Skelly 1992.

Methods	RCT or quasi‐RCT: RCT Setting: orthopaedic unit in a general hospital (St. Joseph’s Hospital, Hamilton, Ontario, Canada) Period: November 1986 – December 1987
Participants	Population: patients ≥ 60 years old admitted for surgical repair of hip fracture Inclusion criteria: patients ≥ 60 years old admitted for surgical repair of hip fracture Exclusion criteria: patients with intractable incontinence, patients admitted with indwelling catheter in place, patients who were unable to provide informed consent Age (mean, SD): A: 78 (8.2); B: 78 (8.6) Number of participants: · Eligible: 76 · Randomised: 67 · Reported: 67 Dropouts (n of participants + reasons): 5 medically unstable, 2 died before surgery, 2 did not have urinary retention after surgery Post‐randomisation: none reported Follow‐up: 5 days
Interventions	Time of intervention: preoperative A (n = 35): indwelling catheter inserted preoperatively and left in place until 48 hours after surgery. If could not void in following 24 hours, intermittent catheterisation performed every 8 hours for 24 hours. If still not able to void indwelling catheter inserted again for 48 hours B (n = 32): intermittent catheter inserted every 6 ‐ 8 hours, with 400 ‐ 600 ml of urine removed each time Catheterisation stopped when residual volume of urine < 150 ml on 2 consecutive occasions Intended duration of catheterisation: A: 48 hours B: 5 postoperative days if normal voiding returned. If not, catheter remained in until voiding returned Voiding defined as > 100 ml and residual volume < 150 ml on 2 consecutive occasions
Outcomes	Primary outcome (symptomatic UTI): not reported Definition of symptomatic UTI: not reported Bacteriuria: A: 11/35; B: 12/32 Definition of bacteriuria: ≥ 10⁵ cfu/ml on postoperative day 5 Postoperative urinary retention : A: 6/35; B: 6/32 Participants voiding on postoperative day 5: A: 13/35; B: 21/32 Days until return of voiding (mean, n): A: 9.4, 35; B: 5.1, 32 Incontinent after postoperative day 5: A: 5/35; B: 2/32 Quality of life: not reported
Sponsorship/Funding	“This work was supported by a research grant from Ontario Ministry of Health”
Notes	Not reported if used prophylactic antibiotics
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	“use of a randomization technique involving block sizes of four” – method of randomisation not given, not clear if selection bias risk is high or low
Allocation concealment (selection bias)	Low risk	“sequentially numbered sealed envelope” – adequate method of allocation concealment, low risk of selection bias
Blinding of participants	High risk	No information given. As indwelling vs intermittent can assume blinding did not occur for participants, therefore high risk of bias
Blinding of personnel	High risk	No information given. As indwelling vs intermittent can assume blinding did not occur for personnel, therefore high risk of bias
Blinding of microbiological outcome assessment	Low risk	Bacteriuria would be assessed by microbiologist who would not know what catheter the participant had
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No information given. Primary outcome is urinary retention, but no method reported on how urinary retention was diagnosed
Incomplete outcome data (attrition bias) All outcomes	Low risk	76 agreed to participate in trial – 9 excluded (5 at physician’s request as medically unstable, 2 died before surgery, 2 did not have urinary retention). 67 randomised. All randomised participants included in analysis
Selective reporting (reporting bias)	Unclear risk	No clear information given in Methods about the outcomes being studied. Unable to access protocol, so uncertainty surrounding reporting bias
Other bias	Low risk	Appears to be free from other sources of bias