Why medical scientists should share data

Our kindergarten teachers and Hippocrates taught us to share. But after school, capitalism takes over among American medical researchers.

Eric Campbell writes in “Data Withholding in Academic Genetics” that it was frequent for investigators to be denied access to data. The reasons for denials include “too much effort” and protecting potential publications of students and themselves.

The respondents expressed only a little worry about protecting the commercial value of results: all data sharing is subject to transfer agreements that limit commercial applications of any of the shared data.

The lack of data sharing may have particularly serious consequences when data take decades to generate in diseases such as Alzheimer’s. Only after autopsy is the dataset complete.

Might the lack of progress during the last 104 years of Alzheimer’s research be related to unwillingness to share data?

Sage Bionetworks, a nonprofit research organization, writes “we’ve witnessed exponential increases in the quantity of molecular and genetic information without concomitant advances in the understanding of biology or disease,” and blames “current systems and practices for sharing datasets.”

Sage president Stephen Friend makes a compelling case for data sharing.

Half-hearted efforts to force researchers to share data include the NIH, which requires a “data sharing plan” for large grants, but nevertheless allows transfer agreements.

Some journals, such as PNAS, insist that data be shared or the authors will be barred from future publication. Again, in practice, it allows transfer agreements but does not review them.

That efforts such as these have not fully succeeded should not be surprising. Piwowar and Chapman report in “Public sharing of research datasets: a pilot study of associations” in 2011 in Journal Infometrics that “being subject to the NIH data sharing plan requirement was not found to correlate with increased data sharing behavior,” and the same applied to studies published in journals that “require” data sharing.

Many scientists in other fields share data.

Physicists even divide their researchers into experimentalists, who generate the data, and theorists who try to figure out what it means. A few years ago physics had been so successful that it was thought not to contain anything else discoverable. But that’s another story.

Medical scientists should share data.

George Lundberg is a MedPage Today Editor-at-Large and former editor of the Journal of the American Medical Association. Eugen Tarnow is an independent researcher publishing in the areas of short term memory, scientific publication ethics, social and organizational psychology and semi-conductor physics.

Originally published in MedPage Today. Visit MedPageToday.com for more health policy news.

Leave a Comment