Mutabazi 2012.

Study characteristics
Methods	Design: single‐center unblinded RCT conducted in Rwanda. Compared PrePex device to surgical circumcision (dorsal‐slit technique). 226 men randomized and allocated according to WHO allocation recommendations (2:1) = (PrePex circumcision: surgical circumcision). Ethical approval: Rwanda National Ethics Committee on 13 January 2011, approval number N0 001/RNEC/2011 Consent: participants provided written informed consent. Duration of study: 17 February 2011 to 8 April 2011
Participants	Inclusion criteria: men ages 21–54 years who were willing to undergo both surgical and PrePex device circumcisions. Exclusion criteria: any contraindication to surgical circumcision (anatomical abnormalities, genital ulcer disease, previous circumcision etc.), HIV or unwilling to consent to an HIV test, and mental illness with unreliable follow‐up. Baseline characteristics: no difference in baseline characteristics. Mean age: 26 years. All were HIV negative.  Circumcision device group mean age was 26 (SD 5.17) years, traditional method mean age was 24 (SD 4.10) years
Interventions	Providers: all surgical circumcisions conducted by surgeons experienced in circumcision. All non‐surgical circumcision procedures were performed by surgeons who participated in prior safety and efficacy studies using PrePex in Rwanda. Intervention group (PrePex circumcision device): 150 participants. Clean skin preparation, anesthetic cream (lidocaine and prilocaine) followed by application of the PrePex device, dry gauze dressing applied and the device removed 7 days later. Follow‐up: assessed 16 hour after procedure, and on the 3rd, 9th, 14th, 21st, 28th, 35th, 42nd, 49th, 56th, 63rd postoperative days Control group (conventional [dorsal slit] surgical circumcision): 76 participants. Sterile skin preparation, injection base penile ring block, dorsal‐slit technique followed by absorbable skin closure and dry dressings. Follow‐up: 7th postoperative day
Outcomes	Primary outcomes: total male circumcision time (skin to skin) Secondary outcomes: procedure preparation time clinical adverse event rates surgical procedure‐related and device‐related incident rates expected adverse effects pain at key time points and during erection on days 3 and 9 loss of productivity patient satisfaction time to complete healing
Funding sources	Government of Rwanda. PrePex devices were provided by Circ MedTech Limited (Tortola, British Virgin Islands).
Declarations of interest	Disclosed no conflicts of interest.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The process for randomization was clearly described. Quote: "A randomized number table was created by the study PI using the RAND algorithm in Excel (Microsoft Corp, Redmond, WA) in which each randomized number was distributed between 1.00000 and 0.00000. To achieve an allocation ratio of 2:1, each random number below 0.33333 was assigned to the surgical arm and each number higher than 0.333334 was assigned to the PrePex arm." Comment: low risk of selection bias.
Allocation concealment (selection bias)	Low risk	The process for allocation concealment was clearly described. Quote: "Study numbers, randomised numbers and their respective treatment arms were then placed in envelopes that were validated to be non‐transparent. This was done by the principal study investigator who was not involved in the enrolment and treatment allocation. It was also done before the enrolment process." Comment: low risk of selection bias.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Blinding of participants and personnel was not possible with this surgical intervention and it was, therefore, unclear whether this may have impacted the performance of either group.
Blinding of outcome assessment (detection bias) Subjective outcomes (severe adverse events, moderate adverse events, minor adverse events, pain, patient satisfaction)	High risk	There were no independent outcome assessors. The lack of blinding was judged to potentially affect the assessment of these subjective outcomes. In addition, the trial design allowed for differential follow‐up schedules for the intervention group (PrePex) within the first visit at 3 days compared to the surgical group who had follow‐up at 7 days. This may introduce some bias as the adverse effects related to wound healing may be expected to be different. High risk of detection bias.
Blinding of outcome assessment (detection bias) Objective outcomes (operative time)	Low risk	As the operative time was considered an objective outcome, the lack of blinding was not considered a risk of detection bias. In addition, the trial put in place additional measures to minimize biased reporting. Quote: "Time intervals were measured using a stopwatch; as a backup, a video of each procedure was recorded."
Incomplete outcome data (attrition bias) All outcomes	Low risk	Attrition for the main outcomes was very low and considered a low risk of attrition bias. It should be noted that there was high attrition (35%) for the outcome 'participant satisfaction' which required a questionnaire to be completed.
Selective reporting (reporting bias)	Low risk	Study registered at ClinicalTrials.gov (NCT01284088) and the prespecified outcomes align with those reported in the published trial.
Other bias	Low risk	No other biases identified.