In January 2014, the patient safety community was stunned to learn that the U.S. Department of Justice had reached a $40 million settlement with CareFusion, a medical products company. 1–3 Included in the settlement was the allegation that CareFusion had paid the co-chair of the Safe Practices committee of the National Quality Forum (NQF), Dr Charles Denham, more than $11 million, allegedly to influence the committee to recommend the company’s products.
Dr Denham is chairman of the Texas Medical Institute of Technology (TMIT), a nonprofit medical research organization centered on patient safety, whose major documentary Chasing Zero: Winning the War on Healthcare Harm was partially funded by CareFusion. He is also chief executive officer of Health Care Concepts, Inc (HCC), a for-profit company (www.safetyleaders.org/disclosures/home.jsp). He has been an advocate of patient safety for more than a decade.
Dr Denham was co-chairman of the NQF Safe Practices committee from 2006 to 2010. The Department of Justice allegations center on reports that Denham lobbied the Safe Practices committee to “use chlorhexidine gluconate 2% and isopropyl alcohol solution as skin antiseptic preparation…” (Safe Practice 22). 4 Because CareFusion’s ChloraPrep product was the only one on the market that contained this specific formulation, the company stood to benefit. In a report published by ProPublica, it is apparent that Dr Denham failed to disclose to NQF the conflicts of interest (COIs) with his for-profit companies and their relationship with CareFusion. 3 In addition, it has also been reported that Denham changed a general recommendation to use chlorhexidine to prevent central line–associated bloodstream infections to specify the ChloraPrep formulation (Safe Practice 21). 5 It should be noted that liability has not been established, and Dr Denham denies any allegation of wrongdoing.
Dr Denham was also editor-in-chief of the Journal of Patient Safety (JPS), and allegations of COI have spread to the Journal. When he was named as the Journal’s new editor in February 2011, 6 his selection was a surprise to many. 5 This included the authors of this editorial, 2 of whom were on the editorial board, and another was an associate editor of the Journal; none were consulted about his appointment. He was hired through an ad hoc process, without a formal search. Dr Denham had not had a traditional academic career. A review of the peer-reviewed literature found only 17 articles published by Dr Denham before 2011, a total of 7 of which he was first author. 5 However, he was a prominent and respected figure in patient safety, served as chairman of the Leapfrog Group Safe Practices Program, was a cofounder of the Global Patient Safety Forum, and was a producer of numerous multimedia productions on the topic. Perhaps, because of this, his lack of academic credentials was overlooked.
Since the Justice Department settlement, questions have been raised about the propriety of Dr Denham’s role as editor-in-chief of the Journal, including the fact that the large majority of his first-authored articles was published in the Journal after he assumed that role in 2011. 3,5 We conducted a review of the articles authored and coauthored by Dr Denham to determine whether there were COIs.
The purpose of this review was to assess whether there were any potential undisclosed COIs, either actual or those that could be perceived, in 10 articles. These conflicts might involve TMIT, Denham’s nonprofit company; HCC, his wholly owned for-profit company; Safety Leaders; or other involvements.
The 10 articles included 8 articles 7–14 published in the Journal after February 2011 and 2 additional articles 15,16 that were published in 2010. Two of the authors of this article (A.W.W and K.T.K.) independently reviewed the articles and COI forms for declared COIs and sources of funding. We also examined the articles’ content for potential COIs. Disagreements between the 2 reviewers were resolved by consensus.
Of the 10 JPS articles reviewed, Dr Denham was the first author of 6 and a coauthor of 4 (Table 1). The 10 articles included original articles, reviews, and editorials. Potential conflicts were identified in 9 of the 10 reviewed articles. There were no significant disagreements between the 2 reviewers.
