Chapter 7 Publication and Dissemination

7.1 Introduction

Research dissemination can be defined as follows: “A planned process that involves consideration of target audiences and the settings in which research findings are to be received and, where appropriate, communicating and interacting with wider policy and health service audiences in ways that will facilitate research uptake in decision- making processes and practice” (Wilson et al. 2010, 5:91).

Traditionally, research dissemination centers on the publication of results in peer reviewed journals and academic conference presentations. Recent initiatives seek to broaden the concept of research dissemination such that the main messages from results are communicated to targeted decision-makers and stakeholders in a way that encourages them to factor the research implications into their work and clinical practice. This module will focus on two aspects of dissemination particularly relevant to Clinical Trials: 1. Data sharing and 2. Publication in scientific journals.

Regardless of the scope, researchers are encouraged to develop a dissemination plan early in the course of all trials. Trials that are closer to later phases of research translation (i.e., implementation) and those supported by certain agencies (e.g., Patient-Centered Outcomes Research Institute (PCORI)) may require more robust dissemination plans in grant applications that go beyond traditional journal publications and meeting presentations.

Resources for dissemination planning beyond journal publication:

7.2 Data Sharing

7.2.1 What

To maximize transparency and investment, policymakers seek to expand public and investigator access to clinical trial data. The extent of data sharing can range from sharing aggregated data describing subject characteristics and outcomes, to providing de-identified patient-level data. Data sharing requirements depend on evolving policies of funding agencies (i.e., NIH) and other regulators. Researchers are strongly encouraged to consult the most up-to-date requirements as early as possible in planning of their clinical trial (see How below).

7.2.2 Why

The intent of data sharing is to maximize the transparency and benefits of research. Moreover, some level of data sharing for clinical trials is often required by regulatory authorities, funding agencies and the ICMJE.

When done appropriately, data sharing:

  • Increases public trust by increasing transparency of research efforts
  • Maximizes the public health impact of research
  • Encourages scientific inquiry and new research efforts
  • Encourages diversity of analysis and opinion
  • Permits the creation of new datasets when data from multiple sources are combined

7.2.3 How

In September 2016 the Department of Health and Human Services (HHS) issued a new regulation and the NIH has issued a new policy to increase the availability of information about clinical trials. The HHS Final Rule describes requirements for registering and submitting summary results information for certain clinical trials to ClinicalTrials.gov. A complimentary NIH policy that applies to all clinical trials funded by NIH, regardless of whether they are subject to the Final Rule was also published on September 2016. The Final NIH Policy was also published in the NIH Guide for Grants and Contracts.

NIH has made available the following resources to help explain these changes.

7.2.3.1 Methods for data sharing

There are numerous approaches to data sharing, each of which has different levels of privacy protection and mechanisms for sharing. Researchers can manage requests for data personally on an individual request basis, or through use of a secure data archive or enclave. Investigators should select a mode of data sharing that is appropriate based on the size, complexity and sensitivity of the data, as well as the volume of data use requests anticipated. See the NIH data sharing guidance for details.

The NIH provides numerous data sharing repositories which make data accessible for reuse. If you are doing an NIH funded trial, consult your Project Officer to discuss options for sharing data.

IFAR has created several sensitive data policies (including data access, encryption, sharing, security, suppression, retention and destruction, etc.) please see these policies on the HSL HUB policy and forms.

7.2.3.2 Human subjects and privacy issues

All data sharing activities must be compliant with HIPAA, local, state and organizational regulations. All compliances and restrictions should be addressed in the data sharing plan section of the funding application. There are two basic approaches to protecting sensitive data: restricting information in the shared dataset, and restricting access to the data. See NIH data sharing guidance for details.

Restricting Information in the Dataset

The plan for data sharing may include stripping a dataset of all potential participant identifiers. Depending on the data and study population, the level of information that will need to be removed from the database varies. Studies involving participants with unusual characteristics (Samples drawn from small geographic areas, rare populations, and linked datasets) will likely involve a greater degree of anonymizing to ensure the privacy of study participants.

Restricting Access to the Dataset

Data archives and enclaves can be set up to restrict access at any level. Data Use Agreements restrict the use of data to those specifically identified in the agreement, and require that data be used only for research purposes.

Please see the IFAR policy on Limited Data Sets and Data Use Agreements on the HSL HUB as well as the HSL IRB SOPs on HIPAA and the use of data in research.

7.3 Special Considerations for Older Subjects

NONE

7.4 Common Pitfalls

  • Not being aware of and planning for up-to-date data sharing requirements by funding agencies, regulators, and ICMJE
  • Failure to develop a clear well documented dataset can undermine the researchers ability to share the data or others to use it.
  • Failure to plan for the time and expense of data sharing

7.4.1 Resources

External

Internal

7.5 Scientific Publications and Presentations

7.5.1 General considerations

All paper and abstracts emanating from your trial should be planned early on in the study to maximize the productivity from the trial and avoid iterative papers. In planning, it is useful to outline the topic of paper, target journal and responsible first author.

Larger trials should have a Publications Steering Committee whose members are usually the lead investigators and establish written guidelines for vetting papers emanating from the trial. This procedure for planning, reviewing and approving all scientific publications (papers and abstracts) prior to submission is to ensure they are accurate (i.e., align with protocol), not duplicative, and respect ICMJE authorship guidelines. Certain funding agencies and industry trial sponsors may have very specific requirements regarding their own review and approval of manuscripts prior to submission.

