Many reports of health research omit important information needed to assess their methodological robustness and clinical relevance. Without clear and complete reporting, it is not possible to identify flaws or biases, reproduce successful interventions, or use the findings in systematic reviews or meta-analyses. The EQUATOR Network (http://www.equator-network.org/) promotes responsible reporting and the use of reporting guidelines to improve the accuracy, completeness, and transparency of health research. EQUATOR supports researchers by providing online resources and training. EQUATOR Oncology, a project funded by Cancer Research UK, aims to support cancer researchers reporting their research through the provision of online resources. In this article, our objective is to highlight reporting issues related to oncology research publications and to introduce reporting guidelines that are designed to aid high-quality reporting. We describe generic reporting guidelines for the main study types, and explain how these guidelines should and should not be used. We also describe 37 oncology-specific reporting guidelines, covering different clinical areas (e.g., haematology or urology) and sections of the report (e.g., methods or study characteristics); most of these are little-used. We also provide some background information on EQUATOR Oncology, which focuses on addressing the reporting needs of the oncology research community.
Health research studies must be reported clearly, accurately, and completely if they are to meaningfully enhance medical knowledge and inform clinical practice. Ambiguous, missing, or misleading information obscures how research was carried out and what was found, limiting critical assessment and impeding further use of published findings. It also wastes the financial and human resources invested in the research (Glasziou et al, 2014).
Readers cannot judge the robustness of the methodology used or the reliability of the findings if study methods are reported badly. Further, other researchers may be prevented from repeating the study (Goodman et al, 2016). Poor reporting of clinical details, inconsistent use of terminology and definitions, insufficient information about interventions, incomplete details of statistical methods, and inconsistent or missing reporting of adverse effects also hamper comparisons of findings across studies, which are necessary to determine the best options for patient care and disease prevention.
Oncology research suffers from the reporting inadequacies that afflict all health research. Table 1 gives some examples of the consequences of poor reporting in different study types in oncology.
Despite early calls by the World Health Organization (Unknown, 1979) and others (Nahum, 1979; Miller et al, 1981) to standardise how the results of cancer treatment studies are reported, many recent studies evaluating the quality of oncology clinical trial reporting have found biased and inconsistent reporting to be very common (Duff et al, 2010; Peron et al, 2012; Vera-Badillo et al, 2016). Basing decisions about patient care on incomplete and misleading research findings may have a profoundly negative impact on patients’ health and wellbeing.
These reporting problems are avoidable. In this paper, we introduce reporting guidelines – simple but effective tools supporting complete and transparent reporting – and highlight guidelines that are specifically useful for oncology research. We introduce the EQUATOR (Enhancing the QUAlity and Transparency Of health Research) Network programme, its online resources, and a new project, EQUATOR Oncology, which is collating oncology-specific resources.
The EQUATOR network
The EQUATOR Network was set up in 2006 to support the implementation of reporting guidelines. Along with its database of reporting guidelines, the EQUATOR programme provides resources and toolkits to help researchers write complete and transparent health research papers. Resources are also provided for journal editors and peer reviewers to help ensure that published research is ‘fit for purpose’, that is, that it provides all of the information needed for its assessment and further use. The Network also organises events and conferences to raise awareness of poor reporting and its consequences, and offers training for researchers and editors to maximise the value of their published research. Ongoing work is expanding the scope of EQUATOR’s resources to also cover guidance for research planning. More information about the EQUATOR Network is available on our website (http://www.equator-network.org/about-us/).
EQUATOR’s scope is primarily studies of humans and pre-clinical animal research. We do not address laboratory research. Related resources for such research tend to target data sharing, such as MIAME for microarray experiments (Brazma et al, 2001; https://fairsharing.org/).
Reporting guidelines provide one solution to the widespread problems in research reporting. They are simple, efficient tools, most often in the form of a checklist, that help researchers to prepare manuscripts that contain all of the information required by readers and those that will use the research report.
Table 2 lists the generic guidelines for the main research study types. Most of these guidelines provide a reporting framework for a whole research paper and list the minimum information that authors should include within the paper so that their study can be fully understood, replicated if desired, and used to inform future research. Some of these guidelines also have extensions, offering additional guidance. The guidelines can also be used by peer reviewers to check that research reports are complete, accurate reflections of the research undertaken (Levine and Kressel, 2016).
The best known guidelines are the CONSORT Statement (checklist shown in Supplementary File 1) for reporting randomised controlled trials (Schulz et al, 2010) and the STROBE Statement (checklist shown in Supplementary File 2) for reporting observational studies (von Elm et al, 2007). Although reporting guidelines should not be used to critically appraise reports of research studies, they prompt authors to report the information needed for a complete critical appraisal.
Generic guidelines exist for reporting most major types of clinical and pre-clinical research. They provide an excellent starting point when writing up any study, including oncology research. Other reporting guidelines that provide guidance on reporting specific aspects of study methods, procedures, or medical conditions, including oncology-specific guidelines, can be found in the database of reporting guidelines on the EQUATOR Network website. The Network systematically collects and classifies all reporting guidelines to help researchers easily find the guidelines relevant to their work.
There is increasing interest in the publication of protocols for research studies. Guidelines are available for preparing protocols for randomised trials (Chan et al, 2013) and systematic reviews (Moher et al, 2015).
The UK EQUATOR Centre was awarded funding from Cancer Research UK to develop oncology-specific resources and activities to enhance the quality and transparency of published oncology research (http://www.equator-network.org/library/equator-oncology/).
