Shen 2008.

Methods	Study design: randomised controlled trial Duration of the study: 2003‐2006 Protocol was published before recruitment of participants: not published Details of trial registration: not registered Funding sources: no benefits in any form were received or will be received from a commercial party related directly or indirectly to the subject of this article
Participants	Place of study: *Zhejiang Province TCM Hospital, Zhejiang TCM University Hospital and a level ‐ an academic trauma centre located in Hangzhou, Zhejiang Province, China Number of participants assigned: 133 (67 in 3D plate group; 66 in superior plate group) Number of participants assessed: 117 (63 in 3D plate group; 54 in superior plate group); 16 participants were lost to follow‐up at 12 months Inclusion criteria: Adults People with completely displaced middle third clavicle fractures, amenable to plate fixation with a minimum of 3 screws in both fragments Exclusion criteria: Open fractures Aged < 18 or > 60 years Fracture in the proximal or distal clavicle Fracture associated with nerve or tendon injuries Multiple injuries Additional fractures in the same or contralateral limb Previous fracture in the injured clavicle Abnormal function in the uninjured side Inflammatory joint disease Cerebrovascular disease or other severe medical illness and the inability to give informed consent or to complete questionnaires Age (mean/range): Total of participants: 44.2/18‐60 years 3D plate group: 43.8/26‐60 years Superior plate group: 44.7/18‐51 years Gender (male/female): 3D plate group: 39/28 Superior plate group: 36/30 Side of injury (dominant/non‐dominant): not reported Classification of injury: The fractures were classified as: Comminuted Spiral
Interventions	Timing of intervention: The operation was usually performed within 2 days of admission Intervention 1 (3D plate group): Open reduction and internal fixation with 3D plate (3.5 mm reconstruction plate). Here the plate was fixed superiorly on the main distal fragment and anteriorly on the main proximal fragments Intervention 2 (Superior plate group): Open reduction and internal fixation with superior plate (3.5 mm reconstruction plate). Here the plate was 'S' shaped and fixed on the superior surface Rehabilitation: *In hospital, participants were instructed on early motion exercises of the fingers, wrist, and elbow. Shoulder immobilisation was applied for 2‐6 weeks, based on the participants' level of comfort. A sling was applied for 6 weeks. The same instructions were used for all the cases Any co‐interventions: not reported
Outcomes	Length of follow‐up: participants underwent clinical and radiological assessment at 4 and 12 months after operation by independent surgeons Loss of follow‐up: 16 participants lost to follow‐up at 12 months: 3D plate group ‐ 4 participants lost to follow‐up: Re‐operation due to delayed union (1 participant) Did not attend for examination (3 participants) Superior plate group ‐ 12 participants lost to follow‐up: Re‐operation due to delayed union (8 participants) Did not attend for examination (4 participants) Primary outcomes Failure of treatment measured by delayed union Secondary outcomes Clinical fracture healing: delayed union and radiographic union Adverse events, measured by: Symptomatic participants (symptoms not specified) Other outcomes assessed: Duration of surgery Mean hospital stay
Notes	* Data obtained by personal contact with the authors
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
Allocation concealment (selection bias)	Low risk	In the operating theatre, the participants were allocated to 1 of 2 treatment groups according to sequentially opened sealed envelopes
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	The participants were blinded to the surgical method (information via personal contact with author) It was not possible blinding the care providers due to the type of treatment
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Follow‐up evaluations were done by the same non‐participating surgeons. All participants had sufficient records and, at follow‐up, all participants were interviewed according to protocol and examined by the non‐participating surgeons, who contributed with similar numbers of cases to each group (P value = 0.29), which should have prevented surgeons from being a confounding variable. The cases were randomised after the initial assessment, so the evaluator was blinded to the group allocation (personal contact)
Incomplete outcome data (attrition bias) All outcomes	High risk	Missing outcome data were not balanced in numbers across intervention groups; more participants in the superior plate group were lost to follow‐up at 12 months (4/67 (6%) in 3D plate group vs. 12/64 (18.1%) in superior plate group). This may have overestimated the benefits of 3D plate
Selective reporting (reporting bias)	High risk	Outcomes of interest in the review are not reported (function, pain, and quality of life)
Other bias	Low risk	The study appeared to be free of other sources of bias