Documenting a publication's process at the scale you work at
Good Publication Practice expects a publication's process to leave a record showing that the process was followed, but no particular commercial system to keep it in. Keeping that record and auditing it can be done in proportion to the work, from a large sponsor with many systems to one person running a small company's medical communications, alongside controlling the versions of drafts and assuring the quality of the manuscript itself.
A record rather than a commercial system
In the view of one of the authors of the 2022 edition of Good Publication Practice, a commercial publication-management system is not mandatory under that edition or any other, and the guideline cannot mandate the use of any commercial tool. There are a number of such systems besides Datavision, and that author expects a tiny biotech with a very small group of publications not to spend all that money on a big system.
The same author considers good enough an archive system that shows the process was followed, that everyone approved and contributed intellectually and that the data were shared with the authors as they should be, kept in the cloud somewhere it can be accessed. The guideline contains a table that shows exactly which steps should be documented.
Small pharmaceutical or biotechnology companies, often limited by resources, may be challenged to maintain an effective work-flow tracking system that ensures both quality control and a consistent method for speeding up work flow and retrieving documents. One cost-effective publication quality control and work-flow tracking system was developed to meet the need for quality and version control and for compliance with all of Good Publication Practice, for scientific documents from conception to final publication. A publication database built in Microsoft Access tracked and maintained every publication, drafts and approvals included; each document type, whether abstract, poster, oral presentation or manuscript, had its own development process set out in clearly designed flowcharts; and when a publication concluded, all its supporting documents were archived in the database for later retrieval. In the experience of the system's developers, smaller, cost-conscious biotechnology and pharmaceutical companies can implement such a system while still complying with Good Publication Practice and without expensive, custom-designed software.
Auditing at the scale of the work
The guideline's authors recommend that people audit what they have done. At Pfizer an audit is a four-day process, because of a large database of studies, multiple different systems and multiple corporate entities to take into account; for the one person who runs all of medical communications at a tiny biotech, auditing the material is a much smaller job.
Guidance on what one person doing publication work alone should be able to do, as opposed to someone in a very big department or company, is minimal, and the question keeps coming up in the guideline authors' continuing work with other groups. Many people are at once the medical editor, the medical writer and the publication professional, and such a person should be able to work in a way that is economical and makes sense for the scope of their work, while someone in a big company should be able to do the same in a logical way.
Version control of drafts
Close collaboration between authors and professional medical writers, with at least the lead author reviewing each step of the writing from the outline through the working drafts to approval of the final paper, requires clear lines of communication and tools for maintaining version control. These tools can be as simple as an agreed file-naming system that incorporates dates and initials, or cloud computing or secure file transfer systems used to share files and maintain a centralised archive of drafts.
Quality assurance of the manuscript
Publications of clinical trial results, whether abstracts, posters, oral presentations or manuscripts, must be objective, accurate and scientifically sound, because the results of clinical trials can affect patient care. Quality-related processes are routine when regulatory documents are prepared, yet in 2010 few medical writing groups in the pharmaceutical industry integrated formal quality-related processes into their work with authors on peer-reviewed publications. Professional medical writers have an obligation to the authors to assure the highest level of publication integrity in the writing assistance they give to authors who serve as a manuscript's guarantors.
It is difficult for authors to spot many errors or inconsistencies in a complicated scientific manuscript, which is not necessarily a reflection on them, and they can benefit significantly from the special expertise of a quality assurance reviewer. Methodological inadequacies, data errors, weaknesses and omissions, inaccurate reporting, unstated study limitations, unsubstantiated conclusions and erroneous literature citations have been reported to be rampant in the medical literature. Quality assurance can improve a manuscript by calling attention to errors, inconsistencies, data misinterpretations, biases and other weaknesses that compromise its quality. Quality is achieved through detailed processes that validate ethical adherence, consistency of data and reporting, standardisation, and uniformity with the requirements of the International Committee of Medical Journal Editors, the World Association of Medical Editors, the Committee on Publication Ethics and the Council of Science Editors, and with journal-level requirements such as CONSORT and independent statistical review. Incorporating quality assurance and quality control into the development of peer-reviewed publications can help to attain the highest level of publication integrity.
In one company's quality control process, built on quality principles, reviewers checked that the data in a publication were accurate and consistent with the source data, that is, the clinical study report. To ensure objectivity, the reviewers were independent of the investigators and primary authors, and their reviews focused primarily on the validity of the data in the publication; the discrepancies they found went back to the investigators or primary authors for correction.