Ghostbusters should only bust ghosts

We are professional medical writers who appreciate the distinctions between ghostwriting and ethical medical writing made in your May editorial1. We welcome the commonsense and balanced approach you put forward in an area that has been generating more heat than light. The article from the April edition of the Journal of the American Medical Association (JAMA)2 about ghost authorship and ghostwriting that you highlighted led to considerable media response and confusion. We wish to discuss ethical medical writing, author accountability and compensation using the clinical trial manuscript as an example. We speak as medical writers only and not on behalf of any company. Medical writers are employed by industry, publishers, academia and others to prepare manuscripts. Like biostatisticians, medical writers are often not acknowledged. That can lead to confusion about the term ghostwriting. The Association of American Medical Colleges (AAMC) Task Force on Industry Funding makes a clear distinction between ghostwriting and ethical collaborative writing. Ghostwriting is “the provision of written material that is officially credited to someone other than the writer(s) of the material. Transparent writing collaboration with attributions between academic and industry investigators, medical writers and/or technical experts is not ghostwriting”3. Medical writers are skilled communications professionals who help the diverse individuals involved in biomedical research to clearly communicate their discoveries. The medical writer plays a vital role in bringing research to publication. Lack of acknowledgement deprives us of well-deserved recognition for our contributions as medical communicators and disenfranchises our profession. Many journals now state that persons who have made substantial contributions to the work but who are not authors should be acknowledged, including names, degrees, affiliations, and indication if compensation has been received. Companies often perform research studies to support product development. They write the study protocol, contract with doctors to run the study, train the doctor's staff in proper data collection, collect the data from all the study sites, lock the database to assure its validity, give the data to the statistician for analysis, examine the statistician's report, consult with the investigators and study sites and write the Clinical Study Report (CSR) for submission to the US Food and Drug Administration (FDA), European Medicines Agency (EMEA) or equivalent body in other countries. The CSR is the source document. It has the tables, graphs, P-values and discussions about the meaning of the data. The manuscript comes from the source document. The authors, generally the investigators and medical monitors, take responsibility for the integrity of the work as a whole, from inception to publication4. Others who make material contributions (e.g., statistician or medical writer) may or may not qualify as authors as defined by the International Committee of Medical Journal Editors' Uniform Requirements for Manuscripts Submitted to Biomedical Journals4. Do the authors sit at the keyboard and write the manuscript? It depends. Producing a manuscript in a timely fashion may conflict with the demands of a busy clinical practice. The style, structure and language of an article for a peer-reviewed journal are very different from those of laboratory reports, clinical summaries or quarterly reports for upper management. The author may speak English as a second language. The author may be disabled. The author or writer must be knowledgeable about the therapeutic area, understand and follow the CONSORT (Consolidated Standards of Reporting Trials; http://www.consort-statement.org) statement5, 6 or other guidelines for how research is reported in a medical journal, be knowledgeable about the statistical analyses, be expert in written English, be detail-oriented (because a decimal point can kill), be experienced in working with investigators, be familiar with the published and unpublished research and be able to pose and respond to questions from all of the stakeholders. This is true whether the person at the keyboard is the named author, a staff writer or an outside vendor. In terms of compensation, writing a clinical research manuscript takes time. Clinical research is time sensitive7. Medical writing assistance may be a form of financial support that the authors have an obligation to acknowledge4, even if it is less costly than taking time away from clinical practice. The cost to write the April JAMA article was, at a minimum, the dollar value of each author's time expended on writing plus the cost of the research time by staff members who sifted through court documents and wrote summaries that the authors used as source documents. The authors, who criticized compensation of investigators and writers, were in fact compensated themselves for writing the article. The American Medical Writers Association (Rockville, MD, USA), European Medical Writers Association (Petersfield, UK) and Australian Medical Writers Association (Chatswood, NSW, Australia) play key roles in educating medical writers about ethical practices, including accountability in authorship and medical writing acknowledgement8, 9. We thank the various researchers for bringing attention to the field of medical writing, albeit in their description of our profession as ethereal medical communicators. It's almost flattering to read that we have the power to spin P-value straw into statistically significant gold. That, however, would be spooky.