The Clinical Divide E4: MSK Conflict of Interest & Unpublished Negative Results

Kevin Campbell, M.D.
OCTOBER 01, 2018
This week we take a look into several interesting issues that have dominated the headlines in both the Wall Street Journal, as well as the New York Times.

First, a scandal erupts in one of the world's greatest cancer centers. Could clinicians and executives have prevented it? Next, science suffers at the hands of research publication bias. What does that mean for data sharing? How can physicians, researchers and healthcare executives work together to solve these problems?



Welcome to the Clinical Divide. I'm Dr. Kevin Campbell. I'm a Duke-trained cardiologist and the CEO of a health data startup called PaceMate. Every week, this Healthcare Analytics News video series examines medicine's most controversial headlines and top technology news.

I work to provide insight into the stories in hopes of helping bring physicians and healthcare executives together, in order to bridge the Clinical Divide.

MSK’S AI SNAFU

Our first story brings us to Memorial Sloan Kettering (MSK) Cancer Center in New York. Outrage exploded against the institution after the New York Times and ProPublica exposed deep conflicts of interest for high-ranking MSK officials.

Here's the gist of it: Some employees founded an artificial intelligence startup called Paige.AI and they got an exclusive deal to use Sloan Kettering's 25 million patient tissue samples and slides. The Cancer Center has a stake in the startup and so do several MSK leaders.

This mess has caused MSK's Chief Medical Officer to resign, and it has angered doctors who worry about patient privacy, HIPAA compliance, and money being made off of their hard work.

>>READ: Yes, Healthcare’s Data Breach Problem Really Is That Bad

This is a sticky one. First of all, it's critical that c-suite executives and physicians work together to promote research and innovation. But we must be wary about crossing the lines and creating significant conflicts that could impact patients and healthcare decisions.

The conflicts of interest here could've really been avoided. Executives need to consider the optics of any situation such as this, where they are using patient materials, patient data, and other privileged information in a way that creates personal financial gain.

They need to collaborate with doctors and researchers to really work hard to develop partnerships with proper checks and balances. If MSK had done that, the doctors might not have felt left out of the process. Patients might not have felt violated, and Sloan Kettering and Paige.AI might have avoided a high-profile scandal.

Communication in these types of situations is critical. Had a partnership ben developed with all parties represented at the negotiating table, this type of scandal could have and most likely would have been avoided, again, illustrating the importance of collaboration between executives, physicians, researchers and patients.

THE UNDERSOLD VALUE OF NEGATIVE TRIAL RESULTS

In our next story, let's talk about negative results. Negative results don't always get published. Let's jump to this issue. What determines which research gets published? A new study in the Journal of Psychological Medicine examined just that question and the investigators found plenty of bias.

The key finding of 105 studies registered with the FDA: 98 percent of those with positive results were published, while only 48 percent of negative studies made it to publication. In some cases, researchers even swapped primary and secondary findings to make their work more positive so that it could make publication.

Publications are what define success in academic medicine, and there's a tremendous amount of pressure to publish in order to advance and be promoted.

Let's put this problem into the context of a healthcare system. Many times executives are looking for reasons to not buy or purchase new technology or institute advanced therapies or research that comes with heavy cost. When negative results are not published, therapies may reach market based on a small number of positive trials. This could be wasteful. It could be in direct opposition to a medical executive's goal of providing the most cost-effective and proven therapies for patients.

It also hurts doctors and patients for obvious reasons. Honestly, when physicians and executives work together, they might be able to cut through the noise and determine what therapy is most effective. Reviewing negative results is just as important as the excitement that surrounds a highly-positive trial.

This study offers us all a lesson in data sharing. Let's not duplicate failures - let's learn from them. Only by airing our failures as well as our successes can we do good, clean research and good work that provides new and better therapies for our patients in a shorter amount of time.

Thanks for joining me for this episode of The Clinical Divide. Until next week, I'm Dr. Kevin Campbell for Healthcare Analytics News.

Get the best insights in healthcare analytics directly to your inbox Register for our daily newsletter.

Related:
The Clinical Divide E02: A New Wearables Landscape & The Big Aspirin Question
The Clinical Divide E01: Civica RX, 23andMe and a New Heart Failure Study

 

SHARE THIS SHARE THIS
62
Become a contributor