To determine whether current care for common shoulder problems in Australian general practice is in keeping with rheumatologist expectations and the best available evidence.
We performed a mailed survey of a random sample of 3500 Australian GPs and an online survey of all 270 rheumatologists in Australia in June 2009. Each survey included four vignettes (first presentation of shoulder pain due to rotator cuff tendinopathy, acute rotator cuff tear in a 45 year-old labourer and early and later presentation of adhesive capsulitis). For each vignette, GPs were asked to indicate their management, rheumatologists were asked to indicate appropriate primary care, and we determined best available evidence from relevant Cochrane and other systematic reviews and published guidelines.
Data were available for at least one vignette for 614/3500 (17.5%) GPs and 64 (23.8%) rheumatologists. For first presentation of rotator cuff tendinopathy, 69% and 82% of GPs and 50% and 56% rheumatologists would order a shoulder X-ray and ultrasound respectively (between group comparisons P = 0.004 and P<0001). Only 66% GPs and 60% rheumatologists would refer to an orthopaedic surgeon for the acute rotator cuff tear. For adhesive capsulitis, significantly more rheumatologists recommended treatments of known benefit (e.g. glucocorticoid injection (56% versus 14%, P<0.0001), short course of oral glucocorticoids (36% versus 6%, p<0.0001) and arthrographic distension of the glenohumeral joint (41% versus 19%, P<0.0001).
Citation: Buchbinder R, Staples MP, Shanahan EM, Roos JF (2013) General Practitioner Management of Shoulder Pain in Comparison with Rheumatologist Expectation of Care and Best Evidence: An Australian National Survey. PLoS ONE 8(4): e61243. https://doi.org/10.1371/journal.pone.0061243
Editor: Monica da Silva Nunes, Universidade Federal do Acre (Federal University of Acre), Brazil
Received: November 9, 2012; Accepted: March 7, 2013; Published: April 16, 2013
Copyright: © 2013 Buchbinder et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The Shepherd Foundation funded this study. RB is supported in part by an Australian National Health & Medical Research Council Practitioner Fellowship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Shoulder pain is a common reason for seeking medical care. The annual prevalence and incidence of care-seeking for shoulder pain is reported to be 2.4% and 1–1.5% respectively , . To encourage consistent and optimal care for patients with shoulder disorders, numerous Cochrane and other systematic reviews have synthesised evidence from the research literature –, with much of this evidence being incorporated into evidence-based summaries, guidelines and recommendations –. Yet the task of incorporating the latest evidence from research studies into clinical practice remains enormous.
Studies that have examined the extent of uptake of best available evidence into routine GP care for shoulder pain suggest a lack of confidence in making specific diagnoses, and an overreliance on early imaging and early specialist referral , , –. For example, a 2006 US survey of primary care physicians reported that less than half of those who completed an on-line questionnaire about appropriate management of specific shoulder conditions gave responses consistent with the treatment algorithms devised by the American Academy of Orthopaedic Surgeons (AAOS) . Forty-four percent chose treatment options for adhesive capsulitis consistent with these algorithms, and only 29% chose the recommended management for chronic rotator cuff symptoms.
Knowledge and beliefs, patient expectation, the availability of allied health services, access to investigations and specialists, receipt of continuing medical education (CME) and special interests may all influence uptake of evidence , . However the direction of effect may sometimes be unexpected. For example, a recent study noted GPs with a special interest in low back pain were apt to have poorer evidence-based beliefs about back pain and poorer management approaches compared with GPs without a special interest .
Recently the Australian National Health & Medical Research Council (NHMRC) have established a Research Translation Faculty tasked with identifying important gaps between research evidence and health policy and practice among 17 major health issues, and recommending to NHMRC the action needed to address the gap . A major health issue in Australia is ‘Arthritis and Musculoskeletal Conditions’, one of nine Australian National Health Priority Areas (NHPAs) .
