Cowan 2012.

Methods	Study design: parallel design Power calculation: provided, assumed 20% reduction in bladder problems Intention‐to‐treat analysis: yes
Participants	Inclusion criteria Setting: radiotherapy booking system used to identify patients, patients had cervical cancer or bladder cancer at 1 centre Country: UK Adults > 18 years with cervical or bladder cancer requiring radiation therapy Number: 128 randomised; 113 analysed (7 in placebo arm, 8 in cranberry arm) Exclusion criteria: NS
Interventions	Treatment group Cranberry juice twice/d; volume (NS); PAC (NS) Control group Matched placebo juice twice/d; volume (NS)
Outcomes	Urinary symptoms
Notes	Exclusions post randomisation: 0
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer based deterministic minimisation algorithm, externally allocated
Allocation concealment (selection bias)	Low risk	Computer algorithm generated a blinded juice pack
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double blinding stated, patients blinded to treatment arm, clinicians blinded
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	For UTI outcome probably low risk, microbiology results independent
Incomplete outcome data (attrition bias) All outcomes	Low risk	Very little missing data
Selective reporting (reporting bias)	Low risk	Urinary symptoms and UTI
Other bias	Low risk	Source of funding: west Research Endowment fund, NHS greater Glasgow and Clyde, Juice and placebo supplied by Ocean Spray