Eur Arch Paediatr Dent DOI 10.1007/s40368-014-0153-9
ORIGINAL SCIENTIFIC ARTICLE
An assessment of the quality of reporting randomised controlled trials published in paediatric dentistry journals S. Rajasekharan • J. Vandenbulcke L. Martens
•
Received: 29 April 2014 / Accepted: 19 September 2014 Ó European Academy of Paediatric Dentistry 2014
Abstract Aim The objectives of this study are to compare the quality of reporting of randomised controlled trials (RCTs) published in 2011 and 2012 within five paediatric dentistry journals. Study design RCTs published in the years 2011 and 2012 were hand-searched by one reviewer. After randomisation and blinding, these journals were independently scored by two blinded reviewers based on the CONSORT 2010 checklist. Methods A total of 59 articles were included for analysis and 70 criteria were scored dichotomously as ‘1’ when reported and ‘0’ when not reported. Descriptive statistics and one-way ANOVA tests were performed. Results The Gwets AC1 Inter rater reliability coefficient was calculated as 0.85 (95 % C.I 0.84–0.86) indicating excellent correlation between the two reviewers. Only 19 articles (32.2 %) reported more than half (35/70) of the expected criteria. Descriptive statistics showed that sections such as introduction, results and discussion were reported better than abstract, materials and methods and other information. One-way ANOVA tests showed no significant difference (p [ 0.05) in the reporting of criteria across different journals and there was also no significant difference between the articles published in 2011 and 2012 (p [ 0.05). A part of this study was presented as an oral presentation in the 24th International Association of Paediatric Dentistry (IAPD) conference held in Seoul, South Korea. S. Rajasekharan (&) J. Vandenbulcke L. Martens Department of Paediatric Dentistry and Special Care, PaeCaMeD Research, Ghent University, UZ Ghent, 1P8, De Pintelaan 185, 9000 Ghent, Belgium e-mail: [email protected]
Conclusion The general quality of reporting of RCTs in paediatric dentistry journals was inadequate. Authors, reviewers and journal guidelines must work together towards a common goal for improving the quality of reporting of RCTs. Keywords
Randomised controlled trial CONSORT
Introduction A randomised controlled trial (RCT) is a quantitative, comparative study where participants of the trial are assigned by chance to treatment or control groups that are followed over time to determine the effect of the intervention. Evidence-based hierarchies place RCTs just below systematic reviews as the highest form of evidence (Concato et al. 2000; Turner et al. 2012; Fleming et al. 2012) that could be achieved from an in vivo/clinical trial (Rosner 2012). RCTs are widely accepted as the ‘‘gold standard’’ for obtaining unbiased estimates of treatment effects (Glasziou et al. 2007). Reporting of a RCT in a transparent manner in all steps from protocol to publication ensures reliability and reproducibility, which are essential for other researchers and practitioners. Clear reporting is indeed a prerequisite to help the readers to judge the quality of the RCT (Williams 2010; Chalmers 1990a). Researchers, authors, reviewers and journals would benefit if a single framework of recommendations and checklist was followed that would enhance the quality of reporting of RCTs (Begg et al. 1996). Consolidated Standards of Reporting Trials (CONSORT) was formed by a group of medical journal editors, clinical trialists, epidemiologists, and methodologists with
123
Eur Arch Paediatr Dent
the aim of developing a new framework for the assessment of the quality of RCT reports (Moher et al. 2001). The CONSORT statement was first published in 1996 (Begg et al. 1996) followed by the revised statement in 2001 (Harrison 2003) and 2010 (Moher et al. 2010). The CONSORT 2010 statement had, in general, simplified and clarified the wording in nine previously existing criteria with addition of subitems to previously existing criteria. The major amendments to the CONSORT 2001 statement were the addition of new criteria such as reporting of how ‘‘blinding’’ was accomplished, statistical analysis of secondary outcomes, whether the trial was registered, where a trial protocol could be found and funding information (Schulz et al. 2010). Currently, the CONSORT checklist to assess quality of reporting of RCTs is endorsed by over 38 % of the core medical journals listed in PubMed (Altman et al. 2012). A RCT report based upon the CONSORT checklist makes it easier for the readers to judge the reliability and relevance of the results of the trial (Moher et al. 1995). The quality of reporting RCTs is regularly assessed in the field of medicine and dentistry using the CONSORT checklist (Turner et al. 2012). In paediatric dentistry, only one study is available by Al-Namankany et al. (2009) that analysed RCTs published between 1985 and 2006 using the CONSORT 2001 checklist. Those authors concluded that the overall quality of reporting of clinical trials was poor and inadequate to allow readers to assess trial validity (AlNamankany et al. 2009). Since the declaration of the revised CONSORT 2010 checklist, no studies analysing RCTs have been published in paediatric dentistry journals. The objective of this study was to compare the quality of reporting of RCTs published in 2011 and 2012 within the journals analysed to test a null hypothesis (Ho) being that the distribution of scores is the same across all the analysed journals.
