Paying more for clinical research than the cost of doing the work may create a conflict of interest that could lead to overzealous recruitment, putting participants and scientific integrity at risk. Thus, although various policies prohibit “finder's fees” simply for recruiting patients, paying the actual costs for research is permissible. Whereas industry-sponsored research routinely pays for the costs of each patient enrolled, the line between reasonable and excessive costs merits more attention. In academic medical centers (AMCs), institutional review boards and conflict of interest committees usually are not involved in reviewing research budgets to determine whether per capita payments are excessive. Also, the costs for clinical services in research are not standardized. Instead, budgets are negotiated both internally, among departments within research institutions, and externally, between researchers and sponsors. Sometimes, rates paid by sponsors exceed what researchers usually receive or are actually paid for particular services, generating a surplus. Nevertheless, the authors see only limited cause for concern because, at the AMCs with which the authors are familiar, any monetary surplus generally remains within the research enterprise to cover unanticipated budget shortfalls or to support research staff in the future during lean times. In addition, the surplus from research budgets is not shared directly with individual investigators. However, further investigation is needed to determine whether practices outside AMCs pose greater concerns.
Mr. Hall is professor of law and public health and director, Center for Bioethics, Health & Society, Wake Forest University, Winston-Salem, North Carolina.
Ms. Friedman was project leader, Duke Clinical Research Institute, Durham, North Carolina.
Ms. King is professor of public health sciences and codirector, Center for Bioethics, Health & Society, Wake Forest University, Winston-Salem, North Carolina.
Dr. Weinfurt is associate professor of psychiatry and behavioral sciences and associate professor of psychology and neuroscience, Duke University, Durham, North Carolina.
Dr. Schulman is professor of medicine and Gregory Mario and Jeremy Mario Professor of Business Administration, Duke University, and associate director, Duke Clinical Research Institute, Durham, North Carolina.
Dr. Sugarman is Harvey M. Meyerhoff Professor of Bioethics and Medicine, Berman Institute of Bioethics, and Department of Medicine, Johns Hopkins University, Baltimore, Maryland.
Correspondence should be addressed to Mr. Hall, Wake Forest University, Public Health Sciences, Medical Center Blvd., Winston-Salem, NC 27157-1063; telephone: (336) 716-9807; e-mail: firstname.lastname@example.org.
Concerns about financial conflicts of interest in clinical research generally focus on financial relationships outside a particular research study, such as consulting fees, patent royalties, and business equity. However, payment for research itself can create a conflict of interest if the payment exceeds the reasonable cost of doing the work.1–3 Excessive payment may generate overzealous recruitment that could pose undue risk to participants. Such payment also compromises the study's scientific integrity if, for example, it results in the enrollment of participants who do not meet inclusion or exclusion criteria. For these reasons, various policies prohibit the payment of “finder's” or “bounty hunting” fees simply for recruiting patients to studies. Largely anecdotal reports of these practices raised considerable alarm a decade ago,4 but following intensive scrutiny, finder's fees are no longer considered legally or ethically acceptable.
Finder's Fees and Per Capita Payments
Banning or avoiding outright bounties for subject recruitment may be well established, but this convention leaves unresolved whether per capita payments for doing the work of clinical research might create inappropriate incentives for researchers and institutions. Each research participant, like each clinical patient, may be an opportunity for a clinical researcher or institution to earn money, sometimes a substantial amount, through honest services dutifully rendered and charged. Such incentives are inherent in any fee-for-service payment model and so are unavoidable, or at least tolerable, if necessary research work is to be done at all. Any potential for abuse that legitimate research payments might create can be policed through other means, such as overseeing the basic safety and integrity of research through institutional review boards (IRBs). Accordingly, when per capita payments cover only the costs of research, they raise no concern, either legally or in the minds of research participants or investigators.5,6 However, if per capita payments exceed reasonable costs, federal prosecutors have targeted research funding as potential violations of criminal “fraud and abuse” laws.3,4
Clearly, then, the line between reasonable and excessive costs of research is critical, both ethically and legally. However, this divide is not clear, and no consistent mechanism currently exists for reviewing the reasonableness of research costs and reimbursements.3 Nor is there any empirical literature of note exploring or documenting the extent to which research payments might exceed reasonable research costs. Therefore, it remains unknown whether and to what extent research sponsor payments exceed reasonable costs.
The reasons for a researcher or institution to accept an excessive payment are obvious, but what are the motivations for sponsors to pay them? Being the first to market a new drug or device can mean large profits, and so the stakes for industry are high. The sooner the Food and Drug Administration (FDA) approves a new drug or device, the more time the company has exclusive rights under its term-limited patent. As such, drug and device companies are under tremendous pressure to enroll enough patients to satisfy FDA standards as quickly as is feasible. Higher per capita payments to investigators may give the sponsor an edge in completing enrollment in trials and obtaining FDA approval more expeditiously.
