As the world scrambles to respond to the Covid-19 pandemic, access to public-health information has never been more urgent or challenging. In this interview, public health legal scholar Amy Kapczynski, the Knight Institute’s current Senior Visiting Research Scholar, assesses whether the public has access to the data and information it needs to evaluate critical health and policy decisions.
At Yale Law School, Kapczynski is faculty co-director of both the Global Health Justice Partnership (with epidemiologist and health activist Gregg Gonsalves) and the Collaboration for Research Integrity and Transparency. In the early stages of the Covid-19 pandemic, she was one of the chief organizers of an open letter to the federal government detailing best practices for responding to the spread of the virus, and she and Gonsalves have published several additional articles proposing policy responses for the present and future.
Our research director Katy Glenn Bass interviewed Kaczynski about the importance of access to particular kinds of information during the pandemic, and other aspects of the ongoing response efforts.
One of your areas of expertise is global public health, including access to information about things like drug studies and data. Can you talk about how that intersects with what you’re observing in this pandemic?
We urgently need new medicines and a vaccine for coronavirus, but we also need reliable knowledge about whether or how well they work. Clinical studies are very complex, and there are structural conflicts of interest endemic to our drug development system. We rely heavily on companies to run trials, and they have high-powered incentives to torque what they study or report to emphasize good news and ignore the bad. For years, scientists have been documenting the problems of this system. Doctors rely on the published data, but studies show that trial sponsors tend to publish studies with positive results, but not publish the ones with negative results. Or, they publish negative or equivocal results, but cast them as if they’re positive. Sometimes – Vioxx is an example here – companies even deliberately design studies to avoid finding deadly side effects that they think might be lurking. We found out about this, but only years after the drug hit the market, when tens of thousands had died. It would be great if the FDA could catch all of these problems in advance, but the agency is underfunded and under a lot of pressure from industry and patient groups. This is why, with others, I’ve been focused on the need for public access to clinical trial data now, asking and sometimes going to court, to try to get researchers access to the complete clinical trial data for drugs in use today.
The stakes couldn’t be higher now – the staggering demand for treatments and vaccines, the hundreds of billions of dollars on the line, and the rush to a cure, all mean that we need access to that data more than ever. Data can help show us whether regulators are doing their jobs too. This is critically important now, especially because President Trump has been promoting unproved cures, and because of the reports of retaliation against officials inside the administration who have insisted on high standards and refused to accommodate politically motivated interference with funding and study design.
Does the public have access to the information it needs right now? What should it have that it doesn’t?
The first thing we need is solidly designed studies for candidate treatments and vaccines – that itself isn’t assured, and companies are doing really crazy things, like throwing a whole handful of drugs at patients with study designs that make it impossible to sort out the true effects of the drugs. The public needs to know, first, that we need to wait for randomized controlled trial results before we know what works. And we need national and international coordination to ensure that the most important candidates are being tested in some coordinated fashion, and that the design of the tests can give us answers. (The National Institutes of Health just announced an initiative to do this, which is an important step. There are probably some intellectual property barriers in the way here, which the government will need to be assertive about addressing.)
Once we have something that we think works, and data is submitted to regulators, the public and expert researchers will only get to see a sliver of that. The agency gives summaries of the data, for example, and companies have to post summary results on a website called clinicaltrials.gov. But, the FDA has been taking the position – wrongly, in most cases – that it cannot release more clinical trial data, because this data constitutes trade secrets or “confidential commercial information.” Especially now, researchers really need access to all of the data (including not just summaries, but things like the “clinical study report” that details all of the data – positive and negative – relied upon for approval, but also clinical trial protocols, adverse event reports, etc.). I’ve been writing about why the agency can and should affirmatively release this kind of data. Recent caselaw has created new challenges for mandatory release under FOIA, so the agency should be doing this voluntarily. But the agency may well fail to act or act too slowly (especially this one, if the president is pressuring them to override the normal rules and cut corners). Congress should step in, to legislate to ensure that we have public access to the data that we need to validate and vaccines and drugs that are developed for Covid-19, and other things too.
More broadly, there are also barriers that claims of confidential commercial information or trade secrecy pose for the public in understanding the impact and responding to coronavirus. One early example: the first drug that went into shortage because of Covid-19 was known to the FDA, but not identified to the public, because the agency viewed it as proprietary information. We have access to almost no public data about Covid virus tests, which could help us understand how well they are really working in practice. These are all part of a similar problem: claims that data is proprietary, and practices that keep data secret, impede what the public and researchers can know, and so our ability to act and shape policy appropriately.
As this crisis has unfolded, we’ve seen reports from various sources that employees of the Centers for Disease Control and Prevention (CDC) get approval before speaking publicly about Covid-19. Is there reason to be concerned here?
Absolutely. This administration – and particularly Trump himself – have repeatedly lied or misled the public about everything from how serious this infection is, to how many tests we have. Inside of these federal agencies there are many good scientists, people whose job it is to collate and publish information and data that the public desperately needs. We cannot shape our response without good data and information, and in a pandemic that means listening to scientists, and not politicians, about questions of science.
How should we think about the balance between making sure there’s a unified government message about the crisis, and ensuring that individual speakers are able to serve as a check on the distortions and politicizations that often occur in centralized messaging efforts?
