Objective. To determine the support services general practitioners (GPs) need when working with patients who have musculoskeletal problems and their ...
Rheumatology 2002;41:503–508
Improving the quality of care of musculoskeletal conditions in primary care C. Roberts, A. O. Adebajo and S. Long1 Department of Medical Education, University of Sheffield, Coleridge House and 1 Institute of General Practice and Primary Care, Community Sciences Building, Northern General Hospital, Herries Road, Sheffield S5 7AU, UK Abstract Objective. To determine the support services general practitioners (GPs) need when working with patients who have musculoskeletal problems and their educational needs in order to deliver an improved service. Methods. GPs (n = 446) on the Sheffield and Barnsley principal lists were sent a questionnaire (53.8% response rate). Semistructured interviews of a purposive sample of 10 GPs were analysed qualitatively to increase understanding of the research objectives. Results. GPs were self-confident in managing common musculoskeletal conditions such as gout (86% of GPs who replied), back pain (69%), osteoarthritis (62%) and sporting injuries (58%) entirely within the surgery. Despite high levels of confidence in diagnosing non-specific pain syndromes, 68% of GPs would refer to a rheumatologist. Most GPs (68%) were happy with their current referral rates to physiotherapists and 65% of GPs in this sample provided a personal injection service. Reduction of inappropriate prescribing of non-steroidal anti-inflammatory drugs would be helped by better patient education materials on treatments (90%) and more resources for the primary care physiotherapy service (85%). Half of the GPs had had specific musculoskeletal training within the last 5 yr. Half of the GPs planned to update their knowledge and skills in the next year, 64% of these preferring a taught interactive course, 50% wanting to sit in with a consultant in clinic and 46% preferring to learn as part of a personal learning plan. Conclusions. GPs feel confident managing the majority of musculoskeletal conditions within the surgery provided they have adequate support in terms of opportunities for appropriate education, particularly joint injection techniques, ongoing consultant support for complex cases with poor outcomes, particularly non-specific pain syndromes, adequate access to physiotherapy, and a multidisciplinary approach to pain control and inappropriate prescribing. KEY WORDS: Musculoskeletal, Continuing professional development, Rheumatology, General practitioners.
Musculoskeletal disorders are a common cause of long-term disability in the UK w1x, making up 15% of the workload of general practitioners (GPs). Epidemiological studies suggest that there is a huge reservoir of patients with significant musculoskeletal disorders who do not consult the health service w2x. Of those who do, inappropriate referrals from primary care are widely considered to be a source of overload on secondary care consultants, for example distracting rheumatologists from patients with inflammatory and connective tissue disorders w3x. New initiatives to manage the demand for musculoskeletal disorders are largely at the primaryu secondary care interface rather than within primary care and include nurse-led secondary care rheumatology
clinics w4x, demand management by prioritizing primary care referrals w5x, an acute rheumatology service to GPs and other clinical units w6x, and triage of orthopaedic referrals by trained physiotherapists w7x. The education and training of GPs may not match demand from patients with musculoskeletal disorders, and this may stem from inadequate teaching at medical school, particularly in clinical skills w8x, evidenced by a low awareness of musculoskeletal disorders amongst junior medical staff w9x. At the vocational training level w10x, a lack of confidence in managing rheumatological conditions is related to a lack of experience and poor teaching, although a third of UK GP registrars had injected a shoulder joint and over three-quarters had aspirated knees during their training. As a consequence, GPs may well manage patients suboptimally; in particular w11x, they are not referring enough patients for physiotherapy, they are prescribing too many
Submitted 11 July 2001; revised version accepted 2 November 2001. Correspondence to: C. Roberts.
