Martina 2013.

Methods	Location: Italy Number of centres: 1 Recruitment period: April 2006 to June 2007 Funding source: Italian Ministry of University and Research Trial design: parallel group RCT
Participants	Inclusion criteria: "Full class II molar relationships, overjet ≥ 6 mm, an age range of 10–13 years for boys and of 9–12 years for girls". Exclusion criteria: "Cervical vertebral maturation stage (CVMS) <2 or >3 (25), lack of parent’s willingness to sign an informed consent form, sella‐nasion to mandibular plane (Me‐Go) angle equal to or greater than the normal value plus a standard deviation, periodontal diseases, orofacial inflammatory conditions, tooth agenesis, congenital syndromes, and previous orthodontic treatment. Age at baseline: range 10 to 13 years for boys and 9 to 12 years for girls Number randomised: 61 Number evaluated: 46
Interventions	Gp A (n = 23): Sander Bite jump appliance Gp B (n = 23): no orthodontic treatment
Outcomes	Dentoalveolar, sagittal, and vertical changes assessed trough cephalometric analysis
Notes	Duration of randomised treatment: 18 months (BJA) and 12 months (control group) Sample size calculation: "The determination of sample size was based upon previous estimates of changes in mandibular length (Pg/OLp) during growth. By setting type I error at 0.05 and type II error at 0.20 (80% power), it was found that at least 19 patients per group were needed to detect an increase in mandibular length ≥ 2.0 mm".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Enrolled patients were allocated to either a treatment (BJA) or control (CTR) group by balanced block randomization using gender as a stratifying factor. A custom‐made Java script was used to generate the randomization procedure by a single investigator (SP) that was not involved in the clinical management of patients and control subjects."
Allocation concealment (selection bias)	Low risk	"The randomization sequence was carefully concealed to the other investigators and was disclosed immediately after obtaining written informed consent. Patients allocated to the BJA group were treated with the BJA, whereas patients allocated to the CTR group did not receive any treatment and acted as passive controls."
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"A single operator (IC), who was blinded to patient allocation (i.e. the allocation was masked to him in the dataset) performed the statistical analyses."
Incomplete outcome data (attrition bias) All outcomes	High risk	Gp A ‐ BJA recruited 31, analysed 23 (loss 25%) Gp B ‐ control recruited 30, analysed 23 (loss 23%) Reasons for discontinuation: BJA group ‐ did not receive allocated intervention (n = 6; 4 not willing to go further in the trial, 2 transferred); lost to follow‐up (n = 2) Control group ‐ lost to follow‐up (n = 7)
Selective reporting (reporting bias)	Low risk	All variables reported
Other bias	Low risk	No other bias found