Materials and methods In the present study, the authors scored the quality of reporting of RCTs in five different paediatric dentistry journals based on the CONSORT 2010 checklist. All RCTs published in the years 2011 and 2012 in five different paediatric dentistry journals were: European Archives of Paediatric Dentistry (EAPD), European Journal of Paediatric Dentistry (EJPD), International Journal of Paediatric Dentistry (IJPD), Journal of Clinical Pediatric Dentistry (JCPD) and Pediatric Dentistry (PD). All manuscripts were hand-searched by one reviewer (LM) to include all articles reporting clinical trials, where the samples were assigned in a randomised manner to
123
either a treatment or control group. The updated CONSORT checklist was published in March 2010; hence articles from the year 2011 were selected for scoring assuming 9 months as a buffer period between acceptance of articles by the journal and their publishing date. The baseline data for article inclusion are shown in Fig. 1. A total of 59 articles were included and they were numbered from 1 to 59 for the purpose of anonymity. They were subsequently randomised using a computer generated table of random numbers. The assessors also ‘‘blinded’’ to the name of the journal, volume, issue, year of publication, name of author, affiliations, correspondence details, settings and locations where the study was carried out and details of ethical committee approval. Questions regarding setting, location, correspondence to authors and ethical committee were scored separately at the end after the blinding was revealed. Two ‘‘blinded’’ reviewers (J.V and S.R) then assessed the articles independently based on an expanded CONSORT checklist shown in Table 1. Conjoint questions in the CONSORT 2010 checklist were split for the purpose of accuracy and scored separately. Certain criteria in the original CONSORT 2010 checklist were omitted as they were not applicable to all the selected studies and their inclusion could result in a bias. The criteria omitted from the check list were 7b (when applicable, explanation of any interim analyses and stopping guidelines), 12b (methods for additional analyses, such as subgroup analyses and adjusted analyses), 17b (for binary outcomes, presentation of both absolute and relative effect sizes is recommended) and 18 (results of any other analyses performed, including subgroup analyses and adjusted analyses, distinguishing pre-specified from exploratory) of the original CONSORT 2010 checklist. No new criteria were added by the authors. All criteria were scored dichotomously as ‘1’ when reported and ‘0’ when not reported. If criteria expected in the materials and methods section were reported in the results section or vice versa, they were still scored as reported. Even if certain criteria from the results section were not directly reported, they were extracted from the tables and figures wherever possible. For the purpose of validation of the checklist, the two reviewers (JV and SR) jointly scored 10 randomly selected RCTs published in any of the five above-mentioned paediatric dentistry journals from 2007 to 2010. This procedure also facilitated better understanding of the criteria and pre-calibration of the reviewers before beginning with the reviewing process. Statistical analysis The articles that were included were scored across 70 criteria using MicrosoftÒ ExcelTM 2013 and the score sheets were analysed using MatlabÒ Version 8.0.0.783. The
Eur Arch Paediatr Dent Fig. 1 Flowchart showing the inclusion of articles analysed in this study on quality of reporting of RCT in paediatric dental journals
Gwets AC1 an inter-rater reliability coefficient was calculated using AgreeStat software version 2011.1 (Advanced Analytics, Gaithersburg, MD, USA). All
statistical analyses were completed using IBMÒ SPSSÒ Statistics Version 21.0.0.0. Descriptive statistics was used to analyse the scores of each article and each section within
123
Eur Arch Paediatr Dent Table 1 Checklist adapted from CONSORT 2010 checklist S. no.
Section
1
Title
2
Abstract
Topic
Modified checklist criteria Identification as a randomised trial in the title Structured summary of trial design, methods, results, and conclusions
3
Authors
Contact details for the corresponding authors
4
Trial design
Description of trial design
5
Methods
Eligibility criteria for participants
6
Settings where the data were collected
7
Interventions intended for each group
8
Specific objective or hypothesis
9
Clearly defined primary outcome for this report
10
How participants were allocated to interventions
11 12
Whether participants were blinded to group assignment Care givers were blinded to group assignment
13
Those assessing the outcomes were blinded to group assignment
14
Results
Number of participants randomised to each group
15
Trial status (recruitment)
16
Number of participants analysed in each group
17
For the primary outcome, a result for each group
18
The estimated effect size and its precision
19
Important adverse events or side effects
20
Conclusions
General interpretation of the results
21
Trial registration
Registration number and name of trial registry
22 23
Introduction
Funding
Source of funding
Background
Scientific background
Objectives
Specific objective or hypothesis
Trial design
Description of trial design (such as parallel, factorial)
24
Explanation of rationale
25 26
Methods
27 28
Allocation ratio Important changes to methods after trial commencement (such as eligibility criteria)
29 30
Reasons to important changes to methods after trial commencement Participants
31
Eligibility criteria for participants Settings where the data were collected
32
Locations where the data were collected
33
Interventions
34 35
Outcomes
The interventions for each group with sufficient details to allow replication How and when they were actually administered Completely defined pre-specified primary and secondary outcome measures, including how they were assessed
36
Completely defined pre-specified primary and secondary outcome measures, including when they were assessed
37
Any changes to trial outcomes after the trial commenced
38 39
123
Reasons for any changes to trial outcomes after the trial commenced Sample size
How sample size was determined
Eur Arch Paediatr Dent Table 1 continued S. no.
Section
Topic
40
Randomization
Sequence generation
Modified checklist criteria Method used to generate the random allocation sequence
41
Type of randomisation
42 43
Details of any restriction (such as blocking and block size) Mechanism used to implement the random allocation sequence (such as sequentially numbered containers)
Allocation concealment mechanism
44
Description of any steps taken to conceal the sequence until interventions were assigned
45
Implementation
Who generated the random allocation sequence
46
Who enrolled participants
47
Who assigned participants to interventions
48
Blinding
If done, who was blinded after assignment to interventions (for example, participants, care providers, those assessing outcomes)
49
How was blinding done
50
If relevant, description of the similarity of interventions
51 52
Results
Statistical methods
Statistical methods used to compare groups for primary and secondary outcomes
Participant flow
For each group, the numbers of participants who were randomly assigned
53
For each group, the numbers of participants who received intended treatment
54
For each group, the numbers of participants who were analysed for the primary outcome
55
For each group, losses and exclusions after randomisation
56
Reasons for losses and exclusions after randomisation
57
Recruitment
58
Dates defining the periods of recruitment and follow-up Why the trial ended or was stopped
59
Baseline data
60
Numbers analysed
61
Outcomes and estimation
A table showing baseline demographic and clinical characteristics for each group For each group, number of participants (denominator) included in each analysis and whether the analysis was by original assigned groups For each primary and secondary outcome, results for each group
62
For each primary and secondary outcome, the estimated effect size
63
For each primary and secondary outcome, its precision (such as 95 % confidence interval)
64
Harms
All important harms or unintended effects in each group (for specific guidance see CONSORT for harms)
Limitations
Trial limitations, addressing sources of potential bias, imprecision, and, if relevant, multiplicity of analyses
66
Generalisability
Generalisability (external validity, applicability) of the trial findings
67
Interpretation
Interpretation consistent with results, balancing benefits and harms, and considering other relevant evidence
Registration
Registration number and name of trial registry
Protocol Funding
Where the full trial protocol can be accessed, if available Sources of funding and other support (such as supply of drugs), role of funders
65
68 69 70
Discussion
Miscellaneous
Self-designed checklist for assessing the quality of reporting of RCTs based on the CONSORT 2010 checklist
an article. One-way an ANOVA test was performed to analyse any difference in distribution of scores within journals and also between the articles published in 2011 and 2012. The level of significance for all tests was set at 0.05.
