Letter to the Editor

Uncontrolled proliferation of the co-authors of clinical and observational studies, and inter-cohort analyses, in the field of clinical management of HIV disease

Roberto Manfredi 1

1 Malattie Infettive, “Alma Mater Studiorum” Università di Bologna, Policlinico S. Orsola-Malpighi, Bologna, Italy

In a previous article [1], we expressed concern about methods of conduction of the most important studies, related to the most significant publications about HIV natural history and its management, which, since 2008, have been based on studies conducted on a consistent number of patients from all over the world, with the main aim to obtain, in relatively short time, statistical significant data (not always synonymous of statistical clinical meaning). Moreover, these mega-studies (often retrospective analyses, or post-hoc experiences) were signed by hundreds of the so called “collaborators”, all of them “elevated” to the rank of true Authors, through the placement of each single name in all major databases and search engines of the international scientific literature, right since 2008 [1].

In the following years (2009-2010) this phenomenon has, if possible, increased exponentially, leading to a multiplication of these “collaborators”, who, until that date, were just listed in a note in the Acknowledgments section of the article, and were not considered as real authors. The current situation seems to lead to a somewhat unacceptable inflation of the author’s role itself in biomedical publications, based on the well-known statement that should be considered as author one that has given a substantial supply to the design of the study, data results and evaluation, while those who contribute to data collection cannot be included among the Authors.

As a specialist in infective diseases, here I would like only to highlight the situation of the most relevant studies in HIV infection, but obviously the same “infection” is contaminating (to use a word common in my field) the whole world of biomedical literature. Even though it is necessary to have wide databases aiming at evaluating the events that happen less frequently, to examine consistent long-term endpoints or pharmacologic safety profiles, nonetheless considerable bias and distortions happen, as a consequence of the extremely high number of the Centres and of the investigators, recruited in a planetary scale [1].

As an example, in Table I are listed the most significant and recent studies (published in 2009-2010) that have also affected the guidelines of the management of HIV-related comorbidities. These studies appear to be burdened by an enormous number of authors and collaborators that clashes especially if compared with the number of patients of cases followed, and with the times of observation and intervention. In these studies about HIV infection and AIDS, the so-called collaborators represent in reality the responsible of the Clinical Centres involved, all over the world, in the study, and in addition some sub-investigators of the belonging to the same Centres, who, in most of the cases, are responsible for the enrolling or the registration of a very little number of patients or events for each Centre or each investigator. For example, a trial published in 2010 has been signed by 13 principle authors and 80 collaborators, all indexed on PubMed-Medline. These authors examined the pharmacokinetic characteristics of an antiretroviral agent administered in association with another anti-HIV drug in no more than 54 patients, followed for 6 months [2]: it is therefore evident that each collaborator has followed an avarage of 0.58 patients, and nonetheless is included among the authors of a relevant scientific publication, included in PubMed with its update in 2008.

Likewise, a total of 204 authors and collaborators analysed the temporal suppression of HIV viraemia as a predictive factor of virologic failure in 451 patients (that is to say, an avarage of 2.21 patients enrolled by each collaborator or author) [3]. A previous trial, published in 2009, examined the role of interleukin-2 in 5,806 patients followed for 7 years, but the total number of authors plus collaborators is 967 (an avarage of 6 patients, thanks to which every collaborator has been listed as author of the prestigious New England Journal of Medicine) [4].

At the same time, all these mega-trials that have an extreme high power in affecting clinical practice, because their results are included in international clinical guidelines about HIV management, can lead to distortions, and in particular the possibility that a sort of hyperinflation of the number of the samples have been made in favour of statistic evaluations: a high number of patients or events or post-hoc analysis are usually made on data coming from hundreds of Centres disseminated all over the world, aiming at reaching a sufficient statistical power, without taking into consideration the possible bias regarding enrolling, measurement, and registration of extremely sensible data, of the time passed in long term studies, and of the inter-human differences about gender, race, BMI, and genetic and pharmacogenomic features, only to cite some of the most evident examples [1].

