When Charles Piller, an investigative correspondent at Science magazine and contributing writer at Stat, started investigating Proove Biosciences, it was before the company sold off its assets as the CEO departed under a criminal probe in August 2017. Piller’s first article about Proove, published October 2016, looked at its dubious gene test for addiction, which Piller said exploited fear surrounding opioids. He’d also written about the lab offering doctors $144,000 a year to push these questionable tests.
Piller, along with Christopher Robinson, professor of law and University of Arizona associate dean for research and innovation, spoke about the science of genetic testing at the Association of Health Care Journalists health journalism conference in Phoenix. The panel offered a blueprint for uncovering and reporting on fraudulent organizations.
Look for eyebrow-raising claims
What piqued Piller’s interest was a claim that seemed to prey on national anxiety about opioids—the company said it could predict who would become addicted to opioids with 93 percent accuracy. Delving into similar credulous claims might prove fruitful.
Examine the science
Lightly regulated industries are prime fodder for investigation. In Piller’s case, the company’s own scientific advisor disavowed the science behind the product, leading experts called it “hogwash,” and the CEO had a bit of a history—he sold supplements based on genetic tests through spas and got in trouble for it. In this instance, Proove was only testing for 12 out of hundreds of genetic variations—and although there are important genetic elements to addiction, the problem is socio-environmental as well as genetic.
Piller recommends looking up studies at clinicaltrials.gov. The site is run by the National Institutes of Health, and anybody conducting human experiments must register on the site and describe the experiment. This allows reporters to glean information on the type of study at play.
Look for evidence of misconduct
Examples of things you might uncover include “pay to play” clinical trials, coercive tactics used to test every patient, doctors, and employees from all levels saying the tests were not useful, internal documents showing contradictory test results, and more. Interviews with employees, doctors, and patients come into play, but Piller offers a word of caution: sources sometimes have their own motives, so it can be a good idea to seek out sources rather than relying on ones that come to you. It’s also smart not to reveal sources to one another. This can help protect you from someone promoting a hidden agenda.
Look for financial conflicts of interest
Physicians often take money from drug and device companies, in the form of royalties, consulting, speakers bureaus, board service, and sometimes even kickbacks, Robertson explained. (Only the latter are illegal, under the federal Anti-Kickback Statute.) These payments were made public due to the Physician Payments Sunshine Act, part of the Affordable Care Act. ProPublica’s Dollars for Docs database includes payments made from August 2013 through December 2015, so users can see which physicians received payment, and from whom. The amounts tend to be substantial.
For more information on the causal effect of financial relationships between companies and health care professionals, see Robertson, et al.’s 2012 paper in the Journal of Law, Medicine, and Ethics:
http://journals.sagepub.com/doi/10.1111/j.1748-720X.2012.00678.x
Look beyond regulation
The FDA may check to see whether tests do what they’re supposed to be doing and that a lab implements it according to instructions, but Robertson points out that the regulation for some categories of medical devices leaves a lot of questions unanswered. Even if a test or device is FDA-approved, individuals can still be given an inaccurate test, or the test could be interpreted improperly.
The tests may not be good value for the money. Many types of tests or devices do not require external review and do not require the reporting of adverse events. Often the FDA does not address clinical validity or impose standards for quality in manufacturing. And, of course, there is the possibility that someone along the line is doing something untoward; for example, using tests that are only cleared for research/investigatory purposes diagnostically.
Fact-check your reporting
Make sure to validate leaked documents, fact-check your own work ruthlessly, and test all of your assumptions. Once you’ve done so, talk to the subjects of the investigation and test the assumptions you have about what’s going on. Be sure to be able to show your work.
Because businesses can sometimes be litigious, make sure to have a lawyer review your reporting before publishing.