Results There were 773 articles from the years 2011 and 2012 considered of which 59 (7.6 %) were selected as RCTs. The Gwets AC1 Inter rater reliability coefficient was
123
Eur Arch Paediatr Dent Table 2 Median, range and interquartile range for all 59 included articles per section and in total
Section (criteria number)
Number of criteria
Median
Range
Inter quartile range
Title (1)
1
0
0 (1–0)
Abstract (2–22)
21
10
8 (14–6)
1 (1–0) 2 (11–9)
Introduction (23–25)
3
3
1 (3–2)
0 (3–3)
Material and methods (26–51)
26
11
19 (26–7)
4 (13–9)
Results (51–64)
13
8
11 (12–1)
2 (8–6)
Discussion (65–67)
3
1
2 (3–1)
1 (2–1)
Other information (68–70)
3
0
2 (2–0)
1 (1–0)
Total of all criteria (1–70)
70
33
36 (57–21)
7 (37–30)
Fig. 2 Histogram depicting the number of articles reporting each checklist criterion
calculated as 0.85 (95 % confidence interval 0.84–0.86) indicating ‘excellent’ correlation between the two reviewers (Fleiss 1981). Descriptive statistics of those 59 articles included for each section and in total is presented in Table 2. Only 19 articles (32.2 %) reported more than half (35/70) of the expected criteria. Sections such as introduction, results and discussion were reported better than abstract, materials and methods and other information. Figure 2 illustrates the number of articles that had reported each individual criterion. Results for each section are discussed below. Title: The title section included only one criterion (criterion 1 of CONSORT checklist) and this was reported in only 15/59 articles (25.4 %) that identified the trial as a RCT in the title. Abstract: The abstract section consisted of 21 criteria (criteria 2–22 of CONSORT checklist). Four criteria were reported in all articles. These were criterion 3 (contact details for the corresponding authors), criterion 7 (interventions intended for each group), criterion 8 (specific objective or hypothesis) and criterion 17 (for the primary
123
outcome, a result for each group). Whether participants (criterion 11), as well as care givers were ‘‘blinded’’ as to group assignment (criterion 12) and the estimated effect size and its precision (criterion 18) were reported in only 1/59 (1.7 %) articles. None of the articles reported criteria 15, 21 and 22 (trial status, registration number and name of trial registry, source of funding, respectively). Introduction: The introduction section required the reporting of three criteria (criteria 23–25 of CONSORT checklist). This section was very well scored by all the articles with 57/59 articles (96.6 %) reporting all the three criteria (scientific background, explanation of rationale and specific objective or hypothesis). Materials and methods: Twenty six criteria (criteria 26–51 of CONSORT checklist) were covered in this section. More than 56/59 articles (94.9 %) reported 6/26 criteria. A high percentage of reporting was seen in criterion 30 (eligibility criteria for participants), criterion 33 (the interventions for each group with sufficient details to allow replication), criterion 34 (how and when the intervention was actually administered), criterion 35 and 36 (completely
Eur Arch Paediatr Dent
defined pre-specified primary and secondary outcome measures, including how and when they were assessed) and criterion 51 (statistical methods used to compare groups for primary and secondary outcomes). On the contrary, only one article reported criterion 37 and 38 (Any changes to trial outcomes after the trial commenced and the reasons for any changes to trial outcomes after the trial commenced, respectively). In the 59 RCTs analysed, randomization, allocation concealment and blinding were reported only in 25 (42.4 %), 8 (13.6 %) and 33 (55.9 %) articles, respectively. Results: This section contained 13 of numbers (criteria 52–64 of the CONSORT checklist). Criterion 52 (for each group, the numbers of participants who were randomly assigned) and criterion 61 (for each primary and secondary outcome, results for each group) were reported in 57/59 articles (96.6 %). Only 2/59 articles (3.4 %) reported criterion 62 (for each primary and secondary outcome, the estimated effect size). Discussion: The discussion section included only three criteria (criteria 65–67 of CONSORT checklist) of which criterion 67 (interpretation consistent with results, balancing benefits and harms, and considering other relevant evidence) was reported in all articles and only 5/59 articles (8.5 %) reported criterion 66 [generalisability (external validity, applicability) of the trial findings]. Other information: This section consisted of information about the registration number and name of trial registry, where the full trial protocol can be accessed, if available and sources of funding and other support (criteria 68–70, respectively, of CONSORT checklist). None of the articles reported a trial registration and only 4/59 articles (6.8 %) reported where the full trial protocol could be accessed. One-way ANOVA tests showed no significant difference (p [ 0.05) in the reporting of criteria across different journals and there was also no significant difference between the articles published in 2011 and 2012 (p [ 0.05).
Discussion The authors would like to emphasize that this study is purely an assessment of the quality of reporting of RCTs and is in no way indicative of the methodological quality of the studies themselves. Poor quality of reporting of RCT could be considered synonymous with poor quality of the trial (Fleming et al. 2012) as it gives rise to unreliable results and misleading conclusions (Chalmers 1990b). Poor reporting of a study could be due to poor methodological quality, while a wellconducted study could also be inadequately reported. It is almost impossible to make a distinction between the two. The results of the present study revealed that the quality of reporting of RCTs in paediatric dentistry is generally
inadequate and suboptimal. The latter findings are in accordance to the results of the study by Al-Namankany et al. (2009) who analysed articles published from 1985 to 2006 in paediatric dentistry journals with and without an impact factor (Al-Namankany et al. 2009). Moreover, these results are also in agreement with studies that analysed dental journals with a high impact factor (Pandis et al. 2010) and journals from other fields of dentistry such as orthodontics (Harrison 2003), periodontology (Faggion et al. 2012), implant dentistry (Cairo et al. 2012), prosthodontics (Jokstad et al. 2002) and dental public health (Marshman and Farid 2010). Four of the five journals included in this study had an impact factor but nevertheless there was no significant difference in the quality of reporting across all journals. This probably indicates the absence of adaptation of the CONSORT 2010 checklist. The present results being similar to the study published in 2009 suggest that there is still scope for improvement in this perspective. The CONSORT checklist emphasises the importance of negative reporting thus giving a clear understanding to any reader of the article over the exact methodology of the study. Negative reporting would mean that a procedure if not performed would still have to be reported. A study would hypothetically score points if it reports that no blinding was done; but on the other hand, a study where blinding had been carried out but not reported would not receive any points. The authors understand the practical difficulties of performing a RCT in a clinical setting and hence it is possible that a few of the CONSORT criteria could not be incorporated in the methodology. However, an explanation of these practical difficulties is an absolute necessity to help the reader to better understand the RCT and also create an awareness among future researchers. Even though negative reporting would result in more transparency of the RCT, negative reporting of all 70 criteria would be tedious and occupy a lot of space in a concise article. A pragmatic approach to this problem could be a disclaimer by the authors stating that the trial was developed according to the CONSORT guidelines and the absence of reporting of any particular criteria is due to its absence in the trial methodology. Not all readers have full access to an article and an abstract is a basis for selecting a research and helps the readership to obtain a quick overview of the complex research. In the present study, the authors found the abstract section to be reported sub-optimally in majority of the articles. A maximum of 14 criteria out of 21 were reported by one article, while all the rest scored at a lower level. The reason for minimal reporting in the abstract section could be attributed to the word limit imposed by a journal’s guidelines. A word limit of 200 or 250 words