503
ß 2002 British Society for Rheumatology
504
C. Roberts et al.
non-steroidal anti-inflammatory drugs (NSAIDS) and are reluctant to use intra- and peri-articular steroid injections. In addition, there is concern by consultant rheumatologists that GPs may be failing to address the importance of disease-modifying drugs (DMARDs) in early rheumatoid arthritis w12x. Whilst there may be agreement about the education needed for the optimal management of individual patients w3x, there is conflicting evidence to guide the optimal performance of GPs over a range of musculoskeletal conditions and still less as to which educational interventions may improve performance. Consequently, service provision remains diverse rather than integrated. For example, both physiotherapy and chiropractic have been shown to reduce symptoms w13x in low back pain, whereas there is inadequate evidence that NSAIDs are superior to simple analgesics such as paracetamol w14x. Nevertheless, in back pain, GPs tend to use NSAIDs in 22% of patients, referring 16% to physiotherapy w15x. Evidence-based guidelines have emerged for managing low back pain, but there is variable adherence by GPs w16x. The evidence on the use of corticosteroid injections administered by the GP may be incomplete. Injections are said to be more effective than physiotherapy for treating a painful stiff shoulder w17x, but the effect only lasts 2–3 months and the injection may need repeating. Whilst the majority of consultant rheumatologists have been involved in teaching GPs w18x, arthritides have made up nearly half of the teaching topics and little attention has been paid to common problems, such as vague soft-tissue pain, back pain, minor injuries and the examination and injection of joints. The present research had three objectives: (i) to identify the learning needs and preferred learning methods of GPs in musculoskeletal medicine in relation to their current ways of working; (ii) to identify innovations which may support primary care management of common musculoskeletal problems; and (iii) to identify ways in which GPs’ perceived barriers to good practice in musculoskeletal medicine might be overcome.
Setting and subjects All 446 GPs on the Sheffield and Barnsley principal lists were sent a questionnaire in September 1999. At the time of the study, in addition to departments of rheumatology and orthopaedics in each centre there was an orthopaedic physician employed at a hospital trust in Sheffield and a metabolic bone unit that took referrals from both Sheffield and Barnsley GPs. Methods Our review of the literature had identified a number of themes pertinent to the research objectives. Open-ended interviews were conducted with a sample of four GPs, two consultant rheumatologists and a sports physician, and were structured so that questions were asked that were relevant to the refining of the developing
questionnaire and ensuring content validity. The draft questionnaire was piloted on 10 randomly selected GPs from an adjacent health authority. The final form requested information on referral patterns, perceived workload and self-rating of the usefulness of a range of educational interventions designed to meet GPs’ possible education needs with regard to musculoskeletal disorders. In addition, five-response Likert-type items were used to explore the self-assessment of confidence in managing common musculoskeletal problems, barriers to accessing physiotherapy services, prescribing nonsteroidals by GPs, and the provision of intra-articular injections. Non-responders were sent one postal reminder after 2 weeks. Responses received after the closing date were not analysed. Statistical analysis was undertaken with SPSS version 10 (SPSS, Chicago, IL, USA). In-depth semistructured interviews were conducted in their places of work with a purposive sample of GPs selected on the basis of the free text in the questionnaires returned by the GPs. All interviews were audio-recorded in full with the subjects’ consent, tapes were transcribed, and thematic analysis was carried out using standard qualitative methods w19x.