To determine whether there is an important evidence-practice gap for management of common shoulder problems in primary care in Australia, we used GP surveys to determine the current pattern of care for shoulder pain in Australian general practice. We also performed a rheumatologist survey to determine Australian rheumatologist expectations of GP care. Australian rheumatologists consider themselves to be experts in the diagnosis and management of rheumatic and musculoskeletal disorders including all forms of arthritis, connective tissue disease, spinal and soft tissue disorders, chronic musculoskeletal pain syndromes and certain metabolic bone disorders, particularly osteoporosis . Therefore the rheumatologist survey would also identify any important evidence-practice gaps among rheumatologists. Responses from both physician groups were compared to the best available evidence, and the level of agreement between GPs and rheumatologist expectations of GP care were assessed. These data may be of use in developing a case for NHMRC action and determining whether general or more targeted approaches may be needed to bridge any detected evidence-practice gaps.
Materials and Methods
We performed a mailed survey of a random sample of Australian GPs identified through the Australasian Medical Publishing Company (AMPCo), the publisher of the Medical Directory of Australia  in June 2009. A cover letter explained that the purpose of the study was to understand how GPs currently manage patients with shoulder pain and that the data obtained would be valuable for informing how the management of this common and challenging problem could be improved. To maintain confidentiality AMPCo mailed individually numbered questionnaires to GPs and the questionnaires did not contain identifying information. Several strategies, based upon a modified Dillman technique , were used to maximise response rate including use of a respondent-friendly design (coloured background fields with white answer spaces and prominent question numbers); a pre-notice letter explaining the reasons for the survey; and a postcard thank you/reminder 3–4 days after sending the questionnaire. GPs were also advised that returned questionnaires would go into a draw for seven $50 book vouchers.
The GP questionnaire was modified from one developed to assess GP management of back pain in Australia ,. It included four vignettes modified from a Canadian study assessing GP management of common musculoskeletal problems , and the US study described above . The vignettes present a 77 year-old female with a six-week history of shoulder pain consistent with rotator cuff tendinopathy; a 45 year-old labourer with clinical features consistent with an acute rotator cuff tear; a 50 year old female with a three week history compatible with adhesive capsulitis; and the same patient presenting three months later with persistent symptoms (Appendix S1). The response options for the scenarios were listed under categories of investigations, consultation management including advice and medications and referrals, and we asked doctors to indicate any they would routinely advocate. We also asked doctors to indicate the likely prognosis and the likelihood that surgery would be required as either very likely, likely, not sure, unlikely or very unlikely. Demographic details and training, membership of societies and associations, special interests, preferences and sources for keeping up to date, and practice characteristics were also collected. The IDs of seven GPs were drawn at random from all returned questionnaires and $50 book vouchers sent to AMPCo for forwarding to these GPs.
We performed an online survey of all currently practicing rheumatologists in Australia as indicated by their membership of the Australian Rheumatology Association (ARA) in June 2009. Responses were confidential. The ARA approved the survey and sent all clinical members a link to SurveyMonkey. To maximise response rate, reminder emails were sent 2 and 6 weeks later. Respondents could leave their name and contact details separately to go into a draw for a raffle for one of three book vouchers valued at $50 each.
The rheumatologists were presented with the same four vignettes and asked to indicate what management they would routinely advocate in primary care. Demographic details, medical training, preferences for keeping up to date, and practice characteristics were also collected.
The Cabrini Health and Monash University Human Research Ethics Committees approved the study.
Determination of Best Evidence-based Care
To determine whether participant responses for each of the scenarios were consistent with the best available evidence we compared responses to relevant Cochrane and other systematic reviews and published guidelines –, , . Two publications were of particular relevance to the Australian context, both written by a multidisciplinary panel and based upon evidence: the NHMRC-endorsed Evidence-based Management of Acute Musculoskeletal Pain Guidelines published in 2003  and Therapeutic Guidelines: Rheumatology version 2 . Therapeutic Guidelines is a widely respected and available resource for GPs in Australia. The first Rheumatology version was released in April 2006 and an update was released in November 2010, both endorsed by multiple organisations including the ARA, The Royal Australian College of General Practitioners and the Australian Physiotherapy Association. Appendix S2 presents a summary of our synthesis of the best available evidence as it applies to the four vignettes.
