Woltz 2017a.

Methods	Study design: multicentre, 2‐arm, parallel‐group, RCT (16 centres) Duration of the study: June 2010‐December 2013 Protocol was published before recruitment of participants: yes. doi: 10.1186/1471‐2474‐12‐196 ‐ BMC Musculoskeletal Disorders 2011;12:196 Details of trial registration: yes. Main ID: NTR 2399. Date of registration: 1 July 2010 Funding sources: the study was financially supported by Fonds NutsOhra, a nonprofit funding organization for improvement in health in the Netherlands. The study authors reported no potential conflict of interest.
Participants	Place of study: 16 teaching and non‐teaching hospitals in the Netherlands, including 4 university hospitals Number of participants assigned: 160 participants (86 surgical; 74 conservative) Number of participants assessed: 154 participants (84 surgical; 70 conservative) Inclusion criteria Aged 18‐60 years A completely displaced middle third clavicle fracture (Robinson type 2B1 or 2B2) Fresh fracture, treatment within 14 days after injury No contraindications for surgery or general anaesthesia Provided informed consent Exclusion criteria Pathological fracture Open fracture Associated neurovascular injury, or suspicion of it Associated head injury Ipsilateral upper extremity fracture at same time 1st presentation > 14 days after injury Pre‐existing impaired shoulder function or previous surgery of the shoulder Inability to comply with follow‐up Age Surgical group (mean/SD): 38/12.7 years Conservative group (mean/SD): 37/12.5 years Gender of participants assigned (male/female) Surgical group: 80/6 Conservative group: 66/8 Classification of injury: fractures were classified according to the Robinson's Classification (Robinson 1998)
Interventions	Timing of intervention: not specified; however, participants with > 14 days after the injury were excluded. Type of surgical intervention: open reduction and plate fixation. There were no restrictions regarding incision, plate location, or type of plate. Type of conservative intervention: simple sling for 2 weeks Rehabilitation Surgical group: sling was used for comfort for 2 weeks; the postoperative exercise protocol was similar to that in the conservative group. Conservative group: in the initial 2 weeks, there is the use of a sling and non‐weight‐bearing exercises given by a physiotherapist. After 6 weeks, full ROM is permitted and strengthening exercises Any co‐interventions: not reported
Outcomes	Length of follow‐up Follow‐up was 1 year Participants were evaluated at 3 and 6 weeks, and at 3 and 12 months Loss to follow‐up: 22 participants were lost to follow‐up at 1 year Surgical group: 9 participants were lost at 1 year Conservative group: 13 participants were lost at 1 year Primary outcomes Function or disability measured by DASH and Constant score Pain measured using VAS Failure of treatment measured by symptomatic non‐union, malunion or other complications (e.g. mechanical failure, refracture and late neurologic complication) Secondary outcomes Other treatment failure measured by: cosmetic result: participant satisfaction with shoulder appearance asymptomatic non‐union Adverse events measured by: infection hardware irritation with removal skin and nerve problems Health‐related quality of life measure by: Short Form‐36
Notes	Besides the 6 participants lost to follow‐up, there were 6 participants (1 in the surgical group and 5 in the conservative group) that were lost to follow up only in the primary outcome, since they didn't have radiographic image to confirm achieved union The type of surgical treatment, relating to plate type, plate position and incision were not pre‐specified. SDs and means of VAS scores were obtained by personal contact with the study authors. Composite adverse events: surgery: wound infection (n = 3); implant removal (n = 14; loss of sensation around scar (n = 15). Conservative: malunion (n = 1); plate removed (n = 1); neurological problems (n = 1). In Analysis 1.14 we used event rates 15 for surgery and 1 for conservative. Cosmetic result events: surgery: unsatisfied with cosmetic result (n = 4). Conservative: unsatisfied (n = 13). Estimates from percentages (5% versus 18%). In Analysis 1.11 we used event rates 4 for surgery and 13 for conservative.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The sequence generation was performed by a computer random number generator using minimisation.
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants and personnel were not blinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcome assessors were not blinded.
Incomplete outcome data (attrition bias) All outcomes	High risk	The authors excluded 6 participants who did not complete follow‐up, and an intention‐to‐treat analysis was performed; however, only 4 participants were lost from the conservative group and 2 from the surgical group. However, there were imbalanced data for availability of function scores. After 1 year, the functional scores of 87% of participants in the surgical group and 78% of participants in the conservative group were available for analysis.
Selective reporting (reporting bias)	Low risk	The study protocol was registered and all of the study's pre‐specified (primary and secondary) outcomes that are of interest for this review were reported in the pre‐specified way.
Other bias	High risk	There were imbalances in numbers allocated into the 2 groups assigned (86 surgical; 74 conservative). There was insufficient data on baseline characteristics to judge whether or not the 2 groups were balanced at baseline.

AO: Arbeitsgemeinschaft fur Osteosynthesefragen; OTA: Orthopaedic Trauma Association; AO/OTA: combination of AO and OTA; ASA: American Society of Anesthesiologists; BESS: British Elbow and Shoulder Society; BUPA: British United Provident Association; DASH: Disability of the Arm, Shoulder, and Hand questionnaire;DCP: dynamic compression plate; LCP: locking compression plate; RCT: randomised controlled trial; ROM: range of movement; SANE: Single Assessment Numeric Evaluation; SD: standard deviation; SEM: standard error of the mean; TEN: titanium elastic nail; UCLA: University of California Los Angeles; VAS: visual analogue scale