In their service development for medically unexplained symptoms (MUS), Röhricht & Elanjithara ^{Reference Röhricht and Elanjithara1} bring much-needed attention to the problem of developing a service that is both effective and one that patients choose to attend. They highlight that a significant proportion may only engage in a collaborative model at a primary care level. One of the first reasons for this is the terminology prevalent in this field. ^{Reference Creed, Kronke, Hennningsen, Gudi, White, Creed, Henningsen and Fink2} The patients find ‘somatoform’ and ‘medically unexplained’ symptoms unsatisfactory terms which have connotations that ‘it is all in the mind’. They wonder if the low referral rate from some general practitioners (GPs) and the non-attendance by nearly a quarter of patients referred is related to this. When developing pilot services for MUS, we chose to call our service the ‘symptom management clinic’ and locate it within GP surgeries, to avoid prejudicing its acceptability by alignment with mental health hospitals or psychological terminology. On auditing our attendees, many said they ‘would not have attended a clinic located with a mental health provider’ and we achieved high user satisfaction ratings for the ease of accessibility and format of the clinic.

We also incorporated the proactive identification that Röhricht & Elanjithara call for. We decided to ‘case find’ and asked GPs in four separate surgeries to identify any patients that had been seen at the surgery more than 10 times in 2 years; had at least two negative diagnostic tests; and were not currently involved with specialist mental health services. We then examined case notes and excluded patients with current diagnostic codes on the GP database. This process was time consuming, although it has future potential to be automated, but it did have the benefit of finding patients who had not been thought by the GP as having MUS but were actually presenting and being referred for repeated investigations without a diagnosis. Similarly, Burton et al ^{Reference Burton, McGorm, Richardson, Weller and Sharpe3} used repeated referrals to secondary care as a guide and found that ‘at least three times in 5 years’ identified MUS patients with high levels of secondary care usage.

In one surgery alone, we identified 17 patients who had 286 out-patient and hospital attendances between them over 2 years with an average cost of £2396/year (range £374–7403). Of these referrals, 13 patients attended a symptom management clinic appointment with a consultant in liaison psychiatry or a consultant clinical neuropsychologist. Involvement of the GP was considered crucial, with a short feedback session with both GP and patient following the clinic to develop a collaborative approach to ongoing management. This also provided a concurrent training benefit for GPs which they valued.

A cost analysis of the patient's healthcare usage before the symptom management clinic and for 2 years following assessment used standard hospital tariff costs and showed a reduction of 48% in secondary care usage alone. We also showed an increase in functioning, as measured by the EuroQol-5D (EQ-5D), and some evidence of a reduction in Hospital Anxiety and Depression Scale (HADS). Around half of the patients went on to access psychotherapy via the improving access to psychological therapies (IAPT) pathway and other established programmes such as pain management, but many remained managed in primary care alone (details available from the author on request).

We look forward to commissioners placing some confidence and resources in these preliminary MUS services to encourage learning and development of methods for improved identification and adequate treatment of this large, neglected and often costly patient group. ^{Reference Andersen, Eplov, Andersen, Hjorth and Health4}