Overcoming Avoidable Research Waste in Carious Lesion Management Research

Abstract

Dental caries is a preventable, yet extremely prevalent, non-communicable oral disease. The clinical manifestations (i.e. carious lesions) and their sequelae bring a significant health burden including loss of function, pain, discomfort, reduced quality of life and tooth loss. This health burden also has a negative social and economic impact on individuals and societies. Although the aetiopathogenesis of the disease is now better understood, with effective preventive strategies identified and developed, uncertainty surrounds how best to manage the lesions when they arise for both short-term and long-term health and wellbeing.

Relevant research has not always resulted in widespread adoption of novel techniques and approaches. This is, in part, due to limitations arising from the avoidable waste resulting from issues with the conduct of the research.

This thesis explores the problem of research waste in carious lesion management research and identifies ways in which this can be mitigated to increase the value of clinical research and, ultimately, improve clinical care for patients.

The thesis begins with a narrative review of dental caries including the aetiopathogenesis and prevention of the disease. The changing epidemiology and current treatment philosophies are also presented. Thereafter, the problem of avoidable waste in healthcare research is explored, within the context of the proposed Global Evidence Ecosystem for Oral Health (GEEOH). Specifically, how the problem of inconsistent outcome selection and reporting in both primary (e.g. Randomised Controlled Trials) and secondary research (Systematic reviews) can be addressed to contribute to the building of an effective and efficient evidence ecosystem.

A systematic assessment of relevant systematic reviews (SRs) is presented in chapter 3. This chapter highlights the growing problem of overproduction of systematic reviews and the failure to appropriately update these reviews. An analysis of updated systematic reviews is presented which shows that several problems arise from the primary literature on which they are based. Inconsistent outcome selection and reporting is one of those problems. Solutions to the issue of overproduction of uninformative reviews are presented, including a coordinated move towards living reviews (LSRs) incorporating Network Meta-Analysis (NMA).

For future results to be meaningful for dental care providers and patients in clinical decision making, the outcomes of importance to these groups needed to be established. The central project in this thesis is the development of an agreed Core Outcome Set (COS) to help harmonise outcome selection and reporting in primary and secondary research in the field of carious lesion management. This COS was developed by following the methodological framework proposed by the Core Outcome Measures in Effectiveness Trials (COMET) initiative.

The systematic review of randomised controlled trials (RCTs) was conducted to complete stage 4(i) identify existing knowledge, of the COMET framework. From this review 76 unique outcomes were identified from 605 RCTs involving over 250,000 patient participants. An analysis of the selection and use of trial outcomes is also presented. The 76 outcomes identified in this stage were taken forward to the two-step consensus process. Two consensus methods were combined to fulfil the COMET initiative’s stages: 4(ii) Filling gaps in knowledge and 4(iii) Eliciting views about important outcomes in a consensus process and 4(iv) Holding a face to face meeting to finalise the recommended core outcome set.

The international e-Delphi survey involved 88 participants (21 patients and 67 dental professionals or researchers) who prioritised 24 outcomes to be taken forward to the face to face meeting.

The face to face consensus meeting was held in Dundee and involved 13 participants (seven patients and six dental professionals and researchers). At this meeting, 23 outcomes were discussed in a modified Nominal Group Technique meeting prior to selecting the core outcome set by anonymous electronic voting. Views on a final outcome were sought from the same group via email after the original meeting. In alphabetical order the seven outcomes agreed by the stakeholder group are: ‘Irreversible Pulpitis’, ‘Oral Function’, ‘Patient Satisfaction’, ‘Quality of Life’, ‘Remineralisation of the Lesion’, ‘Survival of the Restoration’, and ‘Tooth Survival’.

To help reduce outcome heterogeneity and improve the integration between primary and secondary research, this core outcome set should be used for future randomised controlled trials and systematic reviews. Translation of this evidence to help inform patient care will be improved if these can be successfully applied in future research.

The agreed COS is then mapped to the outcomes selected for relevant Cochrane reviews. This analysis shows that none of the relevant Cochrane reviews maps directly to the COS. Furthermore, they also show outcome heterogeneity between Cochrane reviews. This highlights the need to align all stages of the GEEOH, including evidence syntheses, to the agreed COS.

A “new normal” is needed in healthcare research generally, and carious lesion management research specifically, to increase value and reduce waste. A coordinated approach to synchronising methodologies, including outcome selection, in all stages of the GEEOH is recommended. Cochrane are ideally placed to co-ordinate these aspects of the GEEOH. Should the COS developed in this thesis be adopted by Cochrane Oral Health in future living systematic reviews and network meta-analyses, it would represent an important step in harmonising outcome selection and reporting within the GEEOH. Researchers conducting primary research should be encouraged and supported to use standardised methods and outcomes to allow synthesis into living network meta-analyses. Evidence gaps should be addressed by broad global collaborations rather than narrow, competing attempts to be the first to publish results. It is suggested that researchers throughout the GEEOH must become more willing to share knowledge and skills with each other and be less protective of their own interests. Advocates for this coordinated way of working are now needed at each stage of the GEEOH to support this change to research culture.

Date of Award	2020
Original language	English
Awarding Institution	University of Dundee
Supervisor	David Ricketts (Supervisor) & Janet Clarkson (Supervisor)