Baerheim 1994.

Methods	Study design: parallel RCT Study duration: subjects were recruited from February 1990 and followed up for 6 months (final follow‐up date unknown)
Participants	Setting: outpatient Country: Norway Female with 3 or more episodes of distal urinary symptoms during the previous 12 months, with at least 1 episode having been medically verified as a lower UTI. Subjects should have been free of infection at inclusion, and no antibiotic treatment should have been taken during the previous 3 weeks Number: treatment group (25); control group (22) Mean age, 95% CI (years): treatment group (36.9, 33.6 to 40.2); control group (35.1, 30.7 to 39.5) Exclusion criteria: pregnancy at inclusion or during study period, use of a diaphragm, any complicating illness (e.g. diabetes, cancer, urinary tract obstruction)
Interventions	Treatment group Vaginal suppositories containing L. casei v rhamnosus (Gynophilus (R)) At least 7.5 x 108 live L. Casei v rhamnosus per suppository, twice weekly for 26 weeks Control group Placebo vaginal suppositories Solid semisynthetic glycerides (97.3%) and colloidal silica (2.7%) twice weekly for 26 week
Outcomes	Occurrence of distal urinary symptoms defined as dysuria, urinary frequency, and/or suprapubic discomfort Acute lower UTI defined as the presence of all of the following: acute lower urinary symptoms, leucocyturia (≥ 5 leucocytes/HPF), bacteriuria (≥ 10⁴ CFU/mL) of uropathogens, or any amount of Staphylococcus saprophyticus
Notes	Organon A/S, Oslo, Norway provided drug and funding support PI contacted 7 May 2012 to clarify several questions, and responded 10 May 2012: The random number sequence was generated by the pharmaceutical company producing the vaginal suppositories. Allocation concealment was ensured through packages sent to physicians by patient serial number. One patient withdrew from the study early, from the placebo arm, prior to experiencing an event, with no data collected We also requested clarification regarding information sent from physicians who were seen by patients presenting with lower UTI symptoms to investigators. These physicians were provided with dip‐slides and a registration form which was pre‐addressed to the investigators
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information to permit judgement
Allocation concealment (selection bias)	Unclear risk	Insufficient information to permit judgement
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Insufficient information to permit judgement
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: "Double‐blind..." Page 240 ‐ Materials and methods. Not described further
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Quote: "One was later excluded on her own request before she started to use the vaginal suppositories." Page 240 ‐ Results 1/48 lost prior to initiation of therapy. Authors did not report to which arm the patient had been randomised
Selective reporting (reporting bias)	High risk	Main outcome reported with non‐significant difference between groups. Two outcomes described in the methods (compliance and causative pathogen) were not reported and no explanation was given. No protocol published or in a clinical trial registry
Other bias	Unclear risk	Organon A/S, Oslo, Norway provided drug and funding support Limited reporting of harm. Control group had a significant increase in Lactobacilli, intervention group did not ‐ explained as regression to mean. Page 240 ‐ Results