; Publishing Commentary by Authors with Potential Conflicts of Interest: When, Why, and How. Ann Intern Med. 2004;141:73-74. doi: 10.7326/0003-4819-141-1-200407060-00020
Download citation file:
Published: Ann Intern Med. 2004;141(1):73-74.
While some journals do not publish opinion pieces written by authors with declared potential conflicts of interest, Annals sometimes does. In this editorial, we discuss a recent instance that helped us to crystallize our thinking about this prickly topic.
The story began when we received an unsolicited critique of the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), a randomized study of hypertension therapy (1). The author disclosed financial relationships with several pharmaceutical firms that make drugs for hypertension. We thought that the piece raised important, controversial issues about an influential study. Positive and thoughtful comments from external reviewers and a lively, serious discussion among the associate editors and senior editors convinced us that many readers would be likely to find it interesting. We invited the author to revise the commentary on the basis of the reviewers' and editors' suggestions, and then published the revision as a Perspective along with the author's conflict of interest disclosure (2).
Hean T Ong
H T Ong Heart Clinic, 251C Burma Road, Penang 10350, Malaysia
July 16, 2004
Beware bias in drawing conclusions from clinical trials
Your editorial of July 6 is reassuring in detailing the steps taken by the Annals, and other journals, to avoid bias in commentary and opinion pieces (1). Since most large clinical trials are now funded by pharmaceutical companies, an equally pressing concern is the need for vigilance to prevent interpreting data and drawing conclusions to suit the marketing needs of pharmaceutical products.
The LIFE study concluded that "losartan prevents more cardiovascular morbidity and death than atenolol for a similar reduction in blood pressure"(2). Yet only 9% of the study population were on monotherapy with losartan and only 10% were on atenolol monotherapy; the great majority of patients had combination antihypertensive therapy. Furthermore, there was a significantly lower systolic blood pressure in the losartan group at the end of the study period. The concluding statement on losartan's superiority over atenolol is unfair when so few patients were actually involved in a direct comparison of the two drugs, and the blood pressure reduction was not similar statistically in the two treatment groups.
The PROGRESS study involved starting patients on the active treatment group with perindopril, and indapamide is added at the discretion of the treating physicians (3). Compared to placebo, the patients on perindopril alone had no significant reduction in stroke despite having a lower blood pressure level; those on combination treatment saw more blood pressure reduction and a highly significant stroke reduction. Logically then, a case can be made for the importance of blood pressure reduction, or even for the importance of indapamide, given that the reduction in strokes was only seen when indapamide was added to perindopril treatment. The conclusion that "treatment with these two agents should now be considered routinely for patients" is difficult to justify when one of the two agents, perindopril, was shown to have no clinical benefit !
Pharmaceutical companies seem especially efficient in getting the lay media to run impressive headlines putting a spin on trial results that bears little resemblance to the actual facts (4). Medical investigators, authors, reviewers and editors need to be vigilant to ensure that trial conclusions published in academic journals do not sound like releases from the marketing department of the sponsoring pharmaceutical company.
1. The editors. Publishing commentary by authors with potential conflicts of interest: When, Why, and How. Ann Intern Med 2004; 141:73-74. 2. Dahlof B, Devereux RB, Kjeldsen SE, et al for the LIFE study group. Cardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE) : a randomized trial against atenolol. Lancet 2002; 359: 995-1003. 3. PROGRESS Collaborative Group. Randomised trial of a perindopril-based blood pressure lowering regimen among 6105 individuals with previous stroke or transient ischaemic attack. Lancet 2001; 358:1033-41. 4. Preidt R. Newer drug helps prevent heart attack. Forbes.com. http://www.forbes.com/lifestyle/health/feeds/hscout/2004/06/17/hscout519576.html (accessed June 23,2004).