123
Eur Arch Paediatr Dent
would make it almost impossible to report on all 21 criteria requested by the CONSORT checklist. Randomization sequence generation, blinding and allocation concealment are the three important parameters in assessing the risk of bias for RCTs (Higgins et al. 2008). In the present study, 34/59 (57.6 %) articles failed to report randomization sequence generation and 26/59 (44 %) articles missed to report blinding, while 51/59 (86.4 %) did not report allocation concealment thereby signifying a high risk of bias. Unlike ‘‘blinding’’ which could be impossible in certain cases, randomization is always feasible and ensures bias free allocation of subjects (Koletsi et al. 2012). Registration of clinical trials at their inception at a database that is publicly available is of long standing (Moher et al. 2010; DeAngelis et al. 2005) and the World Health Organization (WHO) states that ‘‘the registration of all interventional trials is a scientific, ethical and moral responsibility’’ (Moher et al. 2010). However, none of the 59 RCTs analysed in this study reported trial registration. In certain clinical situations where a huge sample size could not be obtained due to practical difficulties, systematic reviews and meta-analysis could provide valuable information by combining results of individual studies. However, poor reporting of RCTs would make it almost impossible to include/combine such studies within a meta-analysis/systematic review. Considering the huge amount of manpower, time and finance and/or funding invested in every trial, an added effort to report all the CONSORT checklist criteria could provide valuable information to reviewers and future researchers who wish to adopt the trial methodology. The quality of reporting RCTs is dependent on a triad of factors constituting the authors, reviewers and journal guidelines. Plint et al. (2006) and Turner et al. (2012) concluded from their systematic reviews that adoption of the CONSORT checklist by journals is associated with improvement in the quality of reporting of RCTs (Plint et al. 2006; Turner et al. 2012). When the author guidelines for the journals analysed within this study were considered, it was noted that only IJPD requests their authors to conform to the CONSORT guidelines while submitting RCTs. Adequate reporting of the RCT methodology helps the readers to assess and extrapolate the generalizability of the results of a trial to their own patients. A well-reported RCT encourages future researchers to adhere to CONSORT checklist at all stages such as preparing the protocol, executing the study and reporting the trial.
Conclusion The general quality of reporting of RCTs in paediatric dentistry journals was inadequate. There was no significant difference between the quality of reporting of articles
123
published across different journals (Ho retained) and between the quality of reporting of RCTs published in 2011 and 2012 (Ho retained). More paediatric dentistry journals could adopt the CONSORT checklist in their guidelines to the authors who wish to submit RCTs in future. Authors, reviewers and journal guidelines must work together towards a common goal for improving the quality of reporting of RCTs. Acknowledgments
No funding was received for this study.
References Al-Namankany AA, Ashley P, Moles DR, Parekh S. Assessment of the quality of reporting of randomized clinical trials in paediatric dentistry journals. Int J Paediatr Dent. 2009;19:318–24. Altman DG, Moher D, Schulz KF. Improving the reporting of randomised trials: the CONSORT Statement and beyond. Stat Med. 2012;31:2985–97. Begg C, Cho M, Eastwood S, et al. Improving the quality of reporting of randomized controlled trials. The CONSORT statement. JAMA. 1996;276:637–9. Cairo F, Sanz I, Matesanz P, Nieri M, Pagliaro U. Quality of reporting of randomized clinical trials in implant dentistry. A systematic review on critical aspects in design, outcome assessment and clinical relevance. J Clin Periodontol. 2012;39(Suppl 12):81–107. Chalmers I. Underreporting research is scientific misconduct. JAMA. 1990a;263:1405–8. Chalmers I. Underreporting Research Is Scientific Misconduct. JAMA. 1990b;263:1405–8. Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med. 2000;342:1887–92. Deangelis CD, Drazen JM, Frizelle FA, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. Arch Otolaryngol Head Neck Surg. 2005;131:479–80. Faggion JR, Clovis M, Giannakopoulos NN. Quality of reporting in abstracts of randomized controlled trials published in leading journals of periodontology and implant dentistry: a survey. J Periodontol. 2012;83:1251–6. Fleiss JL. Statistical methods for rates and proportions. New York: Wiley; 1981. Fleming PS, Buckley N, Seehra J, Polychronopoulou A, Pandis N. Reporting quality of abstracts of randomized controlled trials published in leading orthodontic journals from 2006 to 2011. Am J Orthod Dentofac Orthop. 2012;142:451–8. Glasziou P, Chalmers I, Rawlins M, McCulloch P. When are randomised trials unnecessary? Picking signal from noise. BMJ. 2007;334:349–51. Harrison JE. Clinical trials in orthodontics II: assessment of the quality of reporting of clinical trials published in three orthodontic journals between 1989 and 1998. J Orthod. 2003;30:309–15 (discussion 297–8). Higgins JPT, Green S, Cochrane Collaboration. Cochrane handbook for systematic reviews of interventions. Chichester: WileyBlackwell; 2008. Jokstad A, Esposito M, Coulthard P, Worthington HV. The reporting of randomized controlled trials in prosthodontics. Int J Prosthodont. 2002;15:230–42. Koletsi D, Pandis N, Polychronopoulou A, Eliades T. What’s in a title? An assessment of whether randomized controlled trial in a
Eur Arch Paediatr Dent title means that it is one. Am J Orthod Dentofac Orthop. 2012;141:679–85. Marshman Z, Farid F. The quality of reporting of randomised controlled trials in dental public health. Community Dent Health. 2010;27:253–6. Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c869. Moher D, Jadad AR, Nichol G, Penman M, Tugwell P, Walsh S. Assessing the quality of randomized controlled trials: an annotated bibliography of scales and checklists. Control Clin Trials. 1995;16:62–73. Moher D, Schulz KF, Altman DG. The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials. Lancet. 2001;357:1191–4. Pandis N, Polychronopoulou A, Eliades T. An assessment of quality characteristics of randomised control trials published in dental journals. J Dent. 2010;38:713–21.
Plint AC, Moher D, Morrison A, et al. Does the CONSORT checklist improve the quality of reports of randomised controlled trials? A systematic review. Med J Aust. 2006;185:263–7. Rosner AL. Evidence-based medicine: revisiting the pyramid of priorities. J Bodyw Mov Ther. 2012;16:42–9. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Ann Intern Med. 2010;152:726–32. Turner L, Shamseer L, Altman DG, et al. Consolidated standards of reporting trials (CONSORT) and the completeness of reporting of randomised controlled trials (RCTs) published in medical journals. Cochrane Database Syst Rev. 2012;11:MR000030. Williams HC. Cars, CONSORT 2010, and clinical practice. Trials. 2010;11:33.
