Blockchain is a word that’s been floating around the NEM recently. No-one seems to know quite what it is, though it brings to mind pleasingly illicit-sounding things like Bitcoin and Silk Road. As with any possibly transformative technology, the full range of what blockchain can do will only become obvious over time. For one potential use, let’s look at a rule change currently before the AEMC – ‘Strengthening protections for consumers on life support’.
Existing life support protections
Normally if you don’t pay your bills for long enough, your electricity gets cut off. This seems reasonable, but not if you need vital electrical equipment like dialysis or an oxygen machine. Businesses such as retailers and distributors are required by law to take life support into account when they are dealing with their customers – not only in disconnections, but in other matters like providing notice for outages.
Under the current system, for a person to be protected under the life support laws, a doctor needs to certify them as sick. The patient or their carer has to convey this information to at least one of their energy providers (retailer or distributor), who then passes it on to the other business.
While this system works well in some respects, it has a few key weaknesses. Working out who does and doesn’t need life support can be expensive and logistically complicated. Because there is no central register for life support both businesses must maintain their own lists, which aren’t always harmonised. Doctors cannot directly access the life support registers at all. Sick people don’t have any way of knowing whether they are listed apart from asking their retailer or distributor – and being on hold with a utility is hardly anybody’s favourite process.
On the flipside, there is no simple process to remove someone from the life support register. Due in part to a lack of up-to-date information, the number of addresses listed as hosting life support has ballooned in recent years, leading to significant additional costs.
As mentioned previously, a rule change is currently before the AEMC (which, for full disclosure, I worked on prior to composing this article). That’s not the subject of this piece. Rather I want to propose how blockchain might be used to better manage energy customers’ health data – and in doing so, explain a little about how the technology works.
Goals of life support
Before we get into the weeds of describing what blockchain actually is, let’s define some parameters for the problem we are trying to solve. An ideal system for managing life support records would achieve the following things:
- It would list all the people who needed life support, and none of those who didn’t. (To stop sick people dying and businesses paying for people who don’t actually need it).
- It would be consistent between different energy businesses and other parties.
- It could be quickly and easily accessed by anyone who needs to – the patient, the doctor, and the relevant electricity businesses. (So that errors and omissions can be quickly found and rectified, saving lives and costs).
- Private medical information could only be accessed by people with the right and need to use it. (To avoid misuse of information).
- If circumstances change, records could be easily and accurately updated and rapidly shared with the relevant people. (For instance if a patient dies, moves house, or makes a full recovery).
How might blockchain address these challenges?
To repeat a question not asked frequently enough – what is blockchain? Perhaps the highest profile use so far has been Bitcoin, a decentralised currency which depending who you ask represents an investment opportunity, a modern day tulip bubble, or the end of the nation state.
In simple terms, Bitcoin is a currency stored in the form of a very long ledger which records every transaction that has ever taken place. You can use the Bitcoin ledger to see who owns what and how this has changed over time. The ledger is shared between every computer on the Bitcoin network – many thousands of people – and is instantly accessible by all those people. (Bitcoin also has a number of well publicised and some would say fundamental flaws – not least its energy footprint).
Let’s now dive into a slightly more technical explanation of how similar technology might be used to manage life support records.
TECHNICAL EXAMPLE AHEAD
Consider the following scenario:
Zaynab diagnoses Hari with breathing problems which mean he needs an oxygen machine. She makes a note of this on a digital ledger, attributing life support status to Hari’s ‘address’ on the network. Importantly, while this address is treated as a consistent identity over time, it’s not necessarily tied to Hari’s actual identity in real life. Let’s call it fegsyf%A&. Without knowing the password (and potentially other security measures), it’s extremely hard for anyone to know that fegsyf%A& = Hari.
The new information about Hari’s life support status is automatically broadcast to everyone else on the network (medical professionals, retailers, distributors and patients), updating the copies of the ledger which they store on their own computers. Meanwhile another doctor, Neel, determines that Liyan no longer needs his sleep apnea machine. This, too, is recorded and broadcast.
After a certain number of such ‘transactions’ or updates in information have taken place, they are grouped together in a ‘block’ and then encoded – or, to make things a bit more dramatic, ‘turned into a secret code’. Let’s call that code $%&A%@A. Reading ‘$%&A%@A’ from the outside, there’s no way you can tell Liyan fixed his sleeping problems but that Hari still needs help to breathe. However, knowing Liyan and Hari’s health records respectively, it’s simple to verify that the encoded block of data really does contain that information.
(The methods by which this encoding is achieved are mathematically and computationally complex. They rely on the following property of numbers: that it’s very difficult to factor large numbers into primes, but easy given a potential solution to check if it’s correct. This is not a fundamental, proven law of mathematics, but is highly secure in the current environment. It is used as the basis for many other widely adopted systems, such as credit card pins.
In future, someone could conceivably invent an algorithm which overcomes this challenge. This would render a huge number of existing security mechanisms moot. Scientists at the University of Hefei are trying to make existing means of encoding secret data (‘cryptography’) even more secure by using the ‘spooky action at a distance’ properties of entangled quantum particles. This is rather exciting, not least because you could potentially build an ansible).
