Like so many of you, I’ve been a bit puzzled by the whole bitcoin thing. What is bitcoin, exactly? Is that what Mario and Luigi were collecting all these years? Do I need to start putting my grant budgets in bitcoin? Also like so many of you, I’m extremely distractible. So, I started looking into what the diddly bitcoin actually is, and I think I’ve learned something that might be of real relevance for medical science.

DISCLAIMER

Everything I’m about to say is coming from someone who doesn’t really know what he’s talking about. Hell, he doesn’t even know enough not to dangle prepositions! Seriously, I might really whiff on the specifics, or even some of the generalities, so any real data/comp-sci wonks should feel free to jump in and correct me.

Don’t Think Bitcoin, Think Blockchain

Wait, aren’t we talking about bitcoin? Yeahyeahyeah, but to understand it, I had to learn about blockchain. It’s a term I’d seen in association with bitcoin, but I had even less of an idea what blockchain is. Turns out, blockchain is the actual thing that makes bitcoin possible.

Blockchain is a specific architecture for a database. Simply stated, blockchain is a publicly accessible, secure, de-centralized ledger. The entire ledger exists on a connected network of peer-to-peer participants and is updated simultaneously on all “nodes.” When the ledger is modified (e.g., a transaction happens), that gets added to the blockchain, and now there’s a permanent record of that transaction, along with all that have happened previously. Verification of all data blocks in the blockchain is required to verify any transaction. In essence, every part must be in agreement about everything that’s happened up to that point before adding the next thing that happens.

The implications of such a database structure are all of the things you’ve heard about bitcoin. The de-centralized nature means that banks aren’t involved. The trust required for any currency to exist in the first place comes from the extensive verification taking place across that de-centralized database. It also means that transactions are, or can be, anonymous. The database architecture gets around the “double spending” problem, something any digital currency must address (your own digital currency has this problem taken care of by banks as the centralized entities). The security is, in part, due to de-centralization. If you wanted to modify a transaction record or whatever, you’d have to modify the entire ledger, everywhere that it exists. It’s the same idea that prevents financial fraud and counterfeiting now – make it so prohibitively difficult to do that no one would bother. (Note that exploration into blockchain and database structures can become a bit of a rabbit hole, with provocative terms like “proof-of-work” and “Byzantine fault tolerance” coming up. If you, too, would like to lose the better part of a Sunday afternoon, have at it.)

As you can imagine, these kinds of computations take a lot of power, which is something I don’t usually think about when I’m firing up Netflix or something. You may have heard the recent dire predictions that bitcoin will require as much power as the entire world’s power usage in some small number of years. But, when you think about it, it starts to make sense. As transactions accumulate, the power needs will grow. That’s probably a significant limitation, at least for now, to what I’m about to discuss.

Blockchain in Medicine and Medical Research

No, we’re not talking about paying your hospital bills in bitcoin or submitting your NIH grant budgets in both U.S. dollars and bitcoin estimates. Just think for a moment about what a de-centralized, accessible, high-fidelity, secure database that’s updated in a reliable way could do for medical research. Imagine a large, multicenter clinical trial that collects a ton of data and a ton of samples. What’s the best way to keep track of all of these precious samples, particularly as different people dip into the supply for different reasons? Current approaches involve barcodes, very large spreadsheets, etc., and these, I have found, are imperfect at best, even within a single center. And that’s just talking about the samples themselves, to say nothing of the linked clinical data.

Let’s go further. Interoperability and portability of health records and electronic health records systems is one of the biggest challenges facing healthcare in the U.S. and worldwide. Taking into account all of the relevant security issues, imagine the savings and the improvements in care that could be realized by not duplicating tests, or by easily tracking controlled prescriptions, or by rapidly accessing the actual results of a given clinical procedure as opposed to someone’s recollection or a barely legible handwritten page that lands on your fax machine about a week after you really needed it. For a really thorough treatment of this idea (admittedly mired a bit in data-speak), check out Blockchains and Electronic Health Records.

Are you starting to see it now?

Of course, there are lots of problems to solve, perhaps chief among them being energy requirements and privacy issues. But the technology is really in its infancy, having only been around since about 2009, with an unknown inventor(s)/developer(s) identified only as Satoshi Nakamoto. As the tech improves from the standpoints of cryptography, coding and database architecture, energy requirements, and energy availability (c’mon, cold fusion!), applications of blockchain that are out of reach currently will be very possible.

I have no idea if this is where medicine and medical research will or should go. But, I think it’s worth considering and discussing. So, okay…now you talk. I’m gonna go play Super Mario Bros., because that’s my typical level of computer sophistication.

UPDATE 4/3/18:

Apparently people are indeed thinking about blockchain for healthcare.  Check this out:

https://m.washingtontimes.com/news/2018/apr/2/unitedhealth-group-joins-blockchain-pilot-project/

UPDATE 10/19/18:

Blockchain in healthcare and health data could be big, big business.  See this article on biobrokers in The Scientist.

About the Author

Screen Name

mark.mcever@vumc.org

About Me

I haven't filled out my About Me yet!