In a pragmatic design, which initially follows the guideline for SP of the DCGP, a study consisting of two phases will be carried out: a Qualification Period of two weeks followed by a randomised controlled trial (RCT) with a 50 week follow-up period (Figure 1). The Qualification period aims to filter out patients with a favorable natural course. During the 2-week Qualification period all patients are advised to start with paracetamol or NSAIDs in maximum dosage on a time contingent base, receive advice regarding activities of daily living, work, hobbies and sports. This advise fits within the first line treatment as recommended in the guideline for SP of the DCGP. Moreover, patients are referred for US of the shoulder to the radiology department of the Maastricht University Medical Centre (MUMC) or Orbis Medical Centre (OMC) in Sittard-Geleen, The Netherlands. Based on the qualification assessment at 2 weeks, patients with insufficient improvement qualify for the RCT. These patients are randomly assigned to the intervention or the control group. The therapies used in both groups are the same except that therapies used in the intervention group will be tailored based on the US results. Primary and secondary outcome measures will be assessed at baseline, 13, 26, 39 and 52 weeks after inclusion. Patient recruitment started in November 2010 and patients will be included until October 2012. The Medical Ethics Committee of the Maastricht University Medical Centre has approved this protocol (NTR2403). This trial is officially called the Maastricht Ultrasound Shoulder pain Trial (MUST).
Patients for this trial will be recruited and treated by 21 GPs, working in 11 general practices, in the Westelijke Mijnstreek, a region in the southern part of the Netherlands. A total of 80 general practitioners received a letter inviting them to participate in this study. Of them, 21 GPs agreed to participate in the study. They attended a 2-hour instruction workshop with their practice assistants. This workshop provided information about the guideline for SP of the DCGP, the inclusion and randomisation procedures, as well as the interventions to be applied. All participating GPs were asked to give the names of their preferred physiotherapy practices. These physiotherapists were additionally invited for the workshop. In total 26 physiotherapists from 12 of the 14 invited physiotherapy practices attended the workshop. The physiotherapists were presented with an evidence based statement regarding subacromial disorders in a separate 1-hour parallel program . Those two practices not represented, received a handout of the presentation and study materials.
The study population will comprise of patients with SP, who are physically active with troublesome pain, and visit their GP with an episode of SP. To be eligibly for recruitment patients have to fulfil the following eligibility criteria: (i) shoulder pain upon abduction with painful arc; (ii) symptoms lasting no longer than three months; (iii) first episode of SP for 12 months; (iv) age between 18 and 65 years. Exclusion criteria will be: (i) consultation or treatment for SP in the past three months; (ii) glenohumeral external rotation range of motion less than 45 degrees as this is a reason to suspect a glenohumeral disorder like osteoarthritis or a frozen shoulder; (iii) history of fractures of the proximal humerus or acromion, dislocation and/or surgery of the affected shoulder; (iv) shoulder complaints caused by rheumatic disease, suspected referred complaints or extrinsic cause; (v) history of depressive or anxiety disorders, or pain catastrophising; (vi) inability to complete a questionnaire independently; (vii) unable to give informed consent (dementia or psychiatric disorders); (viii) involved in disability or liability procedures.
Before randomisation, US of the shoulder is performed by a radiologist with 8 to 20 years of experience in musculoskeletal US at the MUMC or OMC using a protocol-based scanning approach (Additional file 1) [22–26]. US is an accurate diagnostic instrument to diagnose subacromial disorders . The distinguishable disorders are tendinopathy, calcific tendinitis, partial and full thickness tears, and subacromial bursitis.
The advised evidence based, tailored treatment steps are described in Additional file 2 [4, 9, 17, 27–36]. To prevent treatment of supposed asymptomatic pathology, GPs will link US pathology to history and findings from physical examination. In case pathology other than rotator cuff disorders is diagnosed, it will be treated according to this diagnosis (e.g. in cases of signs of rheumatoid arthritis patients are referred to a rheumatologist). If there is no detectable pathology, usual care according to the guideline for SP of the DCGP will be advised . In cases where multiple US findings are present, the most relevant abnormality will be selected by the GP on the basis of the clinical findings. With an explanation, and within the recommendations made in the guideline for SP of the DCGP, GPs are allowed to deviate from the advised treatments steps. The treatments are standardised.
Usual care according to the guideline for SP of the DCGP will be applied in the control group. It consists of a pragmatic, stepwise approach; a wait-and-see policy with advice and analgesia for another 2 weeks; in persisting cases corticosteroid injections and referral to a physiotherapist are advised, depending on the level of pain and functional limitations respectively; referral to a hospital specialist is advised if conservative treatment fails .
Randomisation and allocation
Based on the qualification assessment at 2 weeks, unrecovered patients (measured by the Global Perceived Effect questionnaire; see below) qualify for the RCT and are randomly assigned by central block randomisation (blocks of 4) to the intervention or control group after stratification for age (≥ 50 years). Neither the patient nor the GP can be blinded for the allocated treatment. However, US results are only disclosed to GPs of those patients in the intervention group, as well as those patients themselves. In case a patient is allocated to the intervention group, the GP receives the US result and the advised corresponding treatment strategy. In the control group, neither the patients nor the GPs receive the US results. Their US results will be presented to the GP at the end of patients' follow-up period.
The radiologist performing the US, is not allowed to communicate with the patients about the US findings and results. In case a fracture, septic bursitis or arthritis, or a life-threatening disorder (e.g. tumour) is diagnosed, the radiologist will immediately inform the GP with c.c. to the investigator, and the patient will be excluded and not randomised.
