Healthcare Blockchain Consortia Initiatives

Healthcare Companies are Ramping Up Involvement in Blockchain / DLT Projects in 2019

Part 4: Chosen Protocols

This is the fourth in a series of posts on lessons learned from the series of healthcare blockchain consortia announced over the last year.

This series of posts was inspired by an awareness that something unprecedented is happening in healthcare. The initial newsletter introduced the history and detailed the trend of blockchain-focused enterprise consortia. Newsletter #2 focused on the insurance companies, pharma companies, banks, health systems, and non-traditional healthcare companies who are creating these consortia, and how their efforts represent new utility solutions aimed at solving shared problems. Newsletter #3 discussed the three primary design patterns (data synchronization, data markets, and multi-party business process automation) in these consortia.

Table 1: Enterprise Consortia announced in 2018 and 2019. Updated 10/10/19.

The choice of a proper protocol has a lot to do with the use case. Most use cases in healthcare do not share the same characteristics as cryptocurrency (censorship resistance, public verification, borderless-ness, etc) which was the original intent of Satoshi’s Bitcoin blockchain.

Many of the newer derivatives look more like cloud than Bitcoin. For example, recent initiatives by Amazon and Google use cryptography to solve trust issues related to the use of data assets. Some enterprises and use cases will gravitate towards this end of the spectrum because it seems more familiar, it generally costs less, and it is optimized for a use case that doesn’t need a true blockchain. After all… if you don’t need to run a bunch of nodes, then why would you?

These systems all attempt to solve problems related to data ownership, data governance such as:

• Who owns the data?
• Who enforces the data has not been tampered with
• Who has the authority to change or delete data?
• Who creates and runs the applications on the data layer?
• Who validates the transactions?

Source: https://media.consensys.net/5-reasons-why-enterprise-ethereum-is-so-much-more-than-a-distributed-ledger-technology-c9a89db82cb5

The word “blockchain” has evolved over time and that’s good for those of us in the healthcare space who want to solve old problems in new ways.

The concept of a blockchain was introduced over ten years ago with Satoshi Nakamoto’s vision of an open, public blockchain that enabled the use of Bitcoin. In Nakamoto’s whitepaper, the blockchain was designed to solve cryptocurrency issues related to immutability and censorship resistance using an open, borderless, publicly verifiable ledger with no central authority that could be influenced.

The blockchain that Nakamoto lays out is a network that replicates a single ledger across all nodes. In Bitcoin’s case, new ledger updates are globally broadcast every 10 minutes to all the nodes on the network. New updates are collected into “blocks” of chronological transactions that are cryptographically “chained” to previous blocks stretching back to the first, “genesis” block. Blockchains are radically transparent to all parties, making tampering easily detectible.

By design, the Bitcoin blockchain replicates all ledger data across every node on the network and keeps those nodes in consensus through an energy-intensive, costly, very secure consensus algorithm called Proof of Work.

Bitcoin, which is open source, has inspired many new derivatives blockchains that change some of the characteristics so that the blockchain may be used for applications other than Bitcoin. The first of these derivatives was Ethereum.

Ethereum was introduced roughly five years after Bitcoin. Ethereum shares Bitcoin’s open, public characteristics while appealing to new use cases. It expands on bitcoin by broadening the definition of the digital asset tracked on the ledger and introducing the concept of using a blockchain as a development platform. It was Ethereum that opened up the use of blockchain for the enterprise.

Like the Bitcoin blockchain, Ethereum uses the expensive and energy-intensive Proof of Work algorithm. Traditionally, the cost of using an Ethereum-based protocol tends to be expensive because it is directly tied to the cost of ETH, the cryptocurrency “gas” that powers the Ethereum global computer. Ethereum is also currently very slow, processing around 15 transactions per second. By comparison, Visa processes 24,000 transactions per second. By design, every node on the Ethereum network holds a full copy of the distributed ledger.

These properties related to confidentiality, speed, and cost present significant challenges for enterprise adoption of Ethereum.

Ethereum is an open source protocol, so some companies have “forked” the original protocol and changed some code to account for enterprise needs. These forks of Ethereum designed for the enterprise are referred to as “Enterprise Ethereum” and there are several blockchains sit under this umbrella. Each of these attempt to retrofit Ethereum to improve privacy, increase throughput, or adjust governance controls.

Over the last few years there has been an explosion in Ethereum-based forks including the Parity client (and Substrate for enterprise), Quorum (JP Morgan’s fork of Ethereum for the enterprise) and others.

One well-known example of enterprise Ethereum is the Parity client. Parity is a “light-Ethereum” subprotocol used only by “light clients,” which simply download block headers as they appear and then other parts of the blockchain are subsequently downloaded on command. This structure provides full functionality to securely access the blockchain, but a key difference is that they are not mining, and therefore bypass the expensive consensus process used by traditional Ethereum solutions.

One example of an Enterprise Ethereum project on our list of consortia is MediLedger, a consortium consisting of pharma companies and others who are working to secure the pharma supply chain. Chronicled, the convener of this consortium, was working with Ethereum outside healthcare back in 2016 so Ethereum was a logical choice when MediLedger was introduced in 2017.

According to Suzanne Somerville and Ryan Orr of Chronicled, the choice of Enterprise Ethereum represented “the best combined set of technical attributes, including security, scalability, upgradeability, performance, and suitability to solve our specific customer, industry, and regulatory requirements” for their use case.

