Margit Kooijman

34 known for a subgroup of patients, which is insufficient for a reliable investigation. Therefore, physiotherapists give an indication of the result, so that an outcome measure is known for every patient. In the present study, referred patients and self-referrals achieved the treatment goals to the same extent. Since the outcome is measured in the same subjective manner, it is not expected that the results would be different. Nevertheless, ideally, patient-reported outcome measures should also be studied. In LINH, a diagnosis was not registered for every consultation. Prescriptions, referrals and interventions are calculated for the total care episode of shoulder syndromes and might therefore have been overestimated. However, this calculation only concerned a selection of frequently used prescriptions, referrals and interventions which prevents the inclusion of those actually relating to a diagnosis other than shoulder syndromes. Finally, data is based on two different patient populations. The physiotherapy database is much smaller and, therefore, a longer time period was selected. However, there were no policy changes in the area or indications that the group of patients consulting their GP changed over the period of the study. Nevertheless, it would be interesting to investigate the care process in a multidisciplinary network incorporating the activities of various health care professionals. Conclusions In summary, there are differences in general practice between patients who are referred for physiotherapy and those who are not. Patients who are referred are prescribed less medication and are often referred at the first consultation with their GP. This goes against current guideline for GPs and could result in unnecessary or higher costs. On the other hand, possibly due to the gradual onset of complaints and a wait-and-see policy, for many patients, it takes quite a while before they see a physiotherapist, even though it is suggested that a long duration of complaints could be a predictor for poorer outcomes. When a restricted range of motion is the main problem, it is arguable that patients receive less medication but a quick referral to a physiotherapist. Future research into the long-term cost-effectiveness of an early referral could demonstrate whether this leads to better outcomes and should therefore be the preferred treatment. As clinicians, we should also develop a way of educating patients about shoulder syndromes to prevent them waiting too long before they seek help. However, this can only be cost-effective when the profession sets clear guidelines on indications for physiotherapy, especially since there is debate on the value of clinical diagnostic tests. The classification of pain and functional limitations and adoption of the wait and see policy as used by GPs could be an example or starting point. Perhaps with early detection, a once-off consultation in which advice is given will be sufficient, especially when pain is severe. The consequences of such initiatives for the prognosis of the individual patient as well as cost-effectiveness should be investigated first.