We estimated that responses from at least 103 GP and 103 rheumatologists would be required to detect a difference of 20% between the two groups in the proportion of agreeing with the specified options for investigations, management or referral (5% significance level, 80% power). Allowing for around 80% non-response from GPs, we approached a random sample of 3500 to recruit 670 GPs. We invited all 270 practising rheumatologists in Australia to complete the online survey requiring a response rate of 38% to achieve a total of 103 rheumatologists.
Descriptive statistics were used to summarize the demographics and practice characteristics of study participants. The proportion of responses for the management of each scenario was determined for both the GPs and the rheumatologists. For the prognosis items, we calculated the proportion of respondents who indicated the outcome was either ‘likely’ or ‘very likely’. Comparisons of proportions were analysed with a chi squared (χ2) test of independent samples. Because of the multiplicity of comparisons we have reported p-values <0.05 but have commented only on those with a p-value <0.001. Comparisons between doctors with or without recent CME in musculoskeletal disorders were evaluated in the same way. Comparisons between doctors with and without a special interest in musculoskeletal disorders and/or occupational medicine were also planned but were not performed because of too few numbers.
All statistical analyses were performed using Stata/IC 11.0.
Of the 3500 questionnaires mailed to GPs, 639, including 25 blank forms, were returned. Three GPs did not answer the demographic questions but completed at least one vignette giving a response rate for completing at least one vignette of 17.5% (614/3500). Eighty rheumatologists opened the survey link in the email; one did not see general rheumatology patients and was thus ineligible; nine did not answer any demographic questions or attempt any of the vignettes; the response rate for demographic questions was 26.0% (70/269). Six rheumatologists who completed the demographics did not attempt any of the vignettes. Data were available for at least one vignette from 64 rheumatologists (response rate 23.8%). Those who did not complete any vignettes did not differ from other respondents by gender (p = 0.96), year of graduation (p = 0.20), number of patients seen per week (p = 0.80) or number of patients with shoulder pain per week (p = 0.77).
Demographic and practice characteristics and top preferences for keeping up to date for GPs and rheumatologists are reported in Table 1. For both groups, about two–thirds of respondents were male and median duration of practice was 20 years. Just over a third of GPs had received CME in musculoskeletal disorders in the preceding two years. Journal articles were the top preference for keeping up to date among both GPs and rheumatologists.
Vignette 1: Rotator Cuff Tendinopathy
Tables 2, 3, 4 5 shows the management responses selected by GPs and rheumatologists for each of the vignettes. While the best available evidence suggests that in the absence of red flags no tests are necessary (see Appendix S1), only 10% of GPs and 34% of rheumatologists indicated no tests were necessary (between-group difference p<0.0001)(Table 2). The majority of GPs would order a shoulder X-ray (69%) and/or an ultrasound (82%) while at least half of the rheumatologists indicated shoulder X-ray and/or ultrasound should be performed.
Both GPs and rheumatologists were broadly consistent with Australian recommendations  to provide advice on activity, work modification and home exercise and about three-quarters of both groups would refer for physiotherapy. A third of both groups would prescribe NSAIDs despite the patient having failed a previous 2-week course. Over half the GPs recommended glucocorticoid injection, as did nearly all the rheumatologists (94%). However just under a quarter of GPs and a third of rheumatologists indicated this should be guided by imaging. While there is evidence for the short-term benefit of glucocorticoid injection, image guidance does not appear to provide added benefits (see Appendix S2). Appropriately, only a minority of GPs indicated they would refer to an orthopaedic surgeon or rheumatologist although 41% of rheumatologists indicated referral to rheumatology was appropriate (vs 5% GPs, P<0001).