Results Of the 446 GPs on the Barnsley and Sheffield principal lists, 240 replied following a reminder, giving a 53.8% response rate. Our sample consisted of 131 males and 109 females. Analysis of the non-respondents by gender showed no significant difference from responders (x2 test, P = 0.064). Workload In this survey musculoskeletal disorders made up 18.0% (S.D. 8.4%) (n = 217) of the GPs’ perceived workload. Referrals Although many patients presented ‘without a clear distinct diagnosis’ (GP: W), GPs perceived that musculoskeletal medicine should be based largely within primary care, but that more support was needed. ‘‘90% wof the musculoskeletal servicex is going to stay in primary care and doesn’t need secondary service, but we do need more support, more access to investigations and treatment, and colleagues to work with us. (GP: B)’’ GPs were generally self-confident about managing common musculoskeletal conditions (Table 1). They were happy to manage conditions such as gout (86.3% of GPs who replied), back pain (69.2%), osteoarthritis (61.7%) and sporting injuries (57.6%) entirely within the surgery. They tended to refer patients with early rheumatoid arthritis (34.6%), osteoporosis (18.3%) and polymyalgia rheumatica (17.9%) for a consultant opinion (Table 1), despite knowing the diagnosis. They
Improving primary care of musculoskeletal conditions
505
TABLE 1. Levels of self-confidence (%) of GPs in managing common musculoskeletal problems presenting in the surgery Confident to manage with own skills and knowledge Diagnosis Gout Back pain Osteoarthritis of the knees in the over-60s Minor sports injuries Painful shoulder Polymyalgia rheumatica Generalized jointu muscle aches, pains Osteoporosis Early rheumatoid arthritis
Confident to manage with advice from GP colleagues
Confident to manage with advice from the consultant
Confident of diagnosis, but would refer for management
Not confident about making diagnosis
n
%
n
%
n
%
n
%
n
%
n
240 238 239
86.3 69.2 61.7
207 166 148
7.9 18.8 13.3
19 45 32
2.9 6.7 16.3
7 16 39
2.9 2.1 6.7
7 5 16
0 2.5 1.7
0 6 4
237 238 240 232
56.7 46.3 39.2 29.6
136 111 94 71
21.7 30.4 10.0 22.9
52 73 24 55
12.9 13.8 27.9 27.5
31 33 67 66
3.3 3.8 17.9 9.6
8 9 43 23
4.2 5.0 5.0 7.1
10 12 12 17
239 240
28.8 16.7
69 40
13.8 10.4
33 25
35.8 34.6
86 83
18.3 34.2
44 82
2.9 4.2
7 10
TABLE 2. Patterns of referral to specialties from general practice (%) for common musculoskeletal conditions
Referring condition Early rheumatoid arthritis Polymyalgia rheumatica Generalized jointumuscle aches, pains Gout Painful shoulder Osteoporosis Osteoarthritis of the knees in the over-60s Back pain Minor sports injuries
Rheumatology
Orthopaedic surgeon
Orthopaedic physician
Other consultants
No referral
n
%
n
%
%
n
%
n
%
n
240 240 236
85.4 75.0 68.3
205 180 164
1.3 0.4 0.8
3 1 2
0 1.3 5.4
0 3 13
0 4.2 2.1
0 10 5
13.3 19.2 21.7
32 46 52
240 236 236 239
32.1 10.8 9.6 2.1
77 26 23 5
0 35.4 0 71.7
0 85 0 172
0.4 19.2 4.6 3.8
1 46 11 9
0.8 3.3 68.3 0.8
2 8 164 2
66.7 29.6 15.8 21.3
160 71 38 51
236 238
1.7 1.3
4 3
21.7 5.8
52 14
32.1 34.2
77 82
9.6 8.3
23 20
33.3 68.3
80 119
n
tended to refer early rheumatoid arthritis (85.4%), polymyalgia rheumatica (75%) and non-specific pain syndromes (68.3%) to rheumatologists (Table 2). They were more likely to send patients with back pain (32.1%) to an orthopaedic physician, patients with osteoarthritis of the knees (71.7%) and painful shoulder (35.4%) to an orthopaedic surgeon, and patients with osteoporosis (68.3%) to a metabolic bone clinic. Some GPs perceived that the specialties did not offer a service for some types of patient.
availability of advice to the GP from a consultant may increase self-confidence to manage patients with early rheumatoid disease at the GP practice (34.7%).
‘‘. . . common conditions such as fibromyalgia, pain syndromes, which I don’t think rheumatologists are terribly interested in the management of . . . tend to be sent back to the GP; the message is we can’t do any more, get on with it . . . people deserve more than that, because these are common conditions which cause lots of distress and disability. (GP: W)’’
‘‘It’s not relevant by the time they actually are offered an appointment; they’ve forgotten about it or they prefer to live with it. (GP: P)’’
In early rheumatoid arthritis, for which GPs’ confidence to manage with their own skills was low (16.7%), the
Physiotherapy Most GPs (67.5%) were happy with their own current referral rates to physiotherapists, 73% feeling confident about their own diagnostic skills and 68% of GPs reporting that their physiotherapy waiting list was too long.
