Strengthening And stretching for Rheumatoid Arthritis of the Hand (SARAH). A randomised controlled trial and economic evaluation.

Epidemiology of rheumatoid arthritis

Rheumatoid arthritis (RA) is the most common inflammatory polyarthritis. 1 It is a chronic unpredictable disorder that can cause persistent joint pain, joint damage and long-term disability (especially in the hands and feet). The prevalence of RA is 1.16% in women and 0.44% in men, increasing with age to 5% in those aged over 55 years. 1 It affects approximately 1% of the UK adult population. Five years after diagnosis, 40% of people with RA have relatively normal function (13% in remission), 44% have mild to moderate disability and 16% have marked functional disability. 2 The cause of the disease is unknown but both environmental and genetic factors are believed to contribute.

Rheumatoid arthritis is a whole-body disorder with greater mortality and multisystemic effects. The condition usually starts in the small joints of the hands and feet and later spreads to involve the larger joints. T-cells infiltrate the synovium, resulting in hypertrophy and inflammation of the local area and supporting ligaments. Deformities arise as a result of joint cartilage being eroded, which can then extend into the bone cortex. There are common forms of deformities at the wrist (dorsal ulnar head subluxation),3 metacarpophalangeal (MCP) joints (volar subluxation of proximal phalanges and ulnar drift of fingers), fingers (swan-neck and boutonnière deformities) and thumb [instability at the MCP and interphalangeal (IP) joints]. Tendon rupture can occur as a result of weakening by synovial invasion or abrasion over an irregular bony prominence.

Alongside this process of inflammation and deformity, other common associated problems for the hands and wrists are pain, weakness and restricted mobility resulting in loss of function and social participation. 4–6 RA patients report hand function to be important in their daily lives,5 with at least 70% of patients reporting hand and wrist dysfunction. 7

Pharmacological management of rheumatoid arthritis

Although there are increasingly effective drug treatments,8^,9 the condition has no known cure. Thus, the goals of management are to prevent or control joint damage, loss of function and decrease pain. 10 In order to achieve these goals, combinations of pharmacological, non-pharmacological and surgical treatments are used.

The chief categories of drugs used are analgesics, non-steroidal anti-inflammatory drugs (NSAIDs), disease-modifying anti-rheumatic drugs (DMARDs) (non-biological and biological) and corticosteroids. NSAIDs offer a purely symptomatic treatment, commonly used when the disease flares up. There are a multitude of DMARDs licensed for use and it is now agreed that these should be used early on in the disease process for improved control. 11 They are commonly used in combinations to achieve greater benefit but with no extra harm. Corticosteroids offer a fast-acting solution and are frequently used between changes in DMARD regimes. They can be used in oral, intravenous (IV) and intra-articular forms, depending on how widespread the scope of the problem. The side effects limit the long-term use of these drugs. Biological DMARDs, or biologics, are a relatively recent development in the management of RA. Owing to their greater financial cost, their use is restricted by National Institute for Health and Care Excellence (NICE) guidelines to patients whose disease is not controlled by conventional DMARDs. 12 The main classes of biologics in common use at present are tumour necrosis factor (TNF) inhibitors or ‘anti-TNF’ (including infliximab, etanercept, adalimumab and certolizumab pegol) and rituximab, which works by depleting B lymphocytes. Newer biologics such as abatacept and tocilizumab are used where anti-TNF and rituximab do not work or cause adverse effects.

Non-pharmacological management of rheumatoid arthritis

Current UK clinical guidelines for the management of RA recommend the use of physiotherapy and occupational therapy as adjuncts to drug treatment. 13,14 The three most common components of physiotherapy/occupational therapy for RA hands are exercise therapy, joint protection (JP) advice and provision of functional splinting and assistive devices,15 although we are not aware of any research describing current clinical practice in the NHS.

The use of exercise treatment for RA primarily aims to increase strength, stability and range of motion by tackling rheumatoid cachexia (loss of body cell mass and muscle architecture), disuse atrophy and joint/soft tissue restriction. 16 Additional benefits may include reducing pain and increasing sensory–motor function. 17 A systematic review18 of six randomised controlled trials (RCTs) of the effectiveness of dynamic general exercise programmes in RA concluded that dynamic exercise was effective in improving muscular endurance and strength, without having detrimental effects on disease activity or pain. The number of RCTs that have specifically investigated the effects of exercise on RA hands and wrists is limited to five small studies with short-term follow-up limited to a few months (n = 44, 50, 52, 44 and 67 individuals). 19–23 Each of these studies demonstrated small improvements in hand impairment and/or function with no increase in joint swelling, pain or disease activity.

Adherence to any exercise programme is crucial, as it is suggested that there is a dose–response relationship in both healthy and RA populations for strength and pain. 24 Adherence with short-term supervised exercise programmes is generally high. 25 However longer-term and home-based exercise programmes do not have the same response,26 although data are sparse. There is some evidence to show that a programme incorporating a behaviour-change framework based on the Health Belief Model27 is effective in maintaining long-term adherence in RA patients. 28

Joint protection advice includes pain management advice, planning and pacing activities, regular rest, altering patterns of joint movement and assistive device use in order to minimise pain and fatigue and make tasks easier. This advice may be provided in the form of information leaflets (e.g. Looking After Your Joints When You Have Arthritis29), one-to-one sessions, group interventions or a combination of these. Evidence suggests that, provided appropriate education methods are applied, JP improves function and reduces pain in the short and long term. 28,30

Provision of splinting is widespread in UK clinical practice,31 with the objectives of reducing hand and wrist pain, improving hand function and reduction or prevention of deformity and soft tissue contractures, although evidence of efficacy is unclear. 32 Types of splinting may be categorised into resting or functional depending on the exact requirements of the patient.

Costs of rheumatoid arthritis

The economic cost of RA is substantial for both the individual patient and society as a whole. 33 Patients with poor and declining function from their diagnosis of RA experience much higher costs of care overall. 34 A report by the National Rheumatoid Arthritis Society in 2010 found that the overall cost of RA to the UK economy was almost £8B per annum, with NHS expenditure totalling approximately £700M. 35 To date, no studies evaluating exercise in hand and wrist RA have detailed costs involved or included a cost-effectiveness analysis.

Rationale for Strengthening And stretching for Rheumatoid Arthritis of the Hand trial

Wrist and hand dysfunction as a result of pain, loss of movement and weakness is a common problem in RA. To address this, exercises are currently recommended as part of clinical management of people living with RA with an increasing shift towards more active treatments at an earlier stage. These recommendations are not supported by high-quality evidence.