Vignette 2: Acute Rotator Cuff Tear
Most GPs and rheumatologists appropriately indicated that further imaging and orthopaedic referral was required (Table 3, Appendix S2), and approximately half indicated that surgery was likely. Similar to Vignette 1, only 2% of GPs versus 49% of rheumatologists advocated referral to a rheumatologist (P<0.0001). A third of GPs and 39% of rheumatologists indicated they would prescribe low-dose antidepressants for this vignette. While anti-depressants are widely used to treat chronic painful musculoskeletal conditions , it is unclear what the rationale, if any, would be for acute pain.
Vignettes 3 and 4: Early and Late Presentation of Adhesive Capsulitis
Similar to vignette 1, the majority of GPs and rheumatologists would also routinely order X-rays (73% and 71% respectively) and ultrasound (74% and 58% respectively) for Vignette 3 S2). Almost half of both groups would order blood tests but only a minority specified blood glucose measurement, which would be a reasonable decision as diabetes is a known risk factor for this condition .
Significantly more rheumatologists recommended treatments of known benefit including glucocorticoid injection (56% versus 14% GPs, P<0.0001), a short course of oral glucocorticoids (36% versus 6% of GPs, P<0.0001), home exercises (71% vs 42% for vignette 3, P<0.0001), and referral for arthrographic distension of the glenohumeral joint for the later presentation (41% versus 19% of GPs, P<0.0001). Similar to the other vignettes, significantly less GPs would refer the patient to a rheumatologist for vignettes 3 and 4 and significantly more would refer to orthopaedics.
The majority of GPs and rheumatologists indicated that recovery was likely or very likely to occur within one to two years in both vignettes 3 and 4 in keeping with the reported favourable course of adhesive capsulitis –. However a third of GPs and a quarter of rheumatologists erroneously believed that recurrence was either likely or very likely within 2 years although recurrences have not been reported –.
There were no differences between doctors who had or had not completed CME or postgraduate education about musculoskeletal disorders in the past two years (data not shown).
Based upon national surveys of GPs, the current pattern of care in Australian general practice for four common presentations of shoulder pain is highly variable and often suboptimal. In general, there was a high reliance on imaging, particularly plain radiographs and ultrasound, by both GPs and rheumatologists. For each vignette there was a wide variation in what treatments GPs would prescribe, what referrals would be made and the expected prognosis. This was often mirrored by a similar degree of variation in responses among rheumatologists and there were significant discrepancies between GP responses and rheumatologist expectation of GP care.
Our finding of a high dependence upon plain radiographs and ultrasound in all scenarios is in keeping with previous studies , , , . Although a Canadian multidisciplinary panel did consider a plain radiograph standard of care for vignette 1 in 1998 , no recent guidelines advocate imaging for shoulder pain unless there is a suggestion of serious pathology , –. The results of plain radiographs are unlikely to influence GP management and there are risks associated with radiation exposure.
The diagnostic utility of shoulder ultrasound in primary care is unknown yet in Australia there has been a rapid rise in its use , . The procedure is highly operator-dependent, asymptomatic changes on ultrasonography are common and increase with age , and many observed abnormalities might not require specific treatment . This has led to concerns that ultrasound findings might be misleading and result in inappropriate and/or delay correct diagnosis and management . One study that reviewed a set of ultrasound request forms found that the motivation for the request was often unclear with a third containing no contributory information for the radiologist .
Concomitant with the rise in diagnostic ultrasound for shoulder pain, referral for image-guided injections has risen dramatically in the last decade . Although the exact number of ultrasound-guided shoulder injections being performed in Australia is unknown as the Medicare item number includes other regions of the body, there has been a more than 34-fold increase in the number of ultrasound-guided injections billed to Medicare in the last ten years (3504 per 100,000 population in the 2000–2001 financial year to just over 120,000 per 100,000 population in the 2010–2011 financial year) . Interpretation of this data is also hampered by the fact that government subsidy of anatomically guided joint injections was discontinued in November 2009 which may explain some of the increased use of image-guided injections.