The government does need to have a unified, clear message to be effective, and to guide the public. But that message also has to be accurate, and based on science. The problem comes when government seeks to enlist scientists to distort evidence and violate the norms of professional conduct. One way to think about this is through the lens of Robert Post’s work on democracy and expertise, which he developed in part in response to cases where legislatures were ordering doctors to misstate the risks of abortion. We can’t have an effective democracy if we don’t have forms of expertise. That implies some limits on what governments can force experts to say. If governments force scientists or doctors to lie for political reasons, that undermines what we in a democracy need from both our government and our professions. So it's worrisome that scientists are being pressured to hew to a political line about evidence. I suspect that we will start to see some whistleblowers, like Rick Bright, who was recently demoted from BARDA, and those complaints are important to both discipline the administration and inform the public.
How do the challenges we’ve discussed above impact journalists’ ability to report on this crisis? And researchers’ ability to study it?
Access to data is critical for both. Think about how slow the racial disparities in infection and deaths were to be reported, for example. The data was neither being rigorously collected, nor released to the public. The issues with clinical trial data are similar. We’ve represented both journalists and scientists who want access to clinical study reports because they believe that there are unappreciated dangers or flaws in medicines currently on the market. But existing law makes accessing this data challenging – we get too little data, and it takes too many resources and too much time. That undermines both journalism and science, and puts the public at risk. With Covid-19, every one of us has an exceptionally strong interest in good data. It’s a building block of our individual and collective response.
Many technology developers and firms are developing apps and programs to facilitate contact tracing and otherwise assist with surveillance. Are there civil liberties concerns with this kind of surveillance, as there are with other forms of mass surveillance?
There are definitely reasons to be concerned about privacy as new apps and programs are developed to help us trace and trace those affected by the virus. Some kinds of health research do not require personally identifiable information, or produce aggregated results that have fewer privacy risks. But contact tracing inevitably picks up very sensitive information – geolocation and people you meet, for example. Privacy risks should also be thought of in two registers: there is a risk to individuals when information that is sensitive isn’t handled appropriately. But there is also a more systemic risk, to our sense of self, to who we are as citizens. If we feel we are all being watched all the time, it changes who we are and what we do. Privacy is constitutive, and what we establish about what we’re willing to do with apps here will also shape us over the longer term.
Another public health measure that may raise civil liberties concerns is quarantine and isolation orders. I’m referring not to the broad orders to stay at home that many states now have in place, but more specific orders directed at people who have tested positive for a virus or who have been exposed and are required to remain in a specified place – sometimes a designated facility, other times their homes. These were used in some cases during the Ebola outbreak in 2014, and some of the asymptomatic healthcare workers who were ordered to quarantine filed a challenge to that order in court. Can you talk about that, and the way you look at the concerns on both sides?
With students and colleagues at Yale, I have been involved in representing a Liberian family and several healthcare workers who were wrongly quarantined in 2014. Ebola is a disease that was well understood at the time, based upon decades of research. Experts all agreed that it could not be transmitted by asymptomatic people. It’s a disease that transmits through bodily fluids, once people are sick. Connecticut quarantined our clients even though they had never had contact with anyone with Ebola – posting police outside the door. A whole family with minor children was by law confined to a cold basement for weeks. They were also not given adequate medical supplies or food, nor adequate notice of their legal rights. None of them ever contracted the virus. These measures not only harmed people. They also drove stigma and confusion about the disease, and interfered with the public health response being mobilized by international organization like Doctors Without Borders, as we documented in a report with the ACLU. Though there are very few constitutional cases about quarantine specifically, there are cases about isolation of people with tuberculosis, and other deprivations of liberty (for example, civil commitment). To protect constitutional rights, coercive measures like these (isolation is for the sick, and quarantine for the exposed who we fear may transmit) need to be necessary, based on science, and use the least restrictive means.
What that standard means in any particular instance depends on the disease, and the state of the science. Covid-19 appears to be transmissible asymptomatically, and to be a very serious disease. This is one of those rare circumstances where quarantines can be legally justified, because we can’t just ask people to stay home once they start to feel sick. Isolation of those who are sick is also constitutionally permissible if they are a danger to others. These measures though, it is important to say, would still need to be targeted appropriately, use the appropriate processes, and be updated as knowledge evolves. As yet, we have not seen many individualized quarantine orders in part because we are so badly behind that only mass social distancing will work, and states can’t issue much less enforce individual orders of this kind. But if we are able to flatten the curve, we may well move to a point where such orders become possible.
I hope, however, that they are not widely used. Voluntary measures are almost always better, as public health experience teaches us. If we support people to stay home or in a hospital if they are infected or at risk, people overwhelmingly will. Police aren’t a great way to induce public health compliance or trust – and they are particularly feared in, and dangerous to, vulnerable communities and communities of color. Right now, we have broad orders, but the only enforcement I’ve heard of have been police getting involved to go after people of color in parks. There’s a deep point here about how coercive measures operate in this country. I’d rather have public health workers talking to people about the need for social distancing, and finding out what the barriers for people are, than get the police involved. If we start ordering people to not leave their homes at all on pain of fines or jail, then we also have to provide them with food, medical care, communications, and so forth. I have major concerns about the failure of both process and support, given the way that our public health infrastructures have been weakened over decades. What we saw with the unjustified Ebola quarantines in terms of process failures could be far worse if we take an aggressive carceral approach in a pandemic of this scale.
Katy Glenn Bass is the Knight Institute’s research director.