With previous small-scale studies showing some promise over the short term, there is a clear need for long-term evaluation of an optimised hand exercise programme in a large group of people living with RA. As part of this evaluation, it is important that strategies to maximise programme adherence are incorporated and evaluated. A mixed-methods approach provides rich data that should facilitate understanding of why such a programme does or does not work.

In the Strengthening And stretching for Rheumatoid Arthritis of the Hand (SARAH) trial we evaluate a hand exercise programme that will be acceptable to NHS physiotherapists and occupational therapists based on current available evidence. Such a programme is over and above what is currently provided in the UK NHS. A parallel economic evaluation will enable conclusions to be made about cost-effectiveness.

Inclusion criteria

• RA meeting the American College of Rheumatology (ACR) clinical and immunological criteria (Box 1). 10

• Pain and dysfunction of the hands and/or wrist joints.

• No treatment with a DMARD or having been on a stable DMARD regimen for 3 months or more.

Exclusion criteria

• Age less than 18 years.

• Upper limb joint surgery, or fracture, in the previous 6 months.

• Being on a waiting list for upper limb orthopaedic surgery.

• Pregnancy.

Screening and recruitment

Patients were approached primarily by clinical staff during routine clinic visits. They were provided with a patient information sheet (see Appendix 1) and asked if they would consider participating. If they agreed, a research clinician (physiotherapist, occupational therapist or nurse) then contacted them to book an appointment at a research clinic. At this appointment, the trial information was discussed, the patient had an opportunity to ask questions, eligibility was rechecked and, if appropriate, patients were consented (see Appendix 2). Patients were asked to give written informed consent according to principles of Good Clinical Practice and the Declaration of Helsinki. 36 At the time of consent, outcome assessors collected baseline measures.

In addition, patients on rheumatology consultant or therapy review lists without planned clinic appointments were approached using postal questionnaires. They were sent an information sheet and a brief questionnaire to return with a pre-paid envelope by the research team. The questionnaire asked whether or not they were currently experiencing problems with their hands and/or wrists and about other eligibility criteria. Patients who returned the questionnaire were contacted by the research clinician and, if appropriate, had a research clinic appointment made.

Settings and locations

The trial was run in the hospitals of 17 NHS trusts (names at time of participation):

• Basingstoke and North Hampshire Hospitals NHS Foundation Trust (North Hampshire Hospital)

• Derby Hospitals NHS Foundation Trust (Royal Derby Hospital)

• Dorset Primary Care Trust (Victoria Hospital)

• George Eliot Hospital NHS Trust (George Eliot Hospital)

• Heart of England NHS Foundation Trust (Solihull Hospital)

• Nuffield Orthopaedic Centre NHS Trust (Nuffield Orthopaedic Centre)

• Poole Hospital NHS Foundation Trust (Poole General Hospital)

• Portsmouth Hospitals NHS Trust (Queen Alexandra Hospital)

• Royal Bournemouth NHS Foundation Trust (Christchurch Hospital)

• Royal National Hospital for Rheumatic Diseases NHS Foundation Trust (Bath Royal National Hospital for Rheumatic Diseases)

• South Warwickshire General Hospitals NHS Trust (Warwick Hospital and Stratford-Upon-Avon Hospital)

• Sussex Community NHS Trust (Bognor Regis War Memorial Hospital)

• Winchester and Eastleigh Health Care Trust (Royal Hampshire County Hospital)

• Worcestershire Acute Hospitals NHS Trust (Worcestershire Royal Hospital; Alexandra Hospital, Redditch; Kidderminster Hospital and Treatment Centre)

• Wrightington, Wigan and Leigh NHS Foundation Trust (Wrightington Hospital)

• University Hospitals Coventry and Warwickshire [University Hospital (Walsgrave site) and Rugby St Cross Hospital]

• University Hospitals of Leicester NHS Trust (Leicester Royal Infirmary and Leicester General Hospital).

Sequence generation

The variable block randomisation schedule was prepared by the trial statistician (CM). Randomisation was stratified by centre to control for any confounding factors evident at local recruitment sites, such as therapist effects or local contamination of intervention.

Allocation concealment

The research clinician telephoned the randomisation service, and only once the participant was registered in the trial was the random allocation generated. Hence allocation was concealed. Allocation was faxed or emailed direct to the therapist, and the participant was told either when their first appointment was made or at the appointment itself.

Blinding

The outcome assessor was blind to the group allocation of the participant and was independent of intervention delivery. Participants were requested not to disclose group allocation to the outcome assessor. If an outcome assessor was unblinded, this was recorded. If they remained unblinded, the assessors were asked to guess which allocation they thought the participant was given. The participants and therapists could not be blinded to the group allocation.

Database and data processing

The database was held on Warwick Clinical Trial Unit’s Microsoft SQL Server system (Microsoft Inc., Bellevue, WA, USA) and imposed rules for data entry which include valid range for responses, linked dates and patient identification numbers.

Data were single entered into the database by study personnel. The trial statistician carried out checks of plausibility of values, missing data and form return rates to enable further queries to be resolved prior to freezing data for scheduled DMEC reports and analysis.

For data quality assessment, 10% of all forms at all time points were randomly selected for data checking on a 2-monthly basis. All disagreements found when checking were corrected and any systematic faults found as a result of the checks were also corrected. The levels of accuracy specified were 1% for primary end point data and 5% for secondary end point data. Five time periods were covered and disagreement levels were always below 1% apart from one occasion when, in one 2-month period, a 2% level was found for primary data. Corrections were made and the problem did not recur.

Scoring and missing items

Michigan Hand Outcome Questionnaire scales were scored as described by Chung et al. 58 Individual item scores are summed to give a raw score for each scale. These are converted to a score between 0 and 100 using the MHQ scoring algorithms. Averages of right and left hand are used for those scales where both hands are measured. Scales were calculated if at least 50% of their items were completed. An overall score can be obtained by averaging the score for all scales (with pain scale reversed). At least four scales need to be completed to calculate the overall MHQ score.

The SF-12 was scored as described by Ware et al. 59 We analysed mean component scores for the Mental (Mental Component Score; MCS) and Physical (Physical Component Score; PCS) subscales. Missing responses were dealt with using an imputation algorithm. 60

European Quality of Life-5 Dimensions was used mainly for economic analyses evaluation; methods and results for the economic analyses evaluations are described in Chapter 6.

Further details on the scoring used for impairment measures is given in Appendix 4.

Subgroup analyses

Statistical tests of interaction were used to perform pre-specified subgroup analysis on baseline drug regimen (no DMARD, single DMARD, combination DMARD or biologic DMARD) and disease duration (< 5 years or ≥ 5 years). Post hoc exploratory analyses were also conducted for baseline MHQ overall hand function (< 52.5 points or ≥ 52.5 points), age (< 60 years or ≥ 60 years), type of referral (clinic or review list mail out), gender, baseline ESR (< 16 mm/hour or ≥ 16 mm/hour), baseline CRP (< 6 mg/l or ≥ 6 mg/l), SF-12 PCS (< 34 points or ≥ 34 points) and SF-12 MCS (< 50 points or ≥ 50 points).