Giving steroid injections Of the GPs in this sample, 64.6% provided a personal injection service, with 28.8% prepared to update or enhance their injection skills in the following year. Suggestions for improving the quality of a primary care injection service included regular ‘drop-in’ clinics for teaching injection skills at the local hospital (52.9%),
506
C. Roberts et al.
good access to evidence-based research findings on steroid injections (52.5%), fast-track service to consultants when injection treatment had been tried (51.7%), and locally produced guidelines on the practice of injections (51.1%). Those who did not give injections offered several reasons: little opportunity to learn (64.3%); not enough opportunity for regular practice to keep up skills (60.7%); medicolegal worries (45.3%); and lack of time (42.9%). Reducing the prescribing of NSAIDs GPs recognized that NSAIDs were ‘frequently implicated in ill health’ (GP: W) and ‘do not cure a lot of things’ (GP: S), but the reasons for prescribing NSAIDS were complex. For example, one doctor saw it as part of a gate-keeping role because of lack of other resources, such as physiotherapy. ‘‘. . . a lot of these things wNSAIDSx are given to give a bit of time and hope things settle down, because a lot of these things do settle down with time and rest. (GP: S)’’ In order to reduce the prescribing of NSAIDs, GPs thought some effective strategies might include better patient education materials about treatments (90%), more resources for the physiotherapy service (80.5%), easier access to a chronic pain service (79.3%) and better provision of educational courses on musculoskeletal medicine (77.9%). GP education GPs were happy to identify their learning needs to develop a better service but there was an emphasis on education being multidisciplinary, interactive and centred on primary care. GPs did not want a ‘standard lecture with a secondary care perspective . . . where the opportunities for feedback and discussion are very limited’ (GP: B). Those areas in which GPs were less confident in managing themselves but are nevertheless common would provide the basis of GPs’ future learning needs. These include the management of early rheumatoid arthritis, for which only 16.7% were prepared to manage with their own skill and knowledge, osteoporosis (28.8%), the management of the patient with widespread aches and pains (29.6%), and polymyalgia rheumatica (39.2%). Half of the GPs in this sample had had specific training on musculoskeletal medicine over the last 5 yr and 50% of the GPs planned to update their skills in the following year (Table 3), preferring a taught interactive musculoskeletal course (64.2%), to sit in with a consultant in clinic (49.2%), to do it as part of a personal learning plan (45.9%) or a Primary Care Group study day (31.2%). In addition to learning needs arising from lack of confidence, GPs requested training on injection techniques,
TABLE 3. Preferred learning methods for musculoskeletal disorders (%) for those intending to update knowledge and skills in the next year (n = 109)
Taught course Sit in with consultant colleague in clinic As part of a personal learning planuportfolio As part of a PCGa study day Work with guidelinesuprotocols for diagnosis and treatment Attend lectures by an expert Read educational articles in e.g. Update Taught by more knowledgeable colleagues at the practice Read academic papers in journals, e.g. BMJ Use educational material from Internet Self-teaching in surgery Other methods (not specified) a
%
n
64.2 49.5 45.9 31.2 33.0
70 54 50 34 36
30.3 26.6 23.9
33 29 26
20.2 14.7 14.7 6.4
22 16 16 7
PCG, Primary Care Group.
TABLE 4. Self-declared learning needs of those intending to update within coming year Learning need
n
Joint injections Clinical skills Sports injuries Evidence baseuguidelines Complementary therapy Inflammatory disorders Exploring the role of physiotherapy Miscellaneous Joint pains in the young
47 10 9 7 6 5 5 3 2
refreshing of clinical skills and the management of regional joint problems (Table 4). There was a proportion of GPs who had not updated their skills in the previous 5 yr and did not plan to do so in the coming year (13.8%).
