Versloot 2008.

Methods	Study design: randomised controlled trial, parallel design Location: Netherlands Number of centres: 1 Setting: specialised dental care clinic Recruitment period: not reported Funding source: not reported
Participants	Inclusion criteria: need for 2 subsequent treatment sessions with LA, age between 4 and 11 years and no suspected or known developmental delay Number of participants randomised: 147 (Group 1: 76, Group 2: 71) Number of participants evaluated: 127 (Group 1: 67, Group 2: 60) Number of males/females: 76 males and 71 females Age range: 4 to 11 years Mean age (years): 6.4, SD: 1.7 (Group 1: 6.3, SD: 1.7; Group 2: 6.4, SD: 1.6) No differences found between groups regarding age, gender, experience of LA in the previous 6 months
Interventions	Group 1 (control): LA delivered using a conventional syringe for 2 consecutive appointments Group 2: LA delivered using the wand for 2 consecutive appointments Sites for wand injections were: anterior middle superior alveolar; palatal anterior superior alveolar and for lower teeth periodontal ligament LA was used Sites for conventional injections were: for the maxilla, buccal and palatal; and for lower teeth, mandibular block was used. Topical anaesthetic used in both groups
Outcomes	Children were video taped and all treatments were analysed by 2 independent observers: a psychologist and a third year dental student. Observations were divided into 3 stages: anticipation phase (from the moment child enters surgery to start of LA), during delivery of LA, and after delivery of LA Pain‐related behaviour: rated in 15‐second intervals. 5 behaviours were assessed: body movement muscle tension, crying or screaming, verbal protest and bodily resistance. This was measured prior to and during delivery of LA Distress: measured using Venham's (modified) clinical rating of anxiety and co‐operative behaviour. The scale consists of 6 points: relaxed, uneasy, tense, reluctant, resistant, out of contact or untreatable, from 1 to 6. This was measured prior to and during delivery of LA Self‐reported pain: measured using a modified version of VAS, with11 points from 0 (no pain) to 10 (worst pain possible). 6 faces, expressing different levels of pain/distress, were added for children to choose the face matching their own level of pain/distress. This was completed by children following delivery of LA Dental anxiety: parents completed the parent version of the Dental Subscale of the Children's Fear Survey Schedule (CFSS‐DS). Each item is scored on a 5‐point scale, from 1 to 5. Scores below 32 are considered to be of non‐anxious children. This questionnaire was filled in by parents as the treatment was being carried out ‐ parents were kept in waiting room while child was being treated. Preoperative anxiety only (without comparison to a postoperative measurement of anxiety) is not an outcome for this review, as unsure of when this was undertaken, it was not included in this review
Notes	Consent and ethical approval obtained Observers were trained and there is a reliability analysis The video tapes from the study were evaluated by both observers independently and in case of disagreement a final rating was reached by joint decision
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Each child was randomly assigned to either the Wand (n = 71) or the traditional injection (n = 76) condition based on a randomisation list generated by SPSS (SPSS Inc, 12.0, Chicago, USA)"
Allocation concealment (selection bias)	Unclear risk	Unsure how concealment was achieved, however reference to dentists not knowing what type of LA was to be delivered until they decided which tooth to treat. Quote: "To avoid possible preference of two dentists, they were required to decide on the tooth to be treated before the anaesthetic condition was revealed"
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not possible to blind operators to the intervention. However, this may have introduced bias, as this may have influenced the speed of LA delivery, which was found to be different in both groups ‐ see 'other bias' section Not discussed if children were blinded to intervention. Typically the wand has a 'beeping noise' however, this was not addressed in the discussion
Blinding of outcome assessment (detection bias) All outcomes	High risk	Not reported whether observers were blinded to the intervention. Although it would not be possible to blind the observers due to the different presentation of both syringes, it may have introduced bias in rating the children's behaviour
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Quote: "For 20 children only their first treatment session could be included due to rescheduling of the second appointment." CONSORT flow chart shows that 9 were in the control group and 11 in the intervention group
Selective reporting (reporting bias)	Low risk	All recorded outcomes were reported on within the results section
Other bias	Unclear risk	Different speeds for delivery of LA in control and study groups may have biased results, due to reports of increased speed causing more pain. Furthermore, as the operator was not blinded to the intervention, it is possible that the difference in delivery times might have been biased. By the other hand slow delivery is one of the benefits of the wand, additionally authors report that: "children who are already reacting negatively to an injection seem to be longer in distress with the Wand system", and this may have introduced bias too

ACDAS = Abeer Children Dental Anxiety Scale; ASA = American Society of Anesthesiologists physical status classification system; AV = audiovisual; CI = confidence interval; IANB = inferior alveolar nerve block; LA = local anaesthetic; SD = standard deviation; VAS = visual analogue scale.