Sensitivity analyses

Sensitivity analyses were planned to explore effects of results of adjustment for age, sex, site and any imbalance in baseline characteristics. A repeated measures analysis was also performed.

Ethics committee approval

The SARAH trial was approved by the Oxford C Multicentre Research Ethics Committee in June 2008 (Research Ethics Committee reference 08/H0606/47) and by the research and development department of each participating centre. The final approved study protocol has been published. 42

Trial Steering Committee

A Trial Steering Committee (TSC) was responsible for monitoring and supervising the progress of the SARAH trial. The TSC consisted of three independent experts, a lay member and leading members of the Trial Management Group (TMG). Membership of the TSC is given in Acknowledgements.

Data Monitoring and Ethics Committee

The DMEC was independent of the trial and was tasked with monitoring ethical, safety and data integrity. The trial statistician provided data and analyses requested by the DMEC at each of the meetings. Membership of the DMEC is given in Acknowledgements.

Trial Management Group

A TMG was responsible for the day-to-day management of the trial, consisting of the chief investigator, research fellows, statistician, trial co-ordinator, research nurse and data-entry clerk. The role of the group was to monitor all aspects of the conduct and progress of the trial, to ensure that the protocol was adhered to and to take appropriate action to safeguard participants and the quality of the trial itself.

Clinical guidelines

Current UK guidelines for the management of RA62–64 recommend that access to physiotherapists and occupational therapists should be offered to all people with RA to assess the impact and treat the consequences of the condition. Specifically, the stated aims of therapy include:

• facilitating ADLs (e.g. washing, using the toilet, dressing, cooking, eating, working and leisure)

• improving general health-related fitness and encouraging regular exercise

• providing instruction on a range of strategies to reduce pain and stress on joints while carrying out everyday activities

• providing instruction on specific exercises aimed at enhancing joint flexibility and muscle strength and reducing other functional impairments

• providing information and advice about short-term pain relief provided by electrophysical interventions (e.g. transcutaneous electrical nerve stimulation and wax baths).

Treatment provided by occupational therapists and/or physiotherapists can involve a variety of modalities to achieve these aims. According to surveys of current practice, the three most common components of therapy currently provided for people with RA of the hands are exercise therapy, JP advice and functional splinting. 15,65–67 Hence, current treatment is a balance between the provision of strategies to support and protect joints (including symptomatic relief) and exercise to improve or maintain strength, flexibility and functional ability. However, little guidance is available as to the specifics of any exercise programme.

Evidence base

There is evidence suggesting that exercise improves general muscular endurance and strength without detrimental effects on disease activity or pain in RA. 68 However, few studies have investigated the effect of exercises specifically for the rheumatoid hand. Some improvements in strength, mobility and/or function with no negative effects have been reported,19,20,22,61,69,70 although the long-term effectiveness has not been established.

A systematic review of the literature was performed to establish the evidence base for exercise in RA. This encompassed general exercise programmes designed for the whole body as well as those specifically addressing the hand and upper limb. In particular, exercise programmes from those studies that described the actual intervention in detail were evaluated as part of designing the final SARAH intervention protocol.

General exercise

Several studies have investigated the effects of various types of exercise on different aspects of the patient experience, the majority reporting beneficial responses. 18 Almost all have involved general or ‘whole-body’ programmes focusing on aerobic fitness, strengthening and/or active range-of-movement (ROM) exercises.

In the past, issues of disease activity and potential irritation of symptoms led to a degree of caution in exercise prescription for sufferers of RA. Current evidence suggests that exercise does not appear to have a negative impact on the disease process and may in fact be beneficial,68 including for the small joints of the hands and feet. 71 It should be noted that these studies were not specific to the hands; rather the hands were used during general upper limb exercise activity.

Specific hand exercise

Very few studies have specifically investigated the effect of hand exercises for patients with RA. A systematic review in 200472 found nine studies of sufficient quality from which no definite conclusions could be reached owing to the different designs, outcome measures and exercises utilised. These findings were reinforced in a similar review73 which concluded that evidence was lacking as to the effectiveness of shoulder- and hand-strengthening exercises.

An updated systematic review was undertaken by members of the SARAH trial team, including only RCTs or controlled clinical trials that used quasi-randomised methods of treatment. 74 Five studies were included19–22,61 (Table 2). None of the studies described any significant detrimental effect resulting from exercise – if anything there was a tendency for some measures of disease activity to improve even with intensive exercise. Reported gains in strength following exercise were variable and of unknown clinical significance.

Eight studies reported hand exercise programmes in sufficient detail to enable reproduction (Table 3). They all used various combinations of strength and/or stretching exercises, with only two attempting to differentiate between them in subsequent analyses. 22,61 Three of the studies named the same source for the exercise intervention used,76,77 while a fourth designed their programme following a survey of hand therapists. 61 None of the remaining studies provided any rationale for the exercises used.

Theoretical and physiological considerations

Expert opinion

A crucial part of the development programme was the advice received from clinicians and other experts, including patient groups. Most notably, a consensus meeting was held with specialist hand therapists from across England to gain further understanding of normal practice within typical NHS clinics and to assist in the design of the exercise intervention. The result of this discussion was a standardised protocol for usual care. Agreement was also reached as to what would constitute an effective and practical exercise programme for the hand and upper limb. It needed to be feasible to perform within NHS hand therapy clinics taking into account normal appointment duration and commonly available rehabilitation materials.

A list of upper limb exercises described in the literature, along with others proposed by various hand therapists, was examined in detail and decisions made as to which of these exercises were the most important to include in a programme specifically designed for RA. The selection was based on clinical relevance, a desire to include all functionally relevant movements/muscle actions of the hand and wrist, avoidance of replication and convenience/duration issues, especially with regard to the home exercise component. Initial load and volume, as well as progression and regression strategies, were also agreed.

Pilot study

The acceptability and feasibility of trial procedures and the exercise intervention were tested in a pilot study that ran from June 2009 to February 2010 and involved two senior therapists at two sites. The therapists received a half-day training session and a manual describing the trial and interventions. Sixteen participants were recruited and five were interviewed following their treatment to provide detailed feedback. Patient materials, exercise instructions and some trial procedures were modified following recommendations from therapists and patients as well as observations made by the trial team. Specifically, the assessment form, the Borg scale of perceived exertion of exercises86 and other forms used to document intervention delivery were altered along with instructions for the pinch grip and resisted wrist extension exercises. Information provided to patients at discharge from treatment was also developed following therapist recommendations.

