[This is off-the-cuff live blogging, so apologies for any errors, typos, etc]
First up is Mark Lange from Eli Lilly (who notes that he is here in his personal capacity only!), speaking about “Data Transparency and the Role of the FDA.”
He prefaces his talk by noting that when he refers to “data,” he means raw, patient-level data from clinical trials. Most calls for the transparency of such data, he says, reflect a common theme about lack of trust in the pharmaceutical industry. So we might wonder: why doesn’t the pharmaceutical industry simply accede to that request and make their data available?
Mark notes that industry has several concerns. One important one pertains to data exclusivity. In several (if not all) markets, data exclusivity rights are premised on keeping the relevant data confidential, and posting it publicly would be deemed a waiver of those rights. In addition, data exclusivity prevents generic competitors from free riding, and publishing data could allow them to circumvent the very point of data exclusivity.
Moving to privacy concerns, Mark notes that research subjects’ understanding is that their data will be used for particular purposes and shared with regulators, but not be publicly posted on the Internet for anyone to do with whatever they want. Relatedly, there is the potential for interpretation of public data to be biased; research results may be over-interpreted and analyses may be flawed or even erroneous. Competitors might look for fairly trivial flaws the the data and try to use them to their advantage rather than sincerely trying to advance scientific progress and transparency.
Mark suggests, however, the choice between privacy and transparency is a false one. A better alternative is available — namely, for objective, expert regulators such as the FDA to receive and vet data in ways that address both audiences and both sets of concerns. The FDA is in fact already experienced in doing this. For example, it determines whether research demonstrates that a drug is safe and effective for a particular use through its marketing application approval mechanism, and it determines the accuracy and adequacy of the portrayal of research results in product labeling and product advertisements. And late last year, it was given responsibility for overseeing clinicaltrials.gov, which includes results from all pre-specified primary and secondary outcomes measures from nearly all clinical trials either conducted in the U.S. or intended to be used in support of an application for marketing approval in the U.S. This new responsibility, Mark suggests, could be a powerful tool, depending on how the FDA uses it. For instance, the FDA could exercise authority to monitor and enforce the absence of required results and the inclusion of false or misleading results data.
In concluding, Mark stresses that, when faced with requests for public access to patient-level trial data, we should consider the important role of regulators as trusted intermediaries who can balance competing concerns.Next up is Genevieve Pham Kanter, from the Edmond J. Safra Center for Ethics at Harvard University, who is speaking about “Financial Conflicts of Interest and Voting in FDA Drug Advisory Committees.”
Genevieve is presenting original empirical research she conducted on financial conflicts of interest (COIs) among external experts who vote on FDA decisions. How, if at all, do financial interests align with voting behavior? She notes that prior work has found no relationship between having a financial tie to a sponsor and favoring the sponsor in a vote, but that this work has tended to be based on very small sample sizes.
Her group analyzed data on 400 CDER voting meetings over 15 yrs (1997-2011), including 1200 up-or-down voting questions and 1400 unique people. The most prevalent relationship was consulting (34%), followed by having an ownership interest (25%), having been a paid speaker (19%), and being a member of an advisory board (14%). Many votes, she notes, are unanimous because the science — in particular, the risk-benefit analysis — is clear. So they focused on cases where the evidence is less clear, resulting in split votes (61% of their sample).
Noting, of course, that they show only correlation, not causation, she presented their results, which found a clear “sponsor effect,” in which those whose only relevant financial ties were to the sponsor under review were more likely to vote in favor of that sponsor. Those who had financial ties both to the sponsor and to an industry competitor, however, voted in ways that looked very much like the votes of those who had no financial interests.
The big take away, she concluded to audience laughter, is that “it’s complex.” Policymakers need to be more careful and nuanced with how they manage COIs.
Next up is Patrick O’Leary, a 3L at Harvard Law School and a Student Fellow at the Petrie-Flom Center, who is speaking about “Credible Deterrence: The Park Doctrine and the FDA in the 21st Century.”
Under the Park doctrine, health care executives can be held strictly liable and found guilty of a criminal misdemeanor. Critics of the doctrine say that (1) it can’t deter non-negligent conduct and executives already have ample incentives to ensure compliance; and (2) it’s unfair, since it violates the usual rule in criminal law that liability hinges on a finding of fault.
Patrick argues that the Park doctrine does deter, perhaps uniquely. Individual enforcement tends to be more effective than corporate enforcement because corporations cannot fully indemnify individual employees from liability and choose to absorb the costs of conviction as the cost of doing business. The corporation cannot, for instance, indemnify the individual executive from the stigma of a criminal conviction.
As for the criticism from fairness, Patrick acknowledges that convictions without a finding of fault would be unfair, but argues that that’s not what most Park doctrine cases entail. In fact, the doctrine is rarely used unless regulators believe that the defendant was either herself negligent or had actual knowledge of the violation. Thus, in recent high-profile Park cases (Synthes, Purdue Frederick, KV Pharmaceutical), those making the charging decisions clearly felt that these weren’t true strict liability decisions but rather cases in which the defendants did or should have known about the violations.
The more alarming trend, he says, is HHS OIG’s aggressive use of exclusion, in which they have been willing to exclude officers on almost any basis.
In conclusion, Patrick argues that the risk of abusive use of Park is substantially mitigated for several reasons in its actual implementation, and it does have a unique deterrent effect. As such, he says, the doctrine continues to serve a valuable purpose in FDA’s enforcement arsenal.
The final speaker is Katrice Bridges Copeland, from Penn State Dickinson School of Law, who is speaking about “The Crime of Being in Charge: Executive Culpability and Collateral Consequences.”
Katrice picks up where Patrick left off, focusing on the collateral consequences of criminal conviction, and in particular, HHS OIG’s exclusion of health care executives from federal health care programs for up to 20 years upon a misdemeanor conviction related to fraud. (She notes that there has been some question about whether a misdemeanor conviction without a mens rea requirement counts as “convicted of misdemeanor related to fraud,” but it has been interpreted to include such conviction, so she will assume that is the case for purposes of today’s talk.)
Katrice’s concern is that such exclusion is disproportionate to the harm the executives have actually inflicted. The baseline length of debarment is three years, with discretion to increase that term if aggregating factors are found to have been present. Katrice argues that there shouldn’t be debarment beyond the three-year baseline absent a showing of moral blameworthy conduct that occurred, such as misbranding with the intent to deceive. Generally, corporations can’t employ excluded executives without jeopardizing their own health care activities, so exclusion basically means that these executives can’t work in health care, and reentry thereafter is hard because their experience will be stale.
The problem with exclusions being significantly longer than three years, she says, is that their purpose is to protect federal health care programs from unscrupulous people; the goal is not punishment. Yet when the period of exclusion departs so far from three years based on acts and intent of third parties, it looks more punitive than remedial. If the government is serious about protecting health care programs from these people, then they should target executives with the closest connection to harm. And if the government isn’t going to do that, then just as criminal punishment has to be proportionate to the harm inflicted, so too should exclusion be proportionate to the harm inflicted. The government has sought a 20-year exclusion in a case where there was no knowledge on the part of the executive (later reduced to 12 years). That, she says, seems grossly disproportionate. The question we should be asking is whether acts of the subordinate are fairly attributable to the executive. That might be the case, for example, where the executive received a warning letter or where a compliance officer brought the violation to the executive’s attention, or where they’re the direct supervisor of the subordinate.