Who exactly encodes the transactions will depend on the design of the specific blockchain used. In Bitcoin, different parties compete to be the quickest to do so, and are rewarded with a few units of the currency for their efforts. This, along with another aspect of design known as ‘proof of work’, means that Bitcoin is extremely energy intensive (which may ultimately render the currency unviable).
An alternative method might be to randomise (or ‘pseudo randomise’) the role of encoding the next block between different participants. It’s important that the same party doesn’t get to successively create different blocks as this could undermine the security and trustworthiness of the ledger. Remember, the blockchain is a historical record. Each subsequent block depends to some extent on the contents of the previous ones. If I can falsify the block at time t, this makes it easier for me to falsify t+1, t+2, and so on.
Once each block has been created it is ‘sealed’ off – that is, the person who encoded the last block also ‘does something’ to make sure that its contents cannot be changed retrospectively. For example, they might again rely on the properties of numbers to find a code which, when combined with the contents of the block, yields a particular numerical solution. They then stamp this ‘sealing number’ at the end of the block.
If someone wants to check if the contents of a particular version of the ledger had been falsified, they can combine that version with the sealing number and see if it yields the correct solution. If it doesn’t, the ledger has been compromised.
As soon as the sealing number has been calculated, it is broadcast across the entire network. Everybody – retailers, doctors, distributors, even individual patients – checks it against their own records to see if it yields the right solution. If the majority of the network agrees on a particular number, it becomes designated as ‘correct’. If your ledger doesn’t accord with this version, you’ll need to update and/or replace it so that it accords with the records everybody else is using.
TECHNICAL EXAMPLE CONCLUDES
Well, that was complicated. To make things a little less theoretical, let’s go back to the criteria for an ideal life support register to see how a blockchain-based system might perform.
- The list of who does and doesn’t need life support is shared by everyone, almost instantaneously. This eliminates inconsistency between businesses and other parties.
- It is also updated instantaneously. This prevents dangerous time lags where the retailer (for example) knows that Hari needs life support but the distributor does not.
- It can be verified by anyone on the network. For example, Hari’s doctor (or even Hari himself) can makes sure that his details have been properly added.
- It’s highly accurate. Because the ledger requires consensus between different parties, it would be very difficult for, say, Liyan to deliberately falsify Hari’s records without collaboration from many other participants. Inadvertent mistakes can be quickly identified through comparison with other people’s records.
- It can be made highly secure. While medical records can be accurately attributed to each address on the network, it can be made very difficult for outside parties to match that address with a person in real life. This prevents scenarios where the data is used in nefarious ways (for example, disability discrimination).
Another benefit is that responsibility for maintaining the ledger is shared between many different parties. While many see a benefit in maintaining a centralised register of life support customers, it’s much harder to reach an agreement on who should be responsible for administering that ledger – and who will take the fall if something goes wrong.
A blockchain-based solution makes everyone responsible simultaneously – while also maintaining a clear record of how information was added and edited, which can be audited if need be.
Blockchain-based systems are already being trialled around the world for healthcare uses. In the US, a pilot program called MedRec uses blockchain to exchange data between different hospitals, including blood work, vaccination histories and prescriptions.
In the UK, Google has a subsidiary called DeepMind Health which is currently collaborating with the NHS. One early prototype is ‘Streams’, an app to compile different sources of clinical data and send your doctor an automatic alert if you’re at risk of kidney failure.
As exciting as this is there are caveats, one being privacy concerns. While blockchain-based systems can be made highly private and secure, this is not the same as guaranteeing that they will. For example, DeepMind and the NHS have been criticised for using anonymised patient data for medical research without first notifying patients.
That said, the system currently in place in the NEM requires non-anonymised data to be shared between multiple employees of at least two businesses. It’s hard to see how a blockchain-based system would be inherently less private than what already occurs.
Just as pertinent is the question of who in the NEM might bring such a concept to fruition. PowerLedger is already trialling blockchain in the context of peer-to-peer electricity trading, as an innovative way to attribute generation and consumption between different households. The Australian companies Blackmores and Australia Post are collaborating with Chinese e-commerce business Alibaba on a project to improve traceability of food and healthcare products.
Technological change is often framed as a threat, sometimes justifiably. New systems and inventions displace the old – and disruption isn’t always a benefit to the disrupted. Even when society benefits overall, there are still winners and losers. The Luddites may now be a byword for the ridiculous, but in fact had quite legitimate concerns – decent wages and conditions for those who manned the dark Satanic mills of the Industrial Revolution.
What’s certain is that there are also opportunities. In the case of life support, there might just be a chance to make life better for everyone – improving safety for the sick while saving costs for electricity providers. Jerusalem indeed.
Chris Lim is the director of Embiggen Economics. Prior to founding Embiggen, Chris worked at Australian Energy Market Commission, SGS Economics and Planning, the Centre for Policy Development, and as an academic at the University of New South Wales. She holds degrees in economics/econometrics and in international development.