Can a medical researcher have too many publications?

The most prolific researchers may not be adhering to authorship guidelines

Medical research is a very competitive business, with a low success rate for grants and fellowships. To survive the competition, a researcher needs strong performance indicators, chief of which is the number of publications and associated citations. With publications, more is generally seen as better. However, I argue that very high publication rates should be seen as indicating poor authorship practices and should be discounted in evaluating track record.

The reason is that some researchers are claiming authorship on an extraordinary number of publications. To illustrate, using Publish or Perish software (Harzing, http://www.harzing.com/pop.htm), I did a publication count of the 27 Australian health and medical researchers listed on the Highly Cited Researchers 2014 website (Thomson Reuters, http://www.highlycited.com). In 2014, their median number of publications was 32, with eight individuals having more than 50 publications (more than one per week) and one author having more than 100 publications (more than two per week). I question whether it is possible to meaningfully participate as an author on one or more publications per week.

How many publications are feasible?

The Australian Code for the Responsible Conduct of Research, which is endorsed by the National Health and Medical Research Council (NHMRC), the Australian Research Council and Universities Australia, states:

Attribution of authorship depends to some extent on the discipline, but in all cases, authorship must be based on substantial contributions in a combination of:

• conception and design of the project

• analysis and interpretation of research data

• drafting significant parts of the work or critically revising it so as to contribute to the interpretation.

The right to authorship is not tied to position or profession and does not depend on whether the contribution was paid for or voluntary. It is not enough to have provided materials or routine technical support, or to have made the measurements on which the publication is based. Substantial intellectual involvement is required.1

To get an indication of how many publications per year might be feasible while still adhering to the Code, I conducted a straw poll of 10 full professors at the University of Melbourne who are active in clinical or public health research. I asked them: “Imagine a person who is a full-time researcher funded by an NHMRC Research Fellowship. This person is named as an author on a certain number of publications per year. Let the person’s average number of publications per year be X. What do you think is the maximum feasible value of X if the person adheres to the authorship criteria above?” Answers ranged from 10 to 30, with a mean of 17.5. While this is based only on a straw poll, it has been shown that averaged estimates of experts can be surprisingly accurate.2

Based on these figures, and allowing considerable latitude for differences between disciplines and extremes of productivity, I suggest that it is not plausible that a researcher could author 50 or more publications a year and still adhere to the NHMRC authorship criteria.

Why should we be concerned?

Claiming authorship that is not due is on the spectrum of scientific misconduct. If an author has not contributed at the required level, we cannot trust that there has been adequate oversight of the quality of the work. Unfortunately, such behaviour is likely to be rewarded, except in rare instances where the research is found to be fraudulent or incompetent. There may also be a misuse of power. It has been pointed out that senior researchers can achieve high publication rates by exerting pressure on more junior researchers, who may do most of the work, to be listed as an author.3

What should be done to promote more responsible authorship?

First, I propose that producing an implausibly large number of publications per year should be counted negatively by grant review, appointment and promotion committees. This should be made explicit in the criteria for awarding grants and positions. Sometimes researchers are asked to list their “best X publications”, emphasising quality rather than quantity, but this is not enough, because a person with a large number of publications has more to choose from.

Second, employing institutions should do spot audits on the authorship contributions of staff, particularly those exceeding feasible publication rates for their discipline, similar to what occurs with medical overservicing.

Third, journals could do more to check the role of authors in producing a publication. Some require that each author state what his or her specific contribution is, but this is not universal and editors may not query whether the stated contribution is sufficient.

If action is not taken on this issue by the medical research community, funding bodies, institutions and journals, we are in danger of seeing an escalation of implausible authorship claims as researchers compete with each other for scarce resources.