Even after the FDA approves a new drug or device, companies have an incentive to generously fund “postmarket” or “Phase IV” studies, not only to gather additional information about the product but also to encourage its greater adoption and use. Although gathering as much information as possible about a product arguably should be encouraged, so-called “seeding trials” have been widely criticized for using purported data collection as a thinly disguised pretext to reward doctors for ordering expensive drugs and devices.7 Seeding trials were the target of a notable 2007 legal investigation involving five major orthopedic device manufacturers. The manufacturers agreed, in the end, to avoid such practices in the future. In addition, the Department of Justice has investigated similar allegations against manufacturers of implantable cardiac devices.
Seeding trials illustrate the potential for abuse, but they present a somewhat different issue than our focus in this commentary. In seeding trials, where the research itself is a pretext, any payment presumably is excessive. In this commentary, we are concerned with whether payments exceed their reasonable costs for legitimate research. Per capita payments routinely approach or exceed $10,000 per person for industry-sponsored research.3 Sometimes these payments are for lengthy and complex protocols, but other times they are for simpler office evaluation and data collection of common symptoms over a fairly short time span.
To our knowledge, no academic study has systematically investigated the reasonableness of per capita payments for clinical research. Therefore, we lay the groundwork for further examination by reporting our shared experience from different academic medical settings and from a review of the available literature.
Developing Research Budgets
Research budgets distinguish between fixed costs, which do not vary directly by enrollment, and variable costs, which do. Fixed costs focus on start-up and close-out costs. Variable costs include research staff time for recruiting, enrollment, and data collection, and any medical tests or procedures necessary to successfully implement the protocol.
At the academic medical centers (AMCs) with which we are familiar, research budget negotiations typically are handled by the clinical department conducting the study. Contracts offices review the legal details and decide whether institutional overhead costs are being adequately recouped, but they rarely examine the adequacy or excessiveness of budgets for departmental expenses. IRBs or conflict of interest committees are typically not involved in reviewing the details of research budgets.8 These committees do not tend to view research payments as creating a conflict of interest for the investigator because the payments go to the institution and not the individual.
Conflict of interest reviewers in community research settings sometimes draw a different conclusion, realizing that per capita payments can go to physician researchers individually. Therefore, some community (non-AMC) hospitals do review whether research payments are reasonable, whereas others do not, under the assumption that the payments are intended only to cover legitimate costs. When institutions do review per capita payments, it is difficult for the reviewers to decide whether the payments are excessive, and so usually no objection is made as long as there is a plausible argument that per capita payments only cover costs or are otherwise within reasonable bounds.5
Payments by Industry Sponsors
When negotiating budgets, the goal of industry sponsors is neither to minimize budgets nor to pay excessive rewards. Instead, the objective is to pay the right amount needed to conduct the required work. In general, sponsors seem more concerned with controlling research costs than with finding ways to inappropriately incentivize recruitment.
Sponsors often are willing to pay additional amounts to improve recruitment, but only if the extra expense can be justified in terms of extra costs or efforts rather than as mere finder's fees. We have not heard of any sponsors in recent years paying incentives simply to enroll more patients. After a protocol has been implemented, researchers sometimes receive more money to speed up recruitment when it lags behind the initial goal, but these payments are based on plausible justifications, such as increasing personnel effort for recruitment purposes, or on showing that the actual effort needed exceeds the budgeted effort. Nevertheless, in our collective experience, sponsors or clinical research organizations rarely refuse efforts to negotiate higher budgets.
Professional and Procedure Fees at AMCs
Individual AMCs follow different approaches in determining their desired research budgets. To calculate the expense of professional time by physicians and nurses, AMCs commonly generate an hourly or encounter rate using one of several methods: (1) a calculated hourly rate based on actual salary and fringe benefits and estimated time required per patient, (2) an hourly or encounter rate charged for normal clinical services that is usually higher than actual salary, or (3) either of the first two approaches, plus a research “tax” of about 10% that the sponsor does not necessarily see.3,6
For medical procedures done in research trials, rates are negotiated over a potentially broad range, owing to the large difference between institutions' standard fee schedules (known as “charge masters”) and the deep discounts that are routinely given to public and private insurers. Full charges typically are two to four times greater than what health insurers pay for these services.9 Where in this range research rates are set depends in part on whether internal or external fees are being considered. Internal fees are those used to determine how much each clinical department will charge the researching department's study fund code for the clinical procedures it performs. External fees determine how much the sponsor will pay the researching department for clinical procedures.