Seven of the articles cited and encouraged adoption of the NQF Safe Practices for health care. Another 2 articles 14,15 promoted the TMIT Greenlight program, whose mission includes adoption of the NQF Safe Practices. 11
There were no COIs declared for 5 of the articles with potential COIs (other than being funded or from TMIT in 4 of these articles). These articles were published in 2010 through the first half of 2012 and covered topics of the National Transportation Safety Board for Health Care, 12 Partnership for Patients, 14 The Greenlight Project, 15 and Story Power. 16 In another 2 articles, conflicts were listed on the COI form but were not contained in the published article (other than being from TMIT). 9,10
CareFusion was listed as a declared ongoing relationship for 1 article, 8 as a past relationship on the COI forms (but not in the article) for 3 other articles, 7,9,10 and as a funding source for a fifth article. 11 The HCC was also listed as a funder for this article. 11 Dr Denham was listed as being a consultant with General Electric Co. (GE) and Siemens in an article of medical imaging. 11 The article by Frush et al 9 was also on medical imaging, but no potential conflicts were listed in the published article, although a past relationship with Siemens and GE was declared on the COI form. An article concerning electronic health records (EHRs) 10 also disclosed a past relationship with Siemens (Siemens makes EHRs) and GE only on the COI form.
When a COI exists, or the perception of such a conflict, it could, although it does not always, affect the validity of a recommendation. However, it does have the potential effect of undermining trust by the public. It also undermines the weight given to a recommendation and thus may inhibit its widespread adoption. As stated by Sheldon Krimsky, a medical ethicist, “Conflict of interest is as much an appearance as it is an effect.” 17
In this case, the appearance is that there was a COI. Dr Denham’s for-profit company HCC had a financial interest with CareFusion. In the absence of a convincing alternative explanation from Dr Denham, it appears that he was induced by payments from CareFusion to recommend and promote NQF policies that were favorable to CareFusion.
Although no article published in the JPS directly advocated for the use of 2% chlorhexidine, it also appears that the Journal was used to promote an NQF recommendation that promoted 2% chlorhexidine. At the time, Dr Denham had both editorial control of the JPS and significant input into the development of the NQF product. This is a clear violation of the standards of the Journal.
In addition to COIs as an author, there are potential COIs as an editor, with potential effects on the broader content of the journal. There are a number of ways in which an editor could influence the content of a journal to promote a conflicting interest. These include accepting for publication other articles that would tend to promote the editor’s conflicting interest and rejecting articles that might detract from it, editing articles so as to help support the conflicting interest, as well as soliciting articles and editorials that would tend to support the conflicting interest. We did not review all of CareFusion’s many products or all of the articles published and rejected by the Journal during Dr Denham’s tenure as editor-in-chief. However, there is concern that the JPS could have been used to promote products in addition to 2% chlorhexidine. It is unlikely that the NQF recommendation caused any patient harm. Thus, the main harm was to the readers’ trust in the Journal as an objective source of information.
For years, there was little worry about, or scrutiny of, guidelines for quality of care and patient safety for this type of invidious influence. The field of patient safety has been heavily influenced by other disciplines, such as aviation and human factors engineering, and by patient advocates. It is possible that this multidisciplinary focus or the immaturity of the field has distracted patient safety scholars and practitioners from concern about undue influence. Whatever the reasons, it is clear that this period of innocence has come to an end.
There have been increasing reports that clinical practice guidelines are being influenced by financial COIs and by panel members with a specific point of view. A recent article by Lenzer 18 and the Guideline Panel Review Working Group summarizes the problem, which includes unreported financial ties to industry among guideline developers.
There is a potential trade-off between greater specificity of guidelines and the greater risk for endorsing a specific product. The potential conflict is likely to increase, given the current decline in U.S. government funding for research and concomitant increase in funding of research by industry. 19
Journals for quality and patient safety in general, and the JPS in particular, need to take steps to identify and manage COIs appropriately. This is crucial if the articles they publish are to remain credible and trustworthy. For this reason, there have already been changes at the Journal.
Dr Denham has resigned as editor-in-chief, and after a comprehensive search and evaluation process, Dr Bates has agreed to serve as editor-in-chief. The JPS supports full disclosure of COIs according to the International Committee of Medical Journal Editors (ICJME) standards (http://www.icmje.org/recommendations/). As such, the JPS has taken the several measures to reduce the probability that such undisclosed conflicts will occur again. Measures to be taken to reduce both real and perceived conflict by both authors and editorial staff include the following:
– The Journal will post COI forms online along with the published articles.
– Aspects of the rigorous standards for COI endorsed by the ICJME will be incorporated into the Journal’s operations.
– The Journal will ask reviewers to declare their own potential COIs with any of the authors on the article.
– Editors will publish regular disclosure statements about potential COIs related to the commitments of the journal staff. Guest editors will follow these same procedures.