123
ORIGINAL SCIENTIFIC ARTICLE
An assessment of the quality of reporting randomised controlled trials published in paediatric dentistry journals S. Rajasekharan • J. Vandenbulcke L. Martens
•
Received: 29 April 2014 / Accepted: 19 September 2014 Ó European Academy of Paediatric Dentistry 2014
Abstract Aim The objectives of this study are to compare the quality of reporting of randomised controlled trials (RCTs) published in 2011 and 2012 within five paediatric dentistry journals. Study design RCTs published in the years 2011 and 2012 were hand-searched by one reviewer. After randomisation and blinding, these journals were independently scored by two blinded reviewers based on the CONSORT 2010 checklist. Methods A total of 59 articles were included for analysis and 70 criteria were scored dichotomously as ‘1’ when reported and ‘0’ when not reported. Descriptive statistics and one-way ANOVA tests were performed. Results The Gwets AC1 Inter rater reliability coefficient was calculated as 0.85 (95 % C.I 0.84–0.86) indicating excellent correlation between the two reviewers. Only 19 articles (32.2 %) reported more than half (35/70) of the expected criteria. Descriptive statistics showed that sections such as introduction, results and discussion were reported better than abstract, materials and methods and other information. One-way ANOVA tests showed no significant difference (p [ 0.05) in the reporting of criteria across different journals and there was also no significant difference between the articles published in 2011 and 2012 (p [ 0.05). A part of this study was presented as an oral presentation in the 24th International Association of Paediatric Dentistry (IAPD) conference held in Seoul, South Korea. S. Rajasekharan (&) J. Vandenbulcke L. Martens Department of Paediatric Dentistry and Special Care, PaeCaMeD Research, Ghent University, UZ Ghent, 1P8, De Pintelaan 185, 9000 Ghent, Belgium e-mail: [email protected]
Conclusion The general quality of reporting of RCTs in paediatric dentistry journals was inadequate. Authors, reviewers and journal guidelines must work together towards a common goal for improving the quality of reporting of RCTs. Keywords
Randomised controlled trial CONSORT
Introduction A randomised controlled trial (RCT) is a quantitative, comparative study where participants of the trial are assigned by chance to treatment or control groups that are followed over time to determine the effect of the intervention. Evidence-based hierarchies place RCTs just below systematic reviews as the highest form of evidence (Concato et al. 2000; Turner et al. 2012; Fleming et al. 2012) that could be achieved from an in vivo/clinical trial (Rosner 2012). RCTs are widely accepted as the ‘‘gold standard’’ for obtaining unbiased estimates of treatment effects (Glasziou et al. 2007). Reporting of a RCT in a transparent manner in all steps from protocol to publication ensures reliability and reproducibility, which are essential for other researchers and practitioners. Clear reporting is indeed a prerequisite to help the readers to judge the quality of the RCT (Williams 2010; Chalmers 1990a). Researchers, authors, reviewers and journals would benefit if a single framework of recommendations and checklist was followed that would enhance the quality of reporting of RCTs (Begg et al. 1996). Consolidated Standards of Reporting Trials (CONSORT) was formed by a group of medical journal editors, clinical trialists, epidemiologists, and methodologists with
123
Eur Arch Paediatr Dent
the aim of developing a new framework for the assessment of the quality of RCT reports (Moher et al. 2001). The CONSORT statement was first published in 1996 (Begg et al. 1996) followed by the revised statement in 2001 (Harrison 2003) and 2010 (Moher et al. 2010). The CONSORT 2010 statement had, in general, simplified and clarified the wording in nine previously existing criteria with addition of subitems to previously existing criteria. The major amendments to the CONSORT 2001 statement were the addition of new criteria such as reporting of how ‘‘blinding’’ was accomplished, statistical analysis of secondary outcomes, whether the trial was registered, where a trial protocol could be found and funding information (Schulz et al. 2010). Currently, the CONSORT checklist to assess quality of reporting of RCTs is endorsed by over 38 % of the core medical journals listed in PubMed (Altman et al. 2012). A RCT report based upon the CONSORT checklist makes it easier for the readers to judge the reliability and relevance of the results of the trial (Moher et al. 1995). The quality of reporting RCTs is regularly assessed in the field of medicine and dentistry using the CONSORT checklist (Turner et al. 2012). In paediatric dentistry, only one study is available by Al-Namankany et al. (2009) that analysed RCTs published between 1985 and 2006 using the CONSORT 2001 checklist. Those authors concluded that the overall quality of reporting of clinical trials was poor and inadequate to allow readers to assess trial validity (AlNamankany et al. 2009). Since the declaration of the revised CONSORT 2010 checklist, no studies analysing RCTs have been published in paediatric dentistry journals. The objective of this study was to compare the quality of reporting of RCTs published in 2011 and 2012 within the journals analysed to test a null hypothesis (Ho) being that the distribution of scores is the same across all the analysed journals.
Materials and methods In the present study, the authors scored the quality of reporting of RCTs in five different paediatric dentistry journals based on the CONSORT 2010 checklist. All RCTs published in the years 2011 and 2012 in five different paediatric dentistry journals were: European Archives of Paediatric Dentistry (EAPD), European Journal of Paediatric Dentistry (EJPD), International Journal of Paediatric Dentistry (IJPD), Journal of Clinical Pediatric Dentistry (JCPD) and Pediatric Dentistry (PD). All manuscripts were hand-searched by one reviewer (LM) to include all articles reporting clinical trials, where the samples were assigned in a randomised manner to
123
either a treatment or control group. The updated CONSORT checklist was published in March 2010; hence articles from the year 2011 were selected for scoring assuming 9 months as a buffer period between acceptance of articles by the journal and their publishing date. The baseline data for article inclusion are shown in Fig. 1. A total of 59 articles were included and they were numbered from 1 to 59 for the purpose of anonymity. They were subsequently randomised using a computer generated table of random numbers. The assessors also ‘‘blinded’’ to the name of the journal, volume, issue, year of publication, name of author, affiliations, correspondence details, settings and locations where the study was carried out and details of ethical committee approval. Questions regarding setting, location, correspondence to authors and ethical committee were scored separately at the end after the blinding was revealed. Two ‘‘blinded’’ reviewers (J.V and S.R) then assessed the articles independently based on an expanded CONSORT checklist shown in Table 1. Conjoint questions in the CONSORT 2010 checklist were split for the purpose of accuracy and scored separately. Certain criteria in the original CONSORT 2010 checklist were omitted as they were not applicable to all the selected studies and their inclusion could result in a bias. The criteria omitted from the check list were 7b (when applicable, explanation of any interim analyses and stopping guidelines), 12b (methods for additional analyses, such as subgroup analyses and adjusted analyses), 17b (for binary outcomes, presentation of both absolute and relative effect sizes is recommended) and 18 (results of any other analyses performed, including subgroup analyses and adjusted analyses, distinguishing pre-specified from exploratory) of the original CONSORT 2010 checklist. No new criteria were added by the authors. All criteria were scored dichotomously as ‘1’ when reported and ‘0’ when not reported. If criteria expected in the materials and methods section were reported in the results section or vice versa, they were still scored as reported. Even if certain criteria from the results section were not directly reported, they were extracted from the tables and figures wherever possible. For the purpose of validation of the checklist, the two reviewers (JV and SR) jointly scored 10 randomly selected RCTs published in any of the five above-mentioned paediatric dentistry journals from 2007 to 2010. This procedure also facilitated better understanding of the criteria and pre-calibration of the reviewers before beginning with the reviewing process. Statistical analysis The articles that were included were scored across 70 criteria using MicrosoftÒ ExcelTM 2013 and the score sheets were analysed using MatlabÒ Version 8.0.0.783. The
Eur Arch Paediatr Dent Fig. 1 Flowchart showing the inclusion of articles analysed in this study on quality of reporting of RCT in paediatric dental journals
Gwets AC1 an inter-rater reliability coefficient was calculated using AgreeStat software version 2011.1 (Advanced Analytics, Gaithersburg, MD, USA). All
statistical analyses were completed using IBMÒ SPSSÒ Statistics Version 21.0.0.0. Descriptive statistics was used to analyse the scores of each article and each section within
123
Eur Arch Paediatr Dent Table 1 Checklist adapted from CONSORT 2010 checklist S. no.
