Fjorback 2013.

Methods	Study design: randomised controlled trial
Participants	Diagnosis: chronic multi‐organ BDS Method of diagnosis: participants received a 5‐7 hour bio‐psycho‐social assessment, including a laboratory screening battery, schedules for clinical assessment in neuropsychiatry (SCAN)‐diagnostic interview, as well as a physical and neurological examination. Participants (aged 20‐50 years) with chronic (≥ 2 years) multi‐organ type BDS and moderate‐to‐severe impairment in daily living were eligible for inclusion Exclusion criteria: severe psychiatric morbidity, current alcohol or drug abuse, pregnancy, non‐Scandinavian origin, no informed consent Total number randomised: 120 Age: for intervention group, M = 38 (SD = 9); for control group, M = 40 (SD = 8) Sex: 80% women; 20% men in both study groups, 47 women in intervention group, 48 in control group Severity of symptoms at baseline: all participants in both study groups had a somatisation disorder according to ICD‐10 codes. In the intervention group, 95% had a somatisation disorder according to the DSM‐IV codes (n = 56), in the control group 95% (n = 57). 5% of intervention group (n = 3) and 5% of control group (n = 3) had an undifferentiated somatoform disorder (DSM‐IV codes) Duration of symptoms at baseline: for intervention group M = 12 (SD = 10.6), for control group M = 15 (SD = 12.6) years Setting: primary care physicians and hospital wards referred participants both from rural and urban areas. Intervention took place at the Research Clinic for Funcional Disorders and Psychosomatics in Aarhus University Hospital Location: Arhus region, Denmark Number of treatment centres: 1 Co‐morbidities: 71% of intervention group (n = 42) and 62 % (n = 37) of control group had lifetime psychiatric co‐morbidity. 22% of intervention group (n = 13) and 20% of control group (n =12) had a current major depressive disorder (DSM‐IV codes). 24% of intervention group (n = 14) and 23% of control group (n = 14) had a current anxiety disorder (DSM‐IV codes) Adjunctive therapy/medication: consultation letter; letter to social authorities, when needed; advise to taper off morphine derivatives and benzodiazepines (all also in comparison group)
Interventions	Participants were randomly assigned to either 1. Mindfulness therapy (n = 59) Duration: 8 weekly session and 1 in week 12 (3.5 hours each) during a period of 12 weeks Treatment protocol: participants received information about the nature, course, and treatment of BDS, a treatment manual, 9 group treatment modules based on a mindfulness‐based stress reduction and a cognitive‐behavioural approach (closely following the manual by Jon Kabat Zinn (Fjorback 2013, ref 22, 36, 36). A consultation letter was sent to participants' primary care physicians, and letters to social authorities were sent when needed (see Fjorback 2013 Table 1 for details) Therapist: 2 psychiatrists with experience in BDS and CBT and in psychotherapy of meditation 2. Enhanced TAU (n = 60) (see Fjorback 2013 Figure 1 for details) Duration: 1 x 2‐hour session Treatment protocol: participants received information about the nature, course, and treatment of BDS, a consultation letter to participants' primary care physicians, letters to social authorities when needed, an individual CBT treatment plan according to a manual and an individual psychiatric consultation within the first month after clinical assessment (see Fjorback 2013 Figure 1 for details) Therapist: psychiatrist
Outcomes	Time points for assessment: baseline and 3 months, 9 months, 15 months after baseline Primary outcome: 1. mean change in SF‐36 PCS between baseline and 15 months Secondary outcome: 1. change in other health‐related quality of life measures of the SF‐36 2. illness worry (WI‐8) 3. physical symptoms (SCL‐90‐R Somatisation Subscale) 4. severity of depression and anxiety (SCL‐8) 5. participants reporting improvement (greater than half a SD)
Notes	Date of study: recruitment took place between April 2007 and September 2008, follow‐up until the end of 2009 Funding source: Danish Agency for Science Technology and Innovation, Aase and Ejnar Danielsens Fund, Trygfonden Declarations of interest among the primary researchers: none reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was prepared by a statistician
Allocation concealment (selection bias)	Low risk	They used pre‐defined concealed random numbers tied to consecutive assessments of participants resulting in opaque envelopes numbered in succession containing assigned treatment
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants and personnel were not blinded, given difference in treatment content, intensity, and duration
Blinding of outcome assessment (detection bias) All outcomes	High risk	All outcomes were participant reported, and these were not blinded
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	> 20% no data at follow up, but multiple imputation may have solved this problem
Selective reporting (reporting bias)	Unclear risk	No protocol available. Secondary outcome probability of change > 0.5 SD predefined?
Treatment fidelity	Low risk	Treatment followed a model based on a manual (Fjorback 2013, table 1 and ref 39)
Researcher allegiance	Low risk	No indication that researchers had a preference for 1 of the treatment modalities
Other bias	Low risk	No other sources of bias