At baseline, demographic information will be collected including age, sex and profession, as well as disease specific information regarding the affected side, onset, duration of symptoms, possible cause of complaints, history of shoulder complaints, neck complaints and dominant arm. A number of outcome measures will be collected at baseline, 13, 26, 39 and 52 weeks after inclusion (Table 1).
Primary outcome measure
The primary outcome measure for the clinical effectiveness is the patient-perceived recovery using the Global Perceived Effect questionnaire (GPE); a one-item score concerning recovery following treatment, measured on a seven-point ordinal scale. Patients are considered to be recovered when they report to be much improved or fully recovered. Together with disease-specific functional status measures, this is considered to be an important outcome variable for shoulder complaints.
Secondary outcome measures
Shoulder Pain Score (SPS)
The SPS is a questionnaire to assess pain experienced by patients with shoulder disorders and includes a 24-hour recall frame. The score consists of six pain symptom questions and a 10-point Scale . The SPS has been proved to be a useful instrument for following the course of the disorder over time, and gives an indication when a patient feels cured. Each question receives a maximum of four points. The VAS is also transposed to a four-point scale (0 = 1, 1-3 = 2, 4-6 = 3, 7-10 = 4). The minimum SPS score is seven points, the maximum score 28.
The Shoulder Disability Questionnaire (SDQ)
The SDQ assesses the performance of daily activities. This variable will be assessed by a 16-item questionnaire for functional status limitation in patients with shoulder disorders and assesses the past 24 hours . The 16 questions can be answered with either yes, no or not applicable. The final SDQ-score will be calculated by dividing the number of positive responses by the total number of applicable items, and multiplying this score by 100. Consequently, the SDQ-score can range from 0 to 100 with a higher score indicating more severe disability.
The Euroqol five-item quality of life questionnaire (EQ-5D)
The EQ-5D is one of the most commonly used generic (that is not disease specific) measures used to quantify the health related quality of life in people with musculoskeletal disorders [40, 41]. It is a patient-reported measure that consists of two sections. The first section comprises five questions with three levels of severity in each (1 = no problem, 2 = moderate problem, 3 = severe problem) that covers five dimensions of health: mobility, self-care, usual activities, complaints/discomfort, and anxiety/depression. This generates 243 theoretically possible health states. Calculation of the index score will be performed according to the European recommendations . The second section is a visual analogue scale ranging from 0 (worst imaginable health state) to 100 (best imaginable health state).
Intervention costs, direct and non-health care costs, as well as indirect costs will be collected. A questionnaire composed of 24 questions regarding resource use and expenses in the last three months will be used. In addition, the research team will contact the GPs, physiotherapists, and hospital specialists in case patients have been referred, for treatment costs . Standard unit cost data will be derived from reliable published sources . The costs related to the intervention itself (application of US) will be assessed by detailed cost pricing.
Also in addition, it will be evaluated to what extent patients are blinded for the US findings and results prior to randomization, to what extent the introduction of US influences management decisions, what type of treatment is provided to what type of patient (patient characteristics), and what adherence rates to the initial applied treatment can be shown.
The sample size calculation is based on a recovery rate (measured by GPE) of 60% in the control group  and 80% in the intervention group after 52 weeks, a two sided-alpha of 0.05, a statistical power of 0.80, and a drop-out rate at 52 weeks of 10% [45, 46]. We need 90 patients per study group to detect the difference of 20% between the study groups after 52 weeks; the minimal clinical important difference. With the expectation that the Qualification period filters out 20% of the patients, we need to include 226 patients. With up to 100 new patient encounters per GP per year [1–4] and approximately 25 eligible patients per GP per year, two years of recruitment, 50% consent to the study, and 50% drop-out for other reasons, we need 20 GPs to participate to include the required 226 patients. Based on the reported prevalence, and our experience, we expect to encounter enough patients with symptomatic US pathology to complete recruitment within two years.
The primary analysis will be intention-to-treat and will compare the patient-perceived recovery measured by the Global Perceived Effect questionnaire (GPE) at 52 weeks after randomisation of patients managed by US tailored treatment (intervention group) and those having received care as usual (control group). In order to study the influence of protocol violations on the study outcomes, a per protocol analysis will be performed. Patients with documented deviations from the study protocol (that is no adherence to the treatment steps mentioned in Additional file 2) will be excluded from this analysis.
Continuous variables will be presented as mean ± standard deviation and categorical variables as number (%). The longitudinal trend of primary and secondary parameters will be compared between the intervention and control group using logistic and linear mixed models to take into account the dependency of repeated measurements and nesting structure of data (patients within GP practices). Baseline characteristics that a priori are considered to be possible prognostic factors (fast or gradual development of SP, possible cause of SP, dominant shoulder affected, concomitant neck complaints, and physical work with upper extremities), will be included in the mixed models.
An economic evaluation will be performed from both a health care and societal perspective with a time horizon of 52 weeks. The incremental cost-effectiveness ratios will be expressed as the costs per additionally recovered patient (from a health care perspective) and the costs per Quality Adjusted Life Years (from a societal perspective). Sensitivity analyses will be performed to assess the influence of relevant factors. Finally, bootstrap analysis will be performed to quantify the uncertainty surrounding the incremental costs and effects. Based on these results, a Cost-Effectiveness Acceptability Curve will be constructed to show the probability that the intervention is cost-effective.