Section
1
Title
2
Abstract
Topic
Modified checklist criteria Identification as a randomised trial in the title Structured summary of trial design, methods, results, and conclusions
3
Authors
Contact details for the corresponding authors
4
Trial design
Description of trial design
5
Methods
Eligibility criteria for participants
6
Settings where the data were collected
7
Interventions intended for each group
8
Specific objective or hypothesis
9
Clearly defined primary outcome for this report
10
How participants were allocated to interventions
11 12
Whether participants were blinded to group assignment Care givers were blinded to group assignment
13
Those assessing the outcomes were blinded to group assignment
14
Results
Number of participants randomised to each group
15
Trial status (recruitment)
16
Number of participants analysed in each group
17
For the primary outcome, a result for each group
18
The estimated effect size and its precision
19
Important adverse events or side effects
20
Conclusions
General interpretation of the results
21
Trial registration
Registration number and name of trial registry
22 23
Introduction
Funding
Source of funding
Background
Scientific background
Objectives
Specific objective or hypothesis
Trial design
Description of trial design (such as parallel, factorial)
24
Explanation of rationale
25 26
Methods
27 28
Allocation ratio Important changes to methods after trial commencement (such as eligibility criteria)
29 30
Reasons to important changes to methods after trial commencement Participants
31
Eligibility criteria for participants Settings where the data were collected
32
Locations where the data were collected
33
Interventions
34 35
Outcomes
The interventions for each group with sufficient details to allow replication How and when they were actually administered Completely defined pre-specified primary and secondary outcome measures, including how they were assessed
36
Completely defined pre-specified primary and secondary outcome measures, including when they were assessed
37
Any changes to trial outcomes after the trial commenced
38 39
123
Reasons for any changes to trial outcomes after the trial commenced Sample size
How sample size was determined
Eur Arch Paediatr Dent Table 1 continued S. no.
Section
Topic
40
Randomization
Sequence generation
Modified checklist criteria Method used to generate the random allocation sequence
41
Type of randomisation
42 43
Details of any restriction (such as blocking and block size) Mechanism used to implement the random allocation sequence (such as sequentially numbered containers)
Allocation concealment mechanism
44
Description of any steps taken to conceal the sequence until interventions were assigned
45
Implementation
Who generated the random allocation sequence
46
Who enrolled participants
47
Who assigned participants to interventions
48
Blinding
If done, who was blinded after assignment to interventions (for example, participants, care providers, those assessing outcomes)
49
How was blinding done
50
If relevant, description of the similarity of interventions
51 52
Results
Statistical methods
Statistical methods used to compare groups for primary and secondary outcomes
Participant flow
For each group, the numbers of participants who were randomly assigned
53
For each group, the numbers of participants who received intended treatment
54
For each group, the numbers of participants who were analysed for the primary outcome
55
For each group, losses and exclusions after randomisation
56
Reasons for losses and exclusions after randomisation
57
Recruitment
58
Dates defining the periods of recruitment and follow-up Why the trial ended or was stopped
59
Baseline data
60
Numbers analysed
61
Outcomes and estimation
A table showing baseline demographic and clinical characteristics for each group For each group, number of participants (denominator) included in each analysis and whether the analysis was by original assigned groups For each primary and secondary outcome, results for each group
62
For each primary and secondary outcome, the estimated effect size
63
For each primary and secondary outcome, its precision (such as 95 % confidence interval)
64
Harms
All important harms or unintended effects in each group (for specific guidance see CONSORT for harms)
Limitations
Trial limitations, addressing sources of potential bias, imprecision, and, if relevant, multiplicity of analyses
66
Generalisability
Generalisability (external validity, applicability) of the trial findings
67
Interpretation
Interpretation consistent with results, balancing benefits and harms, and considering other relevant evidence
Registration
Registration number and name of trial registry
Protocol Funding
Where the full trial protocol can be accessed, if available Sources of funding and other support (such as supply of drugs), role of funders
65
68 69 70
Discussion
Miscellaneous
Self-designed checklist for assessing the quality of reporting of RCTs based on the CONSORT 2010 checklist
an article. One-way an ANOVA test was performed to analyse any difference in distribution of scores within journals and also between the articles published in 2011 and 2012. The level of significance for all tests was set at 0.05.