Section III of the ACRIN publication policy provides details related to the roles and responsibilities of publication committees.

7.5.2 Specific reporting guidelines for clinical trials

There are specific reporting guidelines for clinical trials that differ by the trial type which are continually being updated. For example, there are set standards for the flow diagrams presenting subject participation in a clinical trial that differ between traditional parallel design randomized trials, non-inferiority trials, cluster randomized trials, stepped wedge designs etc. Journals generally follow the guidelines described in the Consolidated Standards of Reporting Trials (CONSORT) statement].

Enhancing the QUAlity and Transparency Of health Research Network (EQUATOR is a very useful comprehensible searchable website with up-to-date links to the CONSORT statements and extensions.

7.5.3 Publication of the trial protocol

7.5.3.1 What

Protocol manuscripts report the methodology of your clinical trial. Details in a published protocol paper (e.g., primary and secondary outcome definitions, sample size estimates) need to align with other official communications describing the study’s methodology such as the trial protocol and descriptions in ClinicalTrials.gov. These documents may be compared by journals considering publication of the trial’s results and discrepancies will need to be explained.

7.5.3.2 Why

Reasons to publish a protocol paper include increasing transparency of the trial conduct, providing a detailed reference for future publications emanating from the study, and demonstrating productivity for funding agencies.

7.5.3.3 How

Planning for a protocol manuscript submission should begin early because journals generally only consider publishing protocol papers for studies that have received ethics approval but have not yet concluded patient recruitment.

7.5.3.3.1 Selecting a journal

Examples of journals that publish protocol papers include: Clinical Trials: Journal of the Society of Clinical Trials, Contemporary Clinical Trials, International Journal of Clinical Trials, Journal of Clinical Trials, BMJ, and Trials. Some of these journals allow authors to pay for publication of their protocol paper (e.g., Journal of Clinical Trials) and others have the option of paying for open access should the paper be accepted.

In deciding which journal is the most appropriate to publish your protocol, it is useful to first review prior protocol papers published in specific journals to see which best aligns with your study and examine their formats. All journals will have specific guidelines for preparation which should be carefully consulted and adhered to.

7.5.4 Publication of the results of a clinical trial

7.5.4.1 Timing

The WHO statement on Public Disclosure of Clinical Trial Results states, at a minimum,main findings of clinical trials, positive or negative, are to be submitted for publication in a peer reviewed journal within 12 months of study completion and are to be published through an open access mechanism unless there is a specific reason why open access cannot be used, or otherwise made available publicly at most within 24 months of study completion.

7.5.4.2 Trial registration identification

As per the ICMJE guidelines, top-tier journals require that your trial be registered on ClinicalTrials.gov (or equivalent such as WHO International Clinical Trials Registry Platform (ICTRP)) prior to the enrollment of the first subject, otherwise the paper will be rejected without review. The trial registration number must be included in your manuscript and usually appears at the end of the abstract.

The ICMJE encourages posting of clinical trial results in clinical trial registries. The ICMJE will not consider as prior publication the posting of trial results in any registry that meets the above criteria if results are limited to a brief (500 word) structured abstract or tables (to include patients enrolled, key outcomes, and adverse events).

7.5.4.3 Packaging results

Investigators are cautioned to carefully plan the paper(s) emanating from a single trial to avoid iterative publications, ensure the main trial results are not usurped by early papers, and to adhere with the pre-specified analyses set forth in the trial protocol and as described on ClinicalTrials.gov. One should reserve the main paper to include the pre-specified primary and secondary outcomes. “Outcome switching” (i.e., changing the outcomes reported in results paper with those pre-specified in the protocol and on ClinicalTrials.gov) must be avoided.

Following publication of the main paper, non-iterative auxiliary papers that leverage the rich clinical trials data may be considered.

7.5.4.4 Manuscript preparation

All manuscripts should adhere to the guidelines provided in each journals’ Instructions for Authors which generally follow ICMJE guidelines.

Most journals which have a specific reporting criteria for clinical trials results manuscripts which generally follow the CONSORT statement. CONSORT provides an extremely useful checklist to help guide presentation of every element of a clinical trial manuscript and is a highly recommended resource.

7.5.4.5 Adjunct material

Most top tier journals will request a copy of the trial protocol with the submission of the paper describing the clinical trial results. Editors will compare the details in the protocol, ClinicalTrials.gov and any previously published protocol papers, with those described in the submitted manuscript to ensure they align.

7.5.5 Special considerations for older subjects

NONE

7.5.6 Common pitfalls

  • Failure to develop a publication plan and committee early in the course of the trial.
  • Failure to register the trial on ClinicalTrials.gov prior to enrollment of the first subject.
  • Failure to submit trial results for publication within 12 months of trial completion and failure to publish results within 24 months.
  • Failure to follow ICMJE, journal specific, and CONSORT guidelines for preparing clinical trials publications.
  • Failure to align details of the trial in submitted papers to details as described in the protocol and on ClinicalTrials.gov, particularly pre-specified primary and secondary outcomes
  • Publishing iterative papers or early papers that usurp publication of main findings.
  • Missing opportunities for results dissemination beyond peer reviewed journals and professional conferences.

7.5.7 Resources