Margit Kooijman

40 Introduction Musculoskeletal conditions are highly prevalent in society and are known to have a substantial impact on health related quality of life. Shoulder complaints such as tendinitis or capsulitis are among the most commonly reported of these conditions 1 . The annual consulting incidence in general practice for shoulder symptoms is estimated at 36 per 1000 person years and general practitioners only refer more patients to the physiotherapist for back and neck pain 2 . In physiotherapy practice, about 10% of patients have shoulder complaints, which makes it the most common complaint of the extremities 3 . The majority of shoulder complaints arise gradually and various studies show that many patients already have complaints of long duration when they first present themselves with a clinician 4-6 . Since this is a risk factor for poor treatment results, it is not surprising that unfavourable outcome is reported often 5,6 . In recent years, there has been an increase in research on other prognostic factors but despite the effort, besides longer duration only scores on pain-intensity and function have been consistently linked to outcome 6,7 . Adequate decision-making on treatment is further hampered since physical examination is characterised by the limited validity of the clinical tests to diagnose the origin of symptoms 8,9 . In addition, though the overall evidence is conflicting, numerous psychological factors keep emerging as possible factors contributing to the prognosis of the shoulder complaints. All in all, this shows that delivering adequate treatment is a highly challenging task for physiotherapists and general practitioners as the main caregivers in primary care. To carry out this task well, the Dutch taskforce ‘right care at the right place’ emphasizes the importance of collaboration and integration 10 . The Dutch guideline for general practitioners on the management of shoulder complaints provides some information on this. It suggests referral for physiotherapy in case a wait-and-see policy in combination with use of analgesics fails to reduce complaints sufficiently 11 . In contrast with this advice, previous research shows that the majority of patients was referred during the first consultation with the general practitioner 4 . The World Confederation of Physical Therapists regards the ability to independently determine whether a patient’s condition is suitable for physiotherapy management a core element in their guideline for standards of physiotherapy practice 12 . However, most national guidelines for physiotherapists do not contain statements on keep/refer decision-making, which is true for the situation in the Netherlands as well 13 . The short guideline on subacromial complaints only recommends referring a patient (back) to the GP when treatment fails 14 . Recent research shows that most physiotherapists comply with this guideline except