It has been obvious for some time now that UK telehealth provider Babylon was in deep doo-doo following its quite extraordinary entry into the US market, where it received a massively overpriced valuation after somehow managing to raise $US635 million in venture capital.
In addition to listing on the NASDAQ, the UK company also decided to buy into bricks-and-mortar healthcare, buying up some Californian medical practices for no obvious good reason. By 2022, it was obvious even to the CEO that the US venture, in his own words, was an “unbelievable, unmitigated disaster”.
So disastrous has the global expansion been that the company is now looking to flog off its UK assets, which were its main selling point in the first place, including the GP at Hand telehealth service as well as a triage chatbot. (Unfortunately, Babylon looks likely to withdraw from a 10-year project it first signed in 2016 with the government of Rwanda to provide much needed care, setting up Babyl Rwanda with the help of the Bill and Melinda Gates Foundation.)
It was highly controversial at the time to accept GP at Hand as an NHS-approved “practice”, in that it was allowed to enrol young and healthy Londoners and later Birmingham-based patients to its service while leaving traditional general practices to continue to shoulder the burden.
The alternative argument was these patients were cheap to care for anyway and they wouldn’t be clogging up waiting rooms with minor complaints, and as the BBC’s former tech reporter Rory Cellan-Jones writes, GP at Hand is incredibly popular with its users. It remains to be seen if the company can be salvaged but it’s a lesson for health IT and particularly AI companies in danger of getting too big for their boots.
Coincidently, Pulse+IT’s favourite correspondent Terry Hannan popped up this week with a few references for us to consider in the ongoing debate about the use of AI in healthcare. One was a 2019 book published by the US National Academy of Medicine titled Artificial Intelligence in Health Care: The Hope, the Hype, the Promise, the Peril.
The other was a presentation from California Healthcare Foundation CEO Mark Smith at the American Medical Informatics Association annual symposium in November 2009, entitled Health IT: Hope, Hype, and How to Avoid the Road to Hell, a video of which is sadly no longer online.
Closer to home, it was interesting to see private hospital operator Ramsay Health Care go on a PR blitz this week with news it had signed up with Google Cloud and also revealing some details of its digital strategy. Ramsay has always kept its IT cards very close to its chest, acknowledging EMR vendor MEDiTECH as its patient administration and billing system provider but not much else. We hear that in addition to the pre-admissions module it is trialling, it has also implemented a few other new digital technologies in its administrative and HR departments, and plans on a lot more for its 70-strong hospital system and its specialist doctors.
Also this week eHealth Queensland put out a few feelers for a new radiology information system to replace its legacy QRiS platform, which was first implemented by Cerner and Agfa at Princess Alexandra Hospital back in 1999. While 25 years is several lifetimes in heath IT and it is clear to anyone that its about time to modernise, Queensland does have a very patchy track record with statewide replacements.
There was the infamous, much-ballyhooed (by us) $64 million pathology system replacement project, cancelled in 2020, and equally notorious termination of the $210m HBCIS patient administration system replacement project. The full roll-out of the Cerner ieMR is apparently back on track with budget funding last year, having been paused in 2020, but we don’t hold our breath. And now the QRiS replacement. We’ve used this joke from the movie Airplane! before but all we can say is “Good luck. We’re all counting on you.”
And finally this week, the Australian Digital Health Agency popped its head up with a recruitment drive for a new branch manager for the ambitious but very much needed Connected Care program. Why ADHA uses the term branch manager, as if the person was a minor beancounter at a suburban bank we don’t know, but it is very much a big deal.
The person will be in charge of three massive programs – the national interoperability plan and standards catalogue, which have both been under the purview of ADHA’s Herbert Down, and the building of the new health information exchange (HIE), which receives loads of cash in the budget.
We have to say this newly created position is top notch and should attract a very competitive field. ADHA is also still looking for a new CTO to replace the departed Mal Thatcher, so things are being shaken up at the agency. We’ve got a few stories next week on exactly what the plans are.
That brings us to our poll question for the week.
Was Babylon Health always more about hype than hope?
Vote here and leave your comments below.
Last week we asked: will an aged care transfer summary make a difference? The vast majority of readers said yes: 84 per cent to 16 per cent.
We also asked: If you voted yes, is My Health Record the right vehicle? If no, is there a better way? Here’s what you said.
Babylon health. This crash couldn’t have happened to a nicer company.*
Was Babylon always more about hype than hope? Our readers overwhelmingly said yes: 87.5 per cent to 12.5 per cent.
We also asked, if you said yes, why did venture capitalists fall for it? If no, can the company be salvaged? Here’s what you said:
– Because they’re all #####
– The tech angle captured VC
– The opportunity in digital health is huge, but an execution requires working with current systems, not replacing them. VCs love disruptors so that’s where the money goes.
– Excellent and compelling sales pitch by genius owner
greed….??? and the literature says that only about 1 out of 10 ‘ventures’ actually succeed…
– They only saw dollars and no understanding of the complexity of health care. Babylon was never about AI but was and is based on rule sets re triage ( basic teaching at medical school re patient symptoms) run throu a rules engine.
– Because VCs were persuaded by a combination of very plausible sounding CEO, and FOMO. It was clear for their ML to work effectively they needed access to a much bigger data-set than they could generate themselves – they should have acquired one of the main UK clinical systems’ vendors i.e. EMIS (recently sold for $1.5bn to United Health, TPP (privately owned but an ageing CEO), or Vision – the runt of the litter but still with 25 million patient records to consume.
– Read up on how venture capital partners get incentivized, and you’ll know…