Our finding of a high reliance upon specialist referral is in keeping with some , , but not all , , studies. Although many rheumatologists in our study considered specialist referral appropriate for new presentation of non-traumatic shoulder pain, a lack of GP confidence to diagnose and manage common shoulder complaints may lead to unwarranted or premature specialist referrals , . For example, a national UK study found that patients presenting with shoulder complaints to general practice are not assigned a specific diagnosis . A recent Australian study also found that only 40% of GP referral letters to an orthopaedic clinic offered a diagnosis . GPs and rheumatologists in our study also differed markedly with respect to preferences for specialist referral. This is in keeping with a 1985 US study that found that GPs regarded orthopaedic surgeons more effective than rheumatologists in treating acute shoulder pain and shoulder tendinitis .
Despite the availability of research evidence to inform clinical practice for shoulder pain in primary care, our study indicates an ‘evidence-practice’ gap among both GPs and rheumatologists. Interventions are needed to improve diagnosis and treatment of common shoulder problems in primary care and to reduce inappropriate and costly imaging and specialist referral. Several interventions to change physician management of shoulder pain have shown promising results. Academic detailing has shown favourable results with respect to reducing imaging requests and improving GP knowledge and confidence in managing shoulder problems in one before-after study . A randomised controlled trial found that provision of practical training that included injection techniques to GPs did not improve patient shoulder-specific disability or pain at one year when compared to no training . Another non-randomised controlled trial found that a short skills training course that included shoulder injection technique improved scores on a 5-item knowledge test 3 months after the training and increased use of glucocorticoid injection . Joint consultations with GPs and orthopaedic surgeons were found to reduce referral rates when compared to usual care in a randomised controlled trial . Finally, the development and dissemination of local referral guidelines to a musculoskeletal service in the UK appeared to improve appropriate referrals and reduce waiting times .
The strengths of our study include our robust survey methodology, assurance of de-identified data, and assessment of GP management from the perspective of both Australian rheumatologists and the published literature. On the other hand, we measured stated rather than actual practice. However while the relationship between behavioural simulation and direct measures of clinician behaviour is not perfect , , there is some evidence to support the use of vignettes as a proxy measure for clinician behaviour . While patient preference and/or perceived patient pressure are known to be important influences in clinical practice , this may have been less likely to influence stated management responses to paper vignettes.
Furthermore, the validity of our findings is supported by the demonstrably high use of imaging and image-guided injection for shoulder pain in Australia and its consistency with studies performed elsewhere. Some of the variation in GP and rheumatologist responses we observed could be related to local and/or regional set up of services and access to specialists. However our findings of a general overreliance on imaging and referral would seem to indicate that lack of access was not really a factor in GP stated responses.
Another potential limitation of our study is that the generalisability of results is limited by our low response rate. Unfortunately due to issues of confidentiality we were unable to determine whether there were major differences between responders and non-responders. However in comparing the demographics of respondents with data from the Australian GP workforce study 2001 to 2004, respondents were of similar sex (males: 62% vs 63% in the Australian GP workforce) but older age (mean: approximately 53 years based upon year of graduation vs 48.6 years in the Australian GP workforce), and a lesser proportion were working in rural settings (19.4% versus 26.6%) . As well, while it is possible that GPs who have had CME in musculoskeletal disorders in the past two years may have been more inclined to complete the survey, we found no substantial differences in management between GPs with and without a history of CME.
Another limitation may have been our decision, in keeping with other studies , , , not to ask respondents to stipulate their diagnosis. It is therefore not possible to separate diagnostic from management acumen although it is likely that these skills are highly interdependent. For example, the low uptake of both intra-articular and oral glucocorticoids among GPs for adhesive capsulitis may be a result of misdiagnosis of the vignette, lack of awareness of these treatment approaches, lack of confidence in injection techniques and/or concern about potential harms. It would also be of relevance to obtain the perspective of orthopaedic surgeons.
In summary, there is a mismatch between the stated management of common shoulder problems encountered in primary care by GPs, rheumatologist expectations of GP care and the available evidence. Our results indicate the need for concerted collaborative research and policy endeavours directed towards improving the dissemination and uptake of evidence into practice.
Vignettes for the acute and chronic presentations of shoulder pain.