We are reviewing the literature on the quality of reporting of clinical trials and observational studies in oncology. In the course of our work, we are collating published literature on oncology research methodology and reporting, which we regularly make available through the EQUATOR Oncology Current Awareness Bulletin.
The project focuses on research reporting, but in the next phase will be expanded to include guidance for efficient research planning and design. Ensuring robustness in the planning of any research project is the first condition for obtaining reliable research findings. For example, writing a detailed protocol documenting the study design and all methods forms the basis for the final written research manuscript. Guidelines already exist for preparing some types of research protocol, each linked with corresponding guidelines for reporting study findings. For example, the SPIRIT checklist (Chan et al, 2013) is used to guide the preparation of a protocol for a randomised controlled trial, with much overlap of concepts and structure with the CONSORT checklist (Schulz et al, 2010) for reporting trial findings.
Oncology-specific reporting guidelines
The EQUATOR Network website already offers important resources to help authors write up oncology research studies. In October 2016 our regularly updated database of reporting guidelines included 37 oncology-specific guidelines. These guidelines complement the generic guidelines, offering guidance on reporting aspects of various study types, such as observational studies, prognostic and diagnostic studies, and clinical trials.
Table 3 describes the 37 oncology-specific guidelines by the clinical area, study type, and section of the report that they refer to. Some cover oncology studies in general, while others focus on research in certain diagnostic groups, such as cancer of the lung, liver, breast, kidney, bone, and soft tissue. The fields of haematology, neuro-oncology, urology, and gastroenterology are well-represented.
Use of oncology-specific reporting guidelines
The publication of a reporting guideline will not affect reporting completeness and quality unless researchers working in the field are aware of and use the guideline when they write their manuscripts. We investigated how many times each oncology-specific reporting guideline had been cited by other research papers. We searched the Web of Science Core Collection Science Citation Index Expanded (SCI-EXPANDED) from inception to the present (last search date 3 November 2016). The results of these citation searches are shown in Table 3.
Two guidelines have each been cited more than 1000 times. Both focus on haematology research, one covering the whole study report for trials in a particular disease (acute myeloid leukaemia, (Cheson et al, 2003)) and the other dealing with a particular kind of data (correlating genetic and clinical data (Döhner et al, 2010)). Five have been cited between 100 and 350 times, and 23 have been cited less than 100 times. One guideline had no citations but was only published in 2016, and six guidelines were not found in the Citation Index.
It is likely that many authors who use a reporting guideline do not actually cite it, and that not all research papers that cite a guideline do so because it was used to help write the paper. Nevertheless, the citation numbers give a rough indication of the use of each guideline in the literature. Many factors will influence the differences between the citation rates of specific guidelines, for example the size of the subspecialty within oncology and when the guideline was published.
Journals have an important role to play in improving reporting of research studies by highlighting the use of reporting guidelines in their instructions to authors. Reporting guidelines can also be very helpful for peer-reviewers. The EQUATOR Network has produced a new toolkit to help journals to publish clear and therefore usable research reports (http://www.equator-network.org/toolkits/using-guidelines-in-journals/).
Future work of EQUATOR Oncology
Health research reporting problems have been well-documented in recent years, and oncology research is no exception (Papathanasiou and Zintzaras, 2010; Peron et al, 2012, 2013; Jankova et al, 2015; Maillet et al, 2016; Sivendran and Galsky, 2016). Despite these continuing reporting issues, two specific guidelines for haematology research have been very well cited. The question remains as to why authors do not use existing reporting guidelines. Is it because authors need additional specific guidelines for certain oncology study types, because authors need help to better use existing guidelines (both generic and specific), because they do not think reporting guidelines are worth the effort, or because they are simply unaware that reporting guidelines exist?
EQUATOR Oncology aims to highlight the real-life consequences of poor reporting, to provide resources and to support oncology researchers by helping them to find and use the appropriate reporting guidelines for their research.
Peer-reviewers also play an important role in the process of improving research reporting, and reporting guidelines can be a helpful tool for those reviewing manuscripts prior to publication. However, checking adherence to a reporting guideline can be time-consuming. It may be helpful for guideline developers to produce short lists of items for peer reviewers to focus their attention on.
We will establish an EQUATOR Oncology ‘Advisory Group’ of experts and opinion leaders in oncology research including clinicians, oncologists, methodologists, editors, Cancer Research UK and EQUATOR representatives – who will oversee and inform the development of the project. Based on our findings from literature reviews and advice from our expert advisory group, we will identify issues and develop oncology-specific online resources to help minimise reporting problems and increase the impact of published oncology research.
We will also carry out surveys of oncology researchers, oncology journal editors and peer reviewers to identify their concerns about reporting in journal articles.
Our web resources can help oncology researchers improve their research reporting in manuscripts and will ultimately help to improve the robustness and reliability of the research itself. Only with the expert help of authors, researchers, methodologists, opinion leaders, and journal editors working in cancer research can our online resources fully address the reporting issues that oncology researchers need assistance with. We encourage those involved in cancer research to contact us with suggestions for the development of these resources.
Cancer patients take part in research studies to try to improve their health conditions and for altruistic reasons (Moorcraft et al, 2016). Their contributions to scientific understanding should not be wasted because the research is poorly conducted, inadequately reported, or even not reported at all. EQUATOR Oncology will support oncology researchers to conduct robust research and to produce research papers that are usable, reproducible, and transparent, recognising the important contributions of all patient participants in research.
This work was supported by Cancer Research UK (C5529/A16895). The work was carried out in the Centre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology & Musculoskeletal Sciences, University of Oxford, UK.
About this article
Supplementary Information accompanies this paper on British Journal of Cancer website (http://www.nature.com/bjc)