INTRODUCTION

Declaration of potential conflicts of interest is now required when submitting manuscripts to almost all biomedical journals, but there is much uncertainty regarding their utility for editors, reviewers, and readers.1,2 Although financial involvements are typically emphasized, non-financial competing interests are also important but more difficult to define, measure, and manage.3

Authors are increasingly referring to themselves as ‘unpaid consultants.’ By its literal meaning, this indicates the absence of direct monetary reimbursement. Other benefits from industry involvement may nonetheless constitute actual or perceived conflicts of interest and ought to be disclosed. We systematically studied the term ‘unpaid consultant’ in medical article disclosures by analyzing its use over time, links to vested interests, and association with particular journals and authors.

METHODS

We used Google Scholar to determine occurrence of “unpaid consultant” in biomedical publication disclosures during 1994-2014. Approximately 6% of articles meeting criteria had two or more authors declaring unpaid consultancy; we analyzed results both per article (n = 787, Fig. 1) and per disclosure (n = 865, Table 1).

Fig. 1
figure 1

Articles with unpaid consultant authors. Data indicate unique articles (n = 779) published during 2001–2014 with one or more authors declaring unpaid consultancy to one or more for-profit companies, normalized to control for growth in overall publication volumes. *Histogram bars indicate target article count divided by total MEDLINE citation volume (×106) for that year (see text)

Full size image
Table 1. Breakdown of Unpaid Consultancy by Category
Full size table

Results

Review of each Google Scholar ‘hit’ indicated that most were relevant, inasmuch as authors referred to relationships that were not directly financial but nonetheless suggested important and usually unexplained engagement with for-profit companies. To control for trends in publication volumes, we used annual MEDLINE citation counts (https://www.nlm.nih.gov/bsd/medline_cit_counts_yr_pub.html, last accessed 4 September 2016). As Fig. 1 illustrates, the normalized publication rate of health-related articles with authors disclosing unpaid consultancy increased markedly between 2004 and 2014.

Further analysis of ‘unpaid consultancies’ showed these were overwhelmingly (801/865, 92.6%) associated with for-profit companies (Table 1). Declarations of unpaid consultancy to companies in various categories, including not-for-profit entities (government and NGO), were found to increase comparably during this 10-year period. With very few exceptions, authors provided no detail of these relationships or the associated benefits, either to the companies/entities or themselves.

A wide range of medical journals included declarations of unpaid consultancy; American journals predominated, and some disciplines, notably orthopedics, were overrepresented (data not shown). Some authors made frequent, in most cases identical, disclosures to multiple journals. Additionally, most authors describing themselves as unpaid consultants also disclosed paid relationships with for-profit companies. For example, five of nine authors with 10 or more disclosures of unpaid consultancy also declared financial relationships in 73/152 articles, naming for-profit companies 159 times. While this finding accords with studies showing that individual authors often report multiple conflicts,3 the connection between financial conflicts and the more recent surge of ‘unpaid’ disclosures remains unclear.

DISCUSSION

The use of ‘unpaid consultant’ in disclosures relates to the now prevalent requirement for authors and editors to recognize and manage conflicts of interest. The term may be disingenuous, however, if it fails to identify research funding, conference fees, travel expenses or other benefits from engagement with for-profit industries. Disclosure statements should be as complete and accurate as possible2 to enable editors, reviewers, and readers to appraise the relevance of competing interests to an author’s collection, analysis, and interpretation of data. Declarations of unpaid consultancy appear unhelpful and may do more to conceal than illuminate.

People commonly underestimate their own vulnerability to bias and tend to be better at judging others’ biases than their own.4 Submitting authors have an understandable motivation to portray themselves as willing to disclose possible sources of bias. This motivation, coupled with routine disclosure requirements, may have contributed to the increase in authors describing themselves as ‘unpaid consultants.’ This terminology signals a relationship but almost never explains what it entails, and thus provides little basis to appraise how it may affect an author’s motivation and judgment. Our results are consonant with the view that financial disclosure alone is insufficient, that professional/intellectual conflicts are comparably important, and that standards for disclosure of the latter are currently inadequate.

In conclusion, full and honest disclosure of authors’ competing interests, both financial and non-financial, is essential, but in itself does not assure freedom from bias.2 , 5 In particularly ‘sensitive’ domains, such as treatment guideline preparation, excluding rather than managing conflicts has been recommended.6 More generally, our results indicate the need for further examination of language and meaning in competing interest declarations and suggest that editorial requirements for disclosure be strengthened.