Control arm: usual care only

The design of the usual care intervention was based on evidence from the literature and discussions with hand therapists. Clinical guidelines on the management of RA agree that the goals of management are symptom control, prevention or control of joint damage, reduction of disability, and maintenance or improvement of quality of life. 10,63,64

According to a Cochrane review in 2004, therapy for RA can include a variety of interventions including training of motor function, instruction on JP, advice and instruction in the use of assistive devices and provision of splints. 87 The most commonly provided treatments appear to be prescriptions of assistive devices, orthoses, hand-training instructions and patient education. 88 Evidence of effectiveness varies with relatively strong support for exercise and self-management interventions and modest support for JP programmes, orthoses and comprehensive care interventions. 89

The control arm treatment for the SARAH trial consisted of an initial assessment and advice session with the option of a further two follow-up sessions as necessary. Treatment included the provision of JP information, functional splinting, assistive devices and other general advice as required. In order to evaluate the effects of exercise, participants in the control arm were not prescribed any specific exercises for the upper limb. Apart from this, the content of the control arm was consistent with what is considered usual care according to discussions with specialist hand therapists, unpublished surveys of current practice and clinical guidelines. 31,66,67,87

Experimental arm: hand exercise plus usual care

The experimental arm consisted of the usual care described above plus a hand and upper limb exercise programme (Figure 2), which included seven mobility exercises and four strength exercises against resistance (i.e. therapy putty, Theraband or hand exerciser balls).

FIGURE 2.

Strengthening And stretching for Rheumatoid Arthritis of the Hand exercises.

The experimental intervention involved a total of six sessions, of which the last five were exercise sessions (Figure 3). Participants were provided with an exercise booklet with pictures and instructions describing the programme as well as the resistance materials required. They were asked to perform the programme daily at home between clinic sessions for a period of approximately 12 weeks.

FIGURE 3.

Content of SARAH interventions.

A modified Borg scale was used to set the load (resistance) for the strength exercises based on self-perception of effort. This 10-point version of the original Rate of Perceived Exertion scale86 has been validated for use in regulating the intensity of resistance exercise. 94 For each strength exercise, the level of resistance was determined by the subjects’ rating of perceived effort using the weaker hand in order to avoid overloading the more affected side. The load was purposefully set at a moderate level initially (3–4 on the scale) to permit subsequent progression, enhance motivation and adherence and reduce the possibility of increasing patient symptoms.

Although initial guidelines as to sets and repetitions (volume) of exercise were provided (Table 5), the programme was tailored to the abilities of the participant by progressing the load and volume of each exercise according to each patient’s capabilities. If necessary, the manner of executing a particular exercise could also be modified. For example, joint restrictions may prevent the degree of movement described in a particular exercise, in which case the participant achieved as much movement as possible within their available ROM. The overriding goal was for the participant to get as close as possible to performing the exercise in an ideal manner at a volume and load that was achievable while still providing a stimulus for physiological change. A defined protocol for both strength and ROM exercise progression (or regression) was provided for subsequent sessions according to both patient capability and therapist judgement (see Table 5).

Centre characteristics

Seventeen UK NHS trusts (21 individual departments) participated in the study (Table 6). The recruitment period at each trust varied from 2 to 19 months with resulting variation in the number of participants recruited at each site.

Participating therapists

Forty-eight therapists were involved in providing treatment for both arms of the study (Table 7). Almost two-thirds were occupational therapists and the remainder physiotherapists. All were working in specialist hand therapy departments. The number of therapists involved at each site varied from one to six (see Table 6). The level of experience of each therapist (as determined by Agenda for Change pay banding103) is described in Table 7. The vast majority of therapists were on either band 6 or 7 and the median level of experience was 13 years.

Screening

Screening and recruitment took place between October 2009 and May 2011. A total of 1606 patients were screened in rheumatology clinics and on review lists (Table 8), of whom 512 (32%) were eligible and willing to attend a research clinic appointment (Table 9), with a further 564 (35%) not eligible – Table 10 displays, where known, the reasons for ineligibility. Of those eligible but not willing, no reason was given in the majority of cases (73%) (Table 11).

Recruitment

Of the 512 patients initially given appointments for baseline assessment, six were found to be subsequently ineligible, six did not attend or cancelled and 10 declined to participate, resulting in 490 patients being randomised (see Figure 4). Although the recruitment target of 480 was achieved in April 2011, a further 10 participants were recruited in early May 2011 as bookings made in the later stages before the target was reached were all honoured. The final recruitment total was 490 patients over a 20-month period. The proportion of participants in each arm was equivalent across all centres.

During the study, two participants recruited and allocated to the usual care arm withdrew from the study and also withdrew consent for their data to be used. Therefore, totals for subsequent analysis use a potential maximum of 488.

Baseline characteristics of participants by trial arm

The baseline characteristics of patients recruited to the trial are summarised in Table 12. The randomisation process appears to have been successful, with both study arms well matched in terms of demographic data, primary outcome measure and clinical assessment findings. Baseline RA status was similar in both groups with respect to disease activity (ESR and CRP levels) and disease duration (average ≈ 10 years), with both markers indicating a population with ‘stable’ RA (see Table 12).

Over 80% of participants were recruited from direct clinic referrals, the majority of them white females with a median age of 63 years (range 24–94 years).

Treatment preference was also well matched, with just over half stating no preference in both groups and another 40% preferring the exercise arm.

As part of the initial session with the therapist, participants were asked whether or not they had ever undergone surgery for the upper limb. Of the 430 responses, 144 (33%) replied ‘yes’, with similar proportions in both arms of the trial.

Baseline medications

The most common medications prescribed at baseline were non-biologic DMARDs. Approximately 20% of the cohort was on biologic DMARDs (Table 13). There was a small difference between arms regarding non-biologic DMARDs – the proportion of participants on combinations was slightly higher in the exercise programme arm than in the control arm while the proportion on single DMARDs was slightly lower (29.3% vs. 21.9% and 41.9% vs. 48.8% respectively).

Non-biologic DMARDs are older, less expensive immunosuppressant agents, such as methotrexate, which do not target individual cells or molecules specifically.

Baseline questionnaire scores

The scores for the questionnaires at baseline indicate that both arms of the study were similar (Table 14). The mean overall hand function score of the MHQ (primary outcome) was 52 for both groups out of a possible maximum of 100.

Baseline impairment measurements

Both study arms were well matched in terms of baseline impairment measures (Table 15). On average, participants had moderate deformity, reduced ROM and strength in the wrist and hand with some swollen and tender joints.