– Editors who make final decisions about manuscripts will be instructed to recuse themselves from editorial decisions if they have COIs or relationships that pose potential conflicts related to articles under consideration.
– Other editorial staff members who participate in editorial decisions will provide editors with a current description of their financial interests or other conflicts (as they might relate to editorial judgments) and recuse themselves from any decisions in which a COI exists.
– Editorial staff will be explicitly prohibited from using information gained through working with manuscripts for private gain.
– Future appointments of editors-in-chief, made by the journal owner, will be guided by the evaluation of a panel of independent experts, as recommended by the ICMJE (http://http://www.icmje.org/recommendations/).
– Members of the public will be invited to share information that they have about perceived COIs by authors or editors.
It should be acknowledged that, even if all of the abovementioned processes had been in effect, the undeclared conflicts of Dr Denham may not have been detected. We recognize that it is difficult for an individual reviewer to be aware of all of the potential COIs posed by a given publication. It is even more difficult for a reader to identify the significance of a potential conflict. For example, even if Dr Denham had listed an ongoing consulting relationship with CareFusion, the reader would not be able to infer that an article that supports an NQF recommendation is endorsing a product made by CareFusion. Thus, the successful functioning of the disclosure process relies on the candor of individuals in making their disclosures of potential conflicts. It took a whistleblower to bring the CareFusion story to light. The quality and safety community is relatively close-knit, making it likely that community policing may be a more effective strategy. For this reason, the Journal will encourage all of its editors, as well as readers, to comment online about published articles, including potential COIs.
It is impossible to completely avoid all COIs, and many advances in patient safety will likely come as the result of interventions developed by for-profit companies. That being said, it is clear that patient safety in general and the JPS will need to be much more cognizant of this issue in the future. The Journal’s editors are committed to that.
The Journal’s mission is to publish work that will make patients safer, and that should not be subverted by commercial interests. We regret the breach of trust to the readers of the Journal caused by Dr Denham. It is important for the Journal and for the field of quality and safety that readers have greater trust in the Journal going forward. We believe that the best way to deal with conflicts is for them to be declared and dealt with by greater transparency. Conflicts need to be managed appropriately, and organizations need to have a culture in which people are attentive to conflicts and willing to identify them. We look forward to working with other journals and organizations to help the field of patient safety and quality to mature as well as to implement a useful core set of recommendations for maintaining impartiality.
4. The National Quality Forum. Safe Practices for Better Healthcare 2009 Update: A Consensus Report
. Washington, DC: The National Quality Forum; 2009.
7. Leung AA, Denham CR, Gandhi TK, et al. A safe practice standard for barcode technology. J Patient Saf
. 2014. PubMed PMID: 24618650.
8. Hess EP, Haas LR, Shah ND, et al. Trends in computed tomography utilization rates: a longitudinal practice-based study. J Patient Saf
. 2014; 10: 52–58.
9. Frush D, Denham CR, Goske MJ, et al. Radiation protection and dose monitoring in medical imaging: a journey from awareness, through accountability, ability and action…but where will we arrive?. J Patient Saf
. 2013; 9: 232–238.
10. Denham CR, Classen DC, Swenson SJ, et al. Safe use of electronic health records and health information technology systems: trust but verify. J Patient Saf
. 2013; 9: 177–189.
11. Denham CR, Guilloteau FR. The cost of harm and savings through safety: using simulated patients for leadership decision support. J Patient Saf
. 2012; 8: 89–96.
12. Denham CR, Sullenberger CB 3rd, Quaid DW, et al. An NTSB for health care: learning from innovation: debate and innovate or capitulate. J Patient Saf
. 2012; 8: 3–14.
13. Denham CR. Learn global, act local, and be vocal. J Patient Saf
. 2011; 7: 1–4.
14. Denham CR. The partnership with patients: a call to action for leaders. J Patient Saf
. 2011; 7: 113–121.
15. Denham CR. Greenlight issues for the CFO: investing in patient safety. J Patient Saf
. 2010; 6: 52–56.
16. Quaid D, Thao J, Denham CR. Story power: the secret weapon. J Patient Saf
. 2010; 6: 5–14. doi:10.1097/PTS.0b013e3181d23231. PubMed PMID: .
18. Lenzer J. Why we can’t trust clinical guidelines. BMJ
. 2013; 346: f3830. Erratum in: BMJ