THE BIG PICTURE, HISTORY, CONTEXT
We propose that it is economically viable for the UK's National Health Service (NHS) to become a self-funding entity by selling patient healthcare data to a wide variety of industry actors. We also propose that an individual’s healthcare data record itself can eventually becoming an earning asset that can be passed on to future generations.
In 2004, Jeff Kelley, an American Organisational Development expert working with Freddie McMahon, a UK Knowledge Software expert ‘discovered’ a means of capturing patient behaviour and preference data that is of sufficient value for pharmaceutical companies, insurers and healthcare providers to actually pay for.
They collaborated with IBM, Google and the UK Work Foundation who contributed to, and verified an economic model that could make any large scale healthcare system self-funding. The author was involved in early discussions in the context of using social networks as a platform for testing and piloting the model. Jeff Kelley died and his work was never published.
THE ECONOMICS – ‘HEALTHWEALTH’
What is the data? Think of a patient record as more than a static file or electronic representation of doctors’ notes. Think of it extending to DNA fingerprints, Tesco Clubcard data, medication consumption data (using RFID tagging), exercise data, and most valuably, self-service diagnosis data because it can be gathered in real time. This data, aggregated at scale, provides answers to questions like;
- How many more type 2 diabetics are there?
- If I know the rate at which people are taking up no-smoking treatment should I really launch this new cigarette brand?
- Exactly how many heart drugs should I buy this year?
- Are we over-prescribing antibiotics in children, and by how much?
In fact the combinations of data are almost infinite, as are the questions. What this data really does is allow companies to sense demand, predict and change user behaviour, and they will pay more for this than they do for conventional market data. Enough for the NHS to eventually become completely self-funding.
Both Microsoft and Google have introduced online services aimed at consumers who want to maintain their healthcare data in exchange for a range of benefits. Neither of these initiatives has been successful in terms of scale and economics. Their failure has been largely through lack of trust.
To illustrate the value of consumer data consider this; all the current academic research on large data sets provided by the major social media providers concludes that advertising is not a sustainable revenue model. The reason why Facebook, Bebo (prior to acquisition) and Twitter are free is because they are VC backed, and even YouTube (owned by Google) has applied for broadcast TV rights in its major markets because online ads aren’t generating enough revenue.
Recently both Twitter and Facebook announced the development of what they call ‘sentiment engines’ signalling a radical shift in their business model. Sentiment engines are web services for anyone who wants to extract intelligence from the conversational content of huge numbers of consumers – for a fee. Big companies already have huge budgets for customer insight systems (CRM) and ‘social data’ is just another form of this data. In the next 2 years, between them Facebook and Twitter will hold more valuable (that is, saleable) data about UK citizens than the government itself.
The UK government would have to pay US software companies run by teenagers to test the impact on people of consuming these items: how many burgers is too many? How many beers? How many late night films? How much violence on a video game? How much cough medicine?
NEW MODELS OF ORGANISATION
Some upside-down thinking is needed to create a model of trust such that 60m citizens’ healthcare (anonymised) data can be monetized in a way where there is a continuous exchange of value for all parties. The answer lies in organisational form.
Perversely the clues can be found in financial services. Visa’s organisational model created a network of 22,000 banks, 1 billion card holders, and 16 million merchants, generating $2.4 trillion of transactions at its peak. Only 0.2% of that value was required to operate the Visa platform making it one of the most profitable businesses of all time. The fact is, traditional command and control organisation is not suited to 21st century value creation models. Dee Hock, the founder of Visa created a non-stock, for-profit membership based network with ownership in the form of non-transferable rights of participation. He designed the organisation according to his philosophy: highly decentralized and highly collaborative.
If the participants own the service, have full control and responsibility for the veracity and quality of their data and are able to democratically agree upon how this ‘Health Wealth’ is distributed, then people will trust it. Jeff Kelley suggested the creation of a Public Interest Enterprise based on Open Source models of value exchange and ownership – “The power of us”. Given that healthcare models in the US and Britain are only 10-20 years away from collapse we’re going to have to bet more heavily on the future, than we do on the past. Indeed, such is the value of the nation’s collective health data, that it’s a future national asset, and of major strategic importance to the United Kingdom.