Withdrawals

In total, 25 participants (5%) withdrew from the trial during follow-up, with a greater proportion of these in the experimental arm (n = 17 vs. n = 8 for usual care). Table 18 presents the timings of the withdrawals from the trial by arm. In the exercise programme arm, the greatest proportion of participants withdrew from the trial at the time of their treatment. There was no dominant reason for withdrawal in either arm (Table 19). If participants withdrew from trial treatments, every effort was made to obtain follow-up data from these participants (withdrawal from treatments and withdrawal from trial were differentiated on the event notification form).

Comparisons of those retained compared with those lost to follow-up

There was no evidence of differences in sex or disease duration between those who did and did not complete follow-up at 12 months. Those with 12 months’ follow-up were, on average, 5 years older than those without and had higher baseline scores for MHQ hand function, SF-12 summary scores and EQ-5D-3L health state. These differences were greater in the exercise programme arm. Table 20 shows the characteristics of participants completing and not completing 12 months’ follow-up.

Blinding

Researchers measuring outcome were asked if they knew the allocation of participants or were able to guess; we received 278 responses to this question at 12 months (Table 21). Of these, 173/278 (62%) correctly identified the treatment received and 105/278 (38%) did not. The distribution was similar for usual care and exercise therapy, i.e. the probability that patients would correctly identify the treatment received was marginally higher than would be expected by chance (50%), but this probability was the same in both arms. The knowledge of treatment allocation had no effect in a sensitivity analysis of the primary outcome (p = 0.654 at 4 months, p = 0.407 at 12 months).

Missing data

The proportion of missing data from the self-report questionnaires across all time points was generally low, with the rate exceeding 10% for only one item (diagnosis date at baseline). For the majority of measures, data were missing in less than 5% of cases, and both arms were relatively well-matched across all time points (Table 22).

The pattern of low levels of missing data was repeated for the Research Clinic Assessment forms containing the impairment and disease activity measures (Table 23). Relatively higher levels of missing data for both arms were recorded for the 4- and 12-month follow-ups as compared with the questionnaires, which reflects the fact that these data were not collected for telephone or postal follow-ups.

Treatment attendance rates

Treatments were generally well attended, with 93% of participants completing the usual care treatment and 93% of participants partially or entirely completing the exercise programme (75% completed all six sessions). For the exercise programme, completion of treatment was defined as attendance at assessment and all five treatment sessions, and partial completion was between two and five exercise sessions. For usual care, completion of treatment was defined as attending all booked sessions (completion of treatment in usual care arm may include only attending for assessment session). The proportion of participants not attending any treatment sessions was roughly equal between the arms (Table 24). For the exercise programme there was a gradual attrition of patients as the sessions went on (Table 25). A quality control programme evaluated 38 of the 48 therapists, all of whom were following the protocol satisfactorily.

Timing of delivery of interventions

Table 26 presents median times from randomisation to first and last appointment by arm. Interventions were largely delivered within 3 weeks of randomisation and completed within 3 months and 1 month for the experimental and control interventions respectively. Time to initial appointment was similar in both groups with the differences in treatment arm protocols reflected in the time to their final session.

Type of treatment

Therapists completed treatment logs for each session detailing the types of treatment provided. Table 27 provides a summary of the treatment logs by arm and per session. Note that the usual care arm involved a maximum of three sessions only, although most participants in this arm received only one session. The figures indicate a high level of compliance with the treatment protocols with regard to provision of JP advice, information and completion of exercise arm paperwork.

A similar level of splinting was provided for participants in both arms of the trial. The higher percentages for subsequent sessions in the usual care arm reflect the fact that, for the majority of these participants, follow-up appointments were made with the express purpose of either providing a splint when identified as necessary in the initial session or reviewing and/or modifying those splints that were provided at the first session.

Exercise programme progression

During the therapist training, emphasis was placed on the need to progress the exercise programmes as much as possible during the 3 months of treatment. A treatment log was completed for each session recording the details of the programme provided and allowing an estimation of treatment progression for those completing all six sessions. In this case, progression was defined as an increase in repetitions, sets or resistance between the first and last session. Maintained indicates no changes in exercise parameters between the first and last sessions, whereas regressed is a decrease in repetitions, sets or resistance between the first and last sessions. Table 28 describes the proportion of participants who were progressed, maintained or regressed for each exercise. This illustrates that the vast majority of participants progressed in at least one of the exercise variables (repetitions, sets, and/or resistance). Variation in the number of participants prescribed each exercise is because of the individualised nature of the programme.

Numbers analysed

Data from a total of 488 participants were received at baseline (246 participants in exercise arm; 242 participants in usual care). At 4 months, this was reduced to 454 participants (222 participants in exercise arm; 227 participants in usual care) and at 12 months to 438 participants (216 participants in exercise arm; 222 participants in usual care).

Primary outcome: overall hand function

The primary outcome was the overall hand function subscale of the MHQ at 12 months. On average, both groups’ hand function scores improved over time although within group changes of the exercise programme group were approximately twice those of the usual care group (Figure 5 and Tables 29 and 30).

FIGURE 5.

Scores over time for MHQ overall hand function.

With regard to between group differences, participants allocated to the exercise programme had statistically significantly greater hand function compared with those receiving usual care at both 4 and 12 months. Adjustment for baseline differences between groups for centre, sex, age and medication usage resulted in a small increase in the mean treatment difference at 4 months and a small decrease at 12 months (Table 30).

Other self-report outcomes

Impairment measures

There were inconsistent findings for between-group comparisons of the impairment measures. Some found no difference between the groups whereas others found small improvements in favour of the exercise programme group.

Disease activity

Comparisons of swollen and tender joint counts found a small to moderate effect of the exercise programme compared with the usual care groups in the short term only (Tables 47 and 48). On average, both groups achieve a one-point improvement, however the exercise programme participants achieve this in a shorter time frame (within 4 months) compared with the participants in the usual care group (within 12 months). No significant changes were reported for CRP and ESR levels either within or between groups at all time points indicating a relatively stable RA population that were then unchanged within and between groups.

Participant-rated improvement, benefit and satisfaction

The exercise programme group rated themselves as having greater global improvement in their hands and wrists than the usual care group at both 4 and 12 months (Wilcoxon test, p < 0.0001) (Table 49). They also reported significantly more benefit [p < 0.0001 (both categorical and ordinal) for both time points] and greater satisfaction from trial treatments (4 months: p < 0.0001 categorical, p = 0.0198 ordinal; 12 months: p < 0.0001 categorical, p = 0.347 ordinal) than the usual care group at both 4 and 12 months (Tables 50 and 51).

Hand exercise performance

Participants were encouraged to maintain their exercise routine once the exercise programme had finished. As would be expected, the proportion of participants who reported performing hand and wrist exercises was significantly greater in the exercise programme group than in the usual care group at both 4 and 12 months (p < 0.0001 at 4 months and p = 0.0185 at 12 months) (Table 52).

