Ghent 1988.

Methods	RCT Placebo‐controlled Cross‐over design
Participants	Pruritus: CP Description: participants with PBC Number of participants randomised: 9 Group A (rifampicin ‐ placebo): 4 Group B (placebo ‐ rifampicin): 5 Number of participants evaluable: 9 Withdrawals/dropouts: 0 Reason for dropout: NA Age (years): 45‐64 Sex (male/female): 1/8 Underlying disease(s): primary biliary cirrhosis Participant pool: single‐centre Setting: outpatient Haemodialysis: NA Baseline pruritus assessment: 1 week baseline record of the pruritus VAS was obtained for 5 subjects prior to the study Duration/severity of pruritus: persistent pruritus Baseline parameters: no information
Interventions	Intervention 1: rifampicin (150 mg 2x/d for serum bilirubin level > 51 µmol/L (3 mg/dL); 150 mg 3x/d for serum bilirubin level < 51 µmol/L) Intervention 2: placebo Additional medication: cholestyramine (n = 5) Route of administration: oral Duration of treatment: 2 weeks (2 weeks washout ‐ 2 weeks cross‐over) Follow‐up: no information
Outcomes	Pruritus assessment: VAS (0‐100 mm) Adverse events: none observed Additional outcomes: urine analysis, blood count, serum bilirubin, creatinine, alanine transaminases, aspartate transaminases, alkaline phosphatases, fasting total serum bile acids
Notes	Presentation of study results is inappropriate
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"The order of treatment was random . . .." Method of randomisation not stated
Allocation concealment (selection bias)	Low risk	"The coding of the medication order was done by an independent research pharmacist."
Blinding of participants and personnel (performance bias) All outcomes	Low risk	". . . in a double‐blind manner." Unclear who was blinded
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No information provided
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	All completed and analysed; unclear if used intention‐to‐treat analysis
Selective reporting (reporting bias)	Low risk	No indication of selective reporting
Size of study (possible biases confounded by small size)	High risk	Number of participants randomised: 9
Other bias	Unclear risk	Small number of participants