Blockchains and health care: promising and moving quickly, though no silver bullet

The buzz about opportunities for blockchains in the health care and pharmaceutical industries continues to ramp up. This technology can give us new tools to secure medical information, track pharmaceutical supply chains, and more. While blockchains have promise, they are not a panacea for health care.

Make no mistake: We are excited about the opportunity to use blockchains in our industry. Earlier this year, we launched Elektra Labs, a venture backed by a National Science Foundation I-Corps grant2 to develop a more secure system to share and verify health information. To build our first prototype, ou eam worked with the MIT Media Lab and Ethereum, a blockchain-based platform that was barely a year old.

Although there are many potential applications for health care blockchains, we worry that the heavy jargon and media hype has caused a lot of confusion about how and when to use this technology.

At its core, a blockchain is a digital ledger in which transactions are recorded chronologically and openly. The ledger is decentralized, with multiple copies stored across a wide network; it does not rely upon a central intermediary, like a bank or broker, to verify transactions. Participants who update the ledger are paid for their work using a digital currency or “token,” like bitcoin. Launched in 2009, bitcoin was the first widely adopted digital asset; others have since launched. The bitcoin market cap when we wrote this was $316 billion, three times larger than GlaxoSmithKline.

A large number of blockchain-based projects have launched in the past few years. To keep track, our team created the open-source map of health care-related projects below. 

To dive deeper into the potential for health care blockchains, we’ll explore what this technology can and can’t do.

The promise of blockchains

Blockchains in health care can improve coordination challenges and incentive systems.

The past year has seen a wide adoption of new digital assets, like Ethereum8, which incorporate features like smart contracts. These contracts make money “programmable” by predefining the rules and agreements between parties. For example, say a hospital’s inventory of blood pressure cuffs or insulin decreases to a certain level. A smart contract can automatically reorder the product, pay for the shipment, and record the transaction in a distributed ledger which multiple parties can review. 

Programmable money also makes it possible to create powerful, real-time incentive systems that extend well beyond the supply chain. Some of these can influence behavior. For example, a company could set up a smart contract to release funds as soon as an employee walks a certain number of miles per week, passively tracked via a wearable. A smart contract could also speed up outcome-based reimbursement by releasing bonuses to hospitals that lower their readmissions by a predetermined percentage.

Another opportunity for the use of blockchains in health care revolves around data integrity. The health care industry spends a lot of time talking about the Health Insurance Portability and Accountability Act, encryption, and other aspects of data confidentiality. We strive to make it difficult for adversaries to access protected data. While data confidentiality is essential, data integrity — maintaining the accuracy and consistency of data — is becoming increasingly important.

Put another way: A patient might not care if everyone knows that her blood type is O-positive. But she’d care a lot if someone could go into her health record the day before her surgery, change her blood type to B-negative, and leave no trace. That’s data integrity.

By design, blockchain-based technologies preserve data integrity. These systems are tamper-resistant because multiple, verified copies of the ledger exist in the decentralized network, and participants can compare changes chronologically against their own copies.

Here’s an example of this system at work. Chronicled is building an electronic system that would meet the requirements of the Drug Supply Chain Security Act to identify and trace prescription drugs distributed in the United States. This blockchain-based project would help companies comply with “track and trace” regulations, making the transfer of custody from a pharmaceutical company to a lab, hospital, and patient more secure. Using a decentralized blockchain that makes permanent, time-stamped records removes the risk of having a rogue central system administrator edit, modify, delete, or alter records.

Patient medical records also benefit from improved data integrity. The Estonian government partnered with Guardtime, an enterprise blockchain system, to implement a verifiable tracker any time someone accesses a health care record or makes a change in it. This means recording and reporting whenever an insurance broker — or anyone else — looks at an individual’s secure data online. As The New Yorker reported, in Estonia “peeping at another person’s secure data for no reason is a criminal offense,” and blockchains create a mechanism to track those violations.

Blockchains and misperceptions

Most pure blockchain technologies don’t offer data confidentiality benefits that go above and beyond a conventional toolkit. In fact, depending on the implementation, using a blockchain may result in reduced confidentiality, because an open decentralized ledger could provide data access to even more parties than in the current system.

Furthermore, people often mistakenly believe that blockchains can help with de-identification — removing personal information from a dataset. In its pure form, a blockchain is pseudo-anonymous rather than anonymous. As data sets get bigger and more unique, de-identification becomes a challenging computer science problem and the ability to completely remove personally identifiable information is increasingly becoming a myth. A few years ago, for example, security researcher Yaniv Erlich famously re-identified 50 DNA donors and their relatives from “anonymous” DNA data.

As each individual’s digital footprint is increasingly tracked, society will have to figure out a better system where surveillance is part of the life we live. Perhaps this isn’t a bad thing. As we grow society’s digital footprint, our collective data can be used to enroll the right people in the right clinical trials. We’ll be able to create better drugs and therapies for more targeted groups. Precision medicine will no longer be a dream; it becomes a reality.

Personally identifiable information is good when it comes to drug discovery and not so good when its discovery by or distribution to the wrong people means loss of autonomy and privacy. Life requires us as individuals and as a society to continually draw lines between what’s good and bad. Using blockchains won’t let us dodge these difficult decisions. We will need a healthy governance system to set the rules of the road or, in blockchain parlance, to determine network consensus.

Although cryptocurrencies are enjoying sky-high valuations, most health care blockchain projects are early ideas outlined in white papers. Today’s decentralized tools aren’t yet ready for massive scale. We need a “100x improvement” to host an app that has 1 million to 10 million users. A number of hard computer science and game theory problems must be solved before we can scale blockchain-based apps.

Where do we go from here?

A blockchain is not a silver bullet. Advances in cryptography — techniques for secure communication — will transform the health care industry on a greater scale than blockchains alone. One of the most promising projects is OpenMined, which combines multiple advances in cryptography (secure multi-party computation, homomorphic encryption, and blockchains) along with advances in data science such as federated machine learning. OpenMined is developing artificial intelligence algorithms that can be trained on data it never has full access to. This is a revolutionary breakthrough: the ability to perform calculations on encrypted information without decrypting it first.

Put in practice, something like this could mean that a patient who wants to participate in research or a clinical trial would never have to share a decrypted copy of his or her personal information. This could revolutionize our understanding of medicine while ensuring privacy for the individual.

Health care needs better audit trails, data integrity, and incentive systems. To create this future, we will need to combine blockchain-based tools with other advancements. And that means looking beyond the hype, investing in the technology, and developing a thoughtful governance structure to handle system upgrades and societal conflicts.

Source: Andy Coravos and Sofia Warner - STAT