AMCs usually set the internal research rate at a certain percent of the full charge contained in the institution's charge master. Some departments set rates that rarely vary; others negotiate these rates case-by-case or provide a deeper discount on request. Also, researchers vary considerably in estimating how much work a study requires. One study found that costs estimated by different study sites for conducting the same hypothetical protocol ranged by a factor of 10.2 Another source reports that the difference across institutions between the 25th and 75th percentiles in charges for chest CTs is 73% and for extended blood pressure monitoring is 135%.10
An AMC's research rates determine only internal payments, from the research study budget to the department performing the medical procedure. Amounts built into the research budget that the sponsor pays externally are not necessarily the same. Some researchers charge sponsors the same rate that the institution charges them internally for procedures done in the research project. In other instances, researchers budget clinical procedures at full charges from the charge master, so that the institution charges the project less, and sometimes substantially less, than what the sponsor pays, usually without the sponsor's knowledge.
AMCs that use this approach believe that its purpose is not to make a profit from research. Instead, they feel that clinical research often ends up costing more than is initially budgeted because of unanticipated difficulties. Therefore, the difference between internal and external charges simply provides a cushion to help absorb possible research deficits elsewhere. In the end, any profit produced is not shared with the clinical investigators or their departmental colleagues. Instead, it remains within the institution's research enterprise to offset the cost for research staff who need support in between projects to do unfunded work or to develop new proposals.
Per Capita Payments: Covering Costs or Generating Profits?
Despite the potential concern that clinical investigators are generating large overt profits by conducting industry-sponsored trials, our experience and literature review suggests that this result is not common at AMCs. There is wide variation in pricing practices and an absence of direct oversight, but the overriding goal of research budgeting in AMCs appears to be covering costs and not generating profits. Any surplus that results from a particular project usually goes back into the research group to support staff during lean times or to support unfunded work; it is not paid to individual investigators. We have less knowledge of budgeting practices outside of AMCs, but one systematic study concluded that there is “no empirical evidence of ethical compromise”11 created by clinical trial payments, and a second concluded that “if anything, [investigators] tend to underestimate the time required for [research] activities, rather than overestimating them.”2
Further investigation is needed to determine whether research payments in community-based practices pose greater concerns. It is reasonable to assume that individual physicians in community practice can profit from excess research payments. Also, there is less institutional oversight of research in these settings, where most industry-sponsored research now occurs. For these very reasons, clinical research funds in England are paid mainly to agencies that employ physicians rather than directly to research investigators.6
Institutional arrangements, such as those at AMCs or in England, might be better long-term safeguards for preventing excessive payments than imposing strict scrutiny of negotiated research budgets. Although greater scrutiny might make research payments more accurate from one study to the next, avoiding both excesses and shortfalls, doing so would add another layer of regulatory burden to conducting research. Also, undue scrutiny might cause sponsors to provide too little funding, out of fear of penalty, which itself could pose a threat to conducting important clinical research. If research budgets are not adequate to complete the study, then the entire enterprise might be for naught, meaning that those who endured some level of risk or inconvenience in participating would have done so for no social benefit. Thus, ethical concerns run in both directions—toward payments being either excessive or inadequate. Whether the current approach to determining per capita payments is optimal requires further investigation.
The authors would like to thank Janice S. Lawlor, who assisted with this research.
This research was supported by grant R01HL075538 from the National Heart, Lung, and Blood Institute.
1 Christensen JA, Orlowski JP. Bounty hunting and finder's fees. IRB. May–June 2005;27:16–19.
2 Emanuel EJ, Schnipper LE, Kamin DY, Levinson J, Lichter AS. The costs of conducting clinical research. J Clin Oncol. 2003;21:4145–4150.
4 Lemmens T, Miller PB. The human subjects trade: Ethical and legal issues surrounding recruitment incentives. J Law Med Ethics. 2003;31:398–418.
5 Weinfurt KP, Hall MA, King NM, Friedman JY, Schulman KA, Sugarman J. Disclosure of financial relationships to participants in clinical research. N Engl J Med. 2009;361:916–921.
6 Raftery J, Bryant J, Powell J, Kerr C, Hawker S. Payment to healthcare professionals for patient recruitment to trials: Systematic review and qualitative study. Health Technol Assess. 2008;12:1–128, iii.
7 Sox HC, Rennie D. Seeding trials: Just say “no.” Ann Intern Med. 2008;149:279–280.
8 Hall MA, Weinfurt KP, Lawlor JS, Friedman JY, Schulman KA, Sugarman J. Community hospital oversight of clinical investigators' financial relationships. IRB. January–February 2009;31:7–13.
9 Hall MA, Schneider CE. Patients as consumers: Courts, contracts, and the new medical marketplace. Mich Law Rev. 2008;106:643–689.
11 Glass HE. Do clinical grant payment practices in phase 3 clinical trials influence subsequent clinical investigator prescribing behavior? Dis Manag. Spring 2004;7:77–87.