Unusual Findings in Trials Evaluating Adjuncts to Scaling and Root Planing: Reporting Quality (Part 2).

Abstract

On the topic of adjuncts to scaling and root planing (SRP), numerous randomized clinical trials (RCTs) were published by a single group of authors and frequently reported unusually large effect sizes. A meta-analysis in part 1 of this project failed to explain the causes for these unusual findings. We assessed the reporting quality and trial registration discrepancies to examine the possibility of replicating the work of this research group as well as the overall rigor of the research methodology. This study was preregistered at the Open Science Framework (https://osf.io/4meyd/). The Scopus platform was utilized for the RCT search on SRP adjuncts in intrabony defects in patients with periodontitis as compared with SRP alone. The search analysis was limited from 2010 to 2017, and RCTs on SRP adjuncts published by a single research group were selected for screening and inclusion. RCT registration records were assessed for consistency. Out of 92 studies that were retrieved from Scopus and PubMed, 32 were included for quality assessment per the CONSORT guidelines (Consolidated Standards of Reporting Trials). Results showed that all RCTs were characterized by a low reporting quality. The majority of CONSORT items scored <50%, including critical items (randomization, registration, and blinding). When registration records were compared with published RCTs, several discrepancies were found. The per-protocol follow-up duration was compared against the study's initiation and termination dates. Only 38% of the RCTs presented a follow-up period within the initiation and termination dates. The remaining RCTs showed inconsistent follow-up in comparison with the initiation and termination dates. RCTs by this group were characterized by poor adherence to reporting quality guidelines. Crucial RCT elements, such as randomization, blinding, and primary outcomes, were not reported properly. RCT registration records revealed systematic inconsistencies when compared with RCT publication. Therefore, the unusually large effects reported by this group should be viewed with extreme caution. The included randomized clinical trials were characterized by poor adherence to reporting quality guidelines, missing information about important trial items, and discrepancies between the reports and trial registrations. This quality assessment should guide clinical research and show clinicians that they should be cautious when applying evidence in their clinical practice.