Info for Researchers

PROVENet: a network for high-quality research
PROVENet aims at performing high-quality research in mechanical ventilation in emergency units, intensive care units and in operating rooms, including observational studies, randomized controlled trials, meta-analyses and health-economy analyses. PROVENet initiates research projects based on ideas from researchers within and from outside the network. In every project an original hypothesis is tested, using an adequate study design and sufficient numbers of patients.

Plan of action
PROVENet-initiated studies and trials require a strict action plan:
(1) each project is lead by a Steering Committee that clearly formulates the hypothesis and the primary and secondary endpoints, decides on the type of study to be performed, performs the power calculation, and writes the final study protocol
(2) the Steering Committee submits the study protocol to an Institutional Review Board, usually the one in the hospital of the Principal Investigator of the new research project.
(3) the Steering Committee seeks for participating centers
(4) communications are with the Principal Investigator
(5) finally, the Steering Committee installs a Data Safety Monitoring Board that typically consists of at least four experts in the field, and one Statistician or Epidemiologist

Transparency
Each project is presented at www.clinicaltrials.gov, or a similar site, to disclose exactly which hypothesis is going to be tested, with which endpoints and with how many patients.
Usually, the study protocol and analysis plan are also submitted to a peer-reviewed journal, like TRIALS or JOURNAL OF CLINICAL TRIALS to increase transparency.
Each project has its own website, where major documentation can be found, and downloaded for use locally.

Publications
It is expected that every project will result in at least one scientific report on main findings of a study, following the original analysis plan. Any analysis that was not reported in the original analysis plan will be reported as a posthoc analysis within that report.
The first and usually the main report that comes from a PROVENet-initiated project is written by the Writing Committee for that project, typically a subgroup of the Steering Committee.

It is PROVENet policy to submit the primary manuscript reporting on a randomized controlled trial or observational study with the INTENTION to have it published as a paper by the PROVE Network investigators (e.g.,
as for the first paper reporting on 'PROVHILO', published in Lancet). This means that there will be no names of individual researchers above a publication. The Principal Investigator is mentioned as the contact person, the members of the Steering Committee, the Writing Committee, and ALL local investigators of participating centers are summarized at the end of a manuscript or in the appendix depending journal policy. In this way PubMed can link the names of all investigators to a publication. If a journal does not accept this, another approach will be discussed within the Steering Committee, and an explanation and conclusion will be posted on the website of the project

The maximum number of local investigators per participating center is discussed and decided upon before the start of each project, depending on the efforts to be delivered with each included patient. Notably, we follow the Guidelines of the International Committee of Medical Journal Editors regarding authorships, see below)

Any secondary 
manuscript reporting on a randomized controlled trial or observational study (i.e., a secondary analysis or analysis in a subgroup of patients) may follow another approach. For example, a secondary manuscript can be submitted by a group of investigators having their names in the list of authors, but always referring to 'on behalf of the PROVENet investigators'.

It is PROVENet policy to submit manuscript reporting on meta-analysis to have it published as a paper by the investigators involved in selection and analysis of the papers used in the meta-analysis (e.g., as for a paper reporting on a meta-analysis, published in JAMA). A different approach is followed for meta-analysis that use individual patient data. As an individual patient data meta-analysis requires substantial contributions from the investigators of the original studies, these investigators deserve a place in the list of authors (e.g., as for the first paper reporting on a individual patient data meta-analysis, published in Lancet RM). Notably, here we also follow the Guidelines of the International Committee of Medical Journal Editors, see below).

Guidelines of the International Committee of Medical Journal Editors
According to the guidelines of the International Committee of Medical Journal Editors (http://www.icmje.org, revised in 2013), authorship credit should be based on the following 4 criteria:
(1) substantial contributions to conception or design or the acquisition, analysis, or interpretation of data; and
(2) drafting of the manuscript or revising it critically for important intellectual content; and
(3) final approval of the version to be published; and
(4) agreement to be accountable for all aspects of the manuscript in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Merging databases
Every study will result in a new set of data, and many studies will collect very similar data. By merging different datasets, a large poweful dataset can be obtained to test new hypotheses. Every report that comes from a merged dataset must report it's source. 

PROVENet requires that the study groups producing the original databases are mentioned in any publication that might use their data. PROVENet welcomes any publication that mentions all participating investigators at the end of a report, or in the appendix.
Comments