Other reasons the Chronicled team believes Ethereum is good choice include Ethereum’s Turing complete smart contract framework; Ethereum’s robust developer ecosystem; Ethereum’s traction with other use cases; and Ethereum’s permissive open source license.

To account for the privacy challenges that come with the use of a fully replicated blockchain, MediLedger is innovating with zk-SNARKS, which stands for “Zero-Knowledge Succinct Non-Interactive Argument of Knowledge,” where one can prove possession of certain information, without revealing that information, and without any interaction between the prover and verifier.

Chronicled’s Somerville is excited about the potential for zk-snarks because “with zk-snarks plus smart contracts, we can use the blockchain as a neutral regulator, inspector, or enforcer of industry-wide regulatory rules and business rules. For example, we can create a rule like, ‘only Mfg A is allowed to commission or send SGTINs containing Mfg A’s company prefix’ or ‘an SGTIN can only move forward in the supply chain if it originated from a licensed manufacturer and if all prior transfers between trading partners on the network followed the rules and were valid.’ Used in this way blockchain plays the role of an army of tens of thousands of clip-board carrying inspectors, inspecting and approving every transaction, to ensure that the letter of the regulation is followed. What’s even more exciting, is that with zk-snark, we also gain the benefit of 128-bit encryption applied to the payload of every transaction, so it turns out to be a very strong data privacy solution too.”

zk-SNARKS is still evolving and will be an exciting area to watch in 2020.

Another consortium using Enterprise Ethereum is The Synaptic Alliance. For its data synchronization effort, Synaptic began with Quorum, which is JP Morgan’s fork of Ethereum. A fully replicated ledger is appropriate for this use case because the project is designed to create a shared utility for provider directory data. It also helps that this information is non-sensitive and non-competitive.

Like Ethereum, Fabric is broadcasting globally which means that accommodations must be made to account for enterprise sensitivities specific to privacy and confidentiality. IBM’s answer to this issue is the introduction of private channels between participants, as well as new IBM-hosted, peer-to-peer “private data collections” between parties. Companies are having success with early pilots of Fabric’s privacy-preserving technology. It is too early to understand how these private channels and private data collections will scale for enterprises with hundreds or thousands of private relationships and a need for programmatic use of off-chain data.

The Health Utility Network (HUN) has adopted a version of Fabric upon which the network is now developing a range of use cases. IBM was the convener of this consortium and IBM will be instrumental in evolving the protocol’s use for high volume, data-sensitive, programmatically-focused enterprise healthcare use cases.

Corda is not a traditional blockchain. It is an example of a blockchain derivative distributed ledger technology (DLT) that departs from the open, public, fully-replicated blockchains like Ethereum and Fabric.

Distributed ledgers are designed to serve as an efficient, interoperable network to synchronize with counterparties and business partners. Designed specifically for regulated enterprises, Corda’s first design principle is privacy and confidentiality of transactions. Corda achieves this with a ledger that is not replicated across the entire network. A single transaction is only recorded in the ledger of your specific counterparties or other entities agreed to by the transacting parties. It is an option, but not a requirement for the transacting parties to broadcast globally. There are no “blocks” of transactions. Instead, transactions are cryptographically secured between parties to ensure the validity of interactions.

All parties in the network maintain privacy while also having an interoperable infrastructure that can accommodates many transaction types across the network. Since it’s not a fully replicated blockchain, it is easier for Corda to meet requirements around enterprise privacy, confidentiality, and throughput.

In many ways, protocols are becoming commoditized. So many companies are beginning to focus on creating value above the protocol.

Hashed Health and others are taking a protocol-agnostic approach to building the applications and tooling required to help projects mature in an efficient and meaningful way.

As we work on a variety of protocols with enterprises and consortia, we see trends in the tools necessary to bring emerging design patterns and use cases to life. These tools will help future projects achieve results more efficiently.

At this point in time, flexibility is key. Many of the projects listed in this series have already changed the underlying protocol. For example, the team at Hashed Health has developed on six different protocols amongst our different projects over the last three years.

It is still early. It will be interesting to see how these protocol announcements mature and whether we continue to see significant separation across the spectrum. DLT and blockchain technologies are evolving rapidly. Right now, certain DLT solutions have an edge in terms of scaling privacy- and confidentiality-focused use cases. DLT is currently the better enabler of interesting new business models such as the ProCredEx marketplace for credentialing. Ethereum has the edge in terms of developer communities.

It may be that over time, as the technology matures, more disruptive innovations arise out of true blockchains, especially as you consider how we will begin to enable consumer-focused opportunities. It will be some time before we know for sure.

For now it is clear that Amazon, IBM, VMWare, Microsoft, Google and others are all fighting to create protocol-level offerings for the enterprise. There’s been an acceleration in “as a service” offerings, fully managed offerings, and consulting / design services in 2019 from companies that seek to include high availability, disaster recovery, and other managed service options. New options such as Amazon’s Quantum Ledger Database (QLDB) offers a cryptographically secure serverless database option that will provide guarantees around data. This type of centralized solution comes at a lower price point than operating a bunch of nodes on a network.

DLT also makes it easier to navigate the enterprise procurement process since DLT often looks more like cloud infrastructure than it does Bitcoin.

In the long run, I think the lines between blockchain, DLT, and cloud will all continue to blur. The protocol wars have begun. In the end, most enterprises don’t care about terminology… they just care about solving old problems in new ways.

(I’d like to express my thanks to Stephen Allen, Hashed Health computer science team member / intern all-star for his help researching this series.)