Conceived and designed the experiments: RB MPS. Performed the experiments: RB JFR. Analyzed the data: RB MPS EMS JFR. Contributed reagents/materials/analysis tools: RB JFR. Wrote the paper: RB MPS EMS JFR.
- 1. Linsell L, Dawson J, Zondervan K, Rose P, Randall T, et al. (2006) Prevalence and incidence of adults consulting for shoulder conditions in UK primary care; patterns of diagnosis and referral. Rheumatol 45: 215–221.
- 2. Charles J, Britt H, Fahridin S (2007) Chronic musculoskeletal problems managed in general practice. Aust Fam Phys 36: 392–393.
- 3. Green S, Buchbinder R, Glazier R, Forbes A (1998) Systematic review of randomised controlled trials of interventions for painful shoulder: selection criteria, outcome assessment, and efficacy. Br Med J 316: 354–360.
- 4. Green S, Buchbinder R, Hetrick S (2003) Physiotherapy interventions for shoulder pain. Cochrane Database System Rev 2: CD004258.
- 5. Green S, Buchbinder R, Hetrick S (2005) Acupuncture for shoulder pain. Cochrane Database System Rev 2: CD005319.
- 6. Buchbinder R, Green S, Youd J, Johnston RV, Cumpston M (2008) Arthrographic distension for adhesive capsulitis (frozen shoulder). Cochrane Database System Rev 1: CD007005.
- 7. Buchbinder R, Green S, Youd JM (2003) Corticosteroid injections for shoulder pain. Cochrane Database System Rev 1: CD004016.
- 8. Buchbinder R, Green S, Youd JM, Johnston RV (2006) Oral steroids for adhesive capsulitis. Cochrane Database System Rev 4: CD006189.
- 9. Cumpston M, Johnston RV, Wengier L, Buchbinder R (2009) Topical glyceryl trinitrate for rotator cuff disease. Cochrane Database System Rev 3: CD006355.
- 10. Coghlan JA, Buchbinder R, Green S, Johnston RV, Bell SN (2008) Surgery for rotator cuff disease. Cochrane Database System Rev 1: CD005619.
- 11. van der Heijden GJ, van der Windt DA, de Winter AF (1997) Physiotherapy for patients with soft tissue shoulder disorders: a systematic review of randomised clinical trials. Br Med J 315: 25–30.
- 12. van der Heijden GJ, van der Windt DA, Kleijnen J, Koes BW, Bouter LM (1996) Steroid injections for shoulder disorders: a systematic review of randomized clinical trials. Br J Gen Pract 46: 309–316.
- 13. Favejee M, Huisstede B, Koes B (2010) Frozen shoulder: the effectiveness of conservative and surgical interventions– systematic review. Br J Sports Med 45: 49–56.
- 14. Murphy RJ, Carr AJ. (2010) Shoulder pain. Clin Evid (Online) pii: 1107.
- 15. Australian Acute Musculoskeletal Pain Guidelines Group. (2003) Evidence-based management of acute musculoskeletal pain. Brisbane: Australian Academic Press Pty Ltd.
- 16. Philadelphia Panel (2001) Philadelphia panel evidence-based clinical practice guidelines on selected rehabilitation interventions for shoulder pain. Phys Ther 81: 1719–1730.
- 17. Expert Rheumatology Group. (2010) Therapeutic guidelines: rheumatology. version 2 Melbourne: Therapeutic Guidelines Limited.
- 18. Geraets J, de Jongh A, Boeke A, Buis PA, Spinnewijn WE, et al. (2009) Summary of the practice guideline for shoulder complaints from the Dutch College of General Practitioners. Ned Tijdschr Geneeskd 153: A164.
- 19. Bussières A, Peterson C, Taylor J (2007) Diagnostic imaging practice guidelines for musculoskeletal complains in adults - An evidence-based approach: Introduction. J Manip Physiol Ther 30: 617–683.
- 20. Glazier RH, Dalby DM, Badley EM, Hawker GA, Bell MJ, et al. (1998) Management of common musculoskeletal problems: a survey of Ontario primary care physicians. Can Med Assoc J 158: 1037–1040.
