This article was originally published on June 2.
Researchers at US medical schools who rely heavily on industry support are more likely to have been pressured by sponsors to act unethically and to have firsthand knowledge of integrity breaches at their institutions, according to recently published survey results.
The study also revealed that the most influential faculty members, such as full professors and those with the largest number of publications under their belts, were significantly more likely than their junior colleagues to receive industry support, raising concerns over the influence such researchers might have on research culture.
Among its findings were that 28 percent of researchers for whom industry support was "very important" or "extremely important" reported having been asked by sponsors to either withhold research results, delay publication, present results more favorably, or keep the project secret.
Meantime, an even greater percentage of researchers with heavy industry support reported having "firsthand knowledge" of colleagues that had participated in such conduct.
The findings suggest the possible existence of a "culture of silence" within academia where investigators are aware of ethical breaches in industry-sponsored research but are reluctant to police themselves or colleagues, according to the authors.
The study, which was published in the March issue of Accountability in Research, was conducted by a group led by bioethicist Patricia Tereskerz from the program in ethics and policy in healthcare at the Center for Biomedical Ethics and Humanities at the University of Virginia School of Medicine.
Entitled "Prevalence of Industry Support and its Relationship to Research Integrity," the study marks "the first attempt to acquire updated empirical data about financial arrangements and conflicts of interest between industry and investigators at academic research institutions," the authors wrote.
The team, which also included researchers from the UVA School of Nursing and Center for Survey Research, and the University of Pennsylvania, sent questionnaires to a sample of 1,548 clinical researchers at top US research universities and medical schools. A total of 1,156 were estimated to have qualified for the study.
As part of the study, the authors purposely sent surveys to faculty members at the 33 US universities that received the most research funding over an undisclosed time period, assuming that they would also receive considerable amounts of industry funding.
The study did not specify when the survey was sent out or when results were compiled. The authors used medical-school catalogs and graduate program guides to identify all life sciences departments and graduate programs at the selected institutions.
From each school, the authors randomly selected one department of medicine, another clinical department, and two non-clinical departments. The authors noted that faculty in departments of medicine and their subspecialties were oversamples because they often received more extramural research funds than other departments.
Respondents were asked to report their receipt of financial support in various forms from industry over the five-year period preceding the survey's issuance, and were asked directly about industry pressure to favor the sponsor's interests during that time.
The survey revealed that 66 percent of respondents had received some form of industry support, while 62 percent reported receiving research and publication support from industry entities.
The authors also wrote that 13 percent of respondents held an endowed chair, 14 percent of whom were "funded by industry" in some way. Findings also showed that respondents with 11 years or more of research experience, those with higher academic ranks, and those who published more frequently were "significantly more likely" to receive industry support.
[ pagebreak ]
Survey respondents also reported participating in various types of financial arrangements with industry within the last five years. For instance, 7 percent said they owned an equity interest in a company that supported their research, generally in the pharmaceutical arena. Of these respondents, half owned an equity interest worth between $100,000 and $500,000, the survey results showed.
The survey treated differently questionable actions by sponsors and questionable actions by researchers. For instance, respondents who reported having received industry support were asked whether an industry-connected sponsor had ever made various unethical requests. However, with respect to researchers actually acceding to such requests, respondents were asked only for "firsthand knowledge" of such breaches, mainly to alleviate concerns they might have about admitting to these practices, the authors wrote.
Of the 231 respondents who received industry support for their research and answered survey items about questionable actions, 4 percent reported that a sponsor had asked them to withhold publication of research results; while 13 percent said they had been asked to delay publications.
Meantime, nearly 8 percent had been asked by a sponsor to present research results in a way that favored the sponsor's drug or product; and 7 percent had been asked by a sponsor to keep research results secret.
Far more common, "and of lesser concern," the authors wrote, were reports of sponsors asking for pre-publication review (61 percent) and asking for acknowledgement in the publication (62 percent).
The authors also found that the degree of importance of industry support to a respondent's research significantly correlated to the degree of frequency of unethical sponsor requests. In fact, considering only those respondents who rated their industry support as "very important" or "extremely important," 28 percent had been asked either to withhold research results, delay publication, present results more favorably, or keep the project secret.
Furthermore, for those researchers who relied most heavily on industry support, the rates of "first-hand knowledge" of colleagues' ethical breaches were 25 percent for delaying publication; 17 percent for presenting results to favor the sponsor; and 11 percent for suppressing publication due to pressure from the sponsor.
Another important finding, the authors noted, was that 9 percent of respondents with industry-funded colleagues had firsthand knowledge of compromises to the well-being of human research subjects at their institutions because researchers in their department received industry sponsorship.
"One way to view this finding is that only a small subset of these participants reported this, with nine respondents [of the 9 percent] noting that the compromise was minor," the authors wrote.
However, in a statement, Tereskerz noted that "although such compromises occur infrequently, our concern is that they occur at all. There should be zero tolerance for compromising the well-being of human research participants in any study, regardless of how the study was funded."
The authors also expressed concern with the fact that full professors and those with large numbers of publications were significantly more likely to receive industry support than were their more junior colleagues; and that various financial relationships with industry are both prevalent and substantial in size among those who reported improper conduct.
"That these patterns are seen among those most deeply immersed in industry-supported research, and that these individuals are also the most senior and prolific members of their departments, is a disturbing finding," the authors wrote.
"Because industry-supported respondents were significantly more likely to be the most senior and prolific researchers, their leadership and potential influence over the cultures of individual research units are considerable," they added. "Our data raise the possibility of research cultures in which abuse of research integrity are known but not openly discussed or addressed."
The authors conceded that the primary limitation of their study was the response rate of about 48 percent, "which was to be expected when the questions involve sensitive topics," they wrote.
They add that this issue, along with their "purposive sampling approach," means that they cannot generalize from the sample to the US medical school population. "Additional research with larger samples is needed, and likely will require incentives to increase the percentage of respondents," the authors wrote.
The study was funded by a grant from the National Institute of Neurological Disorders and Stroke of the National Institutes of Health, through its Office of Research Integrity. A copy of the study can be purchased online here.