This was a 2-arm cluster-randomized control trial comparing EL and TSD for their effectiveness in reducing DA in children. This study had the approved by the Ethics Committee of the Shenzhen Nanshan Maternal and Child Health Care Hospital. All parents of the children participated provided written informed consent. The trial was registered in Chinese Clinical Trial Registry (ChiCTR2000028878).
The participants were children aged 7-8 years, selected in September 2018 from the second grade of primary schools in the Nanshan District of the city of Shenzhen, China. At this age, children in Shenzhen routinely receive pit and fissure sealant (PFS) as a prophylaxis for dental caries in permanent molar teeth, which is noninvasive procedure. Although PFS is harmless, children still show anxiety facing the needle and fissure sealant .
The eligibility criteria for second grade classes were: 1) located in Nanshan District, Shenzhen City; 2) had 40 to 50 children aged 7-8 years; 3) agreed to participate. In September 2018, we used a multi-stage random sampling method to recruit the participants. We first used a table of random digits to select six primary schools in Nanshan that had four or more classes met eligibility criteria, and then random selected four eligible classes in each of the six schools. All children in the 24 selected classes were invited to take part in the study, and the informed consent were sent to their parents (Fig. 1). Children who already had a dental visit or refused to participate were excluded.
Randomization and Interventions
In this trial, the class was the unit of randomization. The sampled classes were randomly assigned into the EL or TSD groups in 1:1 ratio in school blocks with the use of a table of random digits (Fig. 1). Cluster randomization was necessary to avoid between-group contamination and it was feasibly conduct school-based intervention. After receiving the signed informed consent from the parents, the intervention was implemented on a class by class basis. Neither the dentists, nurses nor field interviewers who assessed the outcomes were aware of the randomization assignments of the classes, as the measurements were the same for both groups. Only the dental nurses who performed the “Tell and Show” procedures were aware the assignments.
At the beginning of intervention, children in the EL group received a lively 40-minute presentation on oral health given by dentists. The standardized oral health education conveyed information about dental caries and their causes, proper toothbrushing and flossing, and the pit and fissure sealant, by using slides, cartoon videos, and dental models. One month later, the children participated in an EL activity in a simulated dental clinic in the classroom. In the activity, the dentists demonstrated the common tools of dentistry (mouth mirror, dental probe, dental handpiece and dental ejector, etc.). Then the children were organized in 4-person-groups for a five-minute role play around a dental bed, with which the children playing a patient, a dentist, a nurse and a parent. While one group of children performed their role play, the other children were in another room watching cartoon videos that delivered oral health education. And another month after that, they visited the dental clinic to receive the PFS treatment.
Children in the TSD group received a conventional 5-minute TSD behavior management session at the time of their visit to the dental clinic for the PFS procedure. The TSD behavior management process used in this study contained three procedures: Tell, a dental nurse explained to the child what the dentist would do during the PFS treatment; Show, the dentist nurse showed the equipment involved; Do, the dentist performed the PFS procedure.
Another dental nurse (not the same one who delivered the TSD) performed the measurements of the children’s blood pressures and pulse rates with an upper-arm electronic sphygmomanometer in the waiting room after a 15-minute quiet sitting period before the PFS treatment, and then remeasured them again after the treatment.
The field interviewers administered the modified children’s fear survey schedule-dental subscale (Modified CFSS-DS)  to the children after the PFS. In this study, we defined the DA as a sum score of Modified CFSS-DS equal or more than 38. The modified CFSS-DS was translated into Chinese from the original English version and was combined with a facial image scale (FIS) . It had a Cronbach's alpha of 0.85 and a test-retest reliability ICC of 0.73 .
The primary outcome was the prevalence of DA. The secondary outcomes were mean differences of change in systolic blood pressures (SBP), diastolic blood pressures (DBP) and pulse rates (PR) in the measurements taken before and after PFS.
With an enrollment of 12 class per trial group, 40 participants per class, a rate of cohort retention of 80%, a rate of PFS of 90%, and an estimated intraclass correlation coefficient (0 − 0.01), the initial design yielded 80% power to detect a 10% difference in the prevalence of DA between TSD and EL groups, with a two-sided alpha level of 0.05.
The intervention effects were estimated by means of mixed effect logistic regression for prevalence of DA and linear mixed-effects model for continuous variations (DA score, SBP, DBP and PR), which included a random cluster effect of class. The primary predictor was group (EL versus TSD) and included covariates of gender (male or female), school, and baseline value (for analysis of change of SBP, DBP or PR only), which were correlated with the dependent variable. All analyses were performed with the R project (Version 3.6.1 Patched for x64 Window system), and a P value <0.05 was considered statistically significant.