Discussion This research suggests that GPs would challenge the widely held, largely secondary-care view that they were managing musculoskeletal problems poorly and overreferring. However, as there is little published evidence linking perceived competence with actual performance in primary care, the results should be viewed with caution. Additionally, the response rate may have produced sample bias, but falling GP response rates to postal questionnaires are a national problem w20x. If GPs’ self-assessment of their competence is accepted, what emerges is concern over a lack of resources in supporting services, for example in physiotherapy, and a failure of consultant colleagues to provide much support in the management of patients with traditionally poor outcomes, such as patients with non-specific pain syndromes, for which there is a high referral rate. There is favour for multidisciplinary approaches to (i) complex issues, such as reducing the prescribing of inappropriate
Improving primary care of musculoskeletal conditions
NSAIDs, for example by involving community pharmacists, and (ii) pain control in musculoskeletal conditions in primary care, for example by involving complementary therapies. There is a strong desire amongst GPs who provide a personal injection service to maintain and enhance their performance. The specific learning needs of GPs identified in this research need to be addressed by course organizers within short, interactive courses for small groups, aimed at maintaining and developing new knowledge and skills in musculoskeletal medicine. However, GPs need access to a range of learning material according to their personal learning style and preferred format. For example, GPs in this study recognized the need for access to up-to-date research-based evidence on good practice in musculoskeletal medicine. GPs could be encouraged to use such learning material as part of their own continuing professional development through personal learning plans. In addition, GPs may need to gain more awareness of the extensive range of patient educational material w21x and incorporate this into their consultations. GPs consider consultant colleagues to be an important resource for meeting their learning needs, but there is perhaps a disparity between what GPs need to learn about, i.e. managing common conditions well, and what rheumatologists may think is a priority, i.e. managing inflammatory disease and connective tissue disease w3x. Consultant rheumatologists may need to reassess the degree of their speciality interest in inflammatory disease, and place more emphasis on osteoarthritis, metabolic bone disease, low back pain and osteoporosis w22x. In the longer term, there may be more integrated working between primary and secondary care, driven by moves to provide teaching on the locomotor system in a multidisciplinary, community-based way, starting at medical school w23, 24x. Consultants, in conjunction with their units within the hospital trusts, may consider becoming more innovative in providing educational support for GPs who are happy in the management of common musculoskeletal problems in primary care. In particular, GP ‘specialists’ who provide an injection service will benefit from local guidelinesuprotocols and access to local drop-in injection clinics where they can update their skills. A service providing rapid access to consultants for patients in whom injection treatment has been tried appropriately but unsuccessfully may reduce the overall waiting time of patients from their initial presentation in primary care. For those primary care teams that do not offer an injection service, consideration needs to be given to mentorship by local multidisciplinary experts to promote uptake of the necessary training. There may be a place for clinical skills centres where novices can use simulation models to practice injection techniques. There may well be some learning needs for consultants to address for their own professional development. These would include being aware of the range and quantity of common conditions that GPs treat in their surgeries, the management of patients with non-specific
507
musculoskeletal pain syndromes where there are traditionally poor outcomes, and a recognition that many GPs are happy to manage inflammatory rheumatological conditions with advice and support. Innovative ways of providing support have been tried elsewhere, for example telephone helplines w25x. Finally, it is necessary to recognize the need for a multidisciplinary pain relief service in chronic conditions. This research suggests that the barriers to providing a musculoskeletal service in primary care are complex. Primary care trusts will need to consider the most effective configuration of services that meets their patients’ needs and is deliverable in an integrated way. This may cut across the traditional boundaries of orthopaedic surgery, rheumatology and the emerging specialties of orthopaedic medicine and sports medicine. GPs by and large seem willing to support this change and to adapt their own educational priorities to meet the challenge, but will need some direction from education providers.
Acknowledgements This study was supported by a grant from Barnsley Hospital Research Funds. The authors thank Dr N. Fox for suggesstions on study design, and M. Platts for data processing.