Results There were 773 articles from the years 2011 and 2012 considered of which 59 (7.6 %) were selected as RCTs. The Gwets AC1 Inter rater reliability coefficient was
123
Eur Arch Paediatr Dent Table 2 Median, range and interquartile range for all 59 included articles per section and in total
Section (criteria number)
Number of criteria
Median
Range
Inter quartile range
Title (1)
1
0
0 (1–0)
Abstract (2–22)
21
10
8 (14–6)
1 (1–0) 2 (11–9)
Introduction (23–25)
3
3
1 (3–2)
0 (3–3)
Material and methods (26–51)
26
11
19 (26–7)
4 (13–9)
Results (51–64)
13
8
11 (12–1)
2 (8–6)
Discussion (65–67)
3
1
2 (3–1)
1 (2–1)
Other information (68–70)
3
0
2 (2–0)
1 (1–0)
Total of all criteria (1–70)
70
33
36 (57–21)
7 (37–30)
Fig. 2 Histogram depicting the number of articles reporting each checklist criterion
calculated as 0.85 (95 % confidence interval 0.84–0.86) indicating ‘excellent’ correlation between the two reviewers (Fleiss 1981). Descriptive statistics of those 59 articles included for each section and in total is presented in Table 2. Only 19 articles (32.2 %) reported more than half (35/70) of the expected criteria. Sections such as introduction, results and discussion were reported better than abstract, materials and methods and other information. Figure 2 illustrates the number of articles that had reported each individual criterion. Results for each section are discussed below. Title: The title section included only one criterion (criterion 1 of CONSORT checklist) and this was reported in only 15/59 articles (25.4 %) that identified the trial as a RCT in the title. Abstract: The abstract section consisted of 21 criteria (criteria 2–22 of CONSORT checklist). Four criteria were reported in all articles. These were criterion 3 (contact details for the corresponding authors), criterion 7 (interventions intended for each group), criterion 8 (specific objective or hypothesis) and criterion 17 (for the primary
123
outcome, a result for each group). Whether participants (criterion 11), as well as care givers were ‘‘blinded’’ as to group assignment (criterion 12) and the estimated effect size and its precision (criterion 18) were reported in only 1/59 (1.7 %) articles. None of the articles reported criteria 15, 21 and 22 (trial status, registration number and name of trial registry, source of funding, respectively). Introduction: The introduction section required the reporting of three criteria (criteria 23–25 of CONSORT checklist). This section was very well scored by all the articles with 57/59 articles (96.6 %) reporting all the three criteria (scientific background, explanation of rationale and specific objective or hypothesis). Materials and methods: Twenty six criteria (criteria 26–51 of CONSORT checklist) were covered in this section. More than 56/59 articles (94.9 %) reported 6/26 criteria. A high percentage of reporting was seen in criterion 30 (eligibility criteria for participants), criterion 33 (the interventions for each group with sufficient details to allow replication), criterion 34 (how and when the intervention was actually administered), criterion 35 and 36 (completely
Eur Arch Paediatr Dent
defined pre-specified primary and secondary outcome measures, including how and when they were assessed) and criterion 51 (statistical methods used to compare groups for primary and secondary outcomes). On the contrary, only one article reported criterion 37 and 38 (Any changes to trial outcomes after the trial commenced and the reasons for any changes to trial outcomes after the trial commenced, respectively). In the 59 RCTs analysed, randomization, allocation concealment and blinding were reported only in 25 (42.4 %), 8 (13.6 %) and 33 (55.9 %) articles, respectively. Results: This section contained 13 of numbers (criteria 52–64 of the CONSORT checklist). Criterion 52 (for each group, the numbers of participants who were randomly assigned) and criterion 61 (for each primary and secondary outcome, results for each group) were reported in 57/59 articles (96.6 %). Only 2/59 articles (3.4 %) reported criterion 62 (for each primary and secondary outcome, the estimated effect size). Discussion: The discussion section included only three criteria (criteria 65–67 of CONSORT checklist) of which criterion 67 (interpretation consistent with results, balancing benefits and harms, and considering other relevant evidence) was reported in all articles and only 5/59 articles (8.5 %) reported criterion 66 [generalisability (external validity, applicability) of the trial findings]. Other information: This section consisted of information about the registration number and name of trial registry, where the full trial protocol can be accessed, if available and sources of funding and other support (criteria 68–70, respectively, of CONSORT checklist). None of the articles reported a trial registration and only 4/59 articles (6.8 %) reported where the full trial protocol could be accessed. One-way ANOVA tests showed no significant difference (p [ 0.05) in the reporting of criteria across different journals and there was also no significant difference between the articles published in 2011 and 2012 (p [ 0.05).
Discussion The authors would like to emphasize that this study is purely an assessment of the quality of reporting of RCTs and is in no way indicative of the methodological quality of the studies themselves. Poor quality of reporting of RCT could be considered synonymous with poor quality of the trial (Fleming et al. 2012) as it gives rise to unreliable results and misleading conclusions (Chalmers 1990b). Poor reporting of a study could be due to poor methodological quality, while a wellconducted study could also be inadequately reported. It is almost impossible to make a distinction between the two. The results of the present study revealed that the quality of reporting of RCTs in paediatric dentistry is generally
inadequate and suboptimal. The latter findings are in accordance to the results of the study by Al-Namankany et al. (2009) who analysed articles published from 1985 to 2006 in paediatric dentistry journals with and without an impact factor (Al-Namankany et al. 2009). Moreover, these results are also in agreement with studies that analysed dental journals with a high impact factor (Pandis et al. 2010) and journals from other fields of dentistry such as orthodontics (Harrison 2003), periodontology (Faggion et al. 2012), implant dentistry (Cairo et al. 2012), prosthodontics (Jokstad et al. 2002) and dental public health (Marshman and Farid 2010). Four of the five journals included in this study had an impact factor but nevertheless there was no significant difference in the quality of reporting across all journals. This probably indicates the absence of adaptation of the CONSORT 2010 checklist. The present results being similar to the study published in 2009 suggest that there is still scope for improvement in this perspective. The CONSORT checklist emphasises the importance of negative reporting thus giving a clear understanding to any reader of the article over the exact methodology of the study. Negative reporting would mean that a procedure if not performed would still have to be reported. A study would hypothetically score points if it reports that no blinding was done; but on the other hand, a study where blinding had been carried out but not reported would not receive any points. The authors understand the practical difficulties of performing a RCT in a clinical setting and hence it is possible that a few of the CONSORT criteria could not be incorporated in the methodology. However, an explanation of these practical difficulties is an absolute necessity to help the reader to better understand the RCT and also create an awareness among future researchers. Even though negative reporting would result in more transparency of the RCT, negative reporting of all 70 criteria would be tedious and occupy a lot of space in a concise article. A pragmatic approach to this problem could be a disclaimer by the authors stating that the trial was developed according to the CONSORT guidelines and the absence of reporting of any particular criteria is due to its absence in the trial methodology. Not all readers have full access to an article and an abstract is a basis for selecting a research and helps the readership to obtain a quick overview of the complex research. In the present study, the authors found the abstract section to be reported sub-optimally in majority of the articles. A maximum of 14 criteria out of 21 were reported by one article, while all the rest scored at a lower level. The reason for minimal reporting in the abstract section could be attributed to the word limit imposed by a journal’s guidelines. A word limit of 200 or 250 words
