A systematic review of the quality and scope of decision modelling studies in child oral health research.

BACKGROUND: Decision analytic models are often used in economic evaluations to estimate long-term costs and effects of treatment which span beyond the time-frame of a clinical trial, therefore providing a better understanding of the long-term implications of decisions that conventional trial-based economic evaluations fail to provide. This is particularly relevant for considering oral health interventions in children as treatments may affect adult oral health. However, in the field of child oral health there has not been an evaluation of the quality and scope of decision analytical models which extend into adulthood. The aim of this review is to examine the scope and quality of decision modelling studies, with horizons extending into adulthood, within the field of child oral health. METHODS: The following databases were searched: NHS Economic Evaluation Database (CRD York), MEDLINE, EMBASE, CINAHL, Web of Science, Scopus, the Cochrane Library and Econlit. Full economic evaluations, in the field of child oral health, published after 1997 which included a decision model with a horizon that extended beyond the age of 18 years old were included. Included studies were appraised against the Drummond checklist and the Consolidated Health Economic Evaluation Reporting Standards by calibrated reviewers. RESULTS: Four hundred studies were identified, of which nine met the inclusion criteria. Of the nine, eight were cost-effectiveness models. The majority focussed on the prevention or management of dental caries. The mean percentage of applicable Drummond checklist criteria met by the studies in this review was 82% (median = 85%, range = 54-100%). Discounting of costs and performing an incremental analysis were noted as key methodological weaknesses. The mean percentage of applicable CHEERS criteria met by each study was 82% (median = 87%, range = 32-96%). Justifying the type of model, analytical methods used, and sources of funding were most commonly unreported. CONCLUSIONS: There is a paucity of decision analytical models in the field of child oral health. Most of those that are available are of high methodological and reporting quality.

Conscious sedation in children: the need to strengthen the evidence base remains.

Data sources Cochrane Oral Health's Trials Register, Cochrane Central Register of Controlled Trials (CENTRAL), Medline and Embase. The US National Institutes of Health Ongoing Trials Register (ClinicalTrials.gov) and World Health Organisation International Clinical Trials Registry Platform were searched for ongoing trials. Reference lists of eligible studies were checked for additional studies and specialists in the field contacted for any unpublished data. No restrictions were placed on language or publication date.Study selection Studies were selected which met the following criteria: randomised controlled trials of conscious sedation undertaken by a dentist, anaesthetist or one of the dental team comparing two or more drugs/techniques/placebo in children (up to 16 years of age) receiving dental treatment. Crossover trials and studies involving complex surgical procedures were excluded. Data extraction and synthesis Two authors independently selected studies for inclusion, extracted data and assessed for risk of bias. Results were compared and inconsistencies noted, with disagreements resolved by discussion. Where information was unclear or incomplete the authors of trials were contacted for clarification. Results Fifty studies (3704 participants) were included and grouped into placebo-controlled, dosage and head-to-head comparisons. There was wide variation in sedation technique and agent(s) employed across studies (34 different sedatives with or without nitrous oxide). Risk of bias was high for forty studies (81%), low for one study and unclear for the remaining nine studies (18%). Meta-analysis of available data for the primary outcome measure (behaviour) was possible for oral midazolam versus placebo only. There is moderate-certainty evidence from six small clinically heterogeneous studies at high or unclear risk of bias, that oral midazolam in doses between 0.25 mg/kg to 1 mg/kg is associated with more co-operative behaviour compared to placebo. It was not possible to draw conclusions regarding secondary outcome measures (completion of treatment, postoperative anxiety, adverse events) due to inconsistent and/or inadequate reporting.Conclusions There is some moderate-certainty evidence that oral midazolam is an effective sedative for dental treatment in children. Improvements and greater consistency in the design and reporting of future research will enable further evaluation of sedation agents and their potential implications for practice; with it being suggested future trials evaluate experimental regimens in comparison with oral midazolam or inhaled nitrous oxide.