References 1. Martin J, Meltzer H, Eliot D. The prevalence of disability among adults. London: Office of Population Censuses and Surveys, 1998. 2. Urwin M, Symmons D, Allison T et al. Estimating the burden of musculoskeletal disorders in the community: the comparative prevalence of symptoms at different anatomical sites, and the relation to social deprivation. Ann Rheum Dis 1998;57:649–55. 3. Hosie GAC. Teaching rheumatology in primary care. Ann Rheum Dis 2000;59:500–3. 4. Hill J. Patient satisfaction in a nurse-led rheumatology clinic. J Adv Nurs 1997;25:347–54. 5. Hurst NP, Lambert CM, Forbes J, Lochead A, Major K, Lock P. Does waiting matter? A randomized controlled trial of new non-urgent rheumatology outpatient referrals. Rheumatology 2000;39:369–76. 6. Smith EC, Berry H, Scott DL. The clinical need for an acute rheumatology referral service. Br J Rheumatol 1996; 35:389–91. 7. Daker-White G, Carr AJ, Harvey I et al. A randomised controlled trial. Shifting boundaries of doctors and physiotherapists in orthopaedic outpatient departments. J Epidemiol Community Health 1999;53:643–50. 8. Jones A, Maddison P, Doherty M. Teaching rheumatology to medical students: current practice and future aims. J R Coll Physicians 1992;26:41–3. 9. Doherty M, Abawi J, Pattrick M. Audit of medical inpatient examinations: a cry from the joint. J R Coll Physicians 1990;24:115–8. 10. Lanyon P, Pope D, Croft P. Rheumatology education and management skills in general practice: a national study of trainees. Ann Rheum Dis 1995;54:735–9.
508
C. Roberts et al.
11. Davis P, Suarez-Almazor M. An assessment of the needs of family physicians for a rheumatology continuing medical educational program: results of a pilot project. J Rheumatol 1995;22:1762–5. 12. Irvine S, Munro R, Porter D. Early referral, diagnosis, and treatment of rheumatoid arthritis: evidence for changing medical practice. Ann Rheum Dis 1999;58:10–3. 13. Skargen EI, Oberg BE, Carlsson PG, Gade M. Cost and effectiveness of chiropractic and physiotherapy treatment for low back pain and neck pain. Six-month follow-up. Spine 1997;22:2167–77. 14. Dieppe P, Chard J, Faulkner A, Lohmander S. Osteoarthritis. In: Barton S, ed. Clinical evidence: a compendium of the best available evidence for effective healthcare, Issue 4. London: BMJ Publishing Group, 2000:649–73. 15. van Tulder MW, Koes BW, Metsemakers JF, Bouter LM. Chronic low back pain in primary care: a prospective study on the management and course. Fam Pract 1998; 15:126–32. 16. Schers H, Braspenning J, Drijver R, Wensing M, Grol R. Low back pain in general practice: reported management and reasons for not adhering to the guidelines in the Netherlands. Br J Gen Pract 2000;50:640–4. 17. van der Windt DA, Koes BW, Deville W, Boeke AJP, de Jong BA, Bouter LM. Effectiveness of corticosteroid injections versus physiotherapy for treatment of painful
18.
19. 20.
21. 22. 23.
24. 25.
stiff shoulder in primary care: a randomised trial. BMJ 1998;317:1292–6. Badley EM, Lee J. The consultant’s role in continuing medical education of general practitioners: the case of rheumatology. Br Med J Clin Res Ed 1987; 294:100–3. Pope C, May N, eds. Qualitative research in healthcare. London: BMJ Books, 1999. Kaner EF, Haighton CA, McAvoy BR. ‘So much post, so busy with practice—so, no time!’: a telephone survey of general practitioners reasons for not participating in postal questionnaire surveys. Br J Gen Pract 1998; 48:1067–9. Arthritis Research Campaign (ARC) website: http:uu www.arc.org.uku (accessed 15 December 2000). Amor B. What competence does a rheumatologist need?: an international perspective. Ann Rheum Dis 2000; 59:580–2. Doherty M, Woolf A for the EULAR Standing Committee on Education and Training. Guidelines for rheumatology undergraduate core curriculum. Ann Rheum Dis 1999;58:133–5. Doherty M, Lanyon P. Rheumatology: what should all doctors know? Ann Rheum Dis 2000;59:409–13. McCabe C, McDowell J, Cushnaghan J et al. Rheumatology telephone helplines: an activity analysis. Rheumatology 2000;39:1390–5.