Parallely, in the same studies there is, as previously described, the phenomenon of the inflation of the number of the so called co-authors (in reality, nothing more than collaborators), that introduces other distortions during the evaluation of the scientific production of each investigator. Unavoidable consequences are then possible in the evaluation of academic curricula and in the consequent recognitions, for the obtainment of institutional funds, or external sponsorships, or the prestige of the institutions themselves.

In reality, the problem of the proliferation (sometimes inadequate) of the authorship is a well known phenomenon since twenty years and concerns all the fields of the research and clinical medicine, but nonetheless is and remains a scarcely debated topic, as emerged from the few articles published [5-9].

In my opinion, all the involved authorities, starting form scientific associations and biomedical journals and publishers, should become aware of this increasing problem and of the implicit distortion and scientific, technical, and ethical related consequences, with the goal of developing some rules that can guarantee the maintenance of the central and critical role of the authorship, as currently intended, and of preventing and avoiding the main bias related to the conduction of mega-trials and inter-cohort analyses, often signed by a disproportionate number of co-authors. Probably, it appears correct to list an adequate number of authors that contributed in first person in conceiving, following, writing and discussing the study (the so called “steering committee”). The remaining co-investigators should better be represented using an acronym or a “corporate” denomination, that comprehends in a single abbreviation all the co-investigators that took part in the study, that can be then listed in a specific appendix, or in the standard Acknowledgement section, as it happened before 2008 (and as reported in the recent HIV-Causal Collaboration study which, even if not listing any author, allowed the inclusion in PubMed of even 1.179 presumed “co-authors”) [10].

Scientific and practical consequences related to the publication and diffusion of these mega-trials deserve the maximum attention of all the scientific community.

References

1. Manfredi R. Studi multicentrici internazionali, analisi inter-coorte, proliferazione incontenibile del numero degli autori: sono queste le rispose attese per affrontare l’infezione da HIV? Recenti Progr Med 2009; 100: 119-23
2. Baroncelli S, Villani P, Weimer LE, Ladisa N, Francisci D, Tommasi C et al; ISS-NIA Group. Raltegravir plasma concentrations in treatment-experienced patients receiving salvage regimens based on raltegravir with and without maraviroc coadministration. Ann Pharmacother 2010; 44: 838-43
3. Reekie J, Mocroft A, Ledergerber B, Beniowski M, Clotet B, van Lunzen J et al; EuroSIDA Study Group. History of viral suppression on combination antiretroviral therapy as a predictor of virological failure after a treatment change. HIV Med 2010; 11: 469-78
4. INSIGHT-ESPRIT Study Group; SILCAAT Scientific Committee, Abrams D, Lévy Y, Losso MH, Babiker A, Collins G, Cooper DA et al. Interleukin-2 therapy in patients with HIV infection. N Engl J Med 2009; 361: 1548-59
5. Schechter AN, Wyngaarden JB, Edsall JT, Maddox J, Relman AS, Angell M et al. Colloquium on scientific Authorship: rights and responsibilities. FASEB J 1989; 3: 209-17
6. King JT Jr. How many neurosurgeons does it take to write a research article? Authorship proliferation in neurosurgical research. Neurosurgery 2000; 47: 435-40
7. Birkhahn RH, Van Deusen SK, Okpara OI, Datillo PA, Briggs WM, Gaeta TJ. Funding and publishing trends of original research by emergency medicine investigators over the past decade. Acad Emerg Med 2006; 13: 95-101
8. Levski ME, Rosin A, Coon TP, Enslow WL, Miller MA. A descriptive analysis of authorship within medical Journals, 1995-2005. South Med J 2007; 100: 371-5
9. Rahman L, Muirhead-Allwood SK. How many orthopedic surgeons does it take to write a research article? 50 years of Authorship proliferation in and internationalization of the orthopaedic surgery literature. Orthopedics 2010; 33: 478
10. HIV-Causal Collaboration. The effect of combined antiretroviral therapy on the overall mortality of HIV-infected individuals. AIDS 2010; 24: 123-37