- 21. Britt H, Miller G, Knox S. (2001) Shoulder syndrome. Canberra: Australian Institute of Health and Welfare. Contract No.: AIHW Cat. No. GEP 7 General Practice Series No.7.
- 22. Solomon DH, Bates DW, Schaffer JL, Horsky J, Burdick E, et al. (2001) Referrals for musculoskeletal disorders: patterns, predictors, and outcomes. J Rheumatol 28: 2090–2095.
- 23. Broadhurst NA, Gialamas A, McElroy HJ, Beilby JJ (2004) How do Australian GPs manage shoulder dysfunction? Aust Fam Phys 33: 861–862.
- 24. Loebenberg MI, Rosen JE, Ishak C, Jazrawi LM, Zuckerman JD (2006) A survey of decision-making processes in the treatment of common shoulder ailments among primary care physicians. Bull Hosp Jt Dis 63: 137–144.
- 25. Johal P, Martin D, Broadhurst N (2008) Managing shoulder pain in general practice - assessment, imaging and referral. Aust Fam Phys 37: 263–265.
- 26. Patel S, Hossain F, Colaco H, El-Husseiny M, Lee MH (2011) The accuracy of primary care teams in diagnosing disorders of the shoulder. J Eval Clin Pract 17: 118–122.
- 27. American Academy of Orthopaedic Surgeons (1999) Clinical Guideline on Shoulder Pain. Rosemont, IL: American Academy of Orthopaedic Surgeons.
- 28. Zwolsman S, te Pas E, Hooft L, Wieringa-de Waard M, van Dijk N (2012) Barriers to GPs’ use of evidence-based medicine: a systematic review. Br J Gen Pract 62: e511–e521.
- 29. Buchbinder R, Staples M, Jolley D (2009) Doctors with a special interest in back pain have poorer knowledge about how to treat back pain. Spine 34: 1218–1226.
- 30. National Health and Medical Research Council. Research Translation Faculty. http://www.nhmrc.gov.au/research-translation/research-translation-faculty (access Mar 2013).
- 31. Australian Institute of Health and Welfare and Commonwealth Department of Health and Family Services. (1997) First report on National Health Priority Areas, full report. Canberra: AIHW and DHFS. http://www.aihw.gov.au/publications/index.cfm/title/121 (accessed Mar 2013).
- 32. Australian Rheumatology Association. About Advanced training in Rheumatology. http://www.rheumatology.org.au/rheumatologists/training.asp (accessed Mar 2013).
- 33. AMPCo (1997) Medical Masterfile of Australia. Sydney (NSW): Australasian Medical.
- 34. Dillman D (1978) Mail and telephone surveys - The total design method. New York: John Wiley & Sons.
- 35. Buchbinder R, Jolley D, Wyatt M (2001) Population based intervention to change back pain beliefs and disability: three part evaluation. Brit Med J 322: 1516–1520.
- 36. Buchbinder R, Jolley D, Wyatt M (2001) 2001 volvo award winner in clinical studies: effects of a media campaign on back pain beliefs and its potential influence on management of low back pain in general practice. Spine 26: 2535–2542.
- 37. American Academy of Orthopaedic Surgeons (2010). American Academy of Orthopaedic Surgeons Clinical Practice Guideline on optimizing the management of rotator cuff problems: Guidelines and evidence report. Rosemont IL: AMerican Academy of Orthopaedic Surgeons.
- 38. Perrot S, Maheu E, Javier R, Eschalier A, Coutaux A (2006) Guidelines for the use of antidepressants in painful rheumatic conditions. Eur J Pain 10: 185–192.
- 39. Bridgman J (1972) Periarthritis of the shoulder and diabetes mellitus. Ann Rheum Dis 31: 69–71.
- 40. Reeves B (1975) The natural history of the frozen shoulder syndrome. Scand J Rheumatol 4: 193–196.
