Outcomes in randomised controlled trials in prevention and management of carious lesions: a systematic review

Inconsistent outcome reporting is one significant hurdle to combining results from trials into systematic reviews. Core outcome sets (COS) can reduce this barrier. The aim of this review was to map outcomes reported in caries prevention and management randomised controlled trials (RCT) as a first step to COS development. We also investigated RCT characteristics and reporting of primary outcomes and sample size calculations.

PubMed, Embase, Web of Knowledge and Cochrane CENTRAL were systematically searched (1 January 1968 to 25 August 2015).

RCTs comparing any technique for prevention or management of caries with another or placebo and RCTs comparing interventions to support patients undergoing treatment of caries (without setting, dentition or age restrictions). Categories were developed through piloting and group consensus and outcomes grouped accordingly.

Of 4773 search results, 764 were potentially relevant, full text was available for 731 papers and 605 publications met the inclusion criteria and were included. For all outcomes across the time periods 1968-1980 and 2001-2010, reporting of outcome 'caries experience' reduced from 39% to 18%; 'clinical performance of the restoration' reporting increased from 33% to 42% although there was a reduction to 22% in 2011-2015. Emerging outcome domains include 'lesion activity' and 'pulp health-related outcomes', accounting for 1% and 0%, respectively, during 1968-1980 and 10% and 4% for 2011-2015. Reporting 'resource efficiency' and 'quality of life measures' have remained at a low level. No publications reported tooth survival independent of an index such as DMFT or equivalent. Primary outcomes were only identified as such in 414 (68%) of the reports.

Over the past 50 years, outcome reporting for trials on prevention and management of carious lesions have tended to focus on outcomes measuring caries experience and restoration material clinical performance with lesion activity and cost-effectiveness increasingly being reported. Patient-reported and patient-focused outcomes are becoming more common (although as secondary outcomes) but remain low in use. The challenge with developing a COS will be balancing commonly previously reported outcomes against those more relevant for the future.

PROSPERO, CRD42015025310 . Registered on 14 August 2015, Trials (Schwendicke et al., Trials 16:397, 2015) and COMET initiative online (COMET, 2017).