Medication usage

There were no significant differences in medication usage between the two groups at the two follow-up time points. There were small numbers of changes to medication over the 12 months (Tables 53–56).

Secondary analyses

Adverse events

During the recruitment and follow-up period, 103 SAEs were reported to the trial team (Table 63). SAEs were classified through discussions with local principal investigators and the trial lead. As much information as possible was requested from the participant when a potential adverse event was noted, particularly with regard to likelihood of cause of trial treatment. No SAEs were deemed both unexpected and related to the trial involvement and therefore were not communicated to the MREC. The categories of SAE, likelihood of relatedness and reasons are summarised in Table 64.

Sampling procedure

The interview study was carried out at four acute NHS trusts taking part in the SARAH trial; one trust yielded no eligible interview candidates. Trial participants were approached to take part in the interviews at their 4-month assessment. Willing participants were given an information sheet and verbal consent obtained for the interview researcher to telephone them if they were eligible for interview. As the assessors were blind to treatment allocation, all participants were informed about the interviews, but only those receiving the trial intervention were eligible for interview. Purposive sampling was done using the 4-month questionnaire, aiming to interview up to 10 participants who rated themselves as having benefited and up to 10 participants who had not benefited from the intervention in order to gain a range of experiences. Recruitment continued for approximately 1 year. Those eligible for interview were contacted by one of the researchers carrying out the interviews (EW and VN) and a convenient time and venue booked. At the beginning of the interview visit the researcher answered any questions arising from the information sheet and took written consent for the interview study (see Appendix 7).

Interviews

Five pilot interviews were carried out during the trial pilot study. These interviews informed the initial interview schedule and are not included in the analysis. Some changes were made to the interview schedule as the study progressed in response to the completed interviews, which is in line with an interpretative phenomenological analysis (IPA) approach (see Appendix 8 for interview schedules). Two researchers carried out the interviews, EW carried out the pilot and first six 4-month interviews and VN the remainder of the 4-month and all the 12-month interviews. Researcher notes were written immediately after the interviews to capture the researchers’ thoughts on the interviews and promote reflexivity across cases. Interviewees were given a pseudonym for anonymity.

Analysis

Analysis was carried out using IPA,104 which has been recommended as a suitable approach to investigate how an individual perceives a situation or to explore the meaning of particular events, experiences and states for the individual. 105 This approach has been used successfully to describe the experiences of patients with musculoskeletal problems such as chronic low back pain. 106 IPA does not require the testing of a hypothesis but allows the researcher to carry out exploratory work, thus reflecting the aims of this study. 105

All interviews were recorded using an Olympus DM10 digital voice recorder (Olympus Corporation, Shinjuku, Tokyo, Japan), transcribed verbatim, anonymised and checked by the interviewer. All transcripts were entered into NVivo, a computer software tool for managing qualitative data (NVivo V9, QSR, Portsmouth, UK).

The first six interviews by EW were used to explore passages of the data from which initial codes were developed. Codes were given definitions which were refined and ‘sharpened’ as the study progressed. EW and VN worked in collaboration to merge similar codes. Coding was also discussed with another experienced qualitative researcher (FT). After six interviews were completed, the 4-month interview codes were examined again by VN and FT and codes were grouped into four main headings. Subthemes and emergent themes across codes were explored with participant number counts. Analysis also involved cross-case comparisons to identify similarities and differences in the data.

Before each 12-month interview the 4-month transcripts and researcher notes were reread and additions made to the 12-month schedules which were specific to the individual. Interesting or ambiguous ideas and phrases were further explored in line with an IPA approach.

Acceptability and adherence to the Strengthening And stretching for Rheumatoid Arthritis of the Hand intervention

Establishing a routine: the key to success

It was very clear from the narratives that establishing a routine was the key to successfully carrying out the SARAH exercises. Some interviewees found that establishing a routine was easier to do than others and they employed different strategies or ways of doing this.

The Strengthening And stretching for Rheumatoid Arthritis of the Hand therapist: from teacher to facilitator

Interviewees were not directly asked about their relationship with the SARAH therapist. However, they offered insights into their interactions and the therapist’s role in helping them to exercise. Eight interviewees spoke about the therapist in the interviews. All their encounters were spoken about in a positive context and suggested that a good rapport was attained. The SARAH therapist was integral to the interviewees’ ability to establish their exercise routine. The therapist was portrayed as a collaborator who helped participants to exercise in the right way, at a time and routine which was negotiated with the individual:

I mean [the SARAH therapist] took a lot of time. I felt she spent a lot of time with me on each session making sure that I understood and could do the exercises.

Sally: 4 months

Interviewees provided examples of how the therapist facilitated their adherence with the exercise programme. Interviewees said that therapists were motivational and reassuring and provided feedback on how they were doing:

Having that weekly contact is certainly a huge motivator and it tends to make it much more of a focus in sort of the week and then when you’re sort of cut loose and you’re left to your own responsibilities you have to be really, really motivated.

Alice; 4 months

During the therapy sessions interviewees were encouraged to discuss their progress (or regression) and the therapist tailored the programme, progressing them when able and giving follow-up advice as needed. The therapist monitored exercises for accuracy and suitability:

The first time you’re shown them and the physiotherapist does them with you and shows you how to do them and then the next time you go back she’s sitting there and we were doing them together and then after that, it was down to see how well had I improved, or not improved and how perhaps there was a technique I could use that would be a little bit different.

Lizzy; 4 months

Four interviewees spoke of the therapist tailor-making their exercises rather than a ‘one size fits all’, which was appreciated:

She’s so lovely and she was so kind to me and the fact that they’re all sort of modified to suit. You know, it wasn’t that you had this goal that if you didn’t achieve X amount of exercises you couldn’t carry on. It was all how much I could do, so once I found out I was quite excited because I thought it could only help me. It couldn’t do any harm.

Alice; 4 months

The therapist provided input throughout the course of the programme but the role of the therapist varied at the different stages and in response to the needs of the patient. Initially exercises were taught but the PEG introduced behavioural elements which the therapist and patient signed up to. The therapist’s role changed from a teacher of exercise to a collaborator and facilitator through shared goal setting and reviewing how the exercises had or had not been carried out. Barriers were explored, problems solved and new goals were set if appropriate.

The type of exercise provision given was new to many interviewees as seeing a therapist for detailed exercise advice was not a common experience. Five people said that they had had varied or sporadic advice about exercising from different disciplines through the rheumatology clinic, however over half felt that they had had no specific advice about exercise:

No one has really given me advice about exercise . . . No one has actually said to me beforehand, ‘You need to do some exercise.’

