.9 ± 26.1 | |
| Week 32 | 60.5 ± 25.5 | 47.0 ± 25.6 | |
| Week 48 | 54.3 ± 30.2 | 55.3 ± 27.6 | |
| M-S | Average | 51.9 ± 60.2 | 37.9 ± 58.6 |
| Week 0 | 55.5 ± 54.2 | 50.6 ± 52.0 | |
| Week 16 | 58.9 ± 59.3 | 53.1 ± 54.9 | |
| Week 32 | 69.7 ± 62.5 | 44.8 ± 54.5 | |
| Week 48 | 64.5 ± 59.3 | 61.4 ± 65.7 | |
| ASES-Pain | Week 0 | 45.1 ± 20.1 | 47.5 ± 18.8 |
| Week 32 | 49.0 ± 20.1 | 56.4 ± 17.6 | |
| Week 48 | 51.5 ± 21.7 | 56.0 ± 15.3 | |
| ASES-Other | Week 0 | 52.0 ± 19.2 | 55.8 ± 21.6 |
| Week 32 | 54.6 ± 20.1 | 62.3 ± 17.4 | |
| Week 48 | 57.1 ± 21.8 | 62.7 ± 14.7 | |
| AIMS-PF | Week 0 | 3.1 ± 1.7 | 2.9 ± 1.6 |
| Week 48 | 3.4 ± 1.7 | 2.7 ± 1.6 | |
| AIMS-Psych | Week 0 | 3.9 ± 2.1 | 3.8 ± 1.8 |
| Week 48 | 3.9 ± 2.0 | 3.7 ± 1.9 | |
| Knowledge | |||
| Disease process | Week 0 | 6.1 ± 1.3 | 6.3 ± 1.4 |
| Week 48 | 6.3 ± 1.4 | 7.0 ± 1.4 | |
| Drug therapy | Week 0 | 3.9 ± 1.5 | 3.5 ± 1.7 |
| Week 48 | 3.6 ± 1.5 | 4.3 ± 1.6 | |
| Exercise and rest | Week 0 | 5.3 ± 1.4 | 5.1 ± 1.9 |
| Week 48 | 5.4 ± 1.6 | 6.3 ± 0.9 | |
| Joint protection and complementary therapy | Week 0 | 3.6 ± 1.3 | 3.4 ± 1.5 |
| Week 48 | 3.7 ± 1.5 | 4.3 ± 1.5 | |
| Overall score | Week 0 | 18.9 ± 3.9 | 19.2 ± 5.1 |
| Week 48 | 19.0 ± 4.9 | 21.9 ± 4.3 | |
| Satisfaction | |||
| Overall satisfaction | Week 0 | 3.2 ± 0.7 | 3.4 ± 0.9 |
| Week 48 | 3.6 ± 0.8 | 4.0 ± 0.8 | |
| Provision of information | Week 0 | 3.1 ± 0.7 | 3.3 ± 0.6 |
| Week 48 | 3.4 ± 0.7 | 4.1 ± 0.6 | |
| Empathy | Week 0 | 3.2 ± 0.7 | 3.3 ± 0.7 |
| Week 48 | 3.4 ± 0.6 | 3.9 ± 0.6 | |
| Technical competence | Week 0 | 3.8 ± 0.5 | 3.9 ± 0.6 |
| Week 48 | 4.0 ± 0.6 | 4.3 ± 0.5 | |
| Attitude to the patient | Week 0 | 3.4 ± 0.6 | 3.7 ± 0.7 |
| Week 48 | 3.7 ± 0.6 | 4.1 ± 0.6 | |
| Access and continuity with care | Week 0 | 3.1 ± 0.6 | 3.2 ± 0.7 |
| Week 48 | 3.2 ± 0.7 | 3.8 ± 0.6 | |
| Overall score | Week 0 | 19.6 ± 3.3 | 20.7 ± 3.7 |
| Week 48 | 21.3 ± 3.5 | 24.2 ± 3.4 |
| Outcome measure . | Follow-up . | JHD group, mean ± s.d. . | CNS group, mean ± s.d. . |
|---|---|---|---|
| Primary outcome | |||
| VAS-pain | Average | 52.6 ± 26.6 | 47.2 ± 26.6 |
| Secondary outcomes | |||
| VAS-pain | Week 0 | 59.1 ± 24.7 | 51.9 ± 24.6 |
| Week 16 | 58.1 ± 24.8 | 56.9 ± 26.1 | |
| Week 32 | 60.5 ± 25.5 | 47.0 ± 25.6 | |
| Week 48 | 54.3 ± 30.2 | 55.3 ± 27.6 | |
| M-S | Average | 51.9 ± 60.2 | 37.9 ± 58.6 |
| Week 0 | 55.5 ± 54.2 | 50.6 ± 52.0 | |
| Week 16 | 58.9 ± 59.3 | 53.1 ± 54.9 | |
| Week 32 | 69.7 ± 62.5 | 44.8 ± 54.5 | |
| Week 48 | 64.5 ± 59.3 | 61.4 ± 65.7 | |
| ASES-Pain | Week 0 | 45.1 ± 20.1 | 47.5 ± 18.8 |
| Week 32 | 49.0 ± 20.1 | 56.4 ± 17.6 | |
| Week 48 | 51.5 ± 21.7 | 56.0 ± 15.3 | |
| ASES-Other | Week 0 | 52.0 ± 19.2 | 55.8 ± 21.6 |
| Week 32 | 54.6 ± 20.1 | 62.3 ± 17.4 | |
| Week 48 | 57.1 ± 21.8 | 62.7 ± 14.7 | |
| AIMS-PF | Week 0 | 3.1 ± 1.7 | 2.9 ± 1.6 |
| Week 48 | 3.4 ± 1.7 | 2.7 ± 1.6 | |
| AIMS-Psych | Week 0 | 3.9 ± 2.1 | 3.8 ± 1.8 |
| Week 48 | 3.9 ± 2.0 | 3.7 ± 1.9 | |
| Knowledge | |||
| Disease process | Week 0 | 6.1 ± 1.3 | 6.3 ± 1.4 |
| Week 48 | 6.3 ± 1.4 | 7.0 ± 1.4 | |
| Drug therapy | Week 0 | 3.9 ± 1.5 | 3.5 ± 1.7 |
| Week 48 | 3.6 ± 1.5 | 4.3 ± 1.6 | |
| Exercise and rest | Week 0 | 5.3 ± 1.4 | 5.1 ± 1.9 |
