Psychosomatic Therapy, Feasibility and Cost Analysis

Recruitment and signed informed consent:Participating GPs (n=6) were asked to enrol each consecutive eligible patient and include 10 patients with MUS each. The GPs presented written information and an informed consent form to these patients. Patients had the opportunity to read the information to consider participation before signing the consent form.

Randomisation: Participating patients (n=60) are equally randomised over two groups, the intervention group (usual care and psychosomatic therapy delivered by a psychosomatic therapist) and the control group (usual care of the GP). For the randomisation we used a random number table, whereas a central and blinded allocation procedure was applied.For the randomisation we used a random number table, whereas a central and blinded allocation procedure was applied.

Patients in the intervention group were asked to complete a questionnaire at baseline, at the end of the intervention (+/- 3 months)and at 6 and 12 months. Patients in the control group completed a questionnaire at baseline, at 3 (equals end of intervention), 6 and 12 months.

Statistical analysis The number of patients identified and recruited by the GP as well as the number of eligible patients who declined to enter the trial were analysed using descriptive statistics. Furthermore, at each stage of the study, we recorded the number of trial participants who completed the study.

The effects of the intervention on the primary outcome measures (VAS and 4DSQ) and the secondary outcome measures (NHL, SF-36 and MAF) were estimated using analysis of covariance with baseline value as a covariate. Data analysis was be performed according to the intention-to-treat principle. Medical consumption was measured using the Cost Diary for medical consumption. Patients recorded their absenteeism and healthcare consumption, such as consulting a GP, hospital, paramedic facilities and alternative medicine. Medication intake was recorded at the start and the end of the study. We used standardised Dutch prices (guide cost research 2009), whereas medication costs were calculated according to the Dutch Pharmaco Therapeutic Compass. Costs by absenteeism were calculated according to the Friction Cost method.

Difference in cost of medical consumption between the control and intervention group were estimated using analysis of covariance with baseline value as a covariate.

Difference in baseline characteristics were analysed with the student's t-test for continue variables and the chi-squared test or Fisher's exact test for categorical variables. P-values of < 0.05 were considered statistical significant.

Inclusion criteria

patients with medical unexplained symptoms who consulting the general practitioner more than once for one or more of the 12 most common symptoms:

• stomach-/abdominal pain,
• back/neck/shoulder pain,
• pain in arms/legs/joints,
• headache, chest pain,
• dizziness,
• fainting spells,
• palpitations of the heart,
• shortness of breath,
• nausea/indigestion,
• feeling tired,
• sleeping disorder and psychosocial distress

Exclusion criteria

• patients wich chronic diseases,
• palliative care,
• severe psychiatric disorders,
• long-lasting injury treatment
• patients who are not able to speak, read and write Dutch