Think about this. If a patient can derive an income from their medical data, even if it means re-keying it into another platform, they will. All the indications are that Microsoft, Google and a number of ‘up starts’ are creating these platforms.
In December 2007, 23AndMe launched an online service that enables consumers to understand their own genetic information by sending a mouth swab sample for genome mapping. Users know whether their children are predisposed to certain traits or talents — athletics or music or languages — and encourage them to pursue certain paths. “We will, counter intuitively, face even more pressure to conduct our lives carefully, strictly, and cautiously; we'll practice the art of predictive diagnosis and receive a demanding roster of things to avoid, things to do, and treatments to receive — long before there's any physical evidence of disease”. In other words, this service is creating the very behaviour change in patients that governments seek. Google has invested in 23andMe, whose co-founder Anne Wojcicki is married to Google co-founder Sergey Brin.
Last year a major step towards the widespread use of intelligent ‘web agents’ was made when researchers in New York, created a character in the virtual world of Second Life with the reasoning abilities of a child. These web agents are commonly referred to as Chatterbots. The Wikipedia definition;
“A chatterbot (or chatbot) is a type of conversational agent, a computer program designed to simulate an intelligent conversation with one or more human users via auditory or textual methods.”
It is now possible for non-technical subject matter experts to program chatterbots (instead of a computer programmer). So a nurse on the frontline who carries out diabetic screening diagnostics on patients can ‘embed’ her knowledge of clinical procedures in a chatterbot that can be used on the web, interactive TV or mobile phone by any consumer or professional wishing to test themselves or a patient for type 2 diabetes. What would the cost base of NHS Direct look like if most of its conversations with callers could be handled by chatterbots? £75 per hour for a GP, £28/hr for a practice nurse.
HOW DOES IT WORK?
Last year, a group of experts in London carried out an analysis of the potential market for chatterbots based on Gartner research. This research predicts that by 2011, 80% of internet users will have virtual representations of themselves - working or playing online. User-generated scripts will power their chatterbots, enabling conversations with people within their social network or beyond. Dialogue can be text or voice-based, covering any subject from informal socialisation to advice; people will have a Chatterbot as a personal assistant within FaceBook, another as virtual tax advisor and a third specialising in the investor implications of doing business in Shanghai. The group projected that based on these conservative assumptions;
- World population 7.2bn
- PC & mobile devices converged
- $100 PC pervasive
- Smart consumer devices i.e. health
- 4bn connected consumers
- 2.2 chatterbots per consumer
- 1bn users with multiple devices
- 0.5bn organisations (incl. micro)
- 8.8 chatterbots per organisation
..by 2018 there will be.
13.2bn chatterbots
Chatterbot = Social Network Person
…and because chatterbots can carry advertising;
- US$50 revenue for each Chatterbot
- US$660bn
- Variance (-20% to +100%)
Clearly there are many privacy and data protection issues but by using a basic principle of allowing users to control the separation of identity and data, coupled with end user ownership of the organisation that manages it, is the only way. A Creative Commons for health data with government oversight but citizen ownership.
HOW DOES IT START?
It starts with having a healthy disrespect for the impossible.
It starts with the maxim "We are continually faced with great opportunities which are brilliantly disguised as unsolvable problems." - Margaret Mead
It starts with understanding that “It is not because things are difficult that we do not dare; it is because we do not dare that things are difficult.” - Seneca
It starts with taking some time to talk to me.
POST SCRIPT
Serendipitously, on 19th February 2009, a German university announced the development of Assistance for patients: Intelligent sensor networks that remotely monitor important vital signs of patients or measure their activity. Heart patients simply wear a small, lightweight device on their body. The device records movement. A physician can evaluate, for instance, the heart rate in connection with physical effort, thus simplifying the diagnosis.
The data is transferred via a home gateway to an Internet portal where the physician can access the information remotely. A range of optional components can be added to the system to measure other vital data such as blood pressure, weight or blood sugar level.
Recent Comments