Dr H T Ong FRCP(Glas, Edin)
H T Ong Heart Clinic
251C Burma Road
e-mail : firstname.lastname@example.org
David B Neely
August 5, 2004
Pharmaceutical company influence
Thank you for your response to my letter. (1,2). I appreciate the principles you laid out and I appreciate your reluctance to make absolute rules that would exclude experts who have conflicts of interest. Rules that exclude experts in rare conditions or specialized research might compromise the possibility of any expert comment at all; however, for common conditions such as hypertension, there are likely to be many experts, some of whom have less a financial stake in the pharmaceutical companies in question. As we doctors deliberate on this important issue of conflict of interest two major issues emerge. 1) Is expert opinion improperly influenced by the remunerative benefits provided by pharmaceutical companies? 2) If so, what should we do about it?
I worry that you underestimate this influence. For instance, you use the term potential conflict of interest. If a doctor receives tens of thousands of dollars from a particular pharmaceutical company, I am not sure why that is a potential conflict of interest when that doctor is commenting on the drugs marketed by that company.
The startling result of ALLHAT is: there is no medical advantage to using amlodipine or lisinopril instead of chlorthalidone. If anything, chlorthalidone had fewer side effects. (15% vs 18%) If anything, the chlorthalidone group had better BP control (68% vs 61%). If anything, chlorthalidone has better long term outcomes (CHF). Amlodipine costs $76 a month and lisinopril costs $57 a month and chlorthalidone costs $10 a month. Our patients have been spending millions of dollars for unnecessarily expensive medications. The pharmaceutical industry has been reaping the profits. A major reason doctors prescribe amlodipine and lisinopril instead of chlorthalidone is pharmaceutical marketing and advertising. In addition, many of our experts have been paid well to speak about the wonders of CCBs and ACEIs.
Even in the thoughtful article by Messerli (3) that was carefully reviewed by the Editors, evidence of this influence persists. His only discussion related to these "millions of dollars" was "While diuretics are inexpensive"¦their use requires the not-so-inexpensive monitoring of electrolyte, renal and metabolic measures". For most of my patients on diuretics, this monitoring translates to one more chemistry panel a year. He did not mention that amlodipine costs $66 more per month than chlorthalidone. He offers some interesting speculations about why diuretics might make the diagnosis of CHF more difficult. However, he does not mention that this CHF finding is consistent with our long standing concerns about CCBs and systolic dysfunction. Then in the last paragraph, he mentions carcinogenicity of diuretics (and quotes an opinion piece of his own). There have also been studies that have raised concerns about cancer with calcium channel blockers. (4,5)
ALLHAT makes it clear that the cheaper and older drug is as good if not better than the expensive new ones. It seems that "pharmaceutical company influenced" expert opinion has contributed to the preferential prescribing of the expensive new drugs.
My concerns are twofold. The pharmaceutical companies have the financial power to manipulate us. Many doctors enjoy the financial benefits of this manipulation while downplaying its effect on us and our patients. The ANNALS could help discourage our readers and experts from being influenced by pharmaceutical companies by not printing the opinions of experts who have clear financial ties to pharmaceutical companies that are directly related to the the topic upon which they are opining.
This is a difficult and drastic step only worth the effort if the consequences of drug company influence is fully appreciated.
1 Publishing Commentary by Authors with Potential Conflicts of Interest: When, Why, How, The Editors. Ann Intern Med 2004;141:73-74.
2 Neely, D. ALLHAT, or the soft science of the secondary end point. [letter] Ann Intern Med 2004;141:77.
3 Messerli FH. ALLHAT, or the soft science of the secondary end point. Ann Intern Med. 2003;139:777-80. [PMID: 14597462]
4 Pitt B, Byington RP, Furberg CD, et al. Effect of amlodipine on the progression of atherosclerosis and the occurrence of clinical events. PREVENT Investigators. Circulation. 2000; 102: 1503"“1510
5 Pahor M, Furberg CD. Is the use of some calcium antagonists linked to cancer? Evidence from recent observational studies. Drugs Aging. 1998; 13: 99"“108
to gain full access to the content and tools.
Learn more about subscription options.
Register Now for a free account.
Cardiology, Nephrology, Hypertension, Coronary Risk Factors.
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only