123
Eur Arch Paediatr Dent
would make it almost impossible to report on all 21 criteria requested by the CONSORT checklist. Randomization sequence generation, blinding and allocation concealment are the three important parameters in assessing the risk of bias for RCTs (Higgins et al. 2008). In the present study, 34/59 (57.6 %) articles failed to report randomization sequence generation and 26/59 (44 %) articles missed to report blinding, while 51/59 (86.4 %) did not report allocation concealment thereby signifying a high risk of bias. Unlike ‘‘blinding’’ which could be impossible in certain cases, randomization is always feasible and ensures bias free allocation of subjects (Koletsi et al. 2012). Registration of clinical trials at their inception at a database that is publicly available is of long standing (Moher et al. 2010; DeAngelis et al. 2005) and the World Health Organization (WHO) states that ‘‘the registration of all interventional trials is a scientific, ethical and moral responsibility’’ (Moher et al. 2010). However, none of the 59 RCTs analysed in this study reported trial registration. In certain clinical situations where a huge sample size could not be obtained due to practical difficulties, systematic reviews and meta-analysis could provide valuable information by combining results of individual studies. However, poor reporting of RCTs would make it almost impossible to include/combine such studies within a meta-analysis/systematic review. Considering the huge amount of manpower, time and finance and/or funding invested in every trial, an added effort to report all the CONSORT checklist criteria could provide valuable information to reviewers and future researchers who wish to adopt the trial methodology. The quality of reporting RCTs is dependent on a triad of factors constituting the authors, reviewers and journal guidelines. Plint et al. (2006) and Turner et al. (2012) concluded from their systematic reviews that adoption of the CONSORT checklist by journals is associated with improvement in the quality of reporting of RCTs (Plint et al. 2006; Turner et al. 2012). When the author guidelines for the journals analysed within this study were considered, it was noted that only IJPD requests their authors to conform to the CONSORT guidelines while submitting RCTs. Adequate reporting of the RCT methodology helps the readers to assess and extrapolate the generalizability of the results of a trial to their own patients. A well-reported RCT encourages future researchers to adhere to CONSORT checklist at all stages such as preparing the protocol, executing the study and reporting the trial.
Conclusion The general quality of reporting of RCTs in paediatric dentistry journals was inadequate. There was no significant difference between the quality of reporting of articles
123
published across different journals (Ho retained) and between the quality of reporting of RCTs published in 2011 and 2012 (Ho retained). More paediatric dentistry journals could adopt the CONSORT checklist in their guidelines to the authors who wish to submit RCTs in future. Authors, reviewers and journal guidelines must work together towards a common goal for improving the quality of reporting of RCTs. Acknowledgments
No funding was received for this study.
References Al-Namankany AA, Ashley P, Moles DR, Parekh S. Assessment of the quality of reporting of randomized clinical trials in paediatric dentistry journals. Int J Paediatr Dent. 2009;19:318–24. Altman DG, Moher D, Schulz KF. Improving the reporting of randomised trials: the CONSORT Statement and beyond. Stat Med. 2012;31:2985–97. Begg C, Cho M, Eastwood S, et al. Improving the quality of reporting of randomized controlled trials. The CONSORT statement. JAMA. 1996;276:637–9. Cairo F, Sanz I, Matesanz P, Nieri M, Pagliaro U. Quality of reporting of randomized clinical trials in implant dentistry. A systematic review on critical aspects in design, outcome assessment and clinical relevance. J Clin Periodontol. 2012;39(Suppl 12):81–107. Chalmers I. Underreporting research is scientific misconduct. JAMA. 1990a;263:1405–8. Chalmers I. Underreporting Research Is Scientific Misconduct. JAMA. 1990b;263:1405–8. Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med. 2000;342:1887–92. Deangelis CD, Drazen JM, Frizelle FA, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. Arch Otolaryngol Head Neck Surg. 2005;131:479–80. Faggion JR, Clovis M, Giannakopoulos NN. Quality of reporting in abstracts of randomized controlled trials published in leading journals of periodontology and implant dentistry: a survey. J Periodontol. 2012;83:1251–6. Fleiss JL. Statistical methods for rates and proportions. New York: Wiley; 1981. Fleming PS, Buckley N, Seehra J, Polychronopoulou A, Pandis N. Reporting quality of abstracts of randomized controlled trials published in leading orthodontic journals from 2006 to 2011. Am J Orthod Dentofac Orthop. 2012;142:451–8. Glasziou P, Chalmers I, Rawlins M, McCulloch P. When are randomised trials unnecessary? Picking signal from noise. BMJ. 2007;334:349–51. Harrison JE. Clinical trials in orthodontics II: assessment of the quality of reporting of clinical trials published in three orthodontic journals between 1989 and 1998. J Orthod. 2003;30:309–15 (discussion 297–8). Higgins JPT, Green S, Cochrane Collaboration. Cochrane handbook for systematic reviews of interventions. Chichester: WileyBlackwell; 2008. Jokstad A, Esposito M, Coulthard P, Worthington HV. The reporting of randomized controlled trials in prosthodontics. Int J Prosthodont. 2002;15:230–42. Koletsi D, Pandis N, Polychronopoulou A, Eliades T. What’s in a title? An assessment of whether randomized controlled trial in a
Eur Arch Paediatr Dent title means that it is one. Am J Orthod Dentofac Orthop. 2012;141:679–85. Marshman Z, Farid F. The quality of reporting of randomised controlled trials in dental public health. Community Dent Health. 2010;27:253–6. Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c869. Moher D, Jadad AR, Nichol G, Penman M, Tugwell P, Walsh S. Assessing the quality of randomized controlled trials: an annotated bibliography of scales and checklists. Control Clin Trials. 1995;16:62–73. Moher D, Schulz KF, Altman DG. The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials. Lancet. 2001;357:1191–4. Pandis N, Polychronopoulou A, Eliades T. An assessment of quality characteristics of randomised control trials published in dental journals. J Dent. 2010;38:713–21.
Plint AC, Moher D, Morrison A, et al. Does the CONSORT checklist improve the quality of reports of randomised controlled trials? A systematic review. Med J Aust. 2006;185:263–7. Rosner AL. Evidence-based medicine: revisiting the pyramid of priorities. J Bodyw Mov Ther. 2012;16:42–9. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Ann Intern Med. 2010;152:726–32. Turner L, Shamseer L, Altman DG, et al. Consolidated standards of reporting trials (CONSORT) and the completeness of reporting of randomised controlled trials (RCTs) published in medical journals. Cochrane Database Syst Rev. 2012;11:MR000030. Williams HC. Cars, CONSORT 2010, and clinical practice. Trials. 2010;11:33.
123
