Cariology Clinical Trials: What Are We—and What Should We Be—Looking At?

Randomized control trial (RCT) methodology has compared interventions for the prevention and management of dental caries since the late 1960s. Despite almost 50 years and evidence of significant wastage within the wider biomedical research field, there has been little investigation into what works well and where weaknesses lie. This paper aims to draw attention to areas for improvement within cariology clinical trial methodology by summarizing systematic reviews on interventions and outcomes, and using examples to illustrate some challenges with intervention delivery fidelity, outcome analyses, and intervention co-production. Trial design stage choices are critical to ensure that optimum information is obtained when testing interventions. Intervention choice, outcome choice, and analyses are particularly important, and cariology trials have specific issues associated with them. A systematic search and review of cariology RCTs found 650 RCT reports. Social Network Analysis of interventions revealed a high degree of separation between prevention and management trials, gaps in clinically important comparisons, and a tendency for there to be comparisons within groups; e.g., comparison of interventions within the same, rather than different, levels of invasiveness. Outcomes measured for the same trial reports show: a focus on restoration performance and individual/population caries burden; the growing use of carious lesion activity and economic-related outcomes; and sparse, although an increase in the use of, patient-reported/patient-centered outcomes. Fidelity of adherence to complex interventions can be challenging to measure but is important in interpreting trial findings. Involving target populations in intervention design, delivery, and relating it to the planned rollout, are opportunities to ensure intervention relevance and improved uptake. Outcomes analyses should consider the minimum clinically important differences and outcome relevance measures for the target population. Factors underlying trialists’ comparator and outcome choices need to be identified, and there is a need to ensure that a minimum dataset of outcomes allow for combination and comparisons of trial data for systematic review.