- 41. Binder AI, Bulgen DY, Hazleman BL, Roberts S (1984) Frozen shoulder: a long-term prospective study. Ann Rheum Dis 43: 361–364.
- 42. Hand C, Clipsham K, Rees JL, Carr AJ (2008) Long-term outcome of frozen shoulder. J Should Elbow Surg 17: 231–236.
- 43. Broadhurst N, Baghurst T, MacLaren S (2004) Ultrasound imaging for shoulder pain in general practice. Aust Fam Phys 33: 668–669.
- 44. Awerbuch M (2008) The clinical utility of ultrasonography for rotator cuff disease, shoulder impingement syndrome and subacromial bursitis. Med J Aust 188: 50–53.
- 45. Milgrom C, Schaffler M, Gilbert S, van Holsbeeck M (1995) Rotator-cuff changes in asymptomatic adults. The effect of age, hand dominance and gender. J Bone Joint Surg Br 77: 296–298.
- 46. Medicare Australia (2011). Medicare Item Reports: Medicare Australia.
- 47. Speed CA, Crisp AJ (2005) Referrals to hospital-based rheumatology and orthopaedic services: seeking direction. Rheumatol 44: 469–471.
- 48. van der Windt DA, Koes BW, Boeke AJ, Deville W, De Jong B, et al. (1996) Shoulder disorders in general practice: prognostic indicators of outcome. Brit J Gen Pract 46: 519–523.
- 49. Rao C, Halsey JP, Bukhari MA, Dodds WN, Mitchell W (2005) Referrals to hospital-based rheumatology and orthopaedic services: seeking direction. Rheumatol 44: 1332–1333.
- 50. Felson DT, Meenan RF, Dayno SJ, Gertman P (1985) Referral of musculoskeletal disease patients by family and general practitioners. Arthritis Rheum 28: 1156–1162.
- 51. Broadhurst NA, Barton CA, Rowett D, Yelland L, Matin DK, et al. (2007) A before and after study of the impact of academic detailing on the use of diagnostic imaging for shoulder complaints in general practice. BMC Fam Pract 8: 12.
- 52. Watson J, Helliwell P, Morton V, Adebajo A, Dickson J, et al. (2008) Shoulder acute pain in primary healthcare: is retraining effective for GP principals? SAPPHIRE–a randomized controlled trial. Rheumatol 47: 1795–1802.
- 53. Jansen J, Grol R, van der Vleuten C, Scherpbier AJ, Crebolder HF, et al. (2000) Effect of a short skills training course on competence and performance in general practice. Med Educ 34: 66–71.
- 54. Vierhout WP, Knottnerus JA, van Ooij A, Crebolder H, Popp P, et al. (1995) Effectiveness of joint consultation sessions of general practitioners and orthopaedic surgeons for locomotor-system disorders. Lancet 346: 990–994.
- 55. Godin G, Belanger-Gravel A, Eccles M, Grimshaw J (2008) Healthcare professionals’ intentions and behaviours: A systematic review of studies based on social cognitive theories. Implement Sci 3: 36.
- 56. Eccles M, Hrisos S, Francis J, Kaner E, Dickinson H, et al. (2006) Do self-reported intentions predict clinicians’ behaviour: a systematic review. Implement Sci 1: 28.
- 57. Hrisos S, Eccles MP, Francis JJ, Dickinson H, Kaner E, et al. (2009) Are there valid proxy measures of clinical behaviour? a systematic review. Implement Sci 4: 37.
- 58. Little P, Dorward M, Warner G, Stephens K, Senior J, et al. (2004) Importance of patient pressure and perceived pressure and perceived medical need for investigations, referral, and prescribing in primary care: nested observational study. Br Med J 328: 444.
- 59. Australian Medical Workforce Advisory Committee (2005), The General Practice Workforce in Australia: Supply and Requirements to 2013, AMWAC Report 2005.2, Sydney.
- 60. Glazier R, Dalby D, Badley E, Hawker GA, Bell MJ, et al. (1996) Determinants of physician confidence in the management of musculoskeletal disorders. J Rheumatol 23: 351–356.