Quality assessment of randomised controlled trials in oral and maxillofacial surgery

Abstract

Randomised controlled trials are the best way to study the evaluation of treatments. We have evaluated the quantity and quality of clinical trials in three of the main journals in the specialty of oral and maxillofacial surgery between January 2010 and December 2016, using a scientometric analysis, and evaluation by the Jadad scale. In this period, 303 randomised controlled trials (5% of the total) were identified; the largest number of studies were from Asia (45%) followed by Europe (32%). The subgroup that concerned most studies was oral surgery. The mean score on the Jadad scale was 3.06 points, which means that 32% of the total studies had a low risk of bias. Studies that declared funding and adherence to Consolidated Standards of Reporting Trials (CONSORT) were given significantly higher scores (p < 0.001) than studies that did not. We conclude that randomised controlled trials in oral maxillofacial surgery have evolved in both quality and quantity since previous surveys were published. The quality of trials was related to the presence of funding and adherence to CONSORT.

Introduction

The medical and dental communities consider that randomised controlled trials (RCT), systematic reviews of RCT, and meta-analyses, are the most scientifically stringent ways of comparing the efficacy of one intervention with that of another. Oral and Maxillofacial Surgery (OMFS), which is considered either a medical or a dental specialty depending on the region, has a history of few publications of RCT and, consequently, fewer systematic reviews and meta-analyses.

Several studies have shown that the papers published by OMFS journals contained little high-quality evidence and had a high risk of bias. It was reported that only 1% to 3% of papers in 10-year and 3-year periods were RCT. The quality of the methods used in an RCT is commonly used to assess the estimates of the risk of bias about the effects of treatment. Overall, trials for which methods of randomisation are not stated and allocation is not blinded, report increased effects of treatments.

There are techniques to assist authors with recommendations to improve the reporting of different study designs, which increase the quality and facilitate critical appraisal and interpretation of these studies by reviewers, editors, and readers.

However, published investigations of the quality of such trials in OMFS are still scarce. To the best of our knowledge, since 2012 there have been no studies published that have addressed this topic. This information could be helpful to assess how OMFS has evolved, as well as to describe some important aspects related to the quality of RCT. The aim of this study therefore was to assess the quantity and quality of RCT published between January 2010 and December 2016 in three widely-read journals within the specialty.

Material and methods

A retrospective scientometric analysis was conducted of the British Journal of Oral and Maxillofacial Surgery (BJOMS), the International Journal of Oral and Maxillofacial Surgery (IJOMS), and the Journal of Oral and Maxillofacial Surgery (JOMS) from January 2010 to December 2016. All the papers published were analysed online on Science Direct through the Coordination for the Improvement of Higher Education Personnel – CAPES Periodicals Portal. Letters to the Editor, Annals of Congress, and editorials were excluded.

The following scientometric data were collected from the selected articles: year of publication, journal name, the location of the study, the study design, and specialty areas. When the paper had more than one author at different locations, the address of the corresponding author was taken into account.

To define the types of design, the criteria of the Oxford Centre for Evidence-Based Medicine (CEBM) were applied. The 10 specialty areas were categorised in a way similar to that used by the three journals: dentoalveolar surgery; temporomandibular joint disorders and orofacial pain; orthognathic surgery and dentofacial deformity; maxillofacial trauma, pathology, and oncology; reconstruction and grafts; dental implants; anaesthesia; cleft lip and palate; and those from other fields (for example, otolaryngology, neurosurgery, and periodontology).

The RCT were read in full and assessed by the Jadad scale, described as the best reliability and validity scale for assessing RCT. The scale consists of five questions that can be answered “yes” or “no”, reaching an overall score from 0–5. The studies are scored according to the presence of three key features of the conduct of clinical trials: randomisation, blinding, and allocation of participants. The studies with scores of four and five were considered to have a low risk of bias and appropriate quality. For studies with this design, we also assessed sample size, whether or not there was adequate funding, and whether or not there was adherence to Consolidated Standards of Reporting Trials (CONSORT). Pilot studies were also included in the sample.

The research worker (BT) was trained to categorise the study designs and to use the Jadad scale, and also collected the data. In case of doubt, the paper was submitted to a full-text analysis, discussed with a second more experienced research worker (MLM), and standardised for the assessment method. For the RCT a sample (10% of the total, n = 30) was selected so that the research worker (BT) could re-evaluate the sample to ensure intra-examiner reproducibility. The same sample was also evaluated separately by a second research worker with more experience (MLM), to ensure inter-rater reproducibility. The agreement was 100% for both intra-examiner and inter-rater reproducibility.

Statistical analysis

The data were saved and analysed using IBM SPSS software (version 20.0, IBM Corp). Continuous variables were expressed as mean (SD), and categorical variables as absolute frequency (%). The significance of differences between groups was assessed using an independent t test, and probabilities of 0.05 or less were accepted as significant.

Results

The distribution of papers according to design is presented in Table 1 . There were 303 RCT (5.2%) and the sample sizes ranged from 6 to 1121.

Table 1
Distribution of articles by study design. Data are number.
Study design IJOMS JOMS BJOMS Total
Systematic review (randomised controlled trial) 13 13 5 31
Systematic review 97 52 30 179
RCT 113 140 50 303
Non-randomised controlled trial 29 61 13 103
Observational analytical (cohort study, case control, cross-sectional) 543 929 531 2003
Case series and case reports 427 1036 272 1735
Non-systematic review 28 99 93 220
Non-clinical (in vitro study/animal) 246 328 153 727
Technical note 88 178 218 484
Others 29 12 37 78
Total 1613 2848 1402 5863
Only gold members can continue reading. Log In or Register to continue

Aug 5, 2020 | Posted by in Oral and Maxillofacial Surgery | Comments Off on Quality assessment of randomised controlled trials in oral and maxillofacial surgery
Premium Wordpress Themes by UFO Themes