Mark; 4 months

Barriers to the Strengthening And stretching for Rheumatoid Arthritis of the Hand exercises

Alongside the interviewees’ descriptions of factors which helped them to carry out the SARAH programme, barriers were also identified. The majority of interviewees described barriers to doing exercise with RA in general, and many of these applied to the doing the SARAH programme. There were some barriers that were specific to carrying out the SARAH programme. Barriers were classified as intrinsic and extrinsic barriers.

Gauging the personal effectiveness of the Strengthening And stretching for Rheumatoid Arthritis of the Hand intervention

We hoped to gain an understanding of the impact the SARAH intervention had on patient symptoms and how interviewees measured the personal effectiveness of the programme.

Ten interviewees felt that the SARAH exercise program had worked for them, although Emily and Lizzy were not sure if it was their medication or a combination of the intervention and medication. All 10 interviewees described how they felt the intervention had worked for them with physical improvements being the most prevalent. An improvement in function was most cited; functional gains were described by some in terms of being able to: do more in the garden (Samantha), do household tasks such as ironing and drying the dishes more easily (Harry), pick things up more easily (Lucy) and decrease their reliance on splints (Alice). Physical improvements were most often in strength or grip and flexibility or increased ROM:

Using the secateurs and also just to pull the stubborn weeds out. You can probably grip those better and pull them out easier and lifting certain things with my hands and so for example, the container off the lawnmower. I could probably do that a bit easier and maybe shopping bags a bit better. What else; maybe a couple of jars with the tin opener.

Samantha; 4 months

Some spoke of improvements in their symptoms with regard to their exercises. Joe stated that since doing the exercises he had had no flare-ups, and Alice said that her hands felt better after exercising.

Psychological constructs were difficult to characterise; some felt the benefit was ‘feeling better’ without further elucidation. Mark and Kate felt that the sessions had given them a focus, and Lizzy spoke of an increased awareness of her hands:

At 12 months, nine respondents still spoke of physical improvements:

I knew there was so much more flexibility in my hands. So the flexibility and, I suppose to a certain extent, the strength was quite significant.

Lucy; 12 months

However, they also spoke of a higher proportion of psychological improvements at 12 months when compared to 4 months. The most notable being in confidence and increased awareness, suggesting it took time for confidence in improved hand function to develop and translate into everyday function. Additional psychological benefits were a sense of empowerment, a hope of slowing their disease progress or maintaining their function.

Samantha, Alice and Lucy spoke of an increased confidence in their hand function:

It’s little things. I’m probably more confident with picking things up and I’m more confident with the manual dexterity. I mean my handwriting is still pretty appalling at times, but I was quite anxious about picking things up and dropping things and that’s less of an issue now.

Alice; 12 months

Alice and Sadler spoke of a confidence in the exercise programme itself, which had increased their confidence to continue with the exercises. Kate thought that it had empowered her to be in control of her RA, the SARAH intervention had given her hope that she could improve or control her symptoms:

Because emotionally and psychologically I feel more at ease with it and it’s about the power of thought for me, whereas before when I was first diagnosed with it I didn’t have any control over it whatsoever, whereas I’m actually doing something that’s practical that might help and ‘might’ is enough. It doesn’t have to be ‘will.’ It might and therefore, if that is in my hands to make that change I can do that. If it’s not in my hands I’m helpless and therefore that’s actually quite a difficult place to be, so the SARAH project for me I think has been really quite valuable because it’s empowered me.

Kate; 12 months

An increased awareness of their hands was spoken about by five interviewees. Participants had become more aware of addressing their limitations (Lucy), the importance of keeping moving and looking after their hands (Lizzy, Samantha) or that it gave them other ways of doing things (Alice):

Yes, they [the exercises] are valuable. Because they’ve made me realise how important it is to keep moving, it would be so easy to say, ‘Well I can’t do that’.

Lizzy; 12 months

Alice, Sadler and Samantha were hopeful that the exercises may possibly slow the rate of progress of their symptoms and talked about the exercises ‘maintaining what they had’:

And I don’t know if 5 years down the line I’m going to be so grateful I did this, and if this is going to help to maintain function and strength then I’m the only one who’s going to benefit from that, so if I don’t put it in I’m not gonna get it out. And it’s so little to ask to do.

Alice; 12 months

Four participants felt that the SARAH intervention had not worked for them: Jean because she hadn’t done the exercises, Susan because of a longstanding flare-up which hadn’t responded to treatment, Bert because he felt his hands were too good prior to randomisation (and he wasn’t expecting any improvement) and Sally because she hadn’t had any noticeable change yet (but was still hoping for change over the rest of the trial period).

Everyone is different: reasons the Strengthening And stretching for Rheumatoid Arthritis of the Hand intervention might be effective for some people and not others

There was recognition from interviewees that different things may work for different people and they were able to offer reasons as to why the SARAH intervention might be effective for some people and not for others. Almost all interviewees gave insights at 4 or 12 months giving rise to four main themes: (1) a positive mindset, (2) the level of disease, (3) reasons to exercise and (4) incorporating exercise into lives:

So I guess it’s all going to be about individuals in terms of how they’re . . . what their mental state is, what their total environment is, what else is going on, but ultimately, it’s that balance between spending the time doing it, either finding the time to do it in my case, putting yourself through the rigours of trying to do it and the potential discomfort that that may bring, versus the long-term benefit. It’s that balance which is ultimately going to be the equation that’s going to make it successful or not.

Sadler; 12 months

Suggested changes to the Strengthening And stretching for Rheumatoid Arthritis of the Hand intervention

At 4 months, half the interviewees said that they would not change anything about the intervention. Three areas of potential change were identified from the remaining seven interviewees, and these seem to be mostly about practical administration of the intervention. Individual comments were given about the intervention content and suggestions included that daily diary sheets could be put in a ring folder with more space to write explanatory comments, the putty exercises could be further modified and the bag catch would have been better with a hook and loop fastening (rather than a clip fastening). Suggestions were given concerning the intervention format and setting. One interviewee suggested a group setting for interaction and accountability, but another felt a group setting may not work. Another interviewee suggested delivering the programme in a community setting for ease of parking.

At 12 months, however, suggestions given by six people all concerned ways in which people could be helped to ‘stay with’ the programme, three advocating groups. Three areas of change emerged. Firstly, more inventive ways were suggested to help to establish a routine and help incorporate the exercises into their lifestyle:

To change habits is quite difficult. [. . .] So I think it would be very good to carefully sit down with somebody and just honestly say where could you fit this in, in your routine, to help them maximise the chance of completing the exercises and saying ‘that exercise is really going to take you that long to do.’ This exercise, you’ve got to find that putty, where is it? . . . So if people haven’t got time to do that, they are not going to do it. So to really help them, with putting in place a plan that would fit in to their routine, and they’re are more likely to get it done.