| Week 48 | 5.4 ± 1.6 | 6.3 ± 0.9 | |
| Joint protection and complementary therapy | Week 0 | 3.6 ± 1.3 | 3.4 ± 1.5 |
| Week 48 | 3.7 ± 1.5 | 4.3 ± 1.5 | |
| Overall score | Week 0 | 18.9 ± 3.9 | 19.2 ± 5.1 |
| Week 48 | 19.0 ± 4.9 | 21.9 ± 4.3 | |
| Satisfaction | |||
| Overall satisfaction | Week 0 | 3.2 ± 0.7 | 3.4 ± 0.9 |
| Week 48 | 3.6 ± 0.8 | 4.0 ± 0.8 | |
| Provision of information | Week 0 | 3.1 ± 0.7 | 3.3 ± 0.6 |
| Week 48 | 3.4 ± 0.7 | 4.1 ± 0.6 | |
| Empathy | Week 0 | 3.2 ± 0.7 | 3.3 ± 0.7 |
| Week 48 | 3.4 ± 0.6 | 3.9 ± 0.6 | |
| Technical competence | Week 0 | 3.8 ± 0.5 | 3.9 ± 0.6 |
| Week 48 | 4.0 ± 0.6 | 4.3 ± 0.5 | |
| Attitude to the patient | Week 0 | 3.4 ± 0.6 | 3.7 ± 0.7 |
| Week 48 | 3.7 ± 0.6 | 4.1 ± 0.6 | |
| Access and continuity with care | Week 0 | 3.1 ± 0.6 | 3.2 ± 0.7 |
| Week 48 | 3.2 ± 0.7 | 3.8 ± 0.6 | |
| Overall score | Week 0 | 19.6 ± 3.3 | 20.7 ± 3.7 |
| Week 48 | 21.3 ± 3.5 | 24.2 ± 3.4 |
VAS-pain: pain measured by visual analogue scale (0–100 scale; 0 = no pain, 100 = maximum pain); M-S: duration of morning stiffness (0–180 scale; measured in minutes); ASES-Pain: Arthritis Self-Efficacy Subscale—‘Pain’; ASES-Other: Arthritis Self-Efficacy Subscale—‘Other symptoms’ (10–100 scale; 10 = minimum self-efficacy, 100 = maximum self-efficacy); AIMS-PF: AIMS physical functioning; AIMS-Psych: AIMS psychological status (0–10 scale; 0 = highest health state, 10 = lowest health state).
Between-group differences in overall knowledge and satisfaction scores at 48-week follow-up are also shown in Table 3. These differences were, in the main, statistically significant at the 5% two-tailed level in both unadjusted and adjusted analyses and so the null hypothesis of no difference between study groups is rejected. Differences were negative indicating that the CNS group had a better outcome than the JHD group. The negative direction of the mean differences was consistent across subscale measures. The magnitude of the difference was generally ‘moderate’ for subscales of the PKQ-OA, and ‘moderate’ or ‘large’ for subscales of the LSQ.
Table 4 demonstrates the management strategies used in the two clinics and it can be seen that the CNS used the services of the multidisciplinary team far more than her medical colleagues. However, the most striking difference is the number of patients in the JHD cohort who additionally consulted their GP (n = 39) compared with those in the CNS cohort (n = 3).
Comparison of outcomes between study groups
| . | . | Differencea . | |||
|---|---|---|---|---|---|
| . | . | Unadjustedb . | Adjustedc . | ||
| Outcome measures . | Follow-up (weeks) . | Mean (95% CI) . | ES (95% CI)d . | Mean (95% CI) . | ES (95% CI)d . |
| Primary outcome | |||||
| VAS-pain | Average | 5.30 (−4.56, 15.2) | 0.20 (−0.17, 0.57) | 1.61 (−5.68, 8.90) | 0.06 (−0.21, 0.33) |
| Secondary outcomes | |||||
| VAS-pain | 16 | 1.20 (−10.2, 12.6) | 0.05 (−0.40, 0.49) | −1.38 (−11.3, 8.56) | −0.05 (−0.44, 0.34) |
| 32 | 13.5 (2.05, 25.0) | 0.53 (0.08, 0.98) | 10.4 (1.06, 19.8) | 0.41 (0.04, 0.77) | |
| 48 | −0.99 (−14.7, 12.7) | −0.03 (−0.51, 0.44) | −2.54 (−13.8, 8.69) | −0.09 (−0.48, 0.30) | |
| M-S | Average | 14.2 (−8.23, 36.5) | 0.24 (−0.14, 0.61) | 10.6 (−8.05, 29.3) | 0.18 (−0.14, 0.49) |
| 16 | 5.80 (−20.0, 31.6) | 0.10 (−0.35, 0.56) | 6.50 (−17.7, 30.7) | 0.11 (−0.31, 0.54) | |
| 32 | 24.9 (−1.50, 51.2) | 0.43 (−0.03, 0.88) | 19.8 (−2.55, 42.2) | 0.34 (0.04, 0.73) | |
| 48 | 3.12 (−26.8, 33.0) | 0.05 (−0.43, 0.53) | 5.62 (−21.2, 32.4) | 0.09 (−0.34, 0.52) | |
| ASES-Pain | 32 | −7.35 (−15.9, 1.15) | −0.39 (−0.85, 0.06) | −4.24 (−11.2, 2.69) | −0.23 (−0.60, 0.14) |
| 48 | −4.45 (−13.2, 4.32) | −0.24 (−0.73, 0.24) | −3.16 (−9.22, 2.91) | −0.17 (−0.51, 0.16) | |
| ASES-Other | 32 | −7.66 (−16.1, 0.82) | −0.41 (−0.86, 0.04) | −2.46 (−10.2, 5.30) | −0.13 (−0.55, 0.28) |
| 48 | −5.57 (−14.2, 3.08) | −0.31 (−0.79, 0.17) | −1.58 (−8.07, 4.91) | −0.09 (−0.45, 0.27) | |
| AIMS-PF | 48 | 0.65 (−0.12, 1.41) | 0.39 (−0.07, 0.85) | 0.43 (0.03, 0.83) | 0.26 (0.02, 0.50) |
| AIMS-Psych | 48 | 0.11 (−0.79, 1.02) | 0.06 (−0.40, 0.52) | −0.17 (−0.66, 0.32) | −0.09 (−0.34, 0.16) |
| Knowledge | |||||
| Disease process | 48 | −0.73 (−1.39, −0.07) | −0.53 (−1.00, −0.05) | −0.59 (−1.15, −0.03) | −0.42 (−0.83, −0.02) |
| Drug therapy | 48 | −0.66 (−1.40, 0.09) | −0.42 (−0.90, 0.06) | −0.74 (−1.42, −0.06) | −0.47 (−0.91, −0.04) |
| Exercise/rest | 48 | −0.91 (−1.49, −0.32) | −0.76 (−1.24, −0.27) | −0.83 (−1.38, −0.28) | −0.69 (−1.15, −0.23) |
| Joint protection/complementary therapy | 48 | −0.64 (−1.35, 0.07) | −0.43 (−0.90, 0.05) | −0.76 (−1.41, −0.10) | −0.51 (−0.94, −0.07) |
| Overall score | 48 | −2.93 (−5.10, −0.75) | −0.64 (−1.11, −0.16) | −2.98 (−4.75, −1.20) | −0.65 (−1.03, −0.26) |
| Satisfaction | |||||
| Overall satisfaction | 48 | −0.36 (−0.75, 0.02) | −0.45 (−0.94, 0.03) | −0.34 (−0.73, 0.04) | −0.43 (−0.91, 0.05) |
| Provision of information | 48 | −0.67 (−0.98, −0.35) | −1.03 (−1.51, −0.54) | −0.57 (−0.86, −0.28) | −0.88 (−1.32, −0.43) |
| Empathy | 48 | −0.56 (−0.85, −0.27) | −0.92 (−1.39, 0.44) | −0.55 (−0.81, −0.29) | −0.90 (−1.33, −0.48) |
| Technical competence | 48 | −0.31 (−0.58, −0.04) | −0.54 (−1.02, −0.07) | −0.27 (−0.55, 0.01) | −0.47 (−0.96, 0.02) |
| Attitude to the patient | 48 | −0.40 (−0.70, −0.10) | −0.65 (−1.13, −0.16) | −0.31 (−0.60, −0.03) | −0.50 (−0.97, −0.05) |
| Access and continuity with care | 48 | −0.66 (−0.98, −0.34) | −0.99 (−1.46, −0.51) | −0.65 (−0.95, −0.35) | −0.97 (−1.42, −0.52) |
| Overall score | 48 | −2.95 (−4.60, −1.31) | −0.86 (−1.33, −0.38) | −2.62 (−4.17, −1.08) | −0.76 (−1.21, −0.31) |
| . | . | Differencea . | |||
|---|---|---|---|---|---|
| . | . | Unadjustedb . | Adjustedc . | ||
| Outcome measures . | Follow-up (weeks) . | Mean (95% CI) . | ES (95% CI)d . | Mean (95% CI) . | ES (95% CI)d . |
| Primary outcome | |||||
| VAS-pain | Average | 5.30 (−4.56, 15.2) | 0.20 (−0.17, 0.57) | 1.61 (−5.68, 8.90) | 0.06 (−0.21, 0.33) |
| Secondary outcomes | |||||
| VAS-pain | 16 | 1.20 (−10.2, 12.6) | 0.05 (−0.40, 0.49) | −1.38 (−11.3, 8.56) | −0.05 (−0.44, 0.34) |
| 32 | 13.5 (2.05, 25.0) | 0.53 (0.08, 0.98) | 10.4 (1.06, 19.8) | 0.41 (0.04, 0.77) | |
| 48 | −0.99 (−14.7, 12.7) | −0.03 (−0.51, 0.44) | −2.54 (−13.8, 8.69) | −0.09 (−0.48, 0.30) | |
| M-S | Average | 14.2 (−8.23, 36.5) | 0.24 (−0.14, 0.61) | 10.6 (−8.05, 29.3) | 0.18 (−0.14, 0.49) |
| 16 | 5.80 (−20.0, 31.6) | 0.10 (−0.35, 0.56) | 6.50 (−17.7, 30.7) | 0.11 (−0.31, 0.54) | |
| 32 | 24.9 (−1.50, 51.2) | 0.43 (−0.03, 0.88) | 19.8 (−2.55, 42.2) | 0.34 (0.04, 0.73) | |
| 48 | 3.12 (−26.8, 33.0) | 0.05 (−0.43, 0.53) | 5.62 (−21.2, 32.4) | 0.09 (−0.34, 0.52) | |
| ASES-Pain | 32 | −7.35 (−15.9, 1.15) | −0.39 (−0.85, 0.06) | −4.24 (−11.2, 2.69) | −0.23 (−0.60, 0.14) |
| 48 | −4.45 (−13.2, 4.32) | −0.24 (−0.73, 0.24) | −3.16 (−9.22, 2.91) | −0.17 (−0.51, 0.16) | |
| ASES-Other | 32 | −7.66 (−16.1, 0.82) | −0.41 (−0.86, 0.04) | −2.46 (−10.2, 5.30) | −0.13 (−0.55, 0.28) |
| 48 | −5.57 (−14.2, 3.08) | −0.31 (−0.79, 0.17) | −1.58 (−8.07, 4.91) | −0.09 (−0.45, 0.27) | |
| AIMS-PF | 48 | 0.65 (−0.12, 1.41) | 0.39 (−0.07, 0.85) | 0.43 (0.03, 0.83) | 0.26 (0.02, 0.50) |
| AIMS-Psych | 48 | 0.11 (−0.79, 1.02) | 0.06 (−0.40, 0.52) | −0.17 (−0.66, 0.32) | −0.09 (−0.34, 0.16) |
| Knowledge | |||||
| Disease process | 48 | −0.73 (−1.39, −0.07) | −0.53 (−1.00, −0.05) | −0.59 (−1.15, −0.03) | −0.42 (−0.83, −0.02) |
| Drug therapy | 48 | −0.66 (−1.40, 0.09) | −0.42 (−0.90, 0.06) | −0.74 (−1.42, −0.06) | −0.47 (−0.91, −0.04) |
| Exercise/rest | 48 | −0.91 (−1.49, −0.32) | −0.76 (−1.24, −0.27) | −0.83 (−1.38, −0.28) | −0.69 (−1.15, −0.23) |
| Joint protection/complementary therapy | 48 | −0.64 (−1.35, 0.07) | −0.43 (−0.90, 0.05) | −0.76 (−1.41, −0.10) | −0.51 (−0.94, −0.07) |
| Overall score | 48 | −2.93 (−5.10, −0.75) | −0.64 (−1.11, −0.16) | −2.98 (−4.75, −1.20) | −0.65 (−1.03, −0.26) |
| Satisfaction | |||||
| Overall satisfaction | 48 | −0.36 (−0.75, 0.02) | −0.45 (−0.94, 0.03) | −0.34 (−0.73, 0.04) | −0.43 (−0.91, 0.05) |
| Provision of information | 48 | −0.67 (−0.98, −0.35) | −1.03 (−1.51, −0.54) | −0.57 (−0.86, −0.28) | −0.88 (−1.32, −0.43) |
| Empathy | 48 | −0.56 (−0.85, −0.27) | −0.92 (−1.39, 0.44) | −0.55 (−0.81, −0.29) | −0.90 (−1.33, −0.48) |
| Technical competence | 48 | −0.31 (−0.58, −0.04) | −0.54 (−1.02, −0.07) | −0.27 (−0.55, 0.01) | −0.47 (−0.96, 0.02) |
| Attitude to the patient | 48 | −0.40 (−0.70, −0.10) | −0.65 (−1.13, −0.16) | −0.31 (−0.60, −0.03) | −0.50 (−0.97, −0.05) |
| Access and continuity with care | 48 | −0.66 (−0.98, −0.34) | −0.99 (−1.46, −0.51) | −0.65 (−0.95, −0.35) | −0.97 (−1.42, −0.52) |
| Overall score | 48 | −2.95 (−4.60, −1.31) | −0.86 (−1.33, −0.38) | −2.62 (−4.17, −1.08) | −0.76 (−1.21, −0.31) |
VAS-pain: pain measured by VAS (0–100 scale; 0 = no pain, 100 = maximum pain); M-S: duration of morning stiffness (0–180 scale; measured in minutes); ASES-Pain: Arthritis Self-Efficacy Subscale—‘Pain’; ASES-Other: Arthritis Self-Efficacy Subscale—‘Other symptoms’ (10–100 scale; 10 = minimum self-efficacy, 100 = maximum self-efficacy); AIMS-PF: AIMS physical functioning subscale; AIMS-Psych: AIMS psychological status subscale (0–10 scale; 0 = highest health state, 10 = lowest health state). aDifference is presented as values for the JHD group minus the CNS group. bUnadjusted (crude) analysis. cAdjusted analysis—controlled for the following covariates: age; gender; baseline pain score; and corresponding baseline value for the outcome of interest (if applicable). dES: effect size (or standardized difference), which is the ratio of the mean difference between groups to the pooled s.d. of the outcome variable (0.2 = ‘small’; 0.5 = ‘moderate’; and 0.8 = ‘large’) [33].
Comparison of outcomes between study groups
| . | . | Differencea . | |||
|---|---|---|---|---|---|
| . | . | Unadjustedb . | Adjustedc . | ||
| Outcome measures . | Follow-up (weeks) . | Mean (95% CI) . | ES (95% CI)d . | Mean (95% CI) . | ES (95% CI)d . |
| Primary outcome | |||||
| VAS-pain | Average | 5.30 (−4.56, 15.2) | 0.20 (−0.17, 0.57) | 1.61 (−5.68, 8.90) | 0.06 (−0.21, 0.33) |
| Secondary outcomes | |||||
| VAS-pain | 16 | 1.20 (−10.2, 12.6) | 0.05 (−0.40, 0.49) | −1.38 (−11.3, 8.56) | −0.05 (−0.44, 0.34) |
| 32 | 13.5 (2.05, 25.0) | 0.53 (0.08, 0.98) | 10.4 (1.06, 19.8) | 0.41 (0.04, 0.77) | |
| 48 | −0.99 (−14.7, 12.7) | −0.03 (−0.51, 0.44) | −2.54 (−13.8, 8.69) | −0.09 (−0.48, 0.30) | |
| M-S | Average | 14.2 (−8.23, 36.5) | 0.24 (−0.14, 0.61) | 10.6 (−8.05, 29.3) | 0.18 (−0.14, 0.49) |
| 16 | 5.80 (−20.0, 31.6) | 0.10 (−0.35, 0.56) | 6.50 (−17.7, 30.7) | 0.11 (−0.31, 0.54) | |
| 32 | 24.9 (−1.50, 51.2) | 0.43 (−0.03, 0.88) | 19.8 (−2.55, 42.2) | 0.34 (0.04, 0.73) | |
| 48 | 3.12 (−26.8, 33.0) | 0.05 (−0.43, 0.53) | 5.62 (−21.2, 32.4) | 0.09 (−0.34, 0.52) | |
| ASES-Pain | 32 | −7.35 (−15.9, 1.15) | −0.39 (−0.85, 0.06) | −4.24 (−11.2, 2.69) | −0.23 (−0.60, 0.14) |
| 48 | −4.45 (−13.2, 4.32) | −0.24 (−0.73, 0.24) | −3.16 (−9.22, 2.91) | −0.17 (−0.51, 0.16) | |
| ASES-Other | 32 | −7.66 (−16.1, 0.82) | −0.41 (−0.86, 0.04) | −2.46 (−10.2, 5.30) | −0.13 (−0.55, 0.28) |
| 48 | −5.57 (−14.2, 3.08) | −0.31 (−0.79, 0.17) | −1.58 (−8.07, 4.91) | −0.09 (−0.45, 0.27) | |
| AIMS-PF | 48 | 0.65 (−0.12, 1.41) | 0.39 (−0.07, 0.85) | 0.43 (0.03, 0.83) | 0.26 (0.02, 0.50) |
| AIMS-Psych | 48 | 0.11 (−0.79, 1.02) | 0.06 (−0.40, 0.52) | −0.17 (−0.66, 0.32) | −0.09 (−0.34, 0.16) |
| Knowledge | |||||
| Disease process | 48 | −0.73 (−1.39, −0.07) | −0.53 (−1.00, −0.05) | −0.59 (−1.15, −0.03) | −0.42 (−0.83, −0.02) |
| Drug therapy | 48 | −0.66 (−1.40, 0.09) | −0.42 (−0.90, 0.06) | −0.74 (−1.42, −0.06) | −0.47 (−0.91, −0.04) |
| Exercise/rest | 48 | −0.91 (−1.49, −0.32) | −0.76 (−1.24, −0.27) | −0.83 (−1.38, −0.28) | −0.69 (−1.15, −0.23) |
| Joint protection/complementary therapy | 48 | −0.64 (−1.35, 0.07) | −0.43 (−0.90, 0.05) | −0.76 (−1.41, −0.10) | −0.51 (−0.94, −0.07) |
| Overall score | 48 | −2.93 (−5.10, −0.75) | −0.64 (−1.11, −0.16) | −2.98 (−4.75, −1.20) | −0.65 (−1.03, −0.26) |
| Satisfaction | |||||
| Overall satisfaction | 48 | −0.36 (−0.75, 0.02) | −0.45 (−0.94, 0.03) | −0.34 (−0.73, 0.04) | −0.43 (−0.91, 0.05) |
| Provision of information | 48 | −0.67 (−0.98, −0.35) | −1.03 (−1.51, −0.54) | −0.57 (−0.86, −0.28) | −0.88 (−1.32, −0.43) |
| Empathy | 48 | −0.56 (−0.85, −0.27) | −0.92 (−1.39, 0.44) | −0.55 (−0.81, −0.29) | −0.90 (−1.33, −0.48) |
| Technical competence | 48 | −0.31 (−0.58, −0.04) | −0.54 (−1.02, −0.07) | −0.27 (−0.55, 0.01) | −0.47 (−0.96, 0.02) |
| Attitude to the patient | 48 | −0.40 (−0.70, −0.10) | −0.65 (−1.13, −0.16) | −0.31 (−0.60, −0.03) | −0.50 (−0.97, −0.05) |
| Access and continuity with care | 48 | −0.66 (−0.98, −0.34) | −0.99 (−1.46, −0.51) | −0.65 (−0.95, −0.35) | −0.97 (−1.42, −0.52) |
| Overall score | 48 | −2.95 (−4.60, −1.31) | −0.86 (−1.33, −0.38) | −2.62 (−4.17, −1.08) | −0.76 (−1.21, −0.31) |
| . | . | Differencea . | |||
|---|---|---|---|---|---|
| . | . | Unadjustedb . | Adjustedc . | ||
| Outcome measures . | Follow-up (weeks) . | Mean (95% CI) . | ES (95% CI)d . | Mean (95% CI) . | ES (95% CI)d . |
| Primary outcome | |||||
| VAS-pain | Average | 5.30 (−4.56, 15.2) | 0.20 (−0.17, 0.57) | 1.61 (−5.68, 8.90) | 0.06 (−0.21, 0.33) |
| Secondary outcomes | |||||
| VAS-pain | 16 | 1.20 (−10.2, 12.6) | 0.05 (−0.40, 0.49) | −1.38 (−11.3, 8.56) | −0.05 (−0.44, 0.34) |
| 32 | 13.5 (2.05, 25.0) | 0.53 (0.08, 0.98) | 10.4 (1.06, 19.8) | 0.41 (0.04, 0.77) | |
| 48 | −0.99 (−14.7, 12.7) | −0.03 (−0.51, 0.44) | −2.54 (−13.8, 8.69) | −0.09 (−0.48, 0.30) | |
| M-S | Average | 14.2 (−8.23, 36.5) | 0.24 (−0.14, 0.61) | 10.6 (−8.05, 29.3) | 0.18 (−0.14, 0.49) |
| 16 | 5.80 (−20.0, 31.6) | 0.10 (−0.35, 0.56) | 6.50 (−17.7, 30.7) | 0.11 (−0.31, 0.54) | |
| 32 | 24.9 (−1.50, 51.2) | 0.43 (−0.03, 0.88) | 19.8 (−2.55, 42.2) | 0.34 (0.04, 0.73) | |
| 48 | 3.12 (−26.8, 33.0) | 0.05 (−0.43, 0.53) | 5.62 (−21.2, 32.4) | 0.09 (−0.34, 0.52) | |
| ASES-Pain | 32 | −7.35 (−15.9, 1.15) | −0.39 (−0.85, 0.06) | −4.24 (−11.2, 2.69) | −0.23 (−0.60, 0.14) |
| 48 | −4.45 (−13.2, 4.32) | −0.24 (−0.73, 0.24) | −3.16 (−9.22, 2.91) | −0.17 (−0.51, 0.16) | |
| ASES-Other | 32 | −7.66 (−16.1, 0.82) | −0.41 (−0.86, 0.04) | −2.46 (−10.2, 5.30) | −0.13 (−0.55, 0.28) |
| 48 | −5.57 (−14.2, 3.08) | −0.31 (−0.79, 0.17) | −1.58 (−8.07, 4.91) | −0.09 (−0.45, 0.27) | |
| AIMS-PF | 48 | 0.65 (−0.12, 1.41) | 0.39 (−0.07, 0.85) | 0.43 (0.03, 0.83) | 0.26 (0.02, 0.50) |
| AIMS-Psych | 48 | 0.11 (−0.79, 1.02) | 0.06 (−0.40, 0.52) | −0.17 (−0.66, 0.32) | −0.09 (−0.34, 0.16) |
| Knowledge | |||||
| Disease process | 48 | −0.73 (−1.39, −0.07) | −0.53 (−1.00, −0.05) | −0.59 (−1.15, −0.03) | −0.42 (−0.83, −0.02) |
| Drug therapy | 48 | −0.66 (−1.40, 0.09) | −0.42 (−0.90, 0.06) | −0.74 (−1.42, −0.06) | −0.47 (−0.91, −0.04) |
| Exercise/rest | 48 | −0.91 (−1.49, −0.32) | −0.76 (−1.24, −0.27) | −0.83 (−1.38, −0.28) | −0.69 (−1.15, −0.23) |
| Joint protection/complementary therapy | 48 | −0.64 (−1.35, 0.07) | −0.43 (−0.90, 0.05) | −0.76 (−1.41, −0.10) | −0.51 (−0.94, −0.07) |
| Overall score | 48 | −2.93 (−5.10, −0.75) | −0.64 (−1.11, −0.16) | −2.98 (−4.75, −1.20) | −0.65 (−1.03, −0.26) |
| Satisfaction | |||||
| Overall satisfaction | 48 | −0.36 (−0.75, 0.02) | −0.45 (−0.94, 0.03) | −0.34 (−0.73, 0.04) | −0.43 (−0.91, 0.05) |
| Provision of information | 48 | −0.67 (−0.98, −0.35) | −1.03 (−1.51, −0.54) | −0.57 (−0.86, −0.28) | −0.88 (−1.32, −0.43) |
| Empathy | 48 | −0.56 (−0.85, −0.27) | −0.92 (−1.39, 0.44) | −0.55 (−0.81, −0.29) | −0.90 (−1.33, −0.48) |
| Technical competence | 48 | −0.31 (−0.58, −0.04) | −0.54 (−1.02, −0.07) | −0.27 (−0.55, 0.01) | −0.47 (−0.96, 0.02) |
| Attitude to the patient | 48 | −0.40 (−0.70, −0.10) | −0.65 (−1.13, −0.16) | −0.31 (−0.60, −0.03) | −0.50 (−0.97, −0.05) |
| Access and continuity with care | 48 | −0.66 (−0.98, −0.34) | −0.99 (−1.46, −0.51) | −0.65 (−0.95, −0.35) | −0.97 (−1.42, −0.52) |
| Overall score | 48 | −2.95 (−4.60, −1.31) | −0.86 (−1.33, −0.38) | −2.62 (−4.17, −1.08) | −0.76 (−1.21, −0.31) |
VAS-pain: pain measured by VAS (0–100 scale; 0 = no pain, 100 = maximum pain); M-S: duration of morning stiffness (0–180 scale; measured in minutes); ASES-Pain: Arthritis Self-Efficacy Subscale—‘Pain’; ASES-Other: Arthritis Self-Efficacy Subscale—‘Other symptoms’ (10–100 scale; 10 = minimum self-efficacy, 100 = maximum self-efficacy); AIMS-PF: AIMS physical functioning subscale; AIMS-Psych: AIMS psychological status subscale (0–10 scale; 0 = highest health state, 10 = lowest health state). aDifference is presented as values for the JHD group minus the CNS group. bUnadjusted (crude) analysis. cAdjusted analysis—controlled for the following covariates: age; gender; baseline pain score; and corresponding baseline value for the outcome of interest (if applicable). dES: effect size (or standardized difference), which is the ratio of the mean difference between groups to the pooled s.d. of the outcome variable (0.2 = ‘small’; 0.5 = ‘moderate’; and 0.8 = ‘large’) [33].
Referrals, conferrals and changes to treatments
| Referrals . | JHD cohort (n = 153 clinic visits) . | CNS cohort (n = 172 clinic visits) . |
|---|---|---|
| Patient consultation with CR | 25 | 10 |
| Conferral with CR | 8 | 12 |
| Physiotherapist | 19 | 22 |
| Occupational therapist | 0 | 7 |
| Appliance officer | 2 | 9 |
| Medical social worker | 0 | 1 |
| Dietician | 0 | 6 |
| Orthopaedic surgeon | 5 | 6 |
| Podiatrist | 1 | 6 |
| Diabetic clinic | 0 | 1 |
| Admission to rheumatology ward | 8 | 11 |
| Patients visits to their GP | 39 | 3 |
| Treatment changes | ||
| IA steroid injections | 30 | 20 |
| Drug therapy | 30 | 33 |
| Miscellaneous | ||
| Non-protocol laboratory tests | 15 | 9 |
| Non-protocol X-rays | 15 | 2 |
| Referrals . | JHD cohort (n = 153 clinic visits) . | CNS cohort (n = 172 clinic visits) . |
|---|---|---|
| Patient consultation with CR | 25 | 10 |
| Conferral with CR | 8 | 12 |
| Physiotherapist | 19 | 22 |
| Occupational therapist | 0 | 7 |
| Appliance officer | 2 | 9 |
| Medical social worker | 0 | 1 |
| Dietician | 0 | 6 |
| Orthopaedic surgeon | 5 | 6 |
| Podiatrist | 1 | 6 |
| Diabetic clinic | 0 | 1 |
| Admission to rheumatology ward | 8 | 11 |
| Patients visits to their GP | 39 | 3 |
| Treatment changes | ||
| IA steroid injections | 30 | 20 |
| Drug therapy | 30 | 33 |
| Miscellaneous | ||
| Non-protocol laboratory tests | 15 | 9 |
| Non-protocol X-rays | 15 | 2 |
CR: consultant rheumatologist.
Referrals, conferrals and changes to treatments
| Referrals . | JHD cohort (n = 153 clinic visits) . | CNS cohort (n = 172 clinic visits) . |
|---|---|---|
| Patient consultation with CR | 25 | 10 |
| Conferral with CR | 8 | 12 |
| Physiotherapist | 19 | 22 |
| Occupational therapist | 0 | 7 |
| Appliance officer | 2 | 9 |
| Medical social worker | 0 | 1 |
| Dietician | 0 | 6 |
| Orthopaedic surgeon | 5 | 6 |
| Podiatrist | 1 | 6 |
| Diabetic clinic | 0 | 1 |
| Admission to rheumatology ward | 8 | 11 |
| Patients visits to their GP | 39 | 3 |
| Treatment changes | ||
| IA steroid injections | 30 | 20 |
| Drug therapy | 30 | 33 |
| Miscellaneous | ||
| Non-protocol laboratory tests | 15 | 9 |
| Non-protocol X-rays | 15 | 2 |
| Referrals . | JHD cohort (n = 153 clinic visits) . | CNS cohort (n = 172 clinic visits) . |
|---|---|---|
| Patient consultation with CR | 25 | 10 |
| Conferral with CR | 8 | 12 |
| Physiotherapist | 19 | 22 |
| Occupational therapist | 0 | 7 |
| Appliance officer | 2 | 9 |
| Medical social worker | 0 | 1 |
| Dietician | 0 | 6 |
| Orthopaedic surgeon | 5 | 6 |
| Podiatrist | 1 | 6 |
| Diabetic clinic | 0 | 1 |
| Admission to rheumatology ward | 8 | 11 |
| Patients visits to their GP | 39 | 3 |
| Treatment changes | ||
| IA steroid injections | 30 | 20 |
| Drug therapy | 30 | 33 |
| Miscellaneous | ||
| Non-protocol laboratory tests | 15 | 9 |
| Non-protocol X-rays | 15 | 2 |
CR: consultant rheumatologist.
Discussion
A number of studies have demonstrated that outcomes from CNS clinics both in the UK and Europe bestow beneficial effects for patients with RA. However, there is a paucity of data as to their value in other rheumatic diseases such as OA. The findings from this study suggest that management by a CNS is non-inferior to that of JHDs with respect to clinical outcomes of OA to the magnitude of the a priori defined margin of clinical significance, which represents a ‘moderate’ effect size. Moreover, clinical outcomes were slightly better in the CNS group than the JHD group, particularly at the 32-week follow-up. The reduction in pain levels in the CNS group compared with the JHD group are of particular note as this coincides with patients’ priorities [22]. This reduction in pain may also have contributed to the greater increases in physical function obtained by the patients in the CNS cohort.
As in our earlier RA studies in secondary care, this research demonstrated that the CNS was significantly better than her medical counterparts at helping patients to improve their knowledge of their condition and increase their self-efficacy. This aspect is particularly important as it has been emphasized as an important core treatment in the recent NICE guidelines for the management of OA in adults [32] and in the Darzi review [33]. CNSs themselves have also cited patient education and the provision of information as one of the most important aspects of their role [6] as information empowers patients and enables them to make choices and take control of their illness [34]. Patients then become partners rather than recipients of care, which is the underpinning philosophy of many nurse-led clinics. However, a self-management programme ‘Challenging Arthritis’ used by Arthritis Care did show reductions in anxiety and self-efficacy [35]. It is interesting to note that the results of nurse-led self-management programme studies for OA in primary care in the UK [36] and the Netherlands [37] have not demonstrated similar positive effects. It is not clear why this should be, but like the Challenging Arthritis programme, the patient education (PE) provided by the CNS in our study was based on self-efficacy and it is not clear whether the two studies demonstrating negative results [36, 37] used this approach. The authors of the study from the Netherlands [37] suggest that one reason for lack of efficacy was the short follow-up; just a three-part programme. However, the primary-based UK study [36] used a home visit plus four 1-h teaching sessions over a 12-month period. This duration is a directly comparable with our study but the latter had approximately double the teaching/consultation time and so this is unlikely to be a reason for the different results between the studies.
The recent Darzi review [33] has placed great emphasis on patient satisfaction. However, the importance of satisfaction with care has been recognized for a number of years in rheumatology, as it can influence whether a person seeks medical advice, adheres to treatment regimens and maintains a continuing relationship with their practitioner [38]. This is particularly important with a chronic disease such as OA as it is likely to last their lifetime [39]. In this study, patients attending CNS clinics were significantly more likely than patients attending JHD clinics to be satisfied with the care they receive, particularly in terms of ‘provision of information’, ‘empathy’ and ‘access and continuity with care’. Again, this replicates the majority of results from RA research [14, 15,39–41] but not all [17].
