Melean 2015.

Methods	Study design: single‐centre, 2‐group, parallel‐design RCT Duration of the study: February 2010‐February 2012 Protocol was published before recruitment of participants: not reported, but no protocol was published. Details of trial registration: not registered Funding sources: none of the study authors received payments or services, either directly or indirectly.
Participants	Place of study: 1 centre from Santiago, Chile Number of participants assigned: 76 participants (34 surgical; 42 conservative) Number of participants assessed: 76 participants (34 surgical; 42 conservative) Inclusion criteria Aged > 18 years Complete displaced fracture of the middle shaft clavicle (without cortical contact) Signed approval to participate in the study Isolated clavicular fracture Participants with labour accidents that were treated under the national workers' insurance laws and compensations Fracture classified 2B1 or 2B2 according to Robinson 1998 Exclusion criteria Fractures in the lateral or medial segment of the clavicle Neurovascular associated injuries Open fractures > 21 days from the accident Age Surgical group (mean/SD): 38.1/13 years Conservative group (mean/SD): 37.2/11.2 years Gender: not reported Classification of injury: fractures were classified according to the Robinson's Classification (Robinson 1998).
Interventions	Timing of intervention: not specified; however, participants with intervention > 3 weeks after injury were excluded Type of surgical intervention: open reduction and plate fixation using the 3.5 mm LCP system in 12 participants and LCP reconstruction plates in 22 participants. Study authors said that different implants were used according to availability at the day of the surgery. Type of conservative intervention: standard sling for 6 weeks Rehabilitation Surgical group: use of a sling for 4 weeks after surgery. Physical therapy started at 3 weeks, with passive ROM and analgesic physiotherapy for 3 weeks, following active ROM and strengthening exercises. Conservative group: physical therapy was started at 4 weeks, with passive ROM and analgesic physiotherapy for 3 weeks, following active ROM and strengthening exercises Any co‐interventions: not reported
Outcomes	Length of follow‐up Follow‐up was 1 year Participants were evaluated at 3, 6 and 12 months Loss to follow‐up: none lost to follow‐up Primary outcomes Function or disability measured by Constant score Treatment failure measured for symptomatic non‐union Secondary outcomes Adverse events measured by: hardware irritation requiring removal Time to return to previous activities (work)
Notes	We contacted the study authors to request data (i.e. SD for Constant score); however, the study authors declined to provide them. Composite adverse events: surgery: discomfort leading to implant removal (n = 4). Conservative: (n = 0). In Analysis 1.14 we used event rates 4 for surgery and 0 for conservative. Cosmetic result events: not distinguished from adverse events
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Used "the method of tables of 4" and those with '0' assigned to conservative treatment and '1' assigned to surgery; unclear if this is random.
Allocation concealment (selection bias)	Unclear risk	Unclear, reports that those with '0' assigned to conservative treatment and '1' assigned to surgery, and these placed in envelopes, but not if the envelopes were sequentially numbered or opaque.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not reported. Blinding of participants was not feasible.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Not reported. As participants were aware of treatment, there was a high risk of detection bias in measurement of function.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The study authors did not explicitly report if any participants were lost to follow up.
Selective reporting (reporting bias)	High risk	Pain (as primary outcome measured by validated participant‐reported measures) and other important secondary endpoints were not evaluated by the study authors.
Other bias	Unclear risk	The trial did not report baseline data thus it is unclear if 2 groups were balanced at baseline for the primary outcomes.