Samantha; 12 months

Secondly, an increase of initial support by health-care professionals may help to establish a routine and then a ‘touching base’ after the end of the therapist input may be useful:

. . . but it needs to be fairly intense at the beginning just to guide the patient through it, what they are doing, what they are achieving or even keep an eye on what they are not achieving.

Bert; 12 months

Thirdly, interviewees gave ideas of other support systems which could be useful such as groups, peer/mentor support, text messages, email or ‘YouTube’ videos to help people stay engaged with the process:

It could have been done in a group but it was, because really you could all do the same exercises together in a group. But it is nicer when you’re one to one with a person because you can talk about other things as well.

Mark; 12 months

I don’t see why you couldn’t have half a dozen people at a table . . . and it might give . . . it might contribute towards the commitment element. That it would bring people together.

Lucy; 12 months

I think it would be useful to know what type of activity people enjoy, some people like groups, some people don’t. [. . .] I’d have hated it. But that’s just me. But other people would see it as a much more social event and would perhaps find support from being with other people. Sometimes it would have been nice to have another person who was on the trial to sort of say, ‘I’m really finding that one difficult’ . . . so that you had people that (you) could contact . . . I know that might not be feasible, but I do think actually that sort of mentor support is useful. Particularly if it’s such an ongoing long-term programme . . .

Alice; 12 months

One thing that’s just going through my mind is the use of the new technology and the internet and is there something, in terms of keeping in touch, that that could be factored in, because clearly it’s not easy to have that time, both from the patients perspective and from a healthcare professional perspective, people coming in, being on the road and coming in and meeting people. Could we use the internet or email or whatever, to actually do that. Is there any, almost like ‘YouTube’ video place that you could have access to, to remind you what you should be doing and things that . . .

Sadler; 12 months

Implications for clinical practice

We have demonstrated that the behavioural elements of the SARAH intervention were successfully implemented and enhanced the more traditional exercise provision role of therapists. The use of behavioural strategies such as completing the PEG may not be familiar to all therapists. However, they were an important aspect of the SARAH exercise intervention so need to be delivered alongside the exercises. The role of the SARAH therapist changed from teacher to facilitator, enabling the patient to gain expertise and subsequently to exercise independently (Figure 7). Therapists who deliver the intervention need to be trained in all elements of the programme but training is straightforward and could be incorporated into current therapy departmental training programmes.

FIGURE 7.

Therapist and patient roles in the SARAH intervention.

Further research

Several participants suggested the exercise programme could be delivered in a group setting, but this needs to be evaluated before it could be recommended.

Different methods of providing on-going support for patients as a way of ensuring long term adherence of the exercise programme were suggested. These ranged from ‘touching base’ visits to the therapist to the use of on-line resources such as ‘YouTube’. Evaluating ways of providing support is another area of further research.

Intervention costs

Participants in both treatment groups attended physiotherapy or occupational therapy appointments during the course of the trial. Assumptions used to cost the therapist time for these appointments are shown in Table 66. For the usual care group (control), the trial protocol specified a maximum of one and a half hours therapy spread over three sessions, although in practice some control patients attended more sessions than this (n = 6 and often for splinting issues). Usual care therapy included JP education and advice, simple mobility exercise for the hand and functional splinting.

Participants in the usual care and exercise group (experimental) received the usual care regimen supplemented with additional sessions of an optimised exercise programme delivered by specially trained physiotherapists and occupational therapists over a 12-week period. Participants were invited to attend six 30- to 45-minute sessions, given an individualised exercise programme to perform at home, and asked to keep a record of their activities. Training time for therapists to provide the optimised exercise programme ranged from 3 to 5 hours depending on site. In the base-case analysis we allocated the full cost of physiotherapist time spent on training (183 hours at £35 per hour) across the trial participants who attended at least one treatment session (n = 473) (see Table 66). This entails a conservative assumption that the training has no residual value. In practice, there would be a one-off cost for training physiotherapists new to the intervention, and then an on-going cost for training due to staff turnover and possibly also ‘top up’ training for existing staff.

Participants in the exercise programme arm who attended at least one session were given additional information and equipment for home exercises (Table 67). Each participant received an exercise booklet with a description of all the exercises they were to perform at home and in clinic, an exercise diary at each session (printed A4 sheet of paper), a PEG, a discharge advice information sheet and a spare exercise diary, a cardboard folder to hold all the paperwork, and a blue bag to keep this and the exercise materials in. Depending on whether or not there were any major changes to the participants’ goals or programme, they may have received between one and five personal PEGs. Exercise materials included putty, balls and Therabands. These were supplied according to each participant’s capabilities and may have changed as they progressed through the programme. The quantity and colour of the material (which reflects resistance level) that each participant received was individualised. In the cheapest case, a participant would have received 100 g of putty, one ball and a 1 m length of Theraband. In the most expensive case, materials would have been changed at every exercise session: 500 g putty, five balls and 4 m of Theraband.

The estimated costs of hand therapy by the number of sessions attended are shown in Table 68. Following our base-case assumptions, the cost for a usual-care patient attending three sessions is £83, compared with £186 for a patient in the intervention group attending all six recommended sessions. Intervention costs were estimated at the individual patient level, based on the recorded number of session that they attended.

The cost of training therapists is allocated across 473, 237 and 0 intervention participants for the worst-case, base-case and best-case scenarios respectively (473 patients attended at least one session in the trial). For simplicity, the cost of consumables is allocated to all intervention patients who attend at least one session. We conducted a simple one-way sensitivity analysis using the worst-case and best-case estimates (see Discussion).

Unit cost of medications

Information about prescribed medications was collected from participants in the baseline, 4-month and 12-month interviews. A list of all medications recorded was collated, and items related to RA were identified for inclusion in the economic analysis (Table 69). This list included NSAIDs, biological and non-biological disease-modifying drugs, analgesics and gastroprotective agents (possibly related to side effects of NSAID treatment). NHS net prices per mg for these medications were obtained from the British National Formulary (BNF). 115 Costs for individual patients were estimated based on their reported dose and frequency if this information was available, or otherwise on an assumed daily dose. This assumed dose was based on the World Health Organization defined daily dose (WHO DDD),116 or on BNF recommended doses if no WHO DDD was available. For the biologic disease-modifying drugs, doses were based on NICE recommended doses, and the cost of injections and infusions was also included based on NICE guidance117 (Table 70).

Unit costs of other health services

The 4- and 12-month patient questionnaires included questions about their use of a range of other NHS services related to their hand arthritis (Table 71). Unit costs were collected for each of these items from national published sources: Department of Health reference costs118 or Personal Social Services Research Unit (PSSRU). 113