The results of the differences in management styles and patient referrals/self-referrals are also of interest, for instance the greater patient dependency on the GP in those attending the JHD clinic. One reason for this disparity could be that the JHDs were detecting more comorbidity than the CNSs and suggesting that patients contacted their GPs. However, this was not the case in this study as the practitioners were required to note all referrals or new diagnoses on the study checklist and scrutiny of these did not support this proposal. It can, however, be postulated that as both groups were well matched in most respects, patients in the CNS clinic felt more able to manage their condition than their counterparts. The fact that the CNS patients had enhanced self-efficacy provides some credence to this notion. Other differences were also revealed. For instance, the JHD requested more non-protocol blood tests, X-rays and IA injections and the CNS referred more patients to different members of the multidisciplinary team. The JHD also asked the consultant rheumatologist to see more than twice the number of patients than the CNS. As these JHDs are still being trained to provide a high-quality clinical rheumatology service, it is expected that they require support from a senior colleague and it is to be applauded that they recognize their need to consult. Overall, given that CNSs and JHDs are from different professions and have different levels of training and expertise all these differences should perhaps be expected.
One unexpected occurrence was the relatively low study completion rate, 14 patients did not attend for their first appointment even though they had been offered and accepted a number of alternatives. It is not possible to say why this occurred as patients were assured that they might withdraw at any stage without providing a reason, but it may be surmised that the patients did not wish to say they did not want to enter the study and felt awkward doing so. Once the patients had experienced care in the clinics they tended to adhere to their appointments with just two self-withdrawals from the CNS cohort and five from the JHD. For analytical purposes, the overall response to follow-up was 78.3% (based on completion of 235 VAS-pain entries out of a possible 300 across all patients over three follow-up time points), with response being lowest at 48 weeks. Thus, dropout exceeded the 15% forecasted, adversely affecting the power of the individual hypothesis tests. The post hoc power for the primary analysis of ‘average’ pain was 74% for the unadjusted analysis and 85% for the adjusted analysis (given observed intraclass correlation coefficients of 0.56 and 0.31, respectively). However, the study size was large enough to reject the null hypothesis (at the designated threshold effect size of −0.5) and provide statistical evidence to suggest that the CNS group is ‘non-inferior’ to the JHD group. Other concerns with dropout is that of attrition bias and the possible impact on the internal validity of the findings (given that response rate was imbalanced in favour of a higher overall return in the CNS group at 82% than the JHD group at 75%), and external validity through ‘selective’ analysis of available data (as opposed to all data across all recruited patients). The authors accept that these biases may impact on the generalizability of the study findings.
The implications from this study are that many patients with OA could be managed by CNSs without the loss of efficacy in terms of clinical outcomes. Indeed, patients are likely to benefit as these results show that patients increase their self-efficacy, knowledge of their condition and satisfaction with their care. At present the majority of CNSs are based in secondary care, but it would be interesting to see if the results could be replicated in a primary setting, where the vast majority of patients are and would like to be seen.
As with most research, this study has limitations. For instance, replication of the study may be difficult as four 30-min appointments in a 12-month period would be difficult to sustain in most rheumatology departments. Participants and practitioners were not blind to clinic groups and this could influence the results. The study group was broad and included patients with OA of any joint and level of severity and the study was not powered to investigate differences in specific OA patient subgroups. The CNS group was cared for by one highly experienced nurse (although she had not previously managed a cohort of OA patients) and outcomes were compared from less experienced junior doctors, but this reflects clinical practice as CNSs tend to stay in the post for many years. It is interesting to speculate whether the results would have been different if the comparator group comprised experienced rheumatologists. However, results from our previous research in RA suggest that they would not [14]. Finally, the study was undertaken in just one location and so it is not clear that the results would be replicated across the country, particularly when it is taken in account that the majority of OA patients are managed in primary rather than secondary care, the latter only being accessed by patients with complicated OA.
Given these caveats, the study also has a number of advantages. The design and methods were appropriate, well-structured and coherent with internationally set guidelines for clinical trials methods (CONSORT statement) and it was adequately powered to answer the questions it set out to investigate. Moreover, the study results were consistent across similar outcome measures and time points, and the results were similarly robust to the method of analysis; unadjusted or adjusted.
Conclusion
All three of the hypotheses formulated for this study can be accepted. The clinical outcome for patients with OA attending a CNS clinic is non-inferior to that obtained in the JHD clinic. Patients attending a CNS clinic gain additional benefit in the form of improved knowledge of their condition and greater satisfaction with the care provided. Thus, the CNS may be viewed as bestowing additional benefit for patients with OA when compared with the JHD in secondary care. It remains to be seen whether the CNS could repeat these results in a primary care setting.

Acknowledgements
Dr Jackie Hill designed the study, wrote the protocol and acted as the CNS in the study. Professor Howard Bird critiqued the study design and protocol, provided the patients and acted as the CR. He also inputted into the manuscript. Dr Martyn Lewis undertook the statistical analysis and inputted into the manuscript. The authors would like to thank Ms Ruth Thorpe for acting as the blind assessor in this study.
Funding: This work was supported by an ARC project grant.
Disclosure statement: The authors have declared no conflicts of interest.


![Comparison of self-reported ‘average’ pain (primary outcome) during follow-up. ‘Superiority’ and ‘equivalence’ margins are defined in accordance with the a priori definition of a significant clinical effect, i.e. a standardized difference in excess of a ‘moderate’ effect size [33].](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/rheumatology/48/6/10.1093/rheumatology/kep049/2/m_kep049f2.jpeg?Expires=1663968718&Signature=JIbT7ZB4TPaFVuhZsdTu7p4lq4QI2g2jFFAcMCXffhxLRslZI-hXJvC4GRmYM9TxWZ~c94yzpjh9RB1U7qYm-ApnAYfhlNvkyk3~7GSlQT~aiOXKVHcwIY~YDhtuvUwRR9l~LxjsQQ-GAMDDBOH0VVzc~WLU-p2Z9Yn-aFslwYuhqqAu6DEDTqQrpxyObvEMPAO5zBizYFRA6P7HKzJUnaqvr04CnclcRisyMVSt9v2pgk3~kjnZtDDT4YwwvvSRb~Gv3wxnppdRLcSv84BF-Ig1QGW1GbBlO8oSUqFBpFcFbXmXyNMYA6qfpsN~62ABkds38